800xA 6.0 System Guide Technical Data and Configuration 2010. 7. 5. · System800xA SystemGuide...

System 800xASystem GuideTechnical Data and Configuration

System Version 6.0.3.1

System 800xASystem Guide

Technical Data and Configuration

System Version 6.0.3.1

Document Number: 3BSE041434-600Document Revision: H

Release: November 2017

NOTICE

This document contains information about one or more ABB products and may include a descriptionof or a reference to one or more standards that may be generally relevant to the ABB products. Thepresence of any such description of a standard or reference to a standard is not a representation thatall of the ABB products referenced in this document support all of the features of the described orreferenced standard. In order to determine the specific features supported by a particular ABB product,the reader should consult the product specifications for the particular ABB product.

ABB may have one or more patents or pending patent applications protecting the intellectual propertyin the ABB products described in this document.

The information in this document is subject to change without notice and should not be construed asa commitment by ABB. ABB assumes no responsibility for any errors that may appear in this document.

Products described or referenced in this document are designed to be connected, and to communicateinformation and data via a secure network. It is the sole responsibility of the system/product owner toprovide and continuously ensure a secure connection between the product and the system networkand/or any other networks that may be connected.

The system/product owners must establish and maintain appropriate measures, including, but notlimited to, the installation of firewalls, application of authentication measures, encryption of data,installation of antivirus programs, and so on, to protect the system, its products and networks, againstsecurity breaches, unauthorized access, interference, intrusion, leakage, and/or theft of data orinformation.

ABB verifies the function of released products and updates. However system/product owners areultimately responsible to ensure that any system update (including but not limited to code changes,configuration file changes, third-party software updates or patches, hardware change out, and so on)is compatible with the security measures implemented. The system/product owners must verify thatthe system and associated products function as expected in the environment they are deployed.

In no event shall ABB be liable for direct, indirect, special, incidental or consequential damages ofany nature or kind arising from the use of this document, nor shall ABB be liable for incidental orconsequential damages arising from use of any software or hardware described in this document.

This document and parts thereof must not be reproduced or copied without written permission fromABB, and the contents thereof must not be imparted to a third party nor used for any unauthorizedpurpose.

The software or hardware described in this document is furnished under a license and may be used,copied, or disclosed only in accordance with the terms of such license. This product meets therequirements specified in EMC Directive 2014/30/EU and in Low Voltage Directive 2014/35/EU.

TRADEMARKS

All rights to copyrights, registered trademarks, and trademarks reside with their respective owners.

Copyright © 2004-2017 by ABB.All rights reserved.

Table of Contents

1 About this System Guide131.1 Document Conventions ............................................................................................

141.2 Terminology .............................................................................................................

2 Introduction152.1 Related Documentation ............................................................................................

172.2 Hardware Selector ....................................................................................................

172.3 Temporary Sales Authorization ................................................................................

3 Dimensioning and Deployment193.1 Overview ..................................................................................................................

213.2 Application Size Limits ............................................................................................

213.2.1 Tag Calculation ..........................................................................................

223.3 System Size Limits ...................................................................................................

233.4 Definitions ................................................................................................................

233.4.1 Node Types ................................................................................................

233.4.2 Node Functions ..........................................................................................

233.4.3 Options ......................................................................................................

233.5 Node Function Deployment .....................................................................................

233.5.1 General Topologies ....................................................................................

273.5.2 Client Deployment .....................................................................................

303.5.3 Engineering Systems .................................................................................

323.5.4 Function Allocation Rules .........................................................................

363.5.5 Restrictions on Node Function Combinations in a System .......................

363.5.6 Restrictions on Node Function Combinations in a Node .........................

373.6 Connect Service Calculation Rules and Limitations ................................................

383.6.1 Load from Operator Client Displays .........................................................

3BSE041434-600 H 5

Table of Contents

383.6.2 Load from History Logs ............................................................................

403.6.3 Load from External Subscription ..............................................................

413.6.4 General Considerations for External Subscription ....................................

423.6.5 Connectivity Server OPC DA/AE Limits ..................................................

433.6.6 Connectivity Server History Parameters ...................................................

433.6.7 Available Functions Per Controller Connectivity ......................................

443.7 Controllers ................................................................................................................

443.8 Multisystem Integration ...........................................................................................

453.8.1 Read Only Subscriber ................................................................................

453.8.2 Supported System Functions .....................................................................

463.8.3 Supported Connects ...................................................................................

463.8.4 Configuration Rules ...................................................................................

483.9 PLC Connect ............................................................................................................

483.9.1 General Characteristics ..............................................................................

503.9.2 Performance Data ......................................................................................

513.9.3 Alarm and Event Performance Data ..........................................................

513.10 Load-Evaluate-Go ....................................................................................................

513.11 Libraries ...................................................................................................................

523.12 Property Transfer ......................................................................................................

533.13 System Networks ......................................................................................................

533.13.1 800xA Network Equipment .......................................................................

563.13.2 Network Configuration ..............................................................................

563.13.3 Control Network Clock Synchronization ..................................................

573.14 System Redundancy .................................................................................................

583.15 System Deployment .................................................................................................

583.15.1 Domain or Workgroup ...............................................................................

583.15.2 Supported Operating Systems ...................................................................

613.15.3 Workstation vs Server Class Version of Windows ....................................

623.15.4 Hardware Infrastructure .............................................................................

653.15.5 Third Party Software .................................................................................

663.15.6 Virtualization .............................................................................................

683.15.7 Virtualization Host Software .....................................................................

693.15.8 System Installation ....................................................................................

6 3BSE041434-600 H

Table of Contents

703.15.9 System Update ...........................................................................................

703.15.10 Deployment in IT Environment .................................................................

703.15.11 Software Media and Licensing ..................................................................

743.15.12 System Deployment Examples ..................................................................

4 Operations814.1 Operator Clients .......................................................................................................

814.1.1 Client Count Limits ...................................................................................

814.1.2 Process Graphics ........................................................................................

824.1.3 Screen Resolution ......................................................................................

824.1.4 Display Call-up Time ................................................................................

834.1.5 Command Response Time ........................................................................

844.2 Alarm Parameter Limits ...........................................................................................

844.3 VideONet Connect ...................................................................................................

854.3.1 Camera to VideONet Server Communication ...........................................

854.3.2 VideONet Server to 800xA Client Communication ..................................

5 Control and I/O875.1 Memory and Execution Performance .......................................................................

875.1.1 Memory Size .............................................................................................

885.1.2 Available Memory ....................................................................................

915.1.3 Execution Performance ..............................................................................

935.1.4 Considerations for AC 800M High Integrity Controller ...........................

945.1.5 Spare Memory Needed for Online Changes .............................................

955.1.6 Comparing Memory Allocations Made with Different Versions .............

955.1.7 Memory Consumption and Execution Times ............................................

1055.2 Online Upgrade .......................................................................................................

1055.2.1 Stoptime .....................................................................................................

1065.2.2 Online Upgrade in a High Integrity Controller .........................................

1075.2.3 Communication Disturbance .....................................................................

1075.3 Hardware and I/O Performance ................................................................................

1085.3.1 Modulebus Response Time and Load ........................................................

1085.3.2 Calculation of Scan Time on the Modulebus ............................................

3BSE041434-600 H 7

Table of Contents

1105.3.3 Calculation of the Modulebus CPU Load ..................................................

1115.3.4 Example Scan Time and CPU load ...........................................................

1115.3.5 ModuleBus Scanning of ABB Drives .......................................................

1125.3.6 Dynamic Data Exchange S800 I/O Connected via CI854A/B ..................

1155.3.7 S100 I/O Response Time and Load ...........................................................

1165.3.8 Drivebus Communication with CI858 Unit ...............................................

1195.3.9 Calculation of I/O Copy Time Estimate for ControlNet with CI865 Unit .

1225.4 Communication ........................................................................................................

1225.4.1 IAC and MMS Communication ................................................................

1265.4.2 MasterBus 300 Network ............................................................................

1265.4.3 INSUM Network .......................................................................................

1275.4.4 800xA for AC 800M Performance ............................................................

1285.5 Supported Hardware and I/O Families .....................................................................

1285.5.1 AC 800M Controllers ................................................................................

1395.5.2 Adaptors for I/O Types ..............................................................................

1445.5.3 I/O Families ...............................................................................................

6 Fieldbus1576.1 MODBUS .................................................................................................................

1576.1.1 MODBUS RTU Master Communication ..................................................

1596.1.2 MODBUS TCP .........................................................................................

1626.2 PROFIBUS ...............................................................................................................

1626.2.1 Device Management PROFIBUS HART System Extensions ...................

1636.2.2 Network Connection through Communication Interface ..........................

1656.2.3 PROFIBUS Connect Service .....................................................................

1666.3 PROFINET IO ..........................................................................................................


1686.4 HART, WirelessHART .............................................................................................

1686.4.1 HART Interfaces ........................................................................................

1696.4.2 HART Device Management Performance .................................................

1706.4.3 HART Connect Service .............................................................................

1716.4.4 WirelessHART ...........................................................................................

8 3BSE041434-600 H

Table of Contents

1716.5 FOUNDATION Fieldbus .........................................................................................


1756.5.2 FOUNDATION Fieldbus Network and Connect Service ..........................

1766.6 IEC 61850 ................................................................................................................


1856.6.2 IEC61850 Connect ....................................................................................

1866.7 EtherNet/IP and DeviceNet ......................................................................................

1866.7.1 General .......................................................................................................

1886.7.2 EtherNet/IP ................................................................................................

1886.7.3 DeviceNet ..................................................................................................

7 Applications1917.1 Application Parameters Summary ............................................................................

1947.2 Scheduling Service ...................................................................................................

1947.3 Softpoint Service ......................................................................................................

1947.4 Calculations ..............................................................................................................

1957.5 Information Management .........................................................................................

1957.5.1 History Services .........................................................................................

1957.5.2 Information Manager Server Capacity ......................................................

1967.5.3 Disk and Memory Capacity Requirements ................................................

1977.5.4 Computing PDL Data Disk Space .............................................................

1987.5.5 Maximum Number of Entries Per Log (Nominal) ....................................

1997.5.6 History Collection Maximum Sample Rate ...............................................

1997.5.7 Fastest Sample Rate and Timestamp Resolution .......................................

1997.5.8 Disk Requirements Per Log Entry .............................................................

2007.5.9 History Objects Miscellaneous Capacities ................................................

2017.5.10 History Archive .........................................................................................

2017.5.11 Display Services MDI (Multiscreen Display Interface) ............................

2027.5.12 Desktop Trends ..........................................................................................

2037.5.13 ODBC Client .............................................................................................

2037.5.14 Open Data Access (ODA)- Excel Data Access (EDA) .............................

2037.6 800xA History .........................................................................................................

2037.6.1 800xA History Server ................................................................................

3BSE041434-600 H 9

Table of Contents

2047.6.2 Embedded Data Collector Node (DCN) ....................................................

2047.6.3 800xA History Archive Server ..................................................................

2057.6.4 800xA History Hardware Sizing ...............................................................

2057.6.5 Data Access from 800xA History ..............................................................

2057.7 Batch Management ...................................................................................................

2077.7.1 Use of Function Phase Driver ....................................................................

2087.7.2 Function Phase Driver - Performance Considerations ..............................

2087.7.3 Server and Client System Sizing ..............................................................

2087.7.4 Use of a Physical Host Server ...................................................................

2107.7.5 Use of Virtualization Host Software ..........................................................

2117.8 PC, Network and Software Monitoring (PNSM) .....................................................

2127.9 Asset Optimization ...................................................................................................

2137.9.1 Control Loop Asset Monitoring (CLAM) .................................................

2147.10 800xA Public Address System .................................................................................

2147.10.1 PAS Connectivity .......................................................................................

2147.10.2 PAS Announcement ...................................................................................

8 Heritage Control Systems2158.1 800xA for AC 100 ....................................................................................................

2168.1.1 Prerequisites and Restrictions ....................................................................

2168.1.2 Maximum System Configurations .............................................................

2178.1.3 Technical Data ...........................................................................................

2218.1.4 Configuration .............................................................................................

2238.2 800xA for Advant Master with 800xA for Safeguard ..............................................


2268.2.2 Technical Data ...........................................................................................

2298.2.3 Performance ..............................................................................................

2318.2.4 Configuration .............................................................................................

2358.3 800xA for DCI .........................................................................................................

2368.3.1 Technical Data ...........................................................................................

2368.3.2 Performance ...............................................................................................

2388.3.3 Recommended Hardware Configuration ...................................................

2388.3.4 Additional Information ..............................................................................

10 3BSE041434-600 H

Table of Contents

2388.4 800xA for MOD 300 ................................................................................................


2418.4.2 Server Switchover Time Performance .......................................................

2418.4.3 Product Capacity ........................................................................................

2428.5 800xA for Symphony Plus Harmony .......................................................................

2438.5.1 Technical Data ...........................................................................................

2458.5.2 Display Exchange Time Performance .......................................................

2458.5.3 Server Switchover Time Performance .......................................................

2468.5.4 Recommended Hardware Configuration ...................................................

2468.5.5 Additional Information ..............................................................................

2478.6 800xA for Melody ....................................................................................................

2478.6.1 Sizing Data ................................................................................................

2488.6.2 Performance ...............................................................................................

2488.7 800xA for Freelance .................................................................................................

2498.7.1 Sizing Details .............................................................................................

9 Process Engineering Tool Integration2519.1 Performance .............................................................................................................

10 Service and Support25310.1 Life Cycle Services .................................................................................................

25310.2 Life Cycle Management Model ................................................................................

25310.2.1 Life Cycle Policy .......................................................................................

25710.2.2 System 800xA Software Support ..............................................................

25810.2.3 Life Cycle Parts Services ...........................................................................

11 Ordering and Delivery26311.1 Ordering Procedure ..................................................................................................

26311.1.1 Ordering Tools ...........................................................................................

26311.1.2 Sales Configurator Wizard 800xA .............................................................

26411.1.3 Business Online .........................................................................................

26511.1.4 Order Box and Wizard Order Module .......................................................

26511.1.5 Ordering and Configuration Procedure .....................................................

3BSE041434-600 H 11

Table of Contents

26611.2 Special Services ........................................................................................................

26611.2.1 License Change ..........................................................................................

26611.2.2 Pre-installed PC .........................................................................................

26611.2.3 Certificate of Quality .................................................................................

26711.2.4 Suppliers Declaration ................................................................................

26711.2.5 Packing Certificate ....................................................................................

12 Warranty26912.1 Extended Warranty - Hardware ...............................................................................

13 Revision History27213.1 Updates in Revision H ..............................................................................................

27213.2 Updates in Revision G ..............................................................................................

INDEX

12 3BSE041434-600 H

Table of Contents

1 About this System Guide

This document contains technical data and performance information for System800xA, version 6.0. The document also specifies configuration and dimensioningrules. This document is intended for use during:

• Sales process for dimensioning work.

• During the engineering and configuration to ensure alignment to the systemcapacity and performance.

The information in this document may be subject to adjustments due to changes madein system revisions. Refer to the appropriate Release Notes for accurate and detailedinformation.

Updated versions of this system guide will be published on myABB/My ControlSystem when required. myABB/My Control System can be reached at this url;https://myportal.abb.com

1.1 Document ConventionsThis publication includes Warning, Caution, and Information where appropriateto point out safety related or other important information. It also includesTip to pointout useful hints to the reader. The corresponding symbols should be interpreted asfollows:

Electrical warning icon indicates the presence of a hazard which could result inelectrical shock.

Warning icon indicates the presence of a hazard which could result in personalinjury.

Caution icon indicates important information or warning related to the conceptdiscussed in the text. It might indicate the presence of a hazard which could resultin corruption of software or damage to equipment/property.

3BSE041434-600 H 13

1 About this System Guide 1.1 Document Conventions

https://myportal.abb.com

Information icon alerts the reader to pertinent facts and conditions.

Tip icon indicates advice on, for example, how to design your project or how touse a certain function

Although Warning hazards are related to personal injury, and Caution hazards areassociated with equipment or property damage, it should be understood that operationof damaged equipment could, under certain operational conditions, result in degradedprocess performance leading to personal injury or death. Therefore, fully complywith all Warning and Caution notices.

1.2 TerminologyRefer to [1] in Table 2.1 for product and domain specific terms and acronyms usedin this document.

Even if the system is well specified there are sometimes use cases which stretch thelimits, especially if several parameters are pushed to their limits at the same time.This means use cases range from clearly and fully supported, over situations wherethe user has to be more accurate in observing system behavior, to cases known notto function properly. Hence, the following terms are used when needed to guide theuser.

Supported; Solutions within the limits (performance, capacity, recommendations)stated in this system guide

Recommended; You should do it in any of these ways. Recommended ways arevalidated, and supported.

Not recommended; The use case, or parameter value, is not fully validated, theremight be situations where unexpected behavior occur. A larger responsibility isexpected from a delivery project if such use cases are applied.

Not supported; The use case is known to have functional issues, it is not validated,and must not be applied.

14 3BSE041434-600 H

1 About this System Guide 1.2 Terminology

2 Introduction

System 800xA is scalable in both functionality and size. A system is configured basedon different functionality combined on different nodes in a system. Different systemfunctions are provided by different computer hardware nodes in a system installation.A node in this context is a computer that has a network address on the System 800xAnetwork.

For more information, refer to Overview.

Any security measures described in this document, for example, for user access,password security, network security, firewalls, virus protection, and so on, representpossible steps that a user of an 800xA System may want to consider based on arisk assessment for a particular application and installation. This risk assessment,as well as the proper implementation, configuration, installation, operation,administration, and maintenance of all relevant security related equipment,software, and procedures, is the responsibility of the user of the 800xA System.

Any safety measures described in this document shall be understood as examplesand general configuration information only. The information herein is not sufficientto set up and configure a Safety system. For that purpose relevant user guides mustbe considered. Proper implementation, configuration, and installation of Safetyreleated products and functions is the responsibility of the user of the 800xASystem.

2.1 Related DocumentationRefer to [2] in Table 2.1 for a list of all documents applicable to the 800xA Systemthat are included on the Release Notes/Documentation media provided with the systemor available from myABB/My Control System. This document is provided in PDFformat and is also included on the Release Notes/Documentation media. This documentis updated with each release and a new file is provided.

Specific documents referred to in this system guide are presented in Table 2.1.

3BSE041434-600 H 15

2 Introduction 2.1 Related Documentation

The asterisk (*) appended to each document number is a wildcard character usedin place of the document revision. The wildcard allows searches in myABB/MyControl System to be independent of revision. All revisions of the document willbe displayed in the search result.

Table 2.1: Reference Documents

DocumentNumberDocument TitleItem

3BSE089190Terminology and Acronyms[1]

3BUA000263*System 800xA Released User Documents[2]

3BSE056141*System 800xA Virtualization with vSphere ESXi[3]

3BSE035041*System 800xA PLC Connect Configuration[4]

3BDD011861*800xA for Freelance Product Guide[5]

2PAA111708*System 800xA Installation and Upgrade Getting Started[6]

3BUA000500*System 800xA Third Party Software[71]

3BUF001091*System 800xA Information Management Getting Started[8]

3BUF001092*System 800xA Information Management Configuration[9]

3BUF001093*System 800xA Information Management Display Services[10]

3BUF001094*System 800xA Information Management Data Access and Reports[11]

3BSE035980*System 800xA Control AC 800M Configuration[12]

3BDD011677*Device Management FOUNDATION Fieldbus Linking Device LD800HSE / LD 800HSE EX

[13]

3BSE030340*800xA for Advant Master Configuration[14]

3BUA000733*800xA for MOD 300 Performance Considerations TechnicalDescription

[151]

3BSE042621*800xA for Advant Master Performance Guideline[161]

3BDS013980*800xA for AC 100/AC 100 OPC Server Product Guide[17]

3BSE078159*System 800xA 6.0, System Guide Summary[18]

2PAA107280*800xA History Installation 6.0[19]

3BSE079455*Control Builder A Version 1.4 Product Guide[20]

16 3BSE041434-600 H

2 Introduction 2.1 Related Documentation

Table 2.1: Reference Documents(Continued)

DocumentNumberDocument TitleItem

3BDS013988*AC 100 OPC Server 6.0 Configuration and Operation[21]

2PAA108438*System 800xA Engineering, Application Change Management[22]

3BSE034463*System 800xA Network Configuration[23]

3BSE066471*Multisystem Integration version compatibility[24]

2PAA111691*System 800xA Licensing Information[25]

3BUS094370System 800xA Life Cycle Policy[261]

3BDD012511*Freelance - Engineering Manual OPC Server F[27]

3BSE085444-417*800xA Networks - NE800 Getting Started Guide[28]

3BUA002942*800xA Batch Management 6.0.x; Batch Server sizing, BestPractices

[29]

3BSE086256*Using System 800xA with Multiple Aspect Servers[30]

3BSE089211*Auxiliary Operator Workplaces in 800xA[31]

3BSE082294*Virtualization with Microsoft Hyper-V[32]

3BNP004865*AC 800M High Integrity Safety Manual[33]

NOTE:Document is not supplied on the 800xA software media. It is available at myABB/My Control System.1.

2.2 Hardware SelectorWhen planning a new project, or making an expansion to an existing installation,finding the right hardware is made easy by using the Hardware Selection. This toolhelps in finding the correct hardware in a few clicks.

The Hardware Selection is found at www.800xahardwareselector.com

2.3 Temporary Sales AuthorizationIn some cases it is required to keep System 800xA Product Management involved insetting up a system. This is for functions which for one reason or the other are critical

3BSE041434-600 H 17

2 Introduction 2.2 Hardware Selector

www.800xahardwareselector.com

in how they are setup and configured. There are also cases where functions andfeatures are critical to the business but which are not available in the product offering.There is a work process defined for this purpose called Temporary Sales Authorization,TSA. Throughout this document it is explicitly stated when a TSA is needed.

The TSA tool is found at https://sites.abb.com/sites/ControlSystemsTSAs/default.aspx

18 3BSE041434-600 H

2 Introduction 2.3 Temporary Sales Authorization

https://sites.abb.com/sites/ControlSystemsTSAs/default.aspx

3 Dimensioning and Deployment

This section describes the dimensioning and deployment of System 800xA.Dimensioning parameters to be followed are stated in this document. However, thereare many use cases and parameters out of control from the product specification thatmay affect the resulting system capacity or performance. The system integrator, whois dimensioning the actual installation, has the responsibility to validate that theanticipated capacity and performance is achieved.

The parameters defined in this system guide, and their maximum allowed values, areused when verifying the functionality and behavior of the system prior to release ofnew versions. For large and very large applications many of these parameters tendto be maximized at the same time, which increases the user's responsibility to validatesystem behavior before going into production. In cases of uncertainty please contact800xA Product Management through the regular support channels.

3.1 OverviewSystem 800xA can be deployed in different ways depending on the size of theapplication. The options available are determined by the number of operatorworkplaces needed, the size of the control application, as well as the required functions,for example history logs, batch plant needs, asset management, the amount of datarequired for external clients, and so on.

Generically System 800xA consists of operator workplaces, engineering workplaces,controllers, and server nodes. Smaller systems (up to 20,000 tags) can be built withall server functionality running in the workplace nodes. In other words, the node countequals the number of workplaces needed.

Larger systems require server functionality to run in separate server nodes. Dependingon the actual size, server functionality can be combined in different ways to optimallyuse the computer capacity. Each 800xA service referred to as a system function, andthere are rules defined how many of these system functions that can run simultaneouslyin one physical server node.

3BSE041434-600 H 19

3 Dimensioning and Deployment 3.1 Overview

Fieldbuses are connected through specific communication modules to the controllers.Normally the application run-time data is communicated to and from the controlapplication through this interface. In particular for Foundation Fieldbus and IEC61850 communication, acyclic data and service and configuration data iscommunicated through a connectivity function that bypasses the controller, and henceis installed in the server layer of the system. Data needed for object face plates, deviceconfiguration and asset management is routed through this communication.

Workplaces and Servers run on Windows operating system.

Virtualization can be used to combine multiple 800xA node functions onto a singleVMware ESX host computer. The total number of physical computers required in aninstallation may then be reduced, but further the management of the system may befacilitated. This also reduces the required space for computers, hardware acquisitioncosts for computers and cabinets, and operating costs (such as energy expenditures).

Operator workplaces can also be virtualized. There are always limitations, in whichcase the 800xA workplace runs virtualized and the physical workplace is implementedby dedicated hardware, so-called thin client hardware.

An engineering system is intended for engineering functions only, with no productionor control operations. The same basic configuration rules apply to an engineeringsystem as to systems intended for production. Engineering results produced in anengineering system are transferred to the production system, or to other engineeringsystems, such as an on site engineering system, using the Export/Import function.

To define the configuration for an 800xA system the following information shouldbe known and available:

• The size of the application - tag count, control loops, I/O count.

• The required speed of the control application.

• Number of controllers and tags per connectivity type.

• Number of active operator workplaces, screens, and the complexity of displays.

• Number of logs, and their sample rate.

• Number of Asset Monitors.

• Amount of data required from external clients.

• Number of connected field devices for the fieldbus protocols FOUNDATIONFieldbus, PROFIBUS and HART.

• Number of Intelligent Electronic Devices (IEDs) to be connected.

20 3BSE041434-600 H

3 Dimensioning and Deployment 3.1 Overview

• Number of data attributes per IED and the expected data change rate.

3.2 Application Size LimitsTable 3.1 shows the tag values and system configuration limits for system levelparameters.

Table 3.1: Tag Values

ValuesTags

60,000Tags (PC-0001)

120,000Tags for large system after Temporary Sales Authorization(TSA) review (PC-0001)

90,000Tags for large system using Load-Evaluate-Go (PC-0367)

20,000Tags in a system where 800xA services run on workstationoperation system (PC-0532)

Other limitations will apply for Heritage System connectivity as described in Section8 Heritage Control Systems.

3.2.1 Tag CalculationA Tag is the DCS system representation of a real life object in the plant, such as avalve or a pump. In System 800xA the tag is represented by an Aspect Object, andthe different real life aspects of the Tag are represented by the Aspects to the AspectObject.

There are many Aspect Objects in a system that does not represent those real lifeobjects, for example library objects. In order to do proper counting of Tags, an AspectObject representing a real life object is one which has one or more Faceplate Aspects.

For very large systems with a Tag count above the limit in Table 3.1 productmanagement must be consulted through the TSA (Temporary Sales Authorization)process.

3BSE041434-600 H 21

3 Dimensioning and Deployment 3.2 Application Size Limits

3.3 System Size LimitsTable 3.2 specifies maximum supported system sizes. Numbers in parenthesis referto redundant configurations. Note the limitation in node count if 800xA serverfunctions are allocated to nodes with a workstation version of Windows.

Table 3.2: Supported System Size

RemarksMaximum

Excluding remote client servers25 (50)Nodes with server functions(PC-0437)

Including remote clients80Nodes with client functions(PC-0437)

16Nodes in a system where800xA services run onworkstation OS(PC-0506)

Remote Clients

Terminal servers or Virtualizationhosts

10Remote client servers (PC-0380)

Per terminal server or Virtualizationhost

10Simple viewing (PC-0439)

Per terminal server or Virtualizationhost

5Full functionality (PC-0438)

Network Nodes

Clients and servers100Per client/server network

Controllers and servers60Per control network (PC-0289)

NOTE: 100 if only PM891 is used

Controllers, clients and servers60Per combined network (PC-0290)

Multisystem Integration

2Subscribers per provider (PC-031)

20Providers per subscriber (PC-030)

22 3BSE041434-600 H

3 Dimensioning and Deployment 3.3 System Size Limits

3.4 Definitions

3.4.1 Node TypesThe node type determines the main use of an 800xA system node as a client node, oras a server node, or as a combined client and server node. This is used to determinewhich software media folders need to be copied locally to the node during installation.

3.4.2 Node FunctionsSystem 800xA functionality consists of a number of selectable functions. The SystemInstaller installs and activates relevant software components on relevant system nodesto support the selected functions.

3.4.3 OptionsSystem functions may have selectable options to determine functional characteristicsor behavior. The System Installer activates options on relevant nodes to reflect theselections.

3.5 Node Function Deployment

3.5.1 General TopologiesSmaller systems, from a single node1 up to the max node count stated in Table 3.3can be built with all server functionality running in the workplace nodes. No othernodes are required, unless the system is required to run as a Domain, in which caseone (or two if redundant) domain controllers is required, see Domain or Workgroup.Such a system will be based completely on workstation hardware and workstationoperating system, see Figure 3.1 and Figure 3.2.

1. Legend to the node functions: Client - Operator or Engineering workplace, AS - Aspect Service,CS1, CS2, CSn - Connect Service, Batch, History, AO, … - Application Service

3BSE041434-600 H 23

3 Dimensioning and Deployment 3.4 Definitions

Figure 3.1: Single Workstation System

Figure 3.2: Workstation based system where all 800xA services run on the client machines1

Larger systems require the 800xA services functionality to run in separate servernodes. Depending on the actual size, service functionality can be combined in differentways to optimally use the computer capacity. Each service represents one nodefunction, and there are rules for how many functions can be put in one server node1

24 3BSE041434-600 H

3 Dimensioning and Deployment 3.5 Node Function Deployment

as in Table 3.3. The server nodes in such a system require server hardware and serveroperating system, see Figure 3.3.

Figure 3.3: Larger system with separate server nodes for 800xA services

For larger systems1, or when network segregation is required for any reason, thecontrol network and the client/server network can be separated. The connectionbetween the two networks is handled by the connectivity function. Note that theconnectivity functions allocated to a server node will determine the number of networkinterfaces that have to be supported by the server hardware. An example layout couldlook like in Figure 3.4.

3BSE041434-600 H 25


Figure 3.4: Large system with separated Client/Server and Control networks

Geographically distributed plants

Very distributed plants where each part is self-contained with respect to its controland operation functions can be built using System 800xA, still keeping the completeinstallation as one system. For example pipe lines, mines, district heating, water/wastewater, or plants where segregation between parts is needed (e.g. power plants withseveral units).

In this system topology there are controllers and operational clients dedicated to andphysically located at each respective part (location) of the plant. At least one aspectservice needs to be located at the same physical location, as well as the requiredconnect services. The solution is referred to as the 'Multiple Aspect Servers"configuration.

This topology requires an approved TSA. Please request the document according toref [30] to get more details about setup, configuration, and limitations.

26 3BSE041434-600 H


Figure 3.5: Example topology with Multiple Aspect Servers

The alternative solution supported by System 800xA is Multisystem Integration (MI,see Multisystem Integration). While MI support good integrity between parts, it stillrequires management of different systems that are connected. Sometimes this is anadvantage from a life cycle perspective, while in other cases there is a wish to haveone single system. In the latter case, Multiple Aspect Servers are usually required inorder to preserve the operation of one plant part if it gets disconnected from the restof the system due to a network failure.

3.5.2 Client Deployment800xA Clients, that is the operator or engineering workplaces, can be deployed eitheras rich clients or as remote clients.

The Rich Clients are 800xA client nodes running on workstation hardware containingall necessary 800xA software. To create a better operator environment there arefan-less, or very silent, rich clients that can be deployed directly at the actualworkplace. Another way to achieve a better operator environment is using a KVM,

3BSE041434-600 H 27


Keyboard-Video-Mouse, solution in which the workplace can be placed in a computerhall while only a small and silent device is present in the operator environment.

Remote clients are deployed in three different ways:

• Remote client towards a workstation.

• Remote clients towards a terminal server.

• Remote client towards a virtualized workstation.

The remote client itself will not run any 800xA software but will connect to the 800xAhost using RDP, Microsoft Remote Desktop Protocol.

Certified solutions are available for the above mentioned deployments.

Remote Clients

Remote clients are most often realized using so called thin clients as a small and silentdevice is preferred in the operator environment. A certified workstation serve thesame purpose if the environmental and maintenance requirements are neglected.

We certify thin clients based on their ability to run RDP and their ability to providesufficient performance.

A remote client towards a workstation in a 1 to 1 relationship is very similar to a richclient running a KVM solution. Depending on the hardware selection the performancecan be the same.

Remote clients connected to a terminal server will have performance limitations asthe servers resources needs to be shared between the different clients. The lack of agraphical processing unit, GPU, as available in the rich clients will affect theperformance of complex graphics. Due to performance considerations the terminalserver should be realized on a physical server not containing any 800xA nodefunctions.

Activity in the other clients connected to the same terminal server will also impactperformance.

This solution however has a maintenance advantage as in most cases only the terminalserver needs to be maintained and kept up to date.

The limitations for remote clients connected to a virtualized host are the same as forthe terminal server. Virtualized clients are limited to two monitors per virtual node.

28 3BSE041434-600 H


The individual virtual clients as well as the server needs to be maintained which takesaway some of the maintenance advantage compared to a terminal server solution.

Auxiliary Clients

An 800xA system can be amended with additional auxiliary clients (operatorworkplaces) from which the plant can be operated in case of a complete serverbreakdown of the main control system.

This configuration is for use cases where operational availability is a concern, and inparticular concern with the client server architecture, where data and operationalcommands pass through an intermediate server layer.

The solution is to install both Aspect- and Connectivity server functionality on a few(currently two) workplace nodes, and thereby making them self-contained so thatthey can work stand-alone in case of a complete server breakdown in the main system.Those clients are fully operational during normal system operation and can be usedas regular operator workplaces.

This topology requires an approved TSA. Please request the document according toref [31] to get more details about setup, configurations, and limitations.

Figure 3.6: An 800xA system with auxiliary clients

3BSE041434-600 H 29


3.5.3 Engineering SystemsEngineering installations can be made with a single node1 up to the max node countand functionality in the related production system. An engineering system is intendedfor engineering and verification/test functions only, with no production or controloperations. The same configuration rules apply to an engineering system as to systemsintended for production.

Engineering results produced in an engineering system are transferred to the productionsystem, or to other Engineering Systems, such as an on-site engineering system, usingApplication Change Management, the Engineering Repository or theAfwExport/Import function.

Soft controllers can be used for test of engineering as for example control logic andprocess display purpose, where physical controllers can be connected in order to testperformance and capacity.

Generally, Engineering Systems cannot be used to automate production as its servicesare not available always.

Figure 3.7: Single node engineering system

30 3BSE041434-600 H


Figure 3.8: Engineering system where the 800xA services are running on the engineeringworkstations

Larger engineering systems can have dedicated server nodes1 for the engineering database (the aspect directory), and connectivity functions to the test environment, thatcan be either soft controllers or regular controller hardware.

Figure 3.9: Large engineering system with multiple clients and separate server machines forthe engineering data base (aspect directory)

Application Change Management

Application Change Management (ACM) is a service (see Figure 3.10), which enablesthe change management capability for engineering solutions in System 800xA. It is

3BSE041434-600 H 31


a configuration management system designed to handle engineering configurationartifacts in form of afw files.

ACM is a client-server based architecture with communication access to System800xA. Engineering Clients from multiple 800xA systems in the same security zonecan connect to a single Application Change Management back-end service in orderto provide configuration management between Engineering and Production Systems.

ACM Server utilizes Microsoft SharePoint Foundation services that internally includesSQL Server database.

The ACM Client act as an interface between the ACM server and system 800xA.

For information on ACM server configuration, refer to [22] in Table 2.1.

Figure 3.10: ACM Topology

3.5.4 Function Allocation RulesTable 3.3 specifies the maximum number of allocations of each function that can beactivated in a system and per node. Numbers in parenthesis refer to redundantconfigurations.

Note that the numbers presented in Table 3.3 are upper limits - depending on systemload, the actual maximum for a given application may be smaller. See further sectionsfor information on performance calculations.

32 3BSE041434-600 H


Table 3.3: Node Function Configuration Rules

Remarks

Maximum Number of Allocations(figures for redundancy in

parenthesis)Node Function

Per NodePer system Workstation

(HW/OS)Server(HW/OS)

See Restrictions on NodeFunction Combinations in aSystem.

4(PC-0441)8 (PC-0440)Total number of nodefunctions per node

(PC-0442, PC-0443, PC-0502)111 (2/3)Aspect ServicesUnder certain conditions asystem can have up to 11aspect service node functions.Subject to approval (throughTSA).(PC-0368)

Connect Services

Max 48 controllers perinstance (PC-0444-PC-0446)

118 (16)AC 800M Connect

Max 80 IEDs per connectservice instance(PC-0486-PC-0489)

12*16 (32)IEC 61850 Connect

*Four IEC 61850 connects, ifno other node functions on theserver.

Max 1,000 devices perinstance (PC-0471-PC-0473)

148 (16)FOUNDATION FieldbusConnect

Max 2,500 devices perinstance (PC-0474-PC-0476)

118 (16)PROFIBUS & HARTConnect

Max 25.000 signal objects perinstance (PC-0468-PC-0470)

113 (6)PLC Connect

* Up to one per CS11*PNSM Server

3BSE041434-600 H 33


Table 3.3: Node Function Configuration Rules(Continued)

Remarks





Max 20 providers persubscriber system(PC-0480-PC-0482)

1520 (40)Multisystem IntegrationSubscriber Service

Max 2 subscribers perprovider system (PC-0483-PC-0485)

111 (2)Multisystem IntegrationProvider Service

See 800xA for AC 100(PC-0465-PC-0467)

112 (4)AC100 Connect

See the relevant SystemGuide section (PC-0462-PC-0464)

114 (8)AC 800F Connect

See 800xA for Advant Masterwith 800xA for Safeguard(PC-0447-PC-0449)

116 (12)Advant Master Connect

See 800xA for DCI(PC-0459-PC-0461)

-14 (8)DCI Connect

See 800xA for Symphony PlusHarmony (PC-0453-PC-0455)

-18 (16)Harmony Connect

See 800xA for Melody(PC-0456-PC-0458)

-112 (24)Melody Connect

See 800xA for MOD 300(PC-0450-PC-0452)

-16 (12)MOD 300 Connect

Only where AC 800M Connectis selected (PC-0477-PC-0479)

-18 (16)TRIO Connect

Applications

See 800xA History116 (12)800xA History EmbeddedData Collector

34 3BSE041434-600 H


Table 3.3: Node Function Configuration Rules(Continued)

Remarks





800xA History Archive nodecan not be combined with anyother node function of System800xA.

-11800xA History Archive

(PC-0490-PC-0492)-14Asset Optimization

One instance of Batch countsas 2 allocations (PC-0493-PC-0495)

-11 (2)Batch Management

(PC-0496-PC-0498)111 (2)Batch Witness

* Up to one per Server andmax per system.

-110*Calculation Services

(PC-0499-PC-0501)-16Information Manager

* Up to one per Server andmax per system.

1110*Scheduler Service

111SMS & e-mail Messaging

* Up to one per Server1110*SoftPoint Services

Sum of local and remoteclients (PC-0028)

1180Operator Workplace

Maximum 20 in engineeringsystem (PC-0011)

1110Engineering Workplace

Other Node Functions

(PC-0504, PC-0505)-11 (2)Domain Controller andDNS

3BSE041434-600 H 35


3.5.5 Restrictions on Node Function Combinations in a SystemThe following node functions cannot coexist in the same system:

• Harmony and Melody connect services.

• Harmony and DCI connect services.

• Multisystem Integration Subscriber Service and Multisystem Integration ProviderService (a system cannot be both subscriber and provider).

3.5.6 Restrictions on Node Function Combinations in a NodeThe Aspect Service function can be combined with other functions as given inthis section. However, for installations which are frequently engineered in runtime and where the size of change is more than 1000 simple objects (2-3 nestedlevels, 5-6 aspects per object) or more than 100 complex objects (7-8 nested levels,10 aspects per object, or more), it is recommended to install the aspect serviceson separate nodes (PC-0535). It is also recommended in systems larger than 10,000tags to install the aspect services on separate nodes when Load-Evaluate-Go (LEG)is used (PC-0536).

The Aspect Services can be combined with other services as shown in Figure 3.3 andFigure 3.4, up to a certain system size as shown in Table 3.4. Larger systems musthave the Aspect Service (services when redundant) running on a separate serverhardware.

Table 3.4: Aspect Service Tag Limit

ValuesAspect Service

40,000Tags when the Aspect Service must run on separate serverhardware (PC-0534)

Additional Restrictions in Connect Combinations

The following node functions cannot coexist on the same node:

• Asset Optimization options Maximo Integration and SAP Integration

• MOD 300 connect service and any other connect service including IEC 61850,or Information Management

• Harmony connect service and any other connect service including IEC 61850

36 3BSE041434-600 H


• DCI connect service and any other connect service including IEC 61850, orMelody Configuration

• Melody configuration service and Melody connect service

• 800xA history Embedded Data Collector and any of Information Management– IM Services, or Information Management – 800xA History Archive

3.6 Connect ServiceCalculationRules and LimitationsThe required amount of connectivity capacity is determined by the data subscriptionvolume. This means the number of clients and screens in the system, how many logsare needed, how much data is subscribed from outside the system itself, and howmuch data is required to a possible subscriber system in case Multisystem Integrationis used.

The first step of calculating how many connect service node functions are needed, isto sum up all the subscription volume, considering the update rate. Faster updatemeans more capacity needed.

The second step is to look at the total controller count in the system. The ConnectService has a numeric limitation to how many controllers it can handle see Table 3.7.If the total number of controllers in the system exceed the count given by the numericlimitation times the number of connect services, additional services are needed. Inother words, the required number of controllers is distributed across the calculatedamount of Connectivity Servers, considering the maximum number of controllers perConnectivity Server. If this number is exceeded, an additional server has to be usedeven if the data need does not require it.

The calculation is done using the 800xA Wizard. It can be used for capacitycalculations in the following cases:

• At initial purchase, where the 800xA Wizard will calculate the total ConnectivityServer need.

• When extending a plant, where the 800xA Wizard will calculate if there is a needto add more Connectivity Servers.

• For checking the capacity of an existing installation.

There is no manual methodology available, but the basic concepts are described inthe following sections to provide an understanding of how the calculations work.

3BSE041434-600 H 37

3 Dimensioning and Deployment 3.6 Connect Service Calculation Rules and Li

3.6.1 Load from Operator Client DisplaysLoad from operator client displays (graphics subscription) mainly depends on thefollowing parameters:

• Number of currently open displays. This is normally the same as the number ofscreens used (number of clients times their screen count).

• Update rate (how fast is a display point updated). The update rate can beconfigured from one second and upwards.

• Complexity of the display (this is a predefined parameter that can be modifiedby the 800xA Wizard user).

3.6.2 Load from History LogsLoad from history logs depends on the following parameters:

• Log count.

• Log rate.

• Whether Information Manager is used or only primary logs. This will also impactthe required throughput. An Information Manager log always assumes a primarylog, but not vice versa.

800xA History collects data in a similar fashion to basic history logs, that is 800xAHistory data collectors utilize a direct opcDA collection mechanism, see Figure 3.11.

38 3BSE041434-600 H


Figure 3.11: 800xA History Data Collection

Data is transferred to Information Manager in blocks at certain intervals using OPCHDA (Figure 3.12). This is called blocking rate. The blocking rate to InformationManager is a function of the collection rate, for example a one-second log would havea blocking rate by default of four minutes. Logs that are collected at a slower ratewill have a proportionally slower blocking rate. The default values for blocking rateare typically sufficient. The stagger utility must be used after an installation isconfigured or changed to balance the load on the 800xA System. It is not recommendedto manually adjust stagger for logs.

3BSE041434-600 H 39


Figure 3.12: Data Transfer to Information Manager

3.6.3 Load from External SubscriptionTable 3.5 lists the capacity parameter values for external subscription.

Table 3.5: External Subscription Values

ValuesExternal Subscription

10,000OPC items from external subscription per surrogate process (PC-0059)

4,000Externally subscribed OPC items throughput (items per second) persurrogate process (PC-0058)

4Max number of surrogate processes (PC-0288)

2,500OPC items subject to new or changed external subscription (PC-0287)

The external subscription will apply a load similar to the graphics subscription. Theload from the external subscription depends on the following parameters:

• Number of OPC items subscribed by the external client.

• Subscription rate.

The OPC server function in System 800xA that serves external clients is called theOPC surrogate. The surrogate reroutes the OPC request to the appropriate ConnectivityService. It is possible to use several OPC surrogate processes (Figure 3.13).

40 3BSE041434-600 H


Figure 3.13: OPC Surrogate Process

The OPC server surrogate is normally running in one of the Aspect Servers, if itcommunicates with the external application via DCOM. If the external applicationhas an access agent that needs to run in the 800xA System it is not recommended toinstall that agent in the Aspect Server. Instead, a separate Application Server shouldbe set up, running the third party access agent and the corresponding surrogate process.Also, for very large external data volumes (>10,000 OPC items), it is recommendedto install the surrogate in a separate Application Server.

3.6.4 General Considerations for External SubscriptionSetting up or changing the subscription will cause a load peak. This happensnormally in the system when calling up a display, etc., but for external clients thereis a potential for making large changes in this sense. Therefore, the amount of itemssubject to a new/changed subscription is limited.

The load impact from Asset Optimization access to Fieldbus Connectivity Serversis based on device count and required scan rates. Usually the scan rate is fairly lowin the order of hours per device, but with a large device count the resulting load maybe significant.

3BSE041434-600 H 41


The controller load impacts the throughput and the responsiveness of the ConnectivityServer. The controller load should not exceed 60-percent total load.

The rate of change for the subscribed signals has a certain impact on the capacityof the Connectivity Server (the OPC server detects if the signal has changed sincethe last sample). The given parameters are relevant for a change rate of around 50percent (half of the values have changed since last sample, or a parameter changeson average every second sample).

When the same value is subscribed with the same update rate from more than oneclient (more than one graphics display, a display and a log, or even graphics andexternal subscription), the Connectivity Service will optimize the requests to thecontroller with significant capacity optimization as result. The 800xA Wizard willconsider this combination effect and compensate for it according to typical usage.This means choosing the same update rate for different applications will pay back incapacity need.

Different update rates can be configured to save Connectivity Server capacity. Forexample, some displays can be set to update each three seconds instead of one second.Logs can be tuned according to process needs and in accordance with the executionof the control algorithm, and so on. The characteristics of the connectivity is not fullylinear, managing several update rates/log rates will decrease the throughput in itemsper second, but there is still a gain in optimizing the update rates.

The CPU power of the server usually is not the determining factor. This is whythe computer type is not a variable (input parameter) in the calculation rules.Recommended RAM size has to be followed.

3.6.5 Connectivity Server OPC DA/AE LimitsTable 3.6 details the Connectivity Server OPC DA/AE rules. These numbers reflectwhat the base system can support. Each connect option will be equal to or less thanthese numbers.

Table 3.6: OPC DA/AE Values

Values

50,000Sum of active OPC DA subscriptions per CS node for AC 800M(PC-0530)

30,000OPC DA throughput (items per sec) per AC 800M ConnectivityServer (PC-0292)

42 3BSE041434-600 H


Table 3.6: OPC DA/AE Values(Continued)

30/secondContinuous event throughput - Connectivity server(PC-0078)

1,000/s for 3seconds, plus 10/sfor 15 minutes

Event burst capacity (PC-0079)

200 events in100ms

Event handling capability per AC800M controller(PC-0080)

<60 secEvent burst recovery time (PC-0081)

3.6.6 Connectivity Server History ParametersThe Connectivity Server History parameters are detailed in section ApplicationParameters Summary

3.6.7 Available Functions Per Controller ConnectivityAll system functions normally work together with all connect services. However,there are certain restrictions in the functionality. Those are listed below. Details canbe found in the referred sections.

• Batch Management

For allowed combinations with connect services, refer Batch Management.

• Load-Evaluate-Go

Supported with AC 800M only. See section Load-Evaluate-Go.

• Device Management HART

Routing of the HART information through the controller is only supported withAC 800M and AC 870P controllers. For all other controllers the HARTMultiplexer Connect option needs to be selected in addition to DeviceManagement HART to collect the data from HART multiplexers.

• Device Management PROFIBUS

Supported together with AC 800M controllers.

• Device Management FOUNDATION Fieldbus

3BSE041434-600 H 43


Supported for AC 800M controllers and is required to configure FOUNDATIONFieldbus networks together with AC 800M.

• 800xA for IEC 61850

Supported for AC 800M, Harmony, Melody, and DCI. See section Restrictionson Node Function Combinations in a Node for limitations in how to combinenode functions on nodes.

3.7 ControllersThe maximum number of AC 800M controllers in an 800xA System is dependent onthe number of Connect Service node functions used, and the required throughput interms of subscribed data volume. Each connect service can handle a certain numberof controllers depending on the amount of data that is subscribed from them, whichin turn depends on the application size, and the size of the CPU selected. This meansthat in a practical case the number of controllers per connect service may be less thanthe specified max, see Table 3.7.

Table 3.7: AC 800 M Values

Values

Max 48 (can be less dependingon data throughput needs)

AC 800M Controllers per Connect Service nodefunction (PC-00112)

10/secondWrite transactions to AC 800M OPC server (PC-0063)

10msController CPU redundancy switch-over time(PC-0313)

The number of I/O channels per controller ranges from a few hundred to over 1,000depending on the CPU type and application.

3.8 Multisystem IntegrationMultisystem Integration (MI) allows viewing and operation of objects configured inone system (provider) from another system (subscriber).

Geographically separated systems with local control rooms can be controlled from acommon control room. MI enables sectioning of a multiline process to increaseintegrity and simplify maintenance. A common setup is to have one subscriber and

44 3BSE041434-600 H

3 Dimensioning and Deployment 3.7 Controllers

several providers, but each provider system can also serve multiple subscriber systemsin order to share a common resource (provider).

A Multisystem Integration subscriber requires one subscriber service per providersystem. Each Multisystem Integration provider requires a provider service. Theprovider and subscriber services can be installed in separate nodes or combined withother services depending on system size.

MI is for operation, not for engineering. All engineering is done locally on eachconnected system.

The central licensing system (CLS) is local to each system. Each system is orderedseparately, and a separate license file is retrieved for each system from the SoftwareFactory. This also means that each system is managed individually regarding updatesand upgrades, as well as initial system installation.

3.8.1 Read Only SubscriberRead only subscriber systems provides read only clients to office users. Subscribertags are only required in the subscriber system. The subscriber tag is required forevery provider tag that needs to be accessed, from graphics, alarms, trends and so on,in the subscriber system.

3.8.2 Supported System FunctionsThere is no limitation as to which system functions can be installed and used in aSubscriber System. The normal configuration rules apply.

The following functional constraint has to be observed for Device ManagementPROFIBUS and HART on a subscriber system:

Device Type Managers (DTMs) can be used on Subscriber Systems only for devicesthat are directly connected to the Subscriber System. For devices connected to theProvider Systems the DTMs will only be installed on the Provider Systems and canbe accessed through Remote Desktop.

For a Provider System certain limitations apply as follows:

• Information Manager can be used. Log consolidation must use the IMconsolidation feature. MI does not transfer any data in between IMs in thedifferent systems.

• Batch can be used, and is local to each provider (or subscriber). Batchcommunication with Control is not supported across multiple systems.

3BSE041434-600 H 45

3 Dimensioning and Deployment 3.8 Multisystem Integration

• PNSM is local to the provider system (or subscriber). There is no support forPNSM across multiple systems.

• If Asset Optimization is used in the provider system it must also be installed inthe subscriber system.

3.8.3 Supported ConnectsThe following connects are fully supported for Multisystem Integration:

Contact ABB technical sales support for advice on connects not listed.

• 800xA for AC800M.

• 800xA for Advant Master.

• 800xA for Symphony Plus Harmony.

• 800xA for AC 870P/Melody.

• 800xA for MOD 300 (ADHOC subscriptions and MOD 300 Displays are notsupported).

• 800xA for Safeguard.

• PLC Connect.

• Foundation Fieldbus.

• IEC 61850.

3.8.4 Configuration RulesThis describes the configuration limits for the MI Function.

Subscriber System

In the subscriber system, all specifications valid for a local system applies. The totaltag count in the subscriber system equals the tags in the subscriber itself, plus tagssummed up from all the provider systems.

Consider the number of provider connections that is possible to have per server (Table3.3 on page 33).

46 3BSE041434-600 H


Provider System

The limiting factor is the load that the clients in the subscriber system will add to theprovider system. Each workplace in the subscriber system is viewed as an externalOPC client to the provider system. If several subscribers are connected to one providerthe load of the all the workplaces in the subscriber systems must be considered.Calculations must be done to ensure that the connectivity server(s) in the providersystem can handle the load.

Communication

It is possible to use a bandwidth down to 512 kilobits/second between a Provider andSubscriber but display call-up times increase drastically when using bandwidths below1 Megabit/second.

Upload/Clean

It is recommended to upload maximum 5000 objects per upload session in order notto increase the load in the Subscriber system to a limit where it will becomeunresponsive. The Clean operation in the subscriber removes all objects uploadedfrom one Provider. Since the impact may be an unresponsive system this is apotentially dangerous operation. Since it is not possible to limit the number of objectremoved the recommendation is to only use Clean when the system is not runningproduction.

Compatibility Between the Releases

Multisystem Integration has a possibility to communicate between providers andsubscribers with different versions. However, the subscriber must have the same orhigher version as the provider with the highest version. Restrictions may apply forspecific revisions, refer to the compatibility matrix [24] in Table 2.1.

Configuration Examples

Figure 3.14 shows an example.

3BSE041434-600 H 47


Figure 3.14: Multisystem Integration Configuration Examples

3.9 PLC Connect

3.9.1 General CharacteristicsRefer to Table 3.8 and Table 3.9 for general characteristics.

Table 3.8: PLC Connect General Characteristics

SpecificationCharacteristic

Refer to Table 3.3 on page 33.Supported System 800xA Configurations

Refer to Table 3.3 on page 33.Maximum number of Connectivity Servers persystem

48 3BSE041434-600 H

3 Dimensioning and Deployment 3.9 PLC Connect

Table 3.8: PLC Connect General Characteristics(Continued)


Refer to Table 3.3 on page 33.Maximum number of PLC Connect signal objectsin one PLC Connect Connectivity Server runningin a separate node (large or medium size system)

Refer to Table 3.3 on page 33.Maximum number of PLC Connect signal objectsin a server optimized or single node configuration(small or single node system)

Refer to Table 3.3 on page 33.Maximum number of serial channels used in onePLC Connect Connectivity Server for dialedcommunication with PLCs via Comli or ModbusSerial

Set of C++ methods for user written applications.PLC Connect Real Time Data Access

Template dll with methods for user writtenapplications in Visual Basic.

Communication Server Pre Treatment

Most properties of the PLC Connect configurationaspects can be accessed via OPC or by the BulkData Manager.

PLC Connect properties

Communication interface for initiating anddisconnecting calls handled by the Dial managerfor dialed communication with PLCs.

Dial Manager Server Access

Redundant pairs of parallel executing PLCConnect Connectivity Servers are supported.

Redundancy

Configuration data is stored in the AspectDirectory. Aspect Server redundancy in 800xAcovers PLC Connect Configuration Data.

Audit of PLC Connect Configuration changes aresupported. Re-authentication and doubleAuthentication is supported for PLC Connectsignals.

Audit Trail

3BSE041434-600 H 49


Table 3.9: PLC Connect Characteristics


Boolean, integer, long integer, real, double, stringPLC Connect signal types

Typically less than 2 secs for a polled 9.6 Kbaudserial protocol.

The time from a change of a value in thecommunication server until an indication on thescreen

Modbus RTU Serial, Modbus TCP/IP, Comli,Sattbus over TCP/IP, and OPC DA client (2.05A)

Built in PLC protocol drivers

3.9.2 Performance DataPLC Connect performance data is listed in Table 3.10.

Table 3.10: PLC Connect Communication Characteristics


Dependant on several parameters:Time from a process change until signal isavailable in PLC Connect communication server. Speed of field-bus used.

Other products used (3rd party OPC servers).Requested update cycle time set in PLCConnect.Total number of signals using the availablesystem capacity.

Typically less than 2 secs for a polled 9.6 Kbaudserial protocol.

Time from a value change in PLC Connectcommunication server until an indication onscreen.

Typically less than 1 sec for a polled 9,6 Kbaudserial protocol.

Time from a manual control in the processgraphics until a value change on a terminal inPLC.

50 3BSE041434-600 H


3.9.3 Alarm and Event Performance DataPLC Connect alarm and event performance data is listed in Table 3.11.

Table 3.11: PLC Connect Communication Characteristics


25 alarms/events/secContinuous alarm load.

1,000 alarms/events every 10 mins (as long asthe continuous load is less than 25/sec)

Burst capabilities.

3.10 Load-Evaluate-GoThe Load Evaluate Go (LEG) function makes it possible to download a revised versionof an application to the controller without it interfering with the running version. Themodified version is started in passive mode and will be executed in parallel with therunning version and an Evaluation Report tool shows any differences in applicationoutputs as for example communication variables, IO and alarm conditions in realtime. The new version is taken in operation on the users demand, once it is provento perform correctly.

The LEG function is available with AC 800M only, to be used only for non-SILmarked applications. Note that application execution as well as memory space in thecontroller is doubled during a LEG session. This has to be considered whendimensioning the controller capacity.

When the Load-Evaluate-Go functionality is used for downloading applications tothe controllers, the system size (Table 3.1) is limited. See also Restrictions on NodeFunction Combinations in a Node.

3.11 LibrariesLibraries makes it possible to structure, package and protect information into higherlevel version controlled units. Libraries allow reuse of information when configuringa customer system for a specific use or industry which makes the engineering processmore effective.

The library vendor is responsible to make sure the library does not affect the capacityor performance of the system. Table 3.12 provides a recommendation for library

3BSE041434-600 H 51

3 Dimensioning and Deployment 3.10 Load-Evaluate-Go

vendors, who stays within limitations and does not cause performance concerns duringengineering activities (export, import, upload and so on):

Table 3.12: Libraries Values

Values

40Max number of types

3,000Max number of objects

10Max depth (hierarchical levels of dependencies)

50 MBStatic size of export

It is recommended to package libraries into system extensions.

3.12 Property TransferThe Property Transfer function is mainly used to transfer data between System 800xAand other systems. The other systems need to expose the data as OPC DA Server.Property Transfer can also be used for transferring data between controllers that lacknative support for data transfer.

Property Transfer can process the data, entering it into a suitable algorithm, deliveringthe results to aspect object properties.

The performance of Property Transfer is highly dependent on the performance of thesystem or controllers that receive the data (Table 3.13).

Table 3.13: Property Transfer Values

Values

1,000OPC items via Property Transfer (PC-0061)

100OPC items per second via Property Transfer (PC-0060)

52 3BSE041434-600 H

3 Dimensioning and Deployment 3.12 Property Transfer

3.13 System NetworksSystem communication in System 800xA is based on Ethernet and TCP/IP networks,which are functionally and, in most cases, also physically built in layers.

The different layers in the network are a logical definition of the different levels ofcommunication. In an installation, several protocols can share the physical media,depending on how the system is deployed.

The Plant Network can be dedicated for process automation purposes or be part ofthe plant intranet already available on a site.

The Client/Server network is used for communication between servers and betweenclient Workplaces and servers. This is a trusted network zone that should be protectedfrom unauthorized access. It is a private IP network that uses static addresses.Unauthorized access to nodes on the Client Server Network can be prevented usingIPSec and by separating the Client Server Network from other networks with firewalls.

The Client/Server Network supports network redundancy using the RNRP protocol(Redundancy Network Routing Protocol) and redundant Ethernet switches.

The Control Network is used for communication between Controllers and betweenControllers and Connectivity Servers. Similar to the client/server network the ControlNetwork is a trusted network zone that should be protected from unauthorized access.It is also a private IP network that uses static addresses. The Control Network is basedon Ethernet using the MMS protocol on top of a TCP/IP protocol stack, plus additionalservices for time distribution, redundancy management, etc. The Control Networksupports network redundancy using the RNRP protocol and redundant Ethernetswitches. Controllers connect to the control network via dual built-in network ports.

Using RNRPs Backbone area concept it is possible to build systems networks withup to 64 network areas (refer to [28] in Table 2.1. , Section 2: Network Redundancyand Routing).

3.13.1 800xA Network EquipmentSystem 800xA Network equipment provides pre-configured industrial networkequipment verified with System 800xA to ensure top quality performance andprotection against cyber threats.

Wired switches (NE800) include a set of rack and DIN mounted switches, RedundantNetwork Routing Protocol (RNRP) routers, and a wide range of modular transceivers(SFP) specifically developed for use in System 800xA.

3BSE041434-600 H 53

3 Dimensioning and Deployment 3.13 System Networks

The ABB Network Configuration Tool (NeCo) is a configuration tool which enablesusers to efficiently configure and maintain 800xA Network Equipment from onecentral location.

Refer to refer to [28], NE800 Getting Staretd Guide, in Table 2.1. for quick start-upguidance to setup and configure a network of System 800xA, using 800xA Networkequipment.

Managed Switches

NE800 are industrial grade managed switches which are primarily intended forclient/server and control networks. These switches can be managed using NeCO,using the web-interface, or using the Command Line Interface (CLI).

Table 3.14: Managed Switches

DescriptionNetwork Switches

DIN-mounted 10 ports managed switchNE810(8 ports - 10/100Mbit RJ45 ports and 2 ports - Gbit SFPports).Redundant 24V DC-power input.

DIN-mounted 19 ports managed switch.NE820(8 ports - 10/100Mbit RJ45 ports, 7 ports - Gbit RJ45 portsand 4 ports - Gbit SFP ports).Redundant 24V DC-power input.

Rack-mounted 19 ports managed switch.NE840(8 ports -10/100Mbit RJ45 ports, 7 ports -GbitRJ45 ports and4 ports - Gbit SFP ports).110/230V AC-power input.

Routers/firewalls

NE800 RNRP routers are industrial grade routers which support RNRP (RedundantNetwork Routing Protocol) and which can also be configured as a firewall. Therouters/firewalls can be utilized to segment a system into network areas and also intosecurity zones. NE870 is primarily intended to be used to separate client/servernetworks and control networks, to separate different control and safety networks, butalso to create RNRP tunnel areas between RNRP networks. NE871 has three ports

54 3BSE041434-600 H


and is therefore primarily intended to connect a client server network to a plant networkor to a DMZ (demilitarized zone), or to create RNRP tunnel areas.

Table 3.15: Routers/firewalls

DescriptionRouter/Firewall

DIN-mounted 11 ports RNRP router and firewall.NE870(3 ports - 10/100/1000Mbit RJ45 ports and 8 ports -10/100Mbit RJ45 ports).Redundant 24V DC-power input.

DIN-mounted 3 ports RNRP router and firewall.NE871(3 ports 10/100/1000Mbit RJ45 ports).Redundant 24V DC-power input.

Lightly Managed Switches

The lightly managed switches are intended for small systems, situations whenmonitoring is not required, and/or when ring redundancy is not needed. These switchescan be partly managed using the DIP switch settings accessible from the top of theunit.

DescriptionSwitch

DIN-mounted 5 ports lightly managed switch, 4 10/100MbitNE801RJ45 ports &100Mbit LC optical port.Redundant 24V DC-power input.

DIN-mounted 5 ports lightly managed switch; 410/100/1000Mbit

NE802

RJ45 ports & amp; 1Gbit SFP.Redundant 24V DC-power input.

3BSE041434-600 H 55


3.13.2 Network ConfigurationThe maximum number of nodes (workstations, servers, and controllers) in one controlnetwork area is given in System Size Limits. The limitation is primarily due to limitednetwork resources in controller nodes.

For larger systems a split into separate client/server and control networks isrecommended, if needed with several control network areas

Also for smaller systems, the network can be split into separate client/server andcontrol networks.

It is recommended to use 100 or 1,000 megabits/second full duplex switched, Fastor Gigabit Ethernet communication between clients and servers. Controllers with theprocessor modules PM85x and PM86x use 10 megabit/second half duplex. PM 89xuses 100 megabit/second full duplex).

Redundant Ethernet networks need to have similar timing properties. The throughputon the primary and the secondary network must be similar so that the networkperformance, in case of a network failure, is the same as when both networks areoperational. The message transport time between any two nodes must not differ morethan 300 milliseconds between the primary and the secondary network.

Refer to [23] in Table 2.1, Section 8: Ethernet and Network Equipment.

3.13.3 Control Network Clock SynchronizationSystem 800xA can use multiple methods for Time Synchronization. How to choosewhich to use is described in the users manual Network Configuration (refer to [23]in Table 2.1.

Table 3.16 shows the clock synchronization accuracy based on the type of controlprotocol.

Table 3.16: Control Network Clock Synchronization

Accuracy Per NodeType of Clock Synchronization

1 msec.High Precision SNTP (PC-0516)

200 msecs.SNTP (PC-0517)

1 msec.CNCP (between AC 800M) (PC-0518)

56 3BSE041434-600 H


Table 3.16: Control Network Clock Synchronization(Continued)

200 msecs.CNCP (AC 800M to Process Portal) (PC-0519)

3 msecs.MB 300 network (PC-0520)

3.14 System RedundancyAn 800xA system can be made redundant at all levels from the I/O to the systemlevel, including networks. Hot standby concepts as well as duality concepts are useddepending on the actual case, and redundancy can be selected independently fordifferent parts, for example redundant communication and redundant server can beused while only selected controllers are redundant. This gives the possibility to tunecost against availability requirements.

Clients (operator console) do not support redundancy within the node itself. Insteadseveral clients should be arranged so that sufficient availability is achieved. Whenclients are run by a terminal server, or as virtual machines, it is important to observethe common points of failure that may occur, and arrange the setup accordingly. Atleast two terminal servers or virtualization hosts are needed to avoid that a singlepoint that can kill all clients.

At switchover of redundant Connectivity- and Aspect servers there will be a shortnon-responsiveness of the clients, while they are connecting to the redundant server.Likewise when they re-connect to the original server when it comes back again. Thistime will depend on several factors, and will be arbitrary different for different clients.Larger systems typically experience the longer time. Aspect Servers in 2oo3 redundantconfiguration typically experience shorter switchover time versus 1oo2 configuration,depending on affinity setting. The normal fault condition is when the server hardwarebreaks and a switchover is triggered. Manually initiated switchover, by for exampleshutting down the server, may cause longer switchover time.

Aspect servers in 2oo3 redundant configuration are capable of determining if writeoperations to the Aspect Directory have been made (majority decision). This wouldoccur if dual network failures happen. If one server is lost the remaining two serverswill still synchronize. The 2oo3 recovers automatically after a single server failure.

In 1oo2 configuration it is not possible to determine which server is correct if bothhave changed their content while being separated (due to for example a dual networkfailure). This can happen if there are applications that make write operations to the

3BSE041434-600 H 57

3 Dimensioning and Deployment 3.14 System Redundancy

Aspect Directory, or if engineering changes are made while the servers were separated.In this case the system will decide itself which server is considered correct.

The switchover times for redundant Connectivity Server and Aspect Server are shownin Table 3.17.

Table 3.17: Redundancy Switchover Time

TypicalperformanceParameter

Typically <2 secs.Switchover time Connectivity Server (PC-0521)

Typically 2-4 secs.Switchover time Aspect Server (1oo2, 2oo3 redundancy) (PC-0522)

3.15 System Deployment

3.15.1 Domain or WorkgroupA System 800xA installation can be configured either as a Domain or as a Workgroup.

A workgroup becomes increasingly difficult to administrate as the number of usersand computers grow. Workgroup should therefore be considered only for small systemswith few users.

For installation as a production system, it is recommended that the 800xA systemforms its own domain, i.e. it should not be part of a larger domain, such as a corporatenetwork domain.

An engineering system is intended for engineering only, i.e. not for production. Withrespect to domain membership there are three possible alternatives for an engineeringsystem:

• The system can be added to an existing domain (e.g. in an office network domain)

• The system can be defined as a new domain added to an existing domain controller

• The system can be defined as a separate domain, with its own domain controllerand DNS server, or as a workgroup (i.e. same as a production system).

3.15.2 Supported Operating SystemsSystem 800xA version 6.0 runs on Windows 8.1 and Windows Server 2012 R2. Fromversion 6.0.3 Windows 10 IoT 2015 LTSB was introduced, and in version 6.0.3.1

58 3BSE041434-600 H

3 Dimensioning and Deployment 3.15 System Deployment

Windows 10 IoT 2016 LTSB and Server 2016 are introduced. The table below givesan overview. Users can select at update from earlier version 6 installations to keepearlier operating systems, or to update the operating system to the latest one.

Table 3.18: Supported operating systems

Operating system supported800xAversion

Windows 8.16.0, 6.0.1,6.0.2 Windows Server 2012 R2

Windows 8.16.0.3Windows Server 2012 R2Windows 10 IoT 2015 LTSB, Windows 10 Enterprise 2015 LTSB

Windows 8.16.0.3.1Windows Server 2012 R2Windows 10 IoT 2015 LTSB, Windows 10 Enterprise 2015 LTSBWindows Server 2016Windows 10 IoT 2016 LTSB, Windows 10 Enterprise 2016 LTSB

It is possible to run a mixed environment, for example some workstation nodes onWindows 8.1, and some on Windows 10. This makes it possible to extend the plantusing the latest operating system while existing nodes are untouched, and/or to makeselective update of the operating systems based on user convenience. However, aredundant pair should always run on the same operating system except during anupdate phase. Support for older operating systems may be taken away prior to theirtechnical end of life. See ref [7] for details about support for 3rd party software.

With Windows 10 Microsoft introduced different branches of the operating system.These branches have different update strategies. For 800xA 6.0.3 the Long TermService Branch (LTSB) is supported. LTSB versions are established at regular intervalsbase lining the functionality current at that time, and from there they are supportedwith security updates and corrections for 10 years. LTSB is supported only in Windows10 Enterprise edition of Windows 10.

Standard OEM workstations are supplied with Windows 10 Professional, which hasanother strategy for updates called Current Branch for Business (CBB). Windows 10Professional is currently not supported for production systems. The reason for this isthat there are functional changes in the CBB, and the impact of the updates may

3BSE041434-600 H 59


become bigger than those on the LTSB. Bigger changes to 800xA software may haveto be made. This may take longer time, during which a system running a CBBoperating system cannot be updated and will hence not be supported by Microsoft,and may potentially become unsecure. Users who want to use standard OEMworkstations shall use Windows 8.1 for their production systems. However,engineering systems can be run with Windows 10 Professional given that the updatestrategy can be accepted.

For System 800xA version 6.0 only 64-bit (x64) operating systems are used to allowfor memory expansions.

More details about the supported operating systems including their updates are listed,refer to [7] in Table 2.1. This document can be found in myABB/My Control System.

The US English version of the operating system is required even if an NLS packagefor translation of 800xA functions into another language is used.

The Server Operating System licenses use the processor and CAL, Client AccessLicense, model.

One server license is required for each node in System 800xA running ServerOperating System. Note that additional server licenses are required if using more than2 CPU's.

In addition a number of Client Access Licenses (CALs) are required. CAL's comesin two flavors, Per Device or Per User.

The Per Device model means all physical devices, in System 800xA terms allworkstation nodes and all server nodes, each represent one CAL. This model isadvantageous when multiple operators share the same operator client (shift operators).The number of CALs required equals the sum of all PC nodes in the system.

The Per User model means each named user accessing the server software representsone CAL. This model may be advantageous if a relatively small number of operatorsoperate the plant from a large number of different clients throughout the plant. Alarger number of operator client nodes can be installed without purchasing CALs foreach and every one of them. The number of CALs required equals the sum of allnamed users accessing servers in the system.

In the per user mode it is not possible to have one single operator account and letall operators use it. This would break the terms of the Microsoft license agreement.

In most installations of System 800xA the Per Device Model is to prefer as the devicesare known at an early stage.

60 3BSE041434-600 H


Comments to Cloning of PCs at Installation Time

It is strictly recommended to perform the 800xA software installation on all nodesusing the System Installer.

In a system, there are usually many nodes that will have the same installation, suchas all clients, or certain server groups. In these cases, efficiency may increase if PCscan be cloned instead of installing them one by one. Hardware independent diskimaging tools may solve this problem by backing up an installed PC and transferringthe image to another PC. Pure disk imaging tools allow making an image backup ofa PC. However, if this image is installed on another PC, any small differences (suchas different vendors for the hard disk or different BIOS versions) may not be correctlydetected by the operating system and could jeopardize the result. In some cases, noissues will be identified during the initial startup and operation, but will show up laterduring regular operation. No third party packages are formally verified with System800xA and therefore proper function cannot be guaranteed. Any usage must be basedon knowledge about the limitations and risks presented by these tools.

Regardless of how this procedure is performed, the end user is responsible toensure that third party software license agreements are not violated.

3.15.3 Workstation vs Server Class Version of WindowsA workstation version of Windows is used for all client nodes.

For server nodes (and combined client/server nodes) either a workstation or a serverversion of Windows can be used. The following rules apply:

• When a workstation version of Windows is used for a server node, the systemsize is limited. For more information, refer Table 3.2.

• Certain features always require a server version ofWindows. These are:

– Domain Controller

– Batch Server (except in single node engineering system)

– 800xA History Archive

– IM Server

– Asset Optimization (in all configurations exceptsingle node system)

– Harmony connect services

3BSE041434-600 H 61


– Melody connect services and/or and Melody Connect Configuration Server

– MOD 300 connect services

– DCI connect services

Note that all servers in a redundant group must use the same operating system.

3.15.4 Hardware Infrastructure

Server and Workstation Hardware

System 800xA is developed to run on generic hardware running Windows operatingsystems. At 800xA version 6.0.3 the System 800xA Workstation is introduced. TheSystem 800xA Workstation is based on defined workstation hardware from majorvendors, and it has Windows 10 Enterprise IoT installed which supports the long termservice branch. The System 800xA Workstation is available from the following portal.

• https://link.arrow.com/abb/customer/account/login/

It is recommended to use the System 800xA Workstation as well as the Serverhardware, which is validated with 800xA software and available through the portal.Optionally an OEM workstation can be upgraded using a Windows 10 Enterpriselicense still assuing the same hardware is used. The picture below describes the optionsavailable to provide hardware and operating systems for an installation.

Figure 3.15: Hardware and Operating Systems delivery chain

62 3BSE041434-600 H


https://link.arrow.com/abb/customer/account/login/

A Server is referred to as a computer running Windows server operating system onhardware intended for server use. A Workstation is referred to as a computer runningWindows client operating system on hardware intended for client use.

Server hardware is of higher quality and is generally less error prone.

Workstation hardware most often contain graphical capabilities lacking in the serverhardware.

800xA is preferably, for best performance and reliability, deployed as a client-serversystem with servers running the server functions and workstations used for clients.

For smaller systems it is possible to build it entirely on workstations as described inNode Function Deployment.

CPU, Memory, and Disk Requirements. The maximum per node numbers in Table3.3 assume 800xA certified hardware. Memory requirements are:

• For up to 4 function allocations: minimum 8 GB.

• For up to 8 function allocations: minimum 16 GB.

Add memory to those figures for the cases below:

• Nodes that combine client and server roles require an additional 8 GB.

• Nodes that host the Aspect Services function require an additional 8 GB.

• Nodes that host the Batch Management function require an additional 8 GB.

• Nodes that host the History functions require an additional 8 GB. In some cases,additional memory can be purchased to increase performance. For InformationManagement History, the extra memory can be used for the oracle database. For800x History, the memory can be used for RTDB. See product specificdocumentation for specific recommendations

• Systems running Configure-deploy / Load-Evaluate-Go require an additional 8GB in all nodes running Engineering workplace or Aspect Services function.

The hardware available in the portal, used for the 800xA workstations and servers,is validated and approved for use with 800xA. Using other hardware should be avoided.If for any reason alternate hardware is used, the user (delivery project, channel) isresponsible to ensure proper operation. Support for such configurations may havelimitations and may also be subject to cost. ABB may validate additional hardwarethat can be applied for special use cases, but which is available through other channelsthan the portal. Such cases could be different environmental specification for example.

3BSE041434-600 H 63


Acquiring validated hardware will give you some options in terms of CPU, memoryand hard disks. Table 3.19 are the recommended minimum requirements.

Table 3.19: Minimum Requirements

RequirementHardware

Intel 4 cores with hyper threading 2GHzServer CPU

Intel 4 cores with hyper threading 2GHzClient CPU(some server functions)

Intel 2 cores with hyper threading 2GHzClient CPU(no server functions)

8 GB min, 16 GB recommendedMemory in server(depending on number and type of function allocationsas per above)

16 GB min, 16 GB recommendedMemory in client(some server functions)

8 GB min, 8 GB recommendedMemory in client(no server functions)

80 GB min, 2 disks and RAID 1 recommendedServer disk(excluding data storage)

80 GB minClient disk

For optimal graphical performance the video card selected should have 512MB/display.That means that a 4-screen solution preferable should have a video card with 2GB.

For more normal graphical performance expectations 256MB/display might besufficient. Going below that will have a negative performance impact.

Additional disks are required to run application servers storing historical data (IM,800xA History). Please refer to those calculations in [9] and [19] in Table 2.1.Generally but depending on system size performance will improve by installingseparate disks for the following node functions:

• Information manager

• 800xA History, and its DCN node function

• Batch

64 3BSE041434-600 H


• Aspect Directory

• Basic History (connectivity servers)

For virtual servers please see [3] in Table 2.1.

Hardware recommendations to be used with virtualized installations is essentially thesame, but selected such that special provisions for disk arrangements, multiple CPUcores, and so on can be fulfilled. For best performance and capacity in virtualenvironments these recommendations should be followed. Note that for some nodefunctions there are limitations in capacity if virtualization is used (e.g History, Batch).Please consider applicable sections of this document.

Refer also to the computer manufacturer's home page for their computers support ofdifferent virtualization software.

Environmental Specifications. The environmental specification provided for therespective PCs by the PC vendor applies. Normally this means workstations andservers need to be installed in the control room or the electrical room. Installation inthe production area or in the field normally requires special measures. Installation inharsh environments like ships and vehicles may require additional mechanicalmeasures.

Network Appliances

System 800xA is developed to run on a generic TCP/IP infrastructure. As a qualityassurance Industrial IT certification is performed on suitable high quality networkappliances from major brands.

Please refer to [23] in Table 2.1 for information on how to configure and dimensionyour network.

3.15.5 Third Party SoftwareEach version of System 800xA is verified with particular versions and revisions ofunderlying third party software. Refer to [7] in Table 2.1.

It is mandatory that the correct versions of the software are used.

In 800xA 6.0, most of the third party software is installed by the System Installer.

3BSE041434-600 H 65


3.15.6 VirtualizationVirtualization can be used in 800xA Systems to combine multiple 800xA server nodesonto a single computer. Virtualization can be used for engineering systems as wellas for production systems.

System 800xA supports hardware virtualization of system nodes. Each virtual machinecorresponds to a system node, and all configuration rules defined in this documentapply. However, since virtualization means that several virtual machines share theresources of one physical machine, the number of node functions needs to besummarized over all virtual nodes that run on the same physical machine.

Virtualization is supported for virtualization of 800xA Server nodes and 800xA clients.It is not recommended to virtualize Terminal Servers due to the high CPU usagerequired in multiple sessions for the rendering of graphics. Since the virtual clientsare more demanding on physical server resources, they must be run in virtual hostservers which are not running virtual 800xA server functions.

Virtualization of 800xA Clients should be restricted to ancillary operator workplaces,and not used for main operator workplaces due to limitations in supporting full graphicfunctionality.

Server Node Virtualization

[3] in Table 2.1 may contain limitations applicable to the virtualized configurations.It is recommended to read this guide prior to offering a virtualized installation.

Essentially, the configuration and dimensioning rules for 800xA server nodes statedin this document apply also when the node is installed as a virtual machine. Giventhat the rules in this System Guide are followed, the performance data stated herewill be fulfilled. [3] in Table 2.1 details how a virtualized system is planned, set up,and taken into operation. Note that there may be exceptions to this. Refer to the systemconfiguration limit Table 3.2 on page 22 through Table 3.6.5 on page 42.

The maximums recommended for virtualization are summarized in Table 3.20:

Table 3.20: Virtualization Parameters

ValueParameter

2Physical CPU per server

4Physical Cores Per CPU

2GB + (8.5GB per VM)Physical RAM per server

66 3BSE041434-600 H


Table 3.20: Virtualization Parameters(Continued)

ValueParameter

(Sum RAM per VM) +(Sum HD per VM) + 250GB

Physical HD

(Sum networks to connect)Physical Network Adaptor Ports

8Virtualized system node functions per physical server

4VM Maximum node functions

4VM vCPU

8GBVM RAM

Based on requirements of thenode function.

VM HD

For disk intensive nodefunctions like IM, 800xAHistory, and in large systems,consider one disk per virtualmachine

Exceptions to the virtual 800xA server maximums are in Table 3.21:

Table 3.21: Exception

ExceptionProduct

Batch Server VM must have 12 GB RAM800xA Batch Server

Note that the ability to run multiple node functions on the same node is significantlyincreased relative to earlier versions of 800xA, eliminating much of the reasonfor using virtualization to reduce the number of servers needed in a system.

Client Virtualization

Client virtualization involves running a client node as a virtual machine running thesame operating system as the rich clients. Each virtual client is considered to be a

3BSE041434-600 H 67


node with client function, see Table 3.2. For more information on Client Virtualization,refer to [3] in Table 2.1.

The following recommended maximums (Table 3.22) should be observed whenplanning a system implementation. Exceeding these values may result in longer calluptimes. This must be evaluated by reproducing the specified configuration beforerecommending the solution to a customer.

Table 3.22: Virtual Client Parameters

ValueParameter

4*Monitors

500*OPC Items per graphic display

0PG2 Animated graphics (e.g. Roll)

1PG2 Trend

2Faceplates

* This provides a maximum recommended total of 2000 OPC Items which aredisplayed across the 4 screens in 4 graphic displays. Exceeding this may result inlonger callup times.

3.15.7 Virtualization Host SoftwareThe primary choice of virtualization software is VMware ESXi. System 800xAcontinuously follows the life cycle of third party software. This is also valid tovirtualization host software. For more information on Virtualization Host Software,refer to [7] in Table 2.1.

There is also a possibility to use Microsoft Hyper Visor (Hyper-V) for virtualizedinstallations. Currently supported is the Microsoft HyperVisor which is part of theWindows Server Operating System (Hyper-V support on Server 2016).

The obvious advantages using Hyper-V are

• There is no additional software layer with its own life cycle.

• Hyper-V is part of the Operating System license, so no extra cost.

For configuration and set-up please refer to the user guide ref [32] in Table 2.1, whichalso contains advice for when to use ESXi and Hyper-V respectively.

68 3BSE041434-600 H


Hyper-V currently does not support 800xA Multisystem Integration, neither providersnor subscribers. Providers are usually small and normally virtualization does not payback. Subscribers are usually larger systems in which case it is recommended to useESXi in case it should be virtualized.

Please note that the usage of Hyper-V is under managed introduction and is subjectto TSA request.

Safety installations are currently not supported on Hyper-V.

There is also other system functionality currently not supported on Hyper-V. Pleaserefere to system release documentation (release notes).

3.15.8 System InstallationBuilding a system is done using the System Installer. This tool allows the user tospecify the different nodes and their function throughout a system. The System Installerdeploys appropriate software on each node, using the 800xA software media. TheSystem Installer also does basic configuration in order to facilitate the set-up.

The System 800xA System Installer installs all relevant 800xA and required 3rd partysoftware on all nodes (with the exceptions listed below), but excluding such thirdparty software that is separately licensed to the end user.

The following third party software is handled separately:

• Microsoft Windows - must be installed separately before running the 800xASystem Installer. See Workstation vs Server Class Version of Windows.

• Excel is needed on clients where Bulk Data Engineering tools or Excel basedreports are used that requires prior installation.

• Word is needed on nodes where Document Manager, Control Builder MDocumentation tool, or PG2 Display Documentation tool is used (optionally onnodes that have the Engineering Workplace function) that requires priorinstallation.

• Oracle and SQL Server are installed by the System Installer on nodes where nodefunctions that require them are located.

The following ABB software is not installed by the System Installer but needs to beinstalled separately after the system has been installed and configured. Proceduresfor installation are described in the respective product documentation.

• System 800xA Language Packs

3BSE041434-600 H 69


• PROFIBUS HART DTM's

• Various libraries (e.g. FF Device Library, PCDL/EL, Probase, etc.)

• 800xA for Freelance

• Heritage system configuration tools

• Third party (i.e. non ABB) connect packages.

• 800xA Smart Client

3.15.9 System UpdateThe System Installer supports automated updates of earlier version 6 installations.From version 6.0.3 update is also supported on-line node by node. Update can bemade using the latest 800xA software media, or a package containing the updatesoftware only can be downloaded.

The update is normally managed from one node in the system, usually an engineeringnode. The update software is distributed to all other nodes, and analysis is made asto what software needs to be installed on each node. All updates need to be installed,there is no possibility to select parts that mitigate certain known issues. It is stronglyrecommended to use the update function and avoid any manual attempts to updatethe system. This will ensure a controlled, complete and consistent update.

3.15.10 Deployment in IT EnvironmentEngineering systems can potentially run on PCs primarily intended as office PCs.However, compatibility concerns with third party software may arise. Since 800xAis verified with a defined set of Third Party Software compatibility with officeenvironments cannot be guaranteed. For these use cases it is recommended to run the800xA software virtualized, for example using VMware Workstation as the hostsoftware.

3.15.11 Software Media and LicensingAll 800xA software is delivered on a software media that is ordered separately. Theright to use the software is controlled by software licensing. For more informationon Licensing, refer to [25] in Table 2.1.

70 3BSE041434-600 H


Overview

The 800xA System licensing is the means to scale a system to fit the actual needs.By offering the system as a base system with options, with scaled tags, logs, etc., thesystem can be cost wise optimized for each use case. When needed, expansions canbe purchased and the system grows with the customer's needs.

It is important to recognize that installation is separate from licensing. Installation isnever prevented due to lack of licensing. However, appropriate license is required inorder to use the software.

The system is identified to the license function by means of either a license donglewhich is purchased separately, or by the MAC address of one of the computers in theinstallation.

The licensed rights are available in the license file, which is retrieved from ABB andinstalled in the system. The license file contains license features which determine thefeatures and the amounts that are granted.

There are four basic types of licensed features:

• System feature - for system-wide functions available across many or all nodesof the system. They are indicated one per system in the price list.

An example of this type is the connect options.

• System feature with scaling - for system-wide features. The scaling of no. oftags, procedures, logs, and so on. are examples of this type of feature.

• Client feature - for client-specific functions applying to a single client node inthe system. Examples of this type are: Operator Workplace, EngineeringWorkplace, and Batch Client.

• Hardware feature - this feature determines which hardware units are granted foruse. Each hardware unit, I/O, communication, and CPUs, comes with a license.As hardware units are purchased, license features become available in the licensefile.

An end user runtime license can be temporarily extended with additional engineeringand/or operator workplaces, in order to get more seats during commissioning. Anengineering license or a channel license is required. It is most convenient to have theadditional license tied to a dongle, which can be moved between systems to becommissioned.

3BSE041434-600 H 71


At upgrade, the current usage of hardware has to be reported back to ABB so that theappropriate licenses can be registered into the license file. A tool is available for thatpurpose, which must be run on the system to be upgraded.

A system can be in one of the following modes.

• Design Mode

Design Mode license is used during the first phase of system design and systemengineering. This is the phase where the bulk of project engineering is made. In thisphase any feature can be used to its full extent. There is no enforcement of individuallicenses, which means users can build and test their applications without being annoyedby license messages. A License Usage Report keeps track of current license usage inthe system.

In this mode, any operator related activities like using Operator Workplace, OpeningFaceplates and so on will be treated as license violations and will trigger a violationmessage. In this mode a semi-transparent text appears; Design Mode, Licensed forEngineering only.

A design license is time limited and expires after 18 months. The time limit is basedon the license file creation date.

• Production Mode

In production mode the system is controlling a process. Licenses are actively monitoredwith the purpose to validate the usage against what has been ordered. Enforcementof individual license features is active, and the user is informed about possible licenseviolations. There is no license violation message displayed in system for the first 90days, based on the license file creation date.

In this mode, system events are generated whenever there is a license violation.

End user licenses are by default set to design mode from start. The user can selectwhen to set the license to production mode. The selection is enabled in the ABBSoftware Factory (SoFa) once the user has ordered a Production Mode License.

To ensure the correct license is expanded both the SID as well as the license numberare requested at expansion orders.

72 3BSE041434-600 H


License Types

There are five different license types available as end-user license and additionallicense types:

End-user License Types:

• Production system license.

This license type is intended for systems that actively control user’s plants. Thelicense is by default in design phase, and can be used for engineering. Once theuser decides so it is transitioned into production phase.

This license is purchased through the 800xA Sales Wizard, and the ABB SoftwareFactory gets the information about its content from the order management system.

• Engineering system license.

This license type is intended for customers who want an engineering system ascomplement to the production system. This license type can never be configuredso that it controls a plant, that is, it remains in the design mode.

This license is purchased by referring to the production license, and it is functionwise a replica of the production system. If a customer has several productionsystems the customer can use a replica of the largest system or have individualengineering license replicas for each system.

This license type is also used by users who develop their own add-on to an 800xAsystem, such as own aspect systems or libraries. In those cases additionaldevelopment licenses are required.

This license is a time limited license.

Additional License Types:

• Channel license

This license type is intended for application engineering by channel partners andABB channels. System is in design mode. This license is time limited to 18months.


• Internal license

3BSE041434-600 H 73


This license type is intended for demo systems, training, and education systemsthat are permanently installed in demo rooms and training facilities. It is alsoused for ABB internal development (R&D). System is in production mode. Thislicense is time limited to 18 months.


• Demo license

This license type is intended for system demonstrations and previews. Systemis in design mode. This license is time limited to 3 months.


Licensing of Revisions

Revisions are available free of charge only to holders of a valid Automation Sentinelagreement. The expiry date of the Automation Sentinel agreement is checked at thetime of installation and a warning appears before the installation starts. If theinstallation is completed, the license system continues to remind the user until alicense file with a valid Automation Sentinel expiry date is installed. The system isoperational even if the license warnings are displayed.

3.15.12 System Deployment Examples

Example 1

The following requirements are assumed.

• 5 Operator workplaces, one of which is also used as engineering workplace.

• 600 Tags.

• Batch, non-redundant.

• IM.

• 2 AC 800M Controllers.

• FOUNDATION Fieldbus with 300 Devices.

• Redundancy is required for Aspect and Connectivity Servers.

Applying the configuration rules defined in previous sections gives the following:

74 3BSE041434-600 H


• 2 AC 800M Controllers as well as 600 Foundation Fieldbus devices are belowof what can be handled by single connect services.In this case they have to be redundant.

• Both IM and Batch require a server operating system.

• Applying the configuration rules all server functionality is deployed on twoservers:

– 1 aspect server node function per server.

– 1 AC 800M Connect services function per server.

– 1 Foundation Fieldbus Connect services function per server.

– 1 Batch Management server function (counts as two node functions) on firstserver.

– 1 Information Manager server function on second server.

– Optional 1 Domain Controller function per server.

• This gives 6 node functions on the first server and 5 node functions on the second.(If workgroup is selected and no domain server is set up there is one less nodefunction per server.)

Number of nodes needed:

• 2 servers using server operating system.

• 5 rich clients, out of which one also is used as engineering workplace. All usingclient operating system.

Domain or workgroup

As Batch and IM both require a server operating system there is no economical reasonto go for a workgroup system. A relatively small configuration like this may workwell as a workgroup, however, if the system is later expanded with more computersand more users, the workgroup will become increasingly difficult to administrate.

Please note that the domain controller(s), in the case the system is not build as aworkgroup system, are not shown in the picture. It is recommended to use separateserver node(s) for the Domain controller mainly for security reasons, even though itis technically feasible to run the DC on an 800xA server node.

Figure 3.16 shows the node layout for the system specified in this example.

3BSE041434-600 H 75


Figure 3.16: System Deployment Node Layout- Example 1

Example 2

The following requirements are assumed.

• 21 Operator workplaces, out of which 8 also are used as engineering workplaces.

• 5 remote workplaces.

• 20,000 Tags.

• Batch.

• IM.

• Asset optimization.

• 20 AC 800M Controllers.

• 2 Advant AC 400 Controllers.

• FOUNDATION Fieldbus with 2,500 Devices.

• Redundancy is required for server functions.

Applying the configuration rules defined in previous sections gives the following:

76 3BSE041434-600 H


• The client count alone is above 16 which gives that this must be a server basedsystem using a domain controller

• Load calculations are needed to determine if 1 or 2 AC800M ConnectivityServices are needed.

Load calculation is performed in the Sales Wizard. Let us for this case assumethat 2 connectivity services are needed. As there can be only one AC 800Mconnectivity service per node 4 physical nodes will be needed to also realize theredundancy.

• 2,500 FOUNDATION Fieldbus devices require one connectivity service withthree OPC Server FF instances connected to three HSE Subnets with max 1000devices each.

• 2 Advant Controllers require one connectivity service.

• 1 Terminal server can support up to 5 clients with full functionality.

• 1oo3 Aspect services is selected as there is at least 4 servers available.

• Applying the configuration rules concept all server functionality is deployed on4 servers.

– 1 aspect server node function per server 1-3.

– 1 AC 800M Connect service function per server 1-4 (2 redundant pairs).

– 1 Advant AC 400 Connect service function per server 1-2 (1 redundant pair).

– 3 Foundation Fieldbus Connect services function per server 3-4 (3 redundantpairs).

– 1 Batch Management server function (counts as two node functions) onserver 3-4 (1 redundant pair).

– 1 Information Manager server function on server 1-2 (1 redundant pair).

– 1 Asset optimization function on server 4.

• This gives 5 node function each on server 1-4.

• 1 separate server is needed to run the terminal server functionality to support the5 remote clients.

• Please note that different services have different requirements for memory andhard disk space and configure the servers accordingly.

3BSE041434-600 H 77


Number of nodes needed:

• 5 servers using server operating system.

• 21 rich clients, out of which 8 also are used as engineering workplaces. All usingclient operating system.

• 5 thin clients.

Note the Domain Controller require another server or possibly two if redundant, notshown in the picture. It is recommended to use separate server node(s) for the Domaincontroller mainly for security reasons, even though it is technically feasible to runthe DC on an 800xA server node.

Figure 3.16 shows the node layout for the system specified in this example.

78 3BSE041434-600 H


Figure 3.17: System Deployment Node Layout- Example 2

3BSE041434-600 H 79


80 3BSE041434-600 H

4 Operations

This topic describes technical data and configuration information for Operations.

4.1 Operator Clients

4.1.1 Client Count LimitsTable 4.1 details the maximum numbers of clients when combined and separate nodetypes are used.

Table 4.1: Values

Values

80Concurrent operator clients per system (PC-0028)

160Operator client screens per system (PC-0029)

10Engineering Clients (PC-0011)

20 with performancelimitations

40Batch clients (PC-0012)

100Smart Client Office Workplaces (PC-0431)

4.1.2 Process GraphicsProcess Graphics provides the tools for creating a graphical user interface forsupervision and operation of a site. The following can be performed in ProcessGraphics:

• Creating building blocks (for example, graphical representation of Tank or Valve),graphic elements that are used for building graphic displays.

• Building graphic displays that provide an overview of the site.

3BSE041434-600 H 81

4 Operations 4.1 Operator Clients

• Configuring faceplates that are used for controlling and monitoring the site.

The former Visual Basic based graphics was replaced in 800xA System Version 5.0Service Pack 2 Revision A to the current graphics package. With the 800xA SystemVersion 6.0, the Visual Basic graphics package is not included in the defaultinstallation.

Visual Basic Process Graphics (VBPG) is not supported in System Version 6.0.

The Process Graphics that replaced VBPG was introduced in 5.0 SP2 Rev A andgraphics migration needs to be done before upgrading to System Version 6.0.

4.1.3 Screen ResolutionThe recommended screen resolution is 1920x1200 or 1920x1080. Alternativesupported resolutions are 1280 x 1024, 1600 x 1200 and 1680 x 1050. It is stronglyrecommended to use the same aspect ratio and if possible the same resolution for allclients in the system.

4.1.4 Display Call-up TimeTable 4.2 shows the call-up time for different types of displays when used on a clientrunning on dedicated hardware.

Table 4.2: Call-up Time Values

Display Call-up TimeGraphic Displays

<=1 secs 1Graphic Display with maximum 800 OPC items (100 objects)

<=5 secsGroup Display with 10 faceplates

<=1 secsFaceplate

<=2 secsExtended Faceplate

<=1secsSFCViewer as Graphic display - SFCViewer up to 20 Steps<=3 secsSFCViewer up to 200 Steps

82 3BSE041434-600 H


Table 4.2: Call-up Time Values(Continued)

<=2 secs typical 2Trend Display, at first call-up of trend with 8 variables

NOTE:1. Graphic display references are cached after the first call up which makes subsequent display call upsfaster. Each display in a system is cached after the first call up which means there is no limitation in thenumber of cached displays. The performance figure reflects a cached display.2. When a trend display contains OPC string values (engineering units), the call-up time will depend onthe OPC server string handling configuration. With the default configuration the call-up time will typicallybe higher.Eight traces is considered a max from an ergonomic point of view, even though this is not a technicallimitation.

For multiple systems (when Multisystem Integration is used), call-up times fordisplays, faceplates, trends, etc. in a subscriber system where provider data issubscribed, the time will be less than 1.2 (+20%) of the corresponding performancefor a single system.

For remote clients the call-up time will be longer, and will depend on networkperformance, firewall performance, as well as how many clients are being served bythe terminal server machine. As an indication, with five clients on a state of the artserver, and under normal conditions for a 100 megabit/second network, call-up timemay be up to 500 milliseconds longer than for regular client nodes.

4.1.5 Command Response TimeThe Command response time (Table 4.3) is given as two values, one being the timefrom command in a faceplate until an I/O signal reacts in the controller, and the otherfrom command in a faceplate until the feedback indication shows up on the faceplate.All values are first time call-up, and with Audit Trail off. Audit Trail on adds anadditional 0.5 - 1 seconds.

Table 4.3: Command Response Time

Response Time (Average)Measurement

<=1 secs.Command to I/O

<=2 secs typicalCommand to feedback indication

3BSE041434-600 H 83


4.2 Alarm Parameter LimitsTable 4.4 details the Alarm Parameters Limits for a a system.

Table 4.4: Alarm Parameter Values

ValuesAlarm Parameter Limits

100Displayed Alarm & Event lists, per system (PC-0050)

50Displayed Unique Alarm & Event lists per system (PC-0052)

8Displayed alarm & event lists per client (PC-0051)

500Displayed Alarm bands per system (PC-0053)

25Displayed Unique Alarm bands per system (PC-0055)

20Displayed Alarm bands per client (PC-0051)

25External Alarms (PC-0056)

4.3 VideONet ConnectThe VideONet Connect for 800xA is used to visualize and manage video camerasconnected to the 800xA system through the VideONet Server. This enables the operatorto have a live view of the process within the Operator Workplace.

The VideONet Connect for 800xA is a system extension to the 800xA system. Thefollowing are the features of VideONet aspects that belong to the VideONet Connectfor 800xA:

• Connection to a VideONet Server.

• Video camera graphic elements for presentation in 800xA graphic displays.

• Video camera faceplates for configuring, recording, and generating log files forthe video.

• Camera View aspect for playback and live view of video.

84 3BSE041434-600 H

4 Operations 4.2 Alarm Parameter Limits

4.3.1 Camera to VideONet Server CommunicationThe IP cameras send data to one or several VideONet servers during continuous videorecording. Communication increases for each camera with recording or active livestream.

One stream to a camera is shared between configured recording and all live streamsto 800xA clients. This means that additional clients to a camera do not increasecommunication.

4.3.2 VideONet Server to 800xA Client CommunicationThe VideONet Server streams the requested video to the 800xA nodes. Communicationincreases for each recorder playback or live stream presented in an 800xA client,through a faceplate, graphic element or camera view.

It is recommended to use a separate network to setup the integrated 800xA andVideONet System, for the following reasons:

• Video consumes large bandwidth; 8-15 IP cameras consume 100 MBits/s.

• Lower risk of disturbances in the process network.

Higher resolutions, frame rates, and image quality requires more bandwidth. Eachvideo stream consumes 1 to 15 MBits/s on the network depending on the cameratype, video format, resolution, and frame rate. Refer Table 4.5 for details.

Table 4.5: Video Resolutions and Bandwidth

Stream BandwidthRecording disk spaceVideo Format(MJPEG @ 80%

quality)@ 25 fps@ 1 fps@ 25 fps@ 1 fps

< 6 Mbit/s< 250 Kbit/s< 2.5 GB/h< 100 MB/h< 30 KB/frameCIF (384x288)

< 16 Mbit/s< 650 Kbit/s< 7 GB/h< 300 MB/h< 80 KB/frameVGA (640x480)

< 32 Mbit/s< 1250Kbit/s

< 14 GB/h< 600 MB/h< 150KB/frame

Full HD (1920x1080)

NOTE:In actual implementation these figures can be as much as 30% lower.

To calculate the bandwidth for a recording, take the average image size (Table 4.6),multiply by 8 and frame rate. The average image size is available in System > StatusVnHistory in the VideONet Server application.

3BSE041434-600 H 85

4 Operations 4.3 VideONet Connect

Table 4.6: Video Performance

DescriptionsParameters

12Maximum number of streams per client*

< 50 MBClient memory allocation per stream, first connected stream

< 5 MB/streamClient memory allocation per stream, streams 2 -50

32Maximum number of active cameras connected to one videoserver*

See aboveTypical Video Stream Bandwidth (server to client)

See aboveTypical Video Stream Bandwidth (camera to server)

64Maximum number of video recordings per video server*

128Maximum number of streams server to clients

16Number of clients per recording

NOTES:* Depending on Client HW, bandwidth the actual number can be lower.

86 3BSE041434-600 H

4 Operations 4.3 VideONet Connect

5 Control and I/O

This topic presents performance and technical data for Control Software and ControlBuilder key functions, configuration, and items.

The information given is valid for AC 800M version 6.0Late changes might affect performance and/or functionality. For information onlate changes and restrictions on the use of the product, please refer to the ReleaseNotes.

5.1 Memory and Execution Performance

5.1.1 Memory SizeFigure 5.1 shows the memory organization. The total physical memory less theexecuting firmware is called “Memory size” by the “SystemDiagnostics” functionblock. This amount of memory is sometimes also called the “heap”.

The memory usage is also displayed in the Control Builder Heap Utilization dialogwhich can be displayed for each controller. The available memory is called “Non-UsedHeap” and the rest is called “Used Shared Heap”.

3BSE041434-600 H 87

5 Control and I/O 5.1 Memory and Execution Performance

Figure 5.1: The Memory Organization

5.1.2 Available MemoryThe amount of free memory in the controller decreases when the controller has startedup, and an empty project has been downloaded from Control Builder M.

The remaining memory is what can be used for application code, and is hereafterreferred as to “Available memory”.

The measurement results in Table 5.1 are made with IAC, but without anyconfigured communication protocols and CEX units. Memory consumptions forused protocols and CEX units have to be added, according to Table 5.2.

Table 5.1: Available RAM Memory and Performance in AC 800M Controller (without ProtocolHandlers)

AvailableMemory (kbytes)

Firmware andan Empty Project

(kbytes)

Total RAM(kbytes)

ExecutionPerformance

FactorController

2154603881920.5PM851

62536035122880.5PM851A

88 3BSE041434-600 H


Table 5.1: Available RAM Memory and Performance in AC 800M Controller (without ProtocolHandlers)(Continued)

AvailableMemory (kbytes)

Firmware andan Empty Project

(kbytes)

Total RAM(kbytes)

ExecutionPerformance

FactorController

2156603681920.5PM856

103376047163840.5PM856A

71479237163840.6PM858

2157603581921.0PM860

103466038163841.0PM860A

71589226163841.0PM861

71559229163841.0PM861A

235219247327671.2PM862

235269241327671.5PM864

235229246327671.5PM864A

235159253327671.5PM865

2223310535327670.9PM865 SM810

2218410584327670.9PM865 SM811

5139014146655352.1PM866

5138914147655352.1PM866A

4655918977655351.8PM867SM812

208985531592621444.5PM891

3BSE041434-600 H 89


Table 5.2: Memory Consumptions of Protocols and CEX Units

PM891PM866PM864Protocol/CEX Unit Next Unit

(kbytes)First Unit(kbytes)

Next Unit(kbytes)

First Unit(kbytes)

Next Unit(kbytes)

First Unit(kbytes)

48414951496MODBUS RTU

46015711572COMLI

45413631365S3964R

64817601861SerialLib

NA148NA174NA167IAC*

NA35NA42NA43UDP

NA44NA46NA53TCP

454542CI853

141673224733247CI854A/B

39511961294CI855

5971310313103CI856

41731317812182CI857

76019641962CI858

96320149367149364CI860

368360358CI862

731407313273148CI865

331753616935166CI867

41147417775174CI868

5976315764153CI869

291893819038185CI871

81606721368208CI872

42155101208101200CI873

NOTE:*In addition, each communication connection requires about 40 kbytes memory.

90 3BSE041434-600 H


5.1.3 Execution PerformanceCyclic CPU load is calculated as a percentage using the following formula.

Cyclic CPU load (%) = 100*(Total execution time / Total intervaltime)

Depending on the amount of code and requested task interval times, applications maydemand up to 70% of CPU capacity (never more)2; the execution of IEC 61131-3code is called Cyclic Load. Should an application require more than 70% of CPUcapacity, the task scheduler automatically increases the task interval times tore-establish a 70% load.

Load balancing can be disabled (see the manual AC 800M Configuration(3BSE035980*). In a High Integrity Controller, load balancing is always replacedby overrun and latency supervision.

Since IEC 61131-3 has higher priority than communication, it is recommended notto have IEC 61131-3 tasks with execution time over 200ms. For AC 800M HighIntegrity it's not recommended to have IEC 61131-3 tasks with execution time over100ms.

Communication handling has the lowest priority in a controller. It is thereforeimportant to consider controller CPU load if the communication handling is vital tothe application. Running close to 100% total load will result in poor capacity andresponse times for peer-to-peer and (OPC Server for AC 800M) communication. Itis recommended that peak total load will be kept below 100%.

Among the communication protocols, the IAC MMS protocol will be the last to beaffected if there is a communication data starvation.

CPU load is also influenced by other factors, such as Modulebus scan interval andthe number of modules on Modulebus (AC 800M), or the scanning of ABB Drives.

The PM860/PM860A and PM861/PM861A processor units have the same internaldesign and the same performance when execution application program.

The PM851/PM851A, PM856/PM856A and PM860/PM860A processor units havethe same design. They differ only in performance when executing an applicationprogram. The execution time in PM851/PM851A and PM856/PM856A isapproximately two times the execution time in PM860/PM860A.

The execution time in PM858 is four times the execution time in PM866.

2. This is not true if load balancing is set to false or if you run with an AC 800M HI. The controller will run untilit is forced to stop.

3BSE041434-600 H 91


The execution time in PM862 is two times the execution time in PM866.

The PM864/PM864A and PM865 processor unit, in single configuration, hasperformance data which theoretically peaks at twice the performance compared tothe PM860/PM860A. The useful sustained performance improvement is, however,a bit lower and dependent on the actual application program but can be expected tobe 10 to 50% compared to PM860/PM860A. The difference in execution performanceis dependent on how much CEX bus accesses, and how much communication isrunning in the controller (both communication running as CEX bus interfaces andcommunication running on the built in ports on the CPU i.e. ModuleBus, Ethernetand RS-232). CEX bus access and communication decreases execution performance.

In redundant configuration the execution performance is lower than in singleconfiguration (reduction is typically less than 10%). Switch over time from primarycontroller to backup controller, in redundant configuration, is less than 10 ms.

The PM866 processor unit has performance data which is approximately 1.4 timesthe performance of PM864/PM864A.

The PM891 processor unit has performance data which is approximately 2 times theperformance of PM866.

AC 800M High Integrity

The execution performance of an AC 800M High Integrity controller (PM865) is60-70% of an AC 800M Process Automation controller (PM864/PM864A). Thenumber of code tabs used in a SIL application has greater influence to the performanceof an AC 800M High Integrity controller compared to the performance of an AC 800MProcess Automation controller. Also the FDRT setting affects the total cyclic load ofa AC 800M High Integrity controller. However, the memory consumptions of functionblocks and control modules are the same on both controllers.

The PM867 High Integrity controller has performance data which is approximately2 times the performance of PM865 High Integrity.

The AC 800M High Integrity controller load is presented with two different values,Cyclic Load and Total System Load. In order to ensure reliable operation these valuesmust be kept within certain limits. For the AC 800M High Integrity controller thefollowing max values shall not be exceeded.

50%Cyclic Load:

90%Total SystemLoad:

92 3BSE041434-600 H


Cyclic Load:

Load generated by the IEC 61131-3 task execution. This load is mainlydetermined by:

• Application properties, that is size and complexity

• Task properties, that is interval time

• Copy in/out of Communication Variables

For further details, see the manual AC 800M Configuration (3BSE035980*).

Total System Load:

The total load for the controller, including the Cyclic Load (see above).The additional load mainly consist of:

• Communication handling (for example Ethernet and serial channels)

• ModuleBus I/O scanning controlled by the ModuleBus scan time setting

• Internal diagnostics (the contribution to the load is dependent on the applicationsize e.g. the number of code tabs and the FDRT Diagnostics Cycle Time setting)

• General housekeeping

5.1.4 Considerations for AC 800M High Integrity ControllerThe maximum total continuous execution time, should in an AC 800M High IntegrityController be limited to approximately 40% of the FDRT setting. Depending on theapplication and the task parameter configuration, the maximum continuous executionin one controller could be reached at different circumstances for example during theexecution of first scan code for one application or the accumulated execution timefor all tasks in the controller.

For typical execution times in a High Integrity Controller, see Table 5.4 and Table5.5.

It is recommended that the VMT (Virtual Machine Test) task shall have the highesttask priority and that no other task has the same priority.

3BSE041434-600 H 93


5.1.5 Spare Memory Needed for Online ChangesAs a general rule, an application should never exceed half the size of the availablememory. The reason for this is the manner in which applications are updated online.

1. The modifications (the difference between the old and the updated application)are downloaded to the controller memory.

2. A new version of the application is created in controller memory, based on theold application and the modifications.

3. The controller switches from the old to the new application.

4. The old application is deleted.

This technique handles all updates in a controlled and efficient way. Free memoryequal to the size of the largest application is required.

If an application comes close to this limit, it should be divided into two parts so thatthey can be updated separately.

One Application in the Controller

There must be spare memory in the available memory in order to be able to makeon-line changes shown in the Figure 5.1. The amount of spare memory must be atleast 20% of available memory, and may require up to 50%.

A minimum of 20% spare available memory may be sufficient, depending on a numberof factors, such as the complexity of the application and the number of defined alarms.

The function block “SystemDiagnostics” reports used memory based on thememory size, not on the available memory, but the dialog “Heap Utilization” willshow the available memory as “Non-Used Heap”.

The function block SystemDiagnostics also presents another figure: the “Maximumused memory”. This figure is presented in actual bytes, and as a percentage of thememory size. This figure is far more useful to look at when determining how closeyou are to being unable to make on-line changes. Several on-line changes must bemade in order to catch the maximum memory need in the controller.

It is still possible to make on-line changes as long as the maximum used memoryvalue is less than 100%.

94 3BSE041434-600 H


More than One Application in the Controller

Less spare memory is needed when there is more than one application in the controller.

The on-line changes are done to one application at the time. This means that if changesare done to more than one application in the controller, these changes will not takeeffect in a synchronized way.

Example: One application requires 50% used memory and 70% maximum usedmemory. If you split this application into two equally smaller applications, it will stillrequire 50% used memory, but only 60% maximum used memory, since the extramemory needed for the on-line changes will be half.

5.1.6 Comparing Memory Allocations Made with Different VersionsFrom the discussions above, you can see that the “used memory” value provided bythe SystemDiagnostics function block cannot be used to compare different versions.

The amount of available memory in the controller varies between versions for anumber of reasons, one being the number of functions implemented in the firmware.

5.1.7 Memory Consumption and Execution TimesMemory is reserved for each function block type defined. When another instance iscreated, the amount of memory reserved for the instance is very small in relation tothe type. This means that the memory consumed by the type itself is of greatimportance.

The following tables show memory consumption and execution time for AC 800MPM864/PM866/PM891 controllers, for a number of common function blocks andcontrol modules.

In the tables the First Object column shows the required memory for the object typeand one function block or control module andNext Object column shows the requiredmemory for every further function block or control module.

3BSE041434-600 H 95


Table 5.3: AC 800M Memory Consumption for Function Blocks and Control Modules

PM891PM866PM864

ObjectLibrary NextObject(kbyte)

FirstObject(kbyte)

NextObject(kbyte)

FirstObject(kbyte)

NextObject(kbyte)

FirstObject(kbyte)

Function Blocks

1,84,30,85,30,64,1AlarmCondBasicAlarmEventLib

2,05,71,57,20,95,4AlarmCondAlarmEventLib

6,354,45,864,15,562,2PidLoopControlBasicLib

7,356,76,867,47,564,0PidLoop3PControlBasicLib

14,066,813,776,511,475,2PidCascadeLoopControlBasicLib

14,270,113,780,514,376,7PidCascadeLoop3PControlBasicLib

2,210,11,824,11,210,7PidSimpleRealControlSimpleLib

16,680,917,293,712,180,7ACStdDriveProcessObjectDriveLib

9,052,38,766,16,053,5UniProcessObjectExtLib

12,560,614,259,08,062,0BiProcessObjectExtLib

11,362,011,559,77,663,5MotorUniProcessObjectExtLib

15,471,914,568,29,973,3MotorBiProcessObjectExtLib

8,651,38,147,35,552,8ValveUniProcessObjectExtLib

28,5112,028,5107,425,5109,2McuExtendedProcessObjectInsumLib

1,12,20,53,90,34,4SignalBasicBoolSignalBasicLib

1,24,80,74,20,45,6SignalBasicInBoolSignalBasicLib

1,98,91,99,41,910,6SignalBasicInRealSignalBasicLib

1,22,61,24,31,26,0SignalBasicOutBoolSignalBasicLib

1,54,81,65,71,66,9SignalBasicOutRealSignalBasicLib

1,44,31,34,11,35,9SignalBasicRealSignalBasicLib

4,226,63,620,82,829,1SignalInBoolSignalLib

96 3BSE041434-600 H


Table 5.3: AC 800M Memory Consumption for Function Blocks and Control Modules(Continued)

PM891PM866PM864


FirstObject(kbyte)

NextObject(kbyte)

FirstObject(kbyte)

NextObject(kbyte)

FirstObject(kbyte)

4,326,03,720,12,728,8SignalOutBoolSignalLib

3,925,53,320,32,329,0SignalSimpleInRealSignalLib

10,660,510,655,27,664,0SignalInRealSignalLib

3,522,12,917,82,017,8SignalSimpleOutReal

SignalLib

9,557,09,052,96,958,9SignalOutRealSignalLib

5,523,14,924,34,924,4SDBoolSupervisionBasicLib

5,929,85,325,24,031,7SDInBoolSupervisionBasicLib

12,847,812,643,610,749,7SDInRealSupervisionBasicLib

5,626,35,224,93,423,6SDLevelSupervisionBasicLib

7,332,76,828,15,834,2SDOutBoolSupervisionBasicLib

12,343,411,839,010,145,1SDRealSupervisionBasicLib

6,435,45,930,84,836,9SDValveSupervisionBasicLib

3,718,73,214,33,320,4StatusReadSupervisionBasicLib

Control Modules

0,912,51,45,81,85,1AlarmCondBasicMAlarmEventLib

2,813,30,95,71,16,8AlarmCondMAlarmEventLib

22,9199,431,4198,323,4209,0PidAdvancedCCControlAdvancedLib

58,9224,764,3229,660,9232,8CascadeLoopControlSolutionLib

52,3214,347,4234,643,8233,1FeedforwardLoopControlSolutionLib

47,7218,448,1233,644,6232,5MidRangeLoopControlSolutionLib

113,7300,6114,5318,8114,4319,0OverrideLoopControlSolutionLib

36,0205,736,4227,436,0229,9SingleLoopControlSolutionLib

3BSE041434-600 H 97


Table 5.3: AC 800M Memory Consumption for Function Blocks and Control Modules(Continued)

PM891PM866PM864


FirstObject(kbyte)

NextObject(kbyte)

FirstObject(kbyte)

NextObject(kbyte)

FirstObject(kbyte)

3,820,13,721,33,620,9AnalogInCCControlStandardLib

3,814,43,616,23,619,3AnalogOutCCControlStandardLib

4,618,34,721,74,622,6Level2CCControlStandardLib



15,6125,516,1145,316,0109,5PidCCControlStandardLib

2,612,22,514,12,616,0PidSimpleCCControlStandardLib

4,244,44,656,54,242,1ThreePosCCControlStandardLib

16,2109,417,8118,016,883,3ACStdDriveMProcessObjectDriveLib

9,278,59,390,29,451,7UniMProcessObjectExtLib

12,687,315,896,512,161,0BiMProcessObjectExtLib

10,887,713,463,910,161,9MotorUniMProcessObjectExtLib

14,497,117,6106,114,371,4MotorBiMProcessObjectExtLib

25,677,710,788,28,851,1ValveUniMProcessObjectExtLib

27,8138,332,8145,428,6111,0McuExtendedMProcessObjectInsumLib

4,151,43,724,44,124,8SignalInBoolMSignalLib

3,851,24,424,73,824,4SignalOutBoolMSignalLib

9,990,211,863,910,064,1SignalInRealMSignalLib

10,284,710,362,210,359,1SignalOutRealMSignalLib

6,160,95,637,56,134,5DetectorBoolSupervisionLib

12,4102,413,280,212,877,1Detector2RealSupervisionLib

98 3BSE041434-600 H


Table 5.4: AC 800M Execution Time for Function Blocks and Control Modules (Continued)

PM891(µs)

PM866(µs)

PM865SIL3(µs)

PM865SIL1-2(µs)

PM865NoSIL(µs)

PM864(µs)ObjectLibrary

Function Blocks

91533353020AlarmCondBasicAlarmEventLib

152350544732AlarmCondAlarmEventLib

915303021ProcessObjectAEAlarmEventLib

142388535031SignalAEAlarmEventLib

162855555538SimpleEventDetectorAlarmEventLib

251515158CTDBasicLib

351515158CTUBasicLib

4818181810CTUDBasicLib

361616169ErrorHandlerBasicLib

241313135F_TrigBasicLib

12216316317529ForcedSignalsBasicLib

3616168PulseGeneratorBasicLib

241313135R_TrigBasicLib

231313135RSBasicLib

241313135SRBasicLib

187445614160SystemDiagnosticsBasicLib

9153520SystemDiagnosticsSMBasicLib

1531545345TimerDBasicLib

1026474634TimerUBasicLib

361616168TOfBasicLib

461515158TOnBasicLib

361615158TPBasicLib

103244492343PidLoopControlBasicLib

3BSE041434-600 H 99


Table 5.4: AC 800M Execution Time for Function Blocks and Control Modules (Continued)(Continued)

PM891(µs)

PM866(µs)

PM865SIL3(µs)

PM865SIL1-2(µs)

PM865NoSIL(µs)


113281540410PidLoop3PControlBasicLib

179428868618PidCascadeLoopControlBasicLib

173486922666PidCascadeLoop3PControlBasicLib

18528771PidSimpleRealControlSimplelib

81137265240191BiSimpleProcessObjBasiclib

5085164149120UniSimpleProcessObjBasiclib

217402872584ACStdDriveProcessObjDriveLib

134248534535320BiProcessObjExtLib

166286723668417MotorBiProcessObjExtLib

134217518480311MotorUniProcessObjExtLib

87131372302197UniProcessObjExtLib

93150337310205ValveUniProcessObjExtLib

226356888534McuExtendedProcessObjInsumLib

251414147SignalBasicBooISignalBasicLib

351514147SignalBasicInBooISignalBasicLib

173680817979SignalBasicInRealSignalBasicLib

351414147SignalBasicOutBoolSignalBasicLib

492013SignalBasicOutRealSignalBasicLib

41223232415SignalBasicRealSignalBasicLib

2441837555SignalBoolSignalLib

1734706446SignalInBoolSignalLib

82145303235210SignalInRealSignalLib

22331266848SignalOutBoolSignalLib

5290159125SignalOutRealSignalLib

100 3BSE041434-600 H



PM891(µs)

PM866(µs)

PM865SIL3(µs)

PM865SIL1-2(µs)

PM865NoSIL(µs)


60112267194154SignalRealSignalLib

345814412284SignalSimpleInRealSignalLib

18306340SignalSimpleOutRealSignalLib

4983192183184127SDBooISupervisionBasicLib

376519518416594SDlnBoolSupervisionBasicLib

109185410426421266SDInRealSupervisionBasicLib

385613513212877SDLevelSupervisionBasicLib

4986220235214129SDOutBoolSupervisionBasicLib

74182351346361230SDRealSupervisionBasicLib

66121280290273165SDValveSupervisionBasicLib

142542424235StatusReadSupervisionBasicLib

Control Modules

1521413930AlarmCondBasicMAlarmEventLib

1016333225AlarmCondMAlarmEventLib

203376697047CVAckISPBasicLib

511242416ErrorHandlerMBasicLib

152518918736ForcedSignalsMBasicLib

3167221328994PidAdvancedCCControlAdvancedLib

709145231542016CascadeLoopControlSolutionLib

546116124741677FeedforwardLoopControlSolutionLib

483106623381505MidrangeLoopControlSolutionLib

1370280560553859OverrideLoopControlSolutionLib

40686318651187SingleLoopControlSolutionLib

3678201104AnalogInCCControlSolutionLib

3BSE041434-600 H 101



PM891(µs)

PM866(µs)

PM865SIL3(µs)

PM865SIL1-2(µs)

PM865NoSIL(µs)


336215687AnalogOutCCControlSolutionLib

28469963Level2CCControlSolutionLib



218419944595PidCCControlSolutionLib

3777141110PidSimpleCCControlSolutionLib

59125239179ThreePosCCControlSolutionLib

103192352352248CO2FireGasLib

67129277262176DelugeFireGasLib

87163324328227FGOutputOrderFireGasLib

55105235225140SprinklerFireGasLib

75133258246205BiSimpleMProcessObjBasicLib

5097174162126UniSimpleMProcessObjBasicLib

237428904594ACStdDriveMProcessObjDriveLib

141235563521333BiMProcessObjExtLib

162306739655401MotorBiMProcessObjExtLib

130220564491318MotorUniMProcessObjExtLib

97148393348211UniMProcessObjExtLib

96145403344205ValveUniMProcessObjExtLib

238378920521McuExtendedMProcessObjInsumLib

1020414227SDLevelAnd4SignalLib

714343419SDLevelBranch4SignalLib

4571202150104SDLevelMSignalLib

816383823SDLevelOr4SignalLib

102 3BSE041434-600 H



PM891(µs)

PM866(µs)

PM865SIL3(µs)

PM865SIL1-2(µs)

PM865NoSIL(µs)


314811310670SignalBoolCalcInMSignalLib

356214514087SignalBoolCalcOutMSignalLib

244410310562SignalInBoolMSignalLib

103181388335246SignalInRealMSignalLib

31501008967SignalOutBoolMSignalLib

88140266210SignalOutRealMSignalLib

88165413334228SignalRealCalcInMSignalLib

97183434362236SignalRealCalcOutMSignalLib

80158372354222Vote1oo1QSignalLib

1730757141VoteBranch4SignalLib

1527676636VotedAnd4SignalLib

1017565425VotedBranch4SignalLib

92179464429256VoteXoo2DSignalLib

90186482451289VoteXoo3QSignalLib

103265520483302VoteXoo8SignalLib

129227416411323Detector1RealSupervisionLib

145280484481373Detector2RealSupervisionLib

2439818152DetectorAndSupervisionLib

2644868757DetectorAnd4SupervisionLib

416612712393DetectorAnd8SupervisionLib

2039788253DetectorBranchSupervisionLib

2445939264DetectorBranch4SupervisionLib

4173133134100DetectorBranch8SupervisionLib

150301525504406DetectorLoopMonitoredSupervisionLib

3BSE041434-600 H 103



PM891(µs)

PM866(µs)

PM865SIL3(µs)

PM865SIL1-2(µs)

PM865NoSIL(µs)


2339808050DetectorOrSupervisionLib

2542858557DetectorOr4SupervisionLib

386411912290DetectorOr8SupervisionLib

69123286267178DetectorRemoteSupervisionLib

2236807950DetectorVoteSupervisionLib

1424605932OrderOrSupervisionLib

1221585829OrderOr4SupervisionLib

5497177168121OutputBooISupervisionLib

85164316317215OutputOrderSupervisionLib

Table 5.5: Execution Time for a Number of Standard Operations and Function Calls

PM891(µs)

PM866(µs)

PM865SIL3(µs)

PM865SIL1-2(µs)

PM865NoSIL(µs)

PM864(µs)Data TypeOperation/Function

0,20,50,60,60,70,7boola:= b or c

0,30,50,60,60,70,6boola:= b and c

0,20,50,70,60,70,7boola:= b xor c

0,30,50,60,60,70,6dinta := b + c

0,21,31,81,81,91,9reala := b + c

0,20,50,60,60,70,6dinta := b - c

0,21,21,71,71,71,7reala := b - c

0,30,50,60,60,70,6dinta := b * c

0,31,21,71,71,81,7reala := b * c

0,50,70,90,91,01,0dinta := b / c

104 3BSE041434-600 H


Table 5.5: Execution Time for a Number of Standard Operations and Function Calls(Continued)

PM891(µs)

PM866(µs)

PM865SIL3(µs)

PM865SIL1-2(µs)

PM865NoSIL(µs)

PM864(µs)Data TypeOperation/Function

0,62,94,14,04,14,0reala := b / c

0,30,60,80,80,80,8dinta:= b <> c

0,31,31,81,81,81,8reala:= b <> c

2,86,39,29,39,68,5string[140]a := b

3,512,020,617,5string[10]a := b + c

2,713,021,717,8string[40]a := b + c

5,820,032,328,1string[140]a := b + c

1,43,25,15,15,14,0string[10]a := AddSuffix (b + c)



1,011,615,515,715,815,4dinta := real_to_dint(b)

0,31,41,81,81,81,9reala := dint_to_real(b)

1,515,523,922,322,122,0timea := real_to_time(b)

0,84,46,16,16,26,1reala := time_to_real(b)

5.2 Online Upgrade

5.2.1 StoptimeThe total stop time of a controller when doing online upgrade is defined as the longesttime the I/O is frozen. Generally, the following factors will have large impact on thetotal stop time.

• The stop time will increase with increased memory usage in the controller.

• The stop time will increase with increased number of alarm objects in thecontroller.

3BSE041434-600 H 105

5 Control and I/O 5.2 Online Upgrade

The following formula can be used to give guidance to what the total stop time in acontroller will be, when doing an online upgrade:

OLUStopTimeLowPrioTask =

2.7 x ∑AllApp ApplicationStopTime + ∑AllTasks FirstScanExecutionTime

ApplicationStopTime andFirstScanExecutionTime can be found in the Task dialogonline, after a successful download with warm restart.

Note 1: The formula is only approximate and the following factors will affect theaccuracy of the formula:

• With increased cyclic load in the controller, the stop time might be larger thancalculated.

• With increased number of string variables, the stop time might be larger thancalculated.

• The actual stop time (compared to the calculated) will decrease if the controlleris divided on several applications compared to only having one largeapplication in the controller.

Note 2: This formula calculates the total freeze time of the I/O connected to thelowest priority task. To calculate freeze time of I/O connected to a higher prioritytask, the sum of first scan execution times shall only include task with same orhigher priority.

5.2.2 Online Upgrade in a High Integrity ControllerDuring Online Upgrade of firmware in the AC 800M HI, the application controllingthe process is stopped for a short period of time. During this time period the outputsignals are not updated, but keep their current values. The duration of this time isdepending of the configured Online Upgrade Handover Limit.For most processes it is possible to find a process state, or a period of time when theconfigured FDRT can be exceeded without creating any hazardous situation. Basedon such a judgement of the process, the Online Upgrade Handover Limit can be setin accordance with the following formula.

Online Upgrade Handover Limit =Max acceptable Output freeze time - 2 x actual Task Interval Time - 2 x ModuleBusscan time.

106 3BSE041434-600 H

5 Control and I/O 5.2 Online Upgrade

If the configured FDRT are to be maintained also during an Online Upgrade session,the Online Upgrade Handover Limit shall be set in accordance with the followingformula:

Online Upgrade Handover Limit =FDRT - 2 x the longest Task Interval Time - 2 x ModuleBus scan time.

The maximum length of the time period the outputs are not updated can be determinedby using the following formula:

Output freeze time = Online Upgrade Handover Limit + 2 x the actual Task IntervalTime + 2 x ModuleBus scan time.

5.2.3 Communication DisturbanceDuring the switchover of plant control, communication will be down for a while.Measurements have been done to show the impact on different communicationprotocols.

All the values listed below are typical values and can vary from system to system.

• The trend values in Process Portal are interrupted for approximately 20 s.

• The alarms are delayed in the range of 20 - 30 s during switch of primarycontroller.

• ComliSBConnect is down for approximately 2 s before and 250 ms after theswitchover.

• MMSConnect is down for approximately 2 s before and 2 s after the switchover.

• SBConnect is down for approximately 2 s before and 250 ms after the switchover.

5.3 Hardware and I/O PerformanceThe technical limitation for the number of I/Os in one AC 800M High IntegrityController is 1344 in single I/O configuration (672 for redundant I/O), when onlyusing digital I/O. If for example 30% of the I/O signals are analog, the maximumnumber of I/Os per AC 800M High Integrity is approximately 1000 in single I/Oconfiguration (500 in redundant I/O).

For the PM865 High Integrity controller, the application complexity howeverrestricts the practical number of modulebus I/O.

3BSE041434-600 H 107

5 Control and I/O 5.3 Hardware and I/O Performance

5.3.1 Modulebus Response Time and LoadModulebus scanning has a considerable influence on CPU load, since I/O copyingon Modulebus is handled by the controller CPU.The scan time increases as modules are added, and at a certain point Modulebusscanning will have high impact on the CPU load.

The Modulebus scanning is mainly part of the total CPU load. But the cyclic CPUload is indirectly affected by the Modulebus interrupts, as the Modulebus scanninghas the highest priority in the AC 800M controller.

The Modulebus scan cycle time can be set in Control Builder. The cycle time mustbe set to suit the module requiring the shortest scan interval. A solution to this problemis to connect I/O variables requiring shorter scan intervals via the CI854A/BPROFIBUS adaptor.

Modulebus using AC 800M High Integrity

The Modulebus scan time should not be set shorter than the process requires. Thiswill add unnecessary load to the CPU, and leave less capacity for task execution,communication, and Safety diagnostics. Also, it is of no use to scan I/O significantlyfaster than the Task Interval Times. For most applications, the default 100 ms issufficient.

5.3.2 Calculation of Scan Time on the ModulebusThe following definitions are used in the calculations:

1. Amount of modules:

• n1 = amount of Drives and DP, DI, DO, AI and AO modules (except AI880A,DI880 and DO880)

• n2 = amount of AI880A, DI880 and DO880 modules

108 3BSE041434-600 H


For the modules below, the following number of modules should be accounted:

• AO845 (redundant) = 2

• DO840 (redundant) = 2

• DO880 (redundant) = 2

• DP820 = 4

• DP840 (single) = 8

• DP840 (redundant) = 9

• ABB Engineered Drives = 3

• ABB Standard Drives = 2

For other redundant modules, only one should be accounted.

2. Scan time for different modules:

t1 = 0.5 ms (scan time for n1)

t2 = 1.3 ms (scan time for n2)

Calculation of Fastest Possible Scan Time

The fastest possible scan time is n1* t1 + n2* t2.

Example:It can never take less than 10 * 0.5 = 5.0 ms to scan 10 non-High Integrity I/O modules.

Modulebus Scantime for SIL3 Tasks

For each SIL3 task the fastest possible Modulebus scan time has to be increased basedon the amount of application connected I/O Channels according to Table 5.6 whichapplies to both PM865 and PM867 High Integrity controller.

The numbers per channel applies both to single and redundant I/O channels.

3BSE041434-600 H 109


Table 5.6: Modulebus Scan Time for SIL3 Tasks

DO880AI880AI880asDIDI880BaseLoad

InCopy

-14 μs perchannel

10 μs perchannel

7 μs perchannel

3000 μs

OutCopy

9 μs perchannels

---2000 μs

Example: Consider a configuration with all channels connected to 1131 Variableswith 24 AI880A (8 channels per IO -> 192 connections), 36 DO880 (16 channels perIO -> 576 connections) and 36 DI880 (16 channels per IO -> 576 connections) andthree SIL3 tasks, where each task is connected to both input and output channels.This would require an additional 192*14 + 576*9 + 576*7 + (3000 + 2000)*3 = 27ms for the fastest possible Modulebus scantime.

5.3.3 Calculation of the Modulebus CPU LoadThe maximum load caused by n1 (non High Integrity I/O) module types:

L1 = 31% (PA and HI controller)

The maximum load caused by n2 (High Integrity I/O) module types:

L2 = 20% (PA controller)L2 = 23% (HI controller)

The L1 and L2 values are valid for all AC 800M processor unit types, except for theones listed below.

PM866, PM862 and PM858: L1= 16% and L2= 10%PM891: L1 = 10% and L2=7%PM867: L1 = 26% and L2 = 19%

The Modulebus scanning causes the following CPU load if the chosen scan cycletime is less or equal to the fastest possible scan time:

Load(fastest) = (n1 / (n1 + n2) * L1 + (n2 / (n1 + n2) * L2)

The following CPU load is caused for other scan cycle times:

110 3BSE041434-600 H


Load(chosen) = Fastest Possible Scan Time / Chosen Scan time * Load(fastest)

The Fastest Possible Scan Time to be used in the formula is without the SIL3 addition.

5.3.4 Example Scan Time and CPU loadAssume that following units are used:

1 AI810: 0.5*1 = 0.5 ms1 redundant DO880: 1.3*2 = 2.6 ms1 redundant DP840: 0.5*9 = 4.5 ms

This means that n1=10 and n2=2, and fastest possible scan cycle time is 8 ms(resolution of 1 ms).

CPU Load for a PA Controller will be: (10/12)*31 + (2/12)*20 = 29.16%

CPU Load for a HI Controller will be: (10/12)*31 + (2/12)*23 = 29.66%

Updating Rate of Data to an Application

The rate in milliseconds at which all channels of an I/O module are updated in thecontroller to be used in the IEC 61131-3 application, as a function of the scan timein milliseconds is as follows:

• For AI, AO and AI843 (except AI880A and other temperature measuring I/Othan AI843) the updating time is equal to number of channels divided by twomultiplied by the scan time.

• For temperature measuring I/O (except for AI843) the updating time is equal tonumber of channels multiplied by the scan time.

• For AI880A the updating time is equal to scan time.

• For Standard Drives the updating time is equal to scan time.

• For Engineered Drives the updating time is equal to scan time multiplied by 12.

• For DI, DO, DP the updating time is equal to scan time.

5.3.5 ModuleBus Scanning of ABB DrivesScanning of ABB Drives on Modulebus also influences CPU load.

3BSE041434-600 H 111


Modulebus Scanning of ABB Engineered Drives (AC 800M)

Scanning of an engineered Drive is distributed over 3 * 12 scan cycles. Three channels(DDS pairs) are scanned in each scan cycle. The first two are always channels 1 and2 (i.e. DDS pairs 10/11 and 12/13); the third will be different for each scan cycle.

Table 5.7: Scan cycles for ABB Engineered Drives DDS Pair 3

DDS Pair 3Scan Cycle

14/151, 5, 9

16/172, 6, 10

18/193, 7 11

20/214

22/238

24/2512

To scan the three DDS pairs each cycle takes 3 * 0.5 = 1.5 ms. It is not possible tohave a scan interval less then 2 ms (=PA controller) / 5 ms =HI controller) for theModulebus scanner. Thus, for one drive the scan time will be 2 ms.

Example

For four drives, the scan time will be 1.5 ms * 4 = 6.0 ms for the DDS pairs 10/11and 12/13, and the scan time for the remaining of the DDS pairs will be1.5 ms * 4 * 12 = 72.0 ms.

ModuleBus Scanning of ABB Standard Drives (AC 800M)

For ABB Standard Drives, all data sets (DDS 1/2 and DDS 3/4) are scanned in eachscan cycle. It takes 2 * 0.5 = 1.0 ms to scan a single Standard Drive.

Example

For four ABB Standard Drives the scan time will be 1.0 ms * 4 = 4.0 ms.

5.3.6 Dynamic Data Exchange S800 I/O Connected via CI854A/BThe transportation of dynamic data between PROFIBUS-DP/DPV1 master and theS800 I/O modules shown in the Figure 5.2.

112 3BSE041434-600 H


Figure 5.2: Transportation of dynamic data between PROFIBUS-DP/DPV1 master and S800I/O modules.

The transportation of dynamic data between PROFIBUS-DP/DPV1 and the ModuleBusis the main task for the Field Communication Interface FCI. The FCI has a dedicatedmemory area where it sends the output values and reads the input values.

3BSE041434-600 H 113


The CPU in the FCI performs the rest of the data transportation. It reads output valuesfrom the memory and writes to the I/O Modules via the ModuleBus and vice versa.

Data Scanning Principles

The data transfer between PROFIBUS-DP/DPV1 and the ModuleBus (3 and 4 in thefigure) is not synchronized. Read and write operations are performed from and to adual port memory in the FCI.

The ModuleBus data is scanned (read or written) (2 and 5 in the figure) cyclically,depending on the I/O module configuration. On one scan all digital modules, 1/4 ofthe analog modules and 1/10 of the slow analog modules (modules for temperaturemeasurement) are scanned. It takes 4 scans to read all analog modules and 10 scansto read all slow analog modules.

In a typical configuration with 3 AI, 2 AO, 3 DI and 2 DO modules the data scantime will be 18 ms.

For calculation of the ModuleBus data scanning in the FCI, see S800 I/O User’s GuideFieldbus Communication Interface PROFIBUS-DP/DPV1 Section 3 Configurationand Chapter Data Scanning.

The internal data scanning (1 and 6 in the figure) on the I/O modules is notsynchronized with the ModuleBus data scanning.

Typical data scanning on S800 I/O modules (filter times not included):

Digital modules 1ms.

Analog modules 10ms.

Slow analog modules 1s.

Data scanning on S800 I/O modules see, S800 User’s Guide Module and terminationUnits Appendix A Specifications.

Calculation of Signal Delay

Signal delay from process to controller and vice versa can be calculated according tofollowing:

Signal delay = Controller scan time + Profibus scan time + FCI scan time + Modulescan time + Filter time.

For example:

114 3BSE041434-600 H


Signal delay digital signal = Controller scan time + Profibus scan time + FCI scantime + Module scan time + Filter time.

Signal delay analog signal = Controller scan time + Profibus scan time + 4 * FCI scantime + Module scan time + Filter time.

Signal delay slow analog signal = Controller scan time + Profibus scan time + 10 *FCI scan time + Module scan time + Filter time.

5.3.7 S100 I/O Response Time and LoadThe response time is the time it takes for a signal to go from the input terminals ona S100 I/O board to the double port memory on the CI856 unit or vice versa for outputsignals. The delay caused by the filtering of the input signals is not included.

The S100 I/O response time is the sum of the following times:Conversion Time + Internal Scan Time + Scan Interval CI856

• Conversion Time = 0.1 ms for DSAI 130/130A. For other I/O boards it can beignored.

• Internal Scan Time = The internal scan time on DSAX 110 and DSAX 110Ais 20 ms for input signals and 8 ms for output signals. For other I/O boards it is0 ms.

• Scan Interval CI856 = The scan interval on the CI856 is set for each I/O boardor I/O channel and is determined by "scan interval" or "update interval" in theI/O hardware editor, under settings tab for selected I/O unit.

Calculation of CI856 CPU Load

For each I/O board the load on CI856 is calculated as:

BoardLoad = (BaseLoad + N*ChannelLoad)/CycleTime

• BoardLoad = the CPU load on the CI856 caused by the board (unit = percent).

• BaseLoad = the base load to handle the board, see Table 5.8 below.

• ChannelLoad = the additional load for each I/O channel used on the board, seeTable 5.8 below.

• N = number of used I/O channels on the board.

• CycleTime = the cycle time or update interval set for the board or I/O channel(unit = 0.1 ms).

3BSE041434-600 H 115


Table 5.8: BaseLoad and ChannelLoad of S100 I/O

ChannelLoadBaseLoadBoard

12520DSAI 130/130A

4020DSAI 130D, DSAI 133/133A

3.57DSAO

035DSDI

045DSDO

2212DSDP 010

3025DSDP 170Function Mode = Pulse25

3025DSDP 170Function Mode = Frequency

6125DSDP 170Function Mode = Pulse + Frequency

1325DSDP 170Function Mode = Pulse light2513

To allow scan task overhead and event treatment, the total load from all I/O boardsshould not exceed 80%.

5.3.8 Drivebus Communication with CI858 UnitData transfer on Drivebus is managed through datasets pairs. For standard drives 2dataset pairs can be used and for Engineered drives up to 8 data set pairs can bedefined.

Dataset Priority

Datasets can be given two priorities, High and Normal. High priority datasets areassociated with the high priority execution table which is scanned every 2 ms. Normalpriority datasets are associated with the normal priority execution table. This table isbuilt-up of several channels (slots). The number of channels depends on the maximumnumber of normal priority Datasets defined in any drives unit on the bus. Every 2 msone of the normal priority table channels is scanned.

116 3BSE041434-600 H


Example Dataset Priority

If the maximum number of low priority datasets defined in a drives unit on the busis 6, the normal priority execution table contains 6 channels, each channel is scannedevery 12th millisecond (2ms * 6=12ms).

Dataset Pairs

The transfer times for dataset pairs, for example, DS10/DS11, includes transferringthe message from the drive to the AC 800M (DS10) and the response message,including return value, back to the drives unit (DS11).

Drivebus (CI858) Response Time and Load

When calculating the response times between drives units and AC 800M on Drivebusthe following has to be considered:

• Application task interval time in the host system, that is PM86x.

• Dataset execution queue and communication handler in the CI858,

• Bus transfer time, including data handling time in the communication ASICs onthe CI858 and in the drives units.

• Drives unit application program.

Drivebus Response Time Formula

#DS_Channels: Max number of normal priority datasets in one drives unit on the bus.

AC 800M Application Program

Application program: Task interval time

High Priority Datasets

High priority dataset execution queue and communication handler: 2 ms

Drivebus transfer time: 1 ms

Inverter system application program:

DS10/11: 2 msDS12/13: 4 ms(Other DS: 10 - 500 ms)

3BSE041434-600 H 117


Normal Prio Datasets

Normal Prio dataset execution queue and communication handler:2 * #DS_Channels

Drivebus transfer time: 1 ms

Inverter system application program:

DS10/11: 2 msDS12/13: 4 msOther DS: 10 - 500 ms

The response time on Drivebus consists of the sum of the following:

TaskInterval + DataSet + DrivebusTransfTime + ApplTime

• TaskInterval = Application task interval

• DataSet = DataSet Execution queue and communication handler

• DrivebusTransfTime = Drivebus transfer time

• ApplTime = Inverter system application time

Example

Consider a Drivebus containing five drive units. Each drives unit is using one highpriority dataset pair (DS10/DS11). One of the drives units is using five normal prioritydataset pairs DS12/DS13 to DS20/DS21. The other drives are using four normalpriority dataset pairs DS12/DS13 to DS18/DS19. In the drives units the applicationprogram is using an update time of 100 ms for the normal priority datasets.

In the AC 800M the high priority datasets are attached to a high priority applicationtask using a task interval time of 10 ms. The normal priority datasets are attached toa normal priority task using a task interval time of 250 ms.

Table 5.9: Response Times for each Dataset

ResponseTime(ms)

InverterSystem

ApplicationTime

DrivebusTransferTime

DataSetExecutionQueue and

Comm. Handler

ApplicationTask

IntervalDataset

1521210DS10/DS11

265412*5250DS12/DS13

118 3BSE041434-600 H


Table 5.9: Response Times for each Dataset(Continued)

ResponseTime(ms)

InverterSystem

ApplicationTime

DrivebusTransferTime

DataSetExecutionQueue and

Comm. Handler

ApplicationTask

IntervalDataset

36110012*5250DS14/DS15

DS16/DS17

DS16/DS17

DS18/DS19

DS20/DS21

5.3.9 Calculation of I/O Copy Time Estimate for ControlNet with CI865UnitEach ControlNet node (200-ACN, 200-RACN and CI865) has its own I/O datamemory that is asynchronously updated.

Different configurations and parameters, depending on the I/O system type that isused, determine the total I/O update time.

To estimate the maximum time, from I/O point status change until it is processed inthe application program, all times from I/O point to Task Interval Time, tti, have tobe added according to the formula below.

3BSE041434-600 H 119


200-ACN

or200-RACN

PM

CI865 I/O Board

tti tcn tn

I/O Copy Time = tti + tcn+ tn + tf

tf

Figure 5.3: I/O Copy Schedule

Remote Series 200 I/O and Rack I/O

The transmission on the ControlNet network, tcn, runs asynchronously with theexecution of the application program and the I/O copy cycles on 200-ACN and200-RACN, and is determined by the network parameters.

tcn for input signals equals the EPR (Expected Package Rate) for the specific node.The EPR is a user definable setting, 5-60ms.

tcn for output signals equals the NUT (Network Update Time) for the specific node.The NUT is a user definable setting, 5-60ms.

Series 200 I/O

The 200-ACN I/O memory is updated cyclically, asynchronously with the executionof the application program. The node update time, tn, is determined by the numberand type of I/O units. The approximate copying times are 0.05ms for digital I/O unitsand 0.2ms for analogue I/O units. There is an overhead of about 2ms for each cycle.

Example 1:A 200-ACN configured with 8 analogue I/O units gives the following node updatetime:

tn ≈ 2+8*0.2 ≈ 3.6ms

Example 2:

120 3BSE041434-600 H


A 200-ACN configured with 8 digital I/O units gives the following node update time:

tn ≈ 2+8*0.05 ≈ 2.4ms

Rack I/O

The 200-RACN I/O memory is updated cyclically, asynchronously with the executionof the application program. The node update time, tn, is determined by the numberand types of connected to 200-RACN.

The copying of the analogue input boards is spread out in time due to the relativelong copying time. One analogue input board is copied each cycle (for example, ifthere are three analog input boards, each one of them will be copied every third cycle).

The approximate copying times are 0.14 ms for digital boards and analogue outputboards and 1.2 ms for analogue input boards. There is an overhead of about 1ms foreach cycle.

Example 1:A 200-RACN is configured with 12 digital boards, 2 analogue output boards and 2analogue input boards. The node update time, tn, for this rack is calculated accordingto the following:

One cycle corresponds to: 1+14*0.14+1*1.2 ms ≈ 4.2ms

Two cycles are needed to copy all analogue input boards, which gives the total nodeupdate time for this node: tn ≈ 2*4.2 ≈ 8.4ms

Example 2:A 200-RACN is configured with 11 digital boards, 2 analogue output boards and 3analogue input boards. The node update time, tn, for this rack is calculated accordingto the following:

One cycle corresponds to: 1+13*0.14+1*1.2 ms ≈ 4.0ms

Three cycles are needed to copy all analogue input boards which gives the total nodeupdate time for this node: tn ≈ 3*4.0 ≈ 12ms

Filter Time

The I/O filter time, tf has to be added for input boards/units.

3BSE041434-600 H 121


5.4 Communication

5.4.1 IAC and MMS CommunicationCommunication performance is affected by bandwidth, message length and cyclicload.

Higher load on the CPU will cause lower throughput in the communication, and lowerload will give higher throughput.

The 10 Mbit/s is an ethernet speed which is in balance with the performance of theAC 800M controller. The maximum data flow to and from the software in an AC 800Mis less than 10 Mbit/s. This means that the data flow for one AC 800M is not limiteddue to its ethernet speed of 10 Mbit/s.

The Ethernet standard allows bandwidth transmission at 10 Mb/s, 100 Mb/s (fastEthernet), and 1000 Mb/s (Gbit Ethernet) and AC 800M supports 10 Mb/s and100 Mb/s (PM891 only).

In a system with several controllers and PCs a switched network should be usedbetween the nodes. If hubs are used instead of switches the number of connectednodes plays an important role for the throughput of the network and a single nodemay get an ethernet throughput which is less than the nominal network speed. Withswitches this is however not the case. Each node gets an ethernet throughput whichis more or less independent of the number of connected nodes. This means that thedata flow in the complete system is also not limited by AC 800M's ethernet speed of10 Mbit/s.

For networks with several switches ABB recommend to use 100 Mbit/s or 1 Gbitbetween switches since those ports need to manage data from several nodes to severalnodes. 10 Mbit/s should only be used on the ports where AC 800M controllers areconnected. Those ports only need to manage data for one node.

The actual communication throughput for a controller thus mainly depends on otherfactors than the ethernet speed, for example the cycle times of the applications andthe CPU load in the controller.

Inter Application Communication (IAC) uses communication variables (CVs). Thecontroller with the application that holds the out communication variable is the Server.The controller with the application that holds the in communication variable is theClient. The communication with IAC is based on cyclic reading only.

122 3BSE041434-600 H

5 Control and I/O 5.4 Communication

Connections Cannot Block Each Other

The controller can handle a number of concurrent communication connections. Allconnections are handled in a round robin fashion. This means that no connection canblock communication for any other connection.

For example this means that it is guaranteed that variable access from one controllerto another can always be executed even if a Control Builder is downloading a verylarge application domain to one of the controllers.

Number of MMS Connections

The controller’s communication stack handles several simultaneous connections.Messages are treated in a round robin fashion that guarantees that no connection isstarved, but the transmission rate through the stack decreases slightly with the numberof active connections. With 403 or less connections the performance decrease peradditional connection is however small. With more than 401 connections the amountof buffers per connection is reduced. This may decrease the performance for theconnections substantially more, at least for connections transmitting much data.

Communication Load

Table 5.10 shows the execution time for transferring one variable of type dwordbetween two AC 800M controllers using IAC or MMS, server or client.It also tells the resulting cyclic and total load for communicating one dword persecond.

The measurements were done by transferring as many dwords as possible in eachtransaction, that is, non-SIL IAC = 350 dwords, SIL IAC = 19 dwords,non-SIL MMS = 166 dwords, SIL MMS = 4 dwords.The task interval time was set to 200ms, and the IAC interval time to 100 ms.

Table 5.10: Load Caused by External IAC and MMS Communication (Continued)

PM867SIL3

PM867SIL2

PM865SIL3

PM865SIL2PM891PM864

IAC Server, Out variables

3.952.017.804.530.080.74Execution time/dword [us]

2.0E-048.3E-054.0E-042.2E-040.0E+003.5E-05Cyclic load /(dword/sec) [%]

3. 20 in PM85x and 100 in PM891.

3BSE041434-600 H 123


Table 5.10: Load Caused by External IAC and MMS Communication (Continued)(Continued)

PM867SIL3

PM867SIL2

PM865SIL3

PM865SIL2PM891PM864

6.3E-045.9E-041.3E-031.2E-031.4E-044.1E-04Total load /(dword/sec) [%]

IAC Client, In variables


1.5E-041.2E-052.3E-049.3E-050.0E+003.2E-06Cyclic load /(dword/sec) [%]


MMS Server, Read


1.7E-031.7E-035.0E-035.0E-030.0E+000.0E+00Cyclic load /(dword/sec) [%]


MMS Client, Read


5.0E-035.0E-031.0E-029.6E-030.0E+000.0E+00Cyclic load /(dword/sec) [%]


Defined Cyclic and Total Load in the table are based on measured values, andthey are not calculated based on Execution Time

Communication Throughput at 40% Cyclic Load

Table 5.11 shows how many dwords an AC 800M can transfer via IAC or MMS, toor from, another controller in case its cyclic load (without communication) is at 40%.

The number of dwords were increased until the controller was considered to bethrottled. The criteria for throttling were set to:

• Max. cyclic load

– HI Controller: 50%

– PA Controller: 70%

• Max. total load

124 3BSE041434-600 H


– HI Controller: 90%

– PA Controller: N/A

• Numbers of transactions/sec is maximized and does not increase anymore

The measurements were done by transferring as many dwords as possible in eachtransaction, that is, non-SIL IAC = 350 dwords, SIL IAC = 19 dwords, non-SIL MMS= 166 dwords, SIL MMS = 4 dwords.The task interval time was set to 200ms, and the IAC interval time to 100 ms.

Table 5.11: External IAC and MMS Communication Throughput at 40% Cyclic Load

PM867SIL3

PM867SIL2

PM865SIL3

PM865SIL2PM891PM864

IAC Server, Out variables

11021425684703101504900Max no of dwordscommunicated

8687595816989Transactions/sec

424143414142Cyclic load [%]

565673735271Total load [%]

IAC Client, In variables



414041414040Cyclic load [%]

625781744966Total load [%]

MMS Server, Read



424245444040Cyclic load [%]

666477805067Total load [%]

MMS Client, Read

3BSE041434-600 H 125


Table 5.11: External IAC and MMS Communication Throughput at 40% Cyclic Load(Continued)

PM867SIL3

PM867SIL2

PM865SIL3

PM865SIL2PM891PM864



425044484040Cyclic load [%]

717985905571Total load [%]

5.4.2 MasterBus 300 NetworkThe MasterBus 300 network can have maximum 100 nodes on a CI855 in a controlarea. The maximum performance is 200 data set per second. Switch over time to aredundant bus is 3 seconds.

5.4.3 INSUM NetworkTable 5.12: INSUM Design Limitations

Limitation

ReasonValueLimitation type

Execution time for IEC 61131-3 application andsystem heap memory

128Number of MCUs per controller

CPU performance on CI857128Number of MCUs per CI857

CPU performance on CI857 and memory onCI857

2Number of Gateways per CI857

CPU performance6Number of CI857 per AC 800M

126 3BSE041434-600 H


Table 5.13: INSUM Communication Interface CI857 Performance

Response time

CommentsConditionResultAction

Time measured inside theIEC 61131-3 application,from the time it sends thefirst command withINSUMWrite to NVDesStateuntil it receives the last statechange with INSUMReceive fromNVMotorStateExt.

Five NVs subscribedper MCU

5-8 s15-16.5 s

Start/stop,- 64 MCUs- 128 MCUs

Time measured on electricalstate signals on the MCUsfrom the time the first MCUstop until the second MCUstop.

Task cycle 250 ms, 66MCUs, five NVssubscribed per MCU

500 msStop one MCU due tochain interlock from otherMCU

5.4.4 800xA for AC 800M PerformanceThe provided information is measured with the products running on a PC with anIntel Xeon®, 2.40 GHz processor and 4Gbyte RAM.

The following settings were applied during the measurements.Graphic Display update rate 0 ms (default), OPC Server cache update rate 1000 ms,Controller task interval time 10 ms and Controller Modulebus scan time 100 ms.

3BSE041434-600 H 127


Display Exchange Time

The OPC Server for AC 800M collects data from controllers via MMS, and makesit available to OPC clients. The table below shows the display exchange time of aprocess display containing 100 display elements.

Table 5.14: Display Exchange Time for Graphic Display Element

Process Graphics 2(SubscriptionTimings)

VB Graphics(Latest Data received)

100 Elements of TypeObject

432 ms400 msValueAnalogInCC

928 ms622 msBarAnalogInCC

518 ms731 msReduced IconAnalogInCC

909 ms2562 msStandard IconAnalogInCC

939 ms986 msReduced IconMotorUniM

1014 ms2291 msStandard IconMotorUniM

1009 ms895 msReduced IconPidCC

1909 ms3592 msStandard IconPidCC

5.5 Supported Hardware and I/O FamiliesFor some hardware units a certain product revision is required, as described inRelease Notes.

5.5.1 AC 800M ControllersFigure 5.4 shows the symbol on the front of a CEX bus unit which indicates supportfor online replacement.

128 3BSE041434-600 H

5 Control and I/O 5.5 Supported Hardware and I/O Families

Figure 5.4: On-line replacement symbol

All communication interface units support firmware download by the ControlBuilder except CI858, which is upgraded with an external tool.

Supported AC 800M modules are shown in the Table 5.15.

Table 5.15: Supported Modules

Online upgrade(only valid forControl BuilderProfessional in

800xA)RedundancyOnline

ReplacementDescriptionUnit

RedundantNon-redundant

N/ANoNoNoController unit PM851 is a 32-bit, Single BoardComputer, which directly connects to the S800I/O system via ModuleBus (one electrical andone optical) and one communication interface.

PM851

PM851 supports a maximum of one CEX busmodule.

N/ANoNoNoThis is a replacement for PM851 having12 Mbyte RAM in total.

PM851A

N/ANoNoNoController unit PM856 is a 32-bit, Single BoardComputer, which directly connects to the S800I/O system via ModuleBus.

PM856

PM856 supports a maximum of twelve CEX busmodules.


PM856A

3BSE041434-600 H 129


Table 5.15: Supported Modules(Continued)





YesNoYesYesController unit PM858 (Redundant and Singular)is a high-performance, 32-bit, Single BoardComputer, which directly connects to the S800I/O system via ModuleBus. The unit has oneoptional Redundancy Control Link for redundantconfiguration. The PM858 processor unit hasperformance data which is 0.5 times theperformance of PM862.PM858 supports a maximum of twelve CEX busmodules

PM858

N/ANoNoNoController unit PM860 is a 32-bit, Single BoardComputer, which directly connects to the S800I/O system via ModuleBus. PM860 is twice asfast as PM856 in executing an applicationprogram.

PM860



PM860A

YesNoYesYes 1Controller unit PM861 (Redundant and Singular)is a 32-bit, Single Board Computer, whichdirectly connects to the S800 I/O system viaModuleBus. The unit has one optionalRedundancy Control Link for redundantconfiguration.

PM861


YesNoYesYes 1This is a replacement for PM861 and can useredundant communication units.

PM861A

130 3BSE041434-600 H







YesNoYesYes 1Controller unit PM862 (Redundant and Singular)is a high-performance, 32-bit, Single BoardComputer, which directly connects to the S800I/O system via ModuleBus. The unit has oneoptional Redundancy Control Link for redundantconfiguration. The PM862 processor unit hasperformance data which is approximately 1.2times the performance of PM861.

PM862


YesNoYesYes 1Controller unit PM864 (Redundant and Singular)is a 32-bit, Single Board Computer, whichdirectly connects to the S800 I/O system viaModuleBus. PM864 is 50% faster than PM861in executing an application program.

PM864


YesNoYesYes 1This is a replacement for PM864 and can useredundant communication units.

PM864A

3BSE041434-600 H 131







YesNoYesYes 1Controller unit PM865 (Redundant and Singular)is a 32-bit, Single Board Computer, whichdirectly connects to the S800 I/O system viaModuleBus. The unit has one optionalRedundancy Control Link for redundantconfiguration, and can use redundantcommunication units.

PM865

PM865 supports a maximum of twelve CEX busmodules.PM865 can be used in an AC 800M HighIntegrity system.

YesNoYesYes 1Controller unit PM866 (Redundant and Singular)is a high-performance, 32-bit, Single BoardComputer, which directly connects to the S800I/O system via ModuleBus. The unit has oneoptional Redundancy Control Link for redundantconfiguration. The PM866 processor unit hasperformance data which is approximately 1.4times the performance of PM864.

PM866


YesNoYesYes 1This is a replacement for PM866.PM866A

132 3BSE041434-600 H







YesNoYesYes 1Controller unit PM867 (Redundant and Singular)is a 32-bit, Single Board Computer, whichdirectly connects to the S800 I/O system viaModuleBus. The unit has one optionalRedundancy Control Link for redundantconfiguration, and can use redundantcommunication units.

PM867

PM867 supports a maximum of twelve CEX busmodules.PM867 can only be used in an AC 800M HighIntegrity system.

YesNoYesYes 1Controller unit PM891 (Redundant and Singular)is a high performance controller, with four timeshigher memory than PM866, and about twotimes faster performance than PM866. PM891is capable of handling applications with highrequirements.

PM891

PM891 connects to the S800 I/O system throughthe optical Modulebus. It can act as astand-alone Process Controller, or as acontroller performing local control tasks in acontrol network.PM891 supports a maximum of twelve CEX busmodules.

N/AN/AN/AYesCEX-bus interconnection unit.BC810BC810 can be used with PM861A, PM862,PM864A, PM865, PM866, PM866A, PM867.

3BSE041434-600 H 133







N/AN/AN/AYesCEX-Bus and RCU-Link Interconnection Unit.BC820BC820 can be used with PM858, PM862,PM866 (PR:F or later), PM866A.

YesNoYesYesMonitor the hardware and software executionof PM865.

SM810

There must always be one running SM modulein an AC 800M High Integrity controller.SM810 supports only SIL1-2 applications. In aredundant PM865 pair, it is only possible toperform online replacement on one of thesupervisory modules at a time.


SM811

SM811 supports both SIL1-2 and SIL3. A SIL3application code runs in this module in parallelwith PM865.There must always be one running SM modulein an AC 800M High Integrity controller.In a redundant PM865 pair, it is only possibleto perform online replacement on one of thesupervisory modules at a time.

134 3BSE041434-600 H








SM812

SM812 supports both SIL1-2 and SIL3. A SIL3application code runs in this module in parallelwith PM867.There must always be one running SM modulein an AC 800M High Integrity controller.In a redundant PM867 pair, it is only possibleto perform online replacement on one of thesupervisory modules at a time.

N/AYes 2NoYesThe CI853 is the RS-232C serial communicationinterface unit for the AC 800M. Two possiblesettings of the serial ports on the CI853 unit arenot valid and must not be used. These are 7data bits, no parity, 1 stop bit or 8 data bits,parity, 2 stop bits.

CI853

CI853 can be used in an AC 800M High Integritysystem.

N/AYes 3NoNoThe CI854 unit is the communication interfacefor PROFIBUS DP/V1 for the AC 800M withredundant PROFIBUS lines and DP/V1communication. It is a master unit and you canconnect up to 124 slaves to the master.However, you cannot connect more than 32units in one segment.

CI854

3BSE041434-600 H 135







Yes 4Yes 3YesYesThe CI854A unit is the communication interfacefor PROFIBUS DP/V1 for the AC 800M withredundant PROFIBUS lines and DP/V1communication. It is a master unit and you canconnect up to 124 slaves to the master.However, you cannot connect more than 32units in one segment.

CI854A

CI854A can be used in an AC 800M HighIntegrity system.

Yes 4Yes 3YesYesThis is a replacement for CI854A.CI854B can be used in an AC 800M HighIntegrity system. (Pending release, date to becommunicated).

CI854B

N/AYes 2NoYesThe CI855 unit is the communication interfacefor MasterBus 300 for the AC 800M. CI855houses two Ethernet ports to support MasterBus300 Network redundancy.

CI855


N/AYes 3NoYesThe CI856 is a communication interface for theS100 I/O system for the AC 800M. Up to fiveS100 I/O racks can be connected to one CI856where each I/O rack can hold up to 20 I/Oboards.

CI856


136 3BSE041434-600 H







N/AYes 5NoYesThe CI857 unit is the communication interfacefor INSUM for the AC 800M.

CI857


N/AYes 3NoYesThe CI858 unit is the communication interfacefor ABB Drives using DDCS protocol for theAC 800M.

CI858

Yes 4Yes 3YesYesThe CI860 unit is the communication interfacefor Fieldbus Foundation HSE for the AC 800M.

CI860

Yes 3Yes 3YesYesThe CI862 is the communication interface to theTRIO blocks (remote I/O) and manages thechannel data for the AC 800M controller. TheCI862 unit handles the I/O configuration and I/Oscanning of up to 30 TRIO blocks.

CI862

N/AYes 3NoYesThe CI865 is the communication interface toSatt I/O on ControlNet for AC 800M.

CI865

Yes 2Yes 2Yes 6YesThe CI867 unit is the MODBUS TCPcommunication interface for the AC 800M. CI867houses two Ethernet ports. One port supportsfull duplex with 100 Mbps speed and one portsupports half duplex with 10 Mbps speed.

CI867


N/AYes 2,7NoYesThe CI868 unit is the IEC 61850 communicationinterface for the AC 800M.

CI868


3BSE041434-600 H 137







Yes 2Yes 2YesYesThe CI869 is the AF 100 communicationinterface for AC 800M.

CI869


N/AYes 2,3NoYesThe CI871 is the PROFINET IO communicationinterface for the AC 800M.

CI871


Yes 2Yes 2YesYesThe CI872 is a MOD5 Communication interfaceon AC 800M that is used to communicate toMOD5 controllers. The CI872 can be used in aAC 800M HI Integrity system.

CI872

Yes 4Yes 2YesYesThe CI873 is the EtherNet/IP and DeviceNetcommunication interface for AC 800M.

CI873


NOTES:Online replacement is only supported in a redundant configuration, the unit to replace MUST NOT be energized.1.During an online upgrade, the communication between the communication interface and the connected sub units areinterrupted.

2.

During an online upgrade, the communication interface sets the outputs of connected I/O units to values specified byOSP control (Output Set as Predetermined).

3.

Full support of online upgrade. One of the redundant communication interface units is always active during the onlineupgrade process.

4.

During an online upgrade, CI857 is disconnected from INSUM Gateway and the connected INSUM devices keep onrunning with the values they have just before the switch.

5.

Module redundancy only. It is not possible to get media redundancy by enabling the second Ethernet port (Ch2).6.For CI868 Firmware Upgrade scenarios applicable during Control Builder project migration from earlier versions toversion 6.0, refer to AC 800M IEC 61850 Engineering and Configuration (9ARD171385*) Manual.

7.

138 3BSE041434-600 H


5.5.2 Adaptors for I/O TypesTable 5.16 shows the supported adaptors for I/O types.

Table 5.16: Adaptors for I/O Types

SOE 2HART 1Can be connected toAdaptor

YesYesYesYesYesYesYesYesYes

!PM851 and PM851APM856 and PM856APM858PM860 and PM860APM861 and PM861A(Single controller only)PM862(Single controller only)PM864 and PM864A(Single controller only)PM866 and PM866A(Single controller only)PM891(Single controller only)

TB820

YesYesYes

YesYesYes

PM851 and PM851APM856 and PM856APM858PM860 and PM860APM861 and PM861APM862PM864 and PM864APM865PM866 and PM866APM867PM891

TB840TB840A

YesYesYesYesYesYesYesYes

YesYesYesYesYesYesYesYes

YesNoCI856DSBC 173A

YesNoCI856DSBC 174

YesNoCI856DSBC 176

NoYesCI854, CI854A, CI854BCI801

NoNoCI854, CI854A, CI854BCI830 3


3BSE041434-600 H 139


Table 5.16: Adaptors for I/O Types(Continued)

SOE 2HART 1Can be connected toAdaptor

NoYesCI854, CI854A, CI854BCI840A


NoYesCI854, CI854A, CI854BCI920A

NoNoCI854, CI854A, CI854B200-APB12

NoNoCI865200-ACN

NoNoCI873200-AENTR

NoNoCI865200-RACN

NoNoCI854, CI854A, CI854BRPBA-01

NoNoCI854, CI854A, CI854BNPBA-12

NoNoCI854, CI854A, CI854BFPBA-01

YesNoCI871RETA-02

YesNoCI871FENA-11

YesNoCI871MNS iS

YesNoCI871PNQ22

NoNoCI873LD800 DN

NOTES:Only valid for Control Builder Professional in 800xA.1.OPC Server for AC 800M must be used for alarms and events.2.CI830 is replaced by CI801 in new installations.3.

Table 5.17 provides a description of the supported adaptors.

Table 5.17: Adaptor Description (Continued)

DescriptionAdaptor

ModuleBus ModemTB820

ModuleBus Modem, primarily for redundant ModuleBus.TB840TB840A

140 3BSE041434-600 H


Table 5.17: Adaptor Description (Continued)(Continued)

DescriptionAdaptor

The DSBC 173A unit is the bus extender slave inserted in the last position of a S100I/O rack.

DSBC 173A

The DSBC 174 unit is the bus extender slave inserted in the last position of a S100I/O rack.

DSBC 174

The DSBC 176 unit is the bus extender slave inserted in the last position of a S100I/O rack.

DSBC 176

The CI801 is a remote PROFIBUS DP-V1 adaptor for S800 I/O units. The CI801does not support redundancy.

CI801

The CI801 can handle up to 24 S800 I/O-units. 12 I/O-units can be directly connectedto the ModuleBus on the CI801, while the remaining I/O-units have to be connectedvia I/O-clusters. Up to 7 I/O-clusters can be connected to one CI801, and thenumbering of I/O-units connected to a cluster will start with 101 for cluster 1, 201for cluster 2 and so on.

The CI840(A) is a remote PROFIBUS DP-V1 adaptor for S800 I/O units, withredundancy capabilities. CI840 supports redundant I/O modules.

CI840CI840A

The CI840(A) can handle up to 24 S800 I/O-units. 12 I/O-units can be directlyconnected to the ModuleBus on the CI840, while the remaining I/O-units have to beconnected via I/O-clusters. Up to 7 I/O-clusters can be connected to one CI840(A),and the numbering of I/O-units connected to a cluster will start with 101 for cluster1, 201 for cluster 2 and so on.

The CI920(A) is a remote PROFIBUS DP-V1 adaptor for S900 I/O units.CI920CI920A

The 200-APB12 unit is a remote PROFIBUS DP slave I/O adaptor for S200 I/O andS200L I/O units. 200-APB12 is connected to the controller via a PROFIBUS DP/V0master unit on the controller system bus. A 200-APB12 unit can have up to eightS200 I/O units. The number of 200-APB12 slaves are, by the DIP switches, limitedto 99.

200-APB12

The 200-ACN is a remote ControlNet I/O adaptor for Series 200 I/O units. 200-ACNis connected to a controller via a CI865 communication interface on the controllersystem bus. 200-ACN units are used as nodes on the Satt ControlNet fieldbus. Each200-ACN unit can handle up to eight Series 200 I/O units.

200-ACN

3BSE041434-600 H 141



DescriptionAdaptor

The 200-AENTR is a remote EtherNet/IP I/O adaptor for Series 200 I/O units.200-AENTR is connected to a controller via a CI873 communication interface onthe controller system bus. Each 200-AENTR unit can handle up to eight Series 200I/O units. 200-AENTR has two Ethernet ports with an in built switch, which meansthat the adaptors can be daisy-chained to the CI873 using cross-wiredEthernet-cables without the need for external switches.

200-AENTR

The 200-RACN unit is a remote Concurrent adaptor for rack based I/O units.200-RACN is connected to a controller via a CI865 communication interface on thecontroller system bus. One or several adaptor 200-RACN units are used as nodes.A maximum of eight I/O-racks are supported on the Satt Concurrent fieldbus.

200-RACN

These PROFIBUS-DP adaptor units are an optional device for ABB drives whichenables the connection of the drive to a PROFIBUS system. The drive is consideredas a slave in the PROFIBUS network. It is possible to:

RPBA-01NPBA-12FPBA-01

• give control commands to the drive (Start, Stop, Run enable, etc.)• feed a motor speed or torque reference to the drive• give a process actual value or a process reference to the PID controller of the drive• read status information and actual values from the drive• change drive parameter values• reset a drive fault.

These Ethernet Adaptor modules are an optional device for ABB drives, whichenables the connection of the drive to a PROFINET IO (PNIO) network. The driveis considered as a PNIO device on the PROFINET IO network, and it is compatiblewith all PNIO controller stations that support PROFINET IO and sub-slots. Throughthe Ethernet Adaptor module, it is possible to:

RETA-02FENA-11

• give control commands• give control commands to the drive (Start, Stop, Run enable, etc.)• feed a motor speed or torque reference to the drive• give a process actual value or a process reference to the PID controller of the drive• read status information and actual values from the drive• change drive parameter values• reset a drive fault.

142 3BSE041434-600 H



DescriptionAdaptor

ABB MNS iS is a motor control center solution that can be used in PROFINET IOnetwork. MNS iS delivers all the functions for control, protection, and monitoring ofmotors and motor starters using software and hardware modules for the specifictasks. MLink, one of the interface modules in MNS iS, serves as the serial gatewayinterface to higher level systems which communicate to all modules throughPROFINET IO.

MNS iS

The PNQ22 Ethernet adaptor module provides PROFINET IO connectivity forABB Universal Motor Controller UMC100 and ABB soft starters PST and PSE.The PNQ22 allows the connection of four devices to PROFINET IO.

PNQ22

The LD 800DN adaptor, which functions as a gateway to connect control levelnetworks with device level networks, provides a router or bridge functionality toconnect EtherNet/IP to DeviceNet. The LD 800DN provides centralized data storagefor data that is shared between the DeviceNet and Ethernet/IP networks.

LD800 DN

The following adaptors are supported (Table 5.18), but only for migration purposes,NOT at new installations.

Table 5.18: Supported Adaptors for Migration

DescriptionAdaptor

The unit CI830 is a remote PROFIBUS DP-V0 I/O adaptor for units. CI830 is connectedto a controller via a PROFIBUS DP-V0 master unit on the controller system bus.

CI830

The CI830 can handle up to 24 S800 I/O-units. 12 I/O-units can be directly connectedto the ModuleBus on the CI830, while the remaining I/O-units have to be connectedvia I/O-clusters. Up to 7 I/O-clusters can be connected to one CI830, and the numberingof I/O-units connected to a cluster will start with 101 for cluster 1, 201 for cluster 2 andso on.CI830 is replaced by CI801 at new installations. CI830 does not have full support forall S800 I/O-types.

3BSE041434-600 H 143


5.5.3 I/O FamiliesAll I/O units may be replaced in a running system. Table 5.19 shows the differentI/O families.

Table 5.19: I/O Families

Connects ToI/O Family

PM851, PM851A, PM856, PM856A, PM858, PM860, PM860A,PM861, PM861A, PM862, PM864, PM864A, PM865, PM866,PM866A, PM867, PM891TB820, TB840, TB840ACI801, CI830, CI840, CI840A

S800 I/O

CI920, CI920AS900 I/O

CI862TRIO I/O

PM851, PM851A, PM856, PM856A, PM858, PM860, PM860A,PM861, PM861A, PM862, PM864, PM864A, PM865, PM866,PM866A, PM867, PM891TB820, CI801, CI830, CI858, RPBA-01, NPBA-12, FPBA-01,RETA-02, FENA-11

ABB Standard Drives

PM851, PM851A, PM856, PM856A, PM858, PM860, PM860A,PM861, PM861A, PM862, PM864, PM864A, PM865, PM866,PM866A, PM867, PM891TB820, CI858, RPBA-01, NPBA-12, FPBA-01, RETA-02,FENA-11

ABB Engineered Drives

CI856S100 I/O

200-APB12, 200-ACN, 200-AENTRS200 I/O, S200L I/O and I/O 200C

200-RACNSatt Rack I/O

144 3BSE041434-600 H


S800 I/O

Table 5.20 shows the different S800 I/Os support.

Table 5.20: S800 I/O

DescriptionName

Analog input unit, 8 inputsAI801



Analog input unit, 4 differential inputsAI820

Analog input unit, galvanic isolated analog input unit, 4 channelsAI825

Analog input unit, 8 RTD inputsAI830 1

Analog input unit, 8 TC inputsAI835 2

Analog input unit, 8 TC inputs, redundant possibilitiesAI843

Analog input unit, 8 inputs, redundant possibilities , HARTAI845

Analog input unit, 8 inputs, SIL certified, redundant possibilities, HARTAI880A 5

Analog input unit, 8 inputs, Intrinsic Safety interfaceAI890

Analog input unit, 8 RTD/TC inputs, Intrinsic Safety interfaceAI893

Analog input unit, 8 inputs , Intrinsic Safety interface, HARTAI895

Analog output unit, 8 outputsAO801

Analog output unit, 8 outputsAO810 3



Analog output unit, 8 outputs, redundant possibilities, HARTAO845

Analog output unit, 8 outputs, Intrinsic Safety interface.AO890

Analog output unit, 8 outputs , Intrinsic Safety interface, HARTAO895

Digital input unit, 16 inputsDI801



3BSE041434-600 H 145


Table 5.20: S800 I/O(Continued)

DescriptionName







Digital input unit, 8 channels with event recording (SoE, Sequence of events)DI825 4


Digital input unit, 16 inputs with event recording (SoE, Sequence of events)DI830 4

Digital input unit, 16 inputs with event recording (SoE, Sequence of events)DI831 4

Digital input unit 16 inputs, redundant possibilities with event recording (SoE,Sequence of events)

DI840 4

Digital input unit, 16 inputs, SIL certified, redundant possibilitiesDI880 5

Digital input unit, 8 inputsDI885 4

Digital input unit, 8 inputs, Intrinsic Safety interfaceDI890

Digital output unit, 16 outputsDO801









Digital output unit 16 outputs, redundant possibilitiesDO840

146 3BSE041434-600 H



DescriptionName

Digital output unit, 16 outputs, SIL certified, redundant possibilitiesDO880 5

Digital output unit, 8 outputs, Intrinsic Safety interfaceDO890

Digital pulse counterDP820

Pulse/Frequency input, 8 inputs, redundant possibilities, supported in CI830 butwithout redundancy

DP840

NOTES:AI830/AI830A.1.AI835/AI835A.2.AO810/AO810V2.3.No support in CI801 and CI840.4.Only in PM862, PM865, PM866A, PM867 and PM891 via TB840/TB840A.5.

S900 I/O

Table 5.21 shows the different S900 I/Os.


DescriptionName1

Analog input unit, 4 inputs, 4-20 mAAI910N/S

Analog input unit, 4 inputs, 4-20 mA, isolatedAI920N/S

Analog input unit, 4 inputsAI921N/S

Analog input unit, 4 inputs, 4-20 mA, HARTAI930N/S

Analog input unit, 4 inputs, 0/4-20 mA, HARTAI931N/S

Analog input unit, 4 inputs, temperature sensorAI950N/S

Analog output unit, 4 outputs, 4-20 mAAO910N/S

Analog output unit, 4 outputs, 4-20 mA, isolatedAO920N2/S

Analog output unit, 4 outputs, 4-20 mA, HARTAO930N/S

Analog input, 1 x 4 channels, activeAI910

Analog input, 4 x 1 channelAI920

3BSE041434-600 H 147



DescriptionName1

Analog input, 1 x 4 channels, active, HARTAI930

Analog input, 1 x 4 channels, passive, HARTAI931

Temperature input, 4 x 1 channel, resistor and thermocoupleAI950

Analog output, 1 x 4 channelsAO910

Analog output, 4 x 1 channel (isolated)AO920

Analog output, 1 x 4 channels, HARTAO930

Digital input unit, 4 inputs, (NAMUR), isolatedDI920N

Digital output unit, 4 outputs, (for solenoid valves)DO910N/S

Digital output unit, 4/6 outputs, dry contacts (relay)DO930N2/S

Digital output unit, 8 outputsDO940N/S

Digital output unit, 16 outputsDO980N/S

Frequency input and counter, 2 blocksDP910

Frequency input and pulse counter, 2 inputsDP910N/S

Bidirectional unit, 8 channels, (programmable) for digital input, 8 inputs,NAMUR/dry contacts or digital output, 8 outputs, for low power valves

DX910N/S

Digital output, 4 x 1 channel (Solenoid driver)DO910

Relay output, 6 x 1 channelDO930

Digital input/output, 1 x 8 channelsDX910

1) Modules referenced without S,B or N in the ending are valid for all three S900 I/O series.2) Available and supported only by N-Series'

148 3BSE041434-600 H


TRIO I/O

Table 5.22 shows the different TRIO I/Os.

Table 5.22: TRIO I/O

DescriptionName

Analog I/O unit, 4 inputs, 2 outputs4In2Out

Analog, current source I/O unit, 4 inputs / 2 outputsCSAnalogIO (CSANALOG)

Analog, current source input unit, 6 inputsCSAnalogIn (CSANAINP)

Analog, current source output unit, 6 outputsCSAnalogOut (CSANAOUT)

Analog, circuit thermocouple input unit, 6 inputsThermocouple (TC)

Analog, circuit resistive temperature detector, 6 inputsRTD

Digital, 16 circuit DC source and sink I/O, configurable as inputsor outputs

16CircuitIO (IO_16CKT)

Digital, 32 circuit DC source and sink I/O, configurable as inputsor outputs

32CircuitIO (IO_32CKT)

Digital, 115 VAC 8 circuit grouped and low leakage I/O,configurable as inputs or outputs

8CircuitGrouped (GRP_8CKT)

Digital, 115 VAC / 125 VDC 8 circuit isolated I/O (4 groups of twoI/O), configurable as inputs or outputs

8CircuitIsolated (ISO_8CKT)

Digital, 115 VAC 16 circuit inputs (two banks of eight inputs)16CircuitIn (IN_16CKT)

Digital, 16 relay outputs (four groups of four independent outputseach)

16CircuitOut (OP_16CKT)

Counter, high speed counter A, four 16 bit up/down countersHighSpeedCounterA (HSC_A)

Counter, high speed counter B, two bi-directional 24 bit up/downcounters

HighSpeedCounterB (HSC_B)

3BSE041434-600 H 149


S100 I/O

Table 5.23 shows the different S100 I/Os supported by AC 800M.


DescriptionName

Bus extender slave inserted in the last position of an S100 I/O rack.DSBC 173A1, DSBC174, DSBC 176

Analog input Unit. 32 channels, resolution 8 bit, Single ended, 0 to ±10V or 0 to ±20 mA

DSAI 110

Analog input board, 16 inputsDSAI 130DSAI 130A

Analog input board, 16 inputs with 4 sets of filter timesDSAI 130D

Analog input board, 32 inputsDSAI 133DSAI 133A

Analog input Unit.16 channels, resolution 8 bit, differential, 0 to ±10 V or0 to ±20 mA

DSAI 135

Analog input Unit.31 (+ 1 ref.) channels for Pt100, 3-wire, resolution 12or 13 bits, -100/ +320 or -200/ + 640°C.

DSAI 145

Galvanic isolation between channel and system.

Analog input unit. 31 (+ 1 ref.) channels for Pt100, 3-wire, resolution 12or 13 bits, -100/ +320, -40/+40 or -200/ + 640°C.

DSAI 146


14 (+ 2 ref.) channels for Pt100, 4-wire, resolution 12 or 13 bits, -100/+320, -40/+40 or -200/ + 640°C.

DSAI 151


Digital input board, 32 inputs, 24VDSDI 110, DSDI 110ADSDI 110AV1

Digital input board, 32 inputs, 24 VDSDI 115

Digital input board, 32 inputs, 24 V non-isolatedDSDI 116

Digital input board, 32 inputs, 48 VDSDI 120, DSDI 120ADSDI 120AV1

Digital input board, 32 inputs, 48 VDSDI 125

150 3BSE041434-600 H



DescriptionName

Digital input board, 32 inputs, 48 V non-isolatedDSDI 126

Digital input unit. 16 channels, 110 V A.C./D.C., controlled by scanningor interrupt, pulse extension. Local time tagging of process events.Galvanic isolation between channel and system.

DSDI 131

Digital input unit. 16 channels, 220 V A.C./D.C., controlled by scanningor interrupt, pulse extension. Local time tagging of process events.

DSDI 141


Digital output board, 32 outputsDSDO 110

Digital output board, 32 outputsDSDO 115

Digital output board, 32 outputs, OSP controlDSDO 115A

Digital output unit. 16 channels, 24-48 V d.c., 1A. Galvanic isolationbetween channel and system.

DSDO 120

Digital output board, 16 relay outputs 24 - 240 V A.C/D.C.DSDO 130

Digital output board, 16 relay outputs 24 - 240 V A.C/D.C.DSDO 131

Digital output unit. 16 channels, 48-240 V D.C. Galvanic isolation betweenchannel and system.

DSDO 140

Analog output board, 4 outputsDSAO 110


Analog output board, 8 outputs, OSP controlDSAO 120A


Analog output board, 16 outputs, OSP controlDSAO 130A

Analog input/output board, 8 inputs 8 outputsDSAX 110DSAX 110A

Absolute binary decoder with hardware strobe, 2 channelsDSDP 010

Positioning control board for one positioning loopDSDP 140B

Loop transducer interface board, 4 channelsDSDP 160

Loop transducer interface board, 4 channelsDSDP 161

3BSE041434-600 H 151



DescriptionName

Pulse counter board, 4 channelsDSDP 170

NOTE:1. Product Revision PR:B or later.

S200 I/O

Table 5.24 shows the different S200 I/Os.


DescriptionName

Dummy I/O unit200-DUTB

Digital input unit, 8 inputs200-IA8

Digital combined unit, 10 inputs and 6 outputs200-IB10xOB6

Digital input unit, 16 inputs200-IB16

Digitally combined unit, 16 inputs and 16 outputs200-IB16xOB16P

Digital input unit, 32 inputs200-IB32

Analog combined unit, 4 inputs and 2 outputs200-IE4xOE2

Analog input unit, 8 inputs200-IE8

Analog input unit, 4 inputs200-IF4I

Digital input unit, 8 inputs200-IM8

Pulse counter board, 2 x 4 inputs200-IP2

Pulse counter board, 4 x 2 inputs200-IP4

Analog input unit, 8 inputs200-IR8

Analog input unit, 8 inputs200-IR8R

Analog input unit, 8 inputs200-IT8

Digital output unit, 8 outputs200-OA8

Digital output unit, 16 outputs200-OB16

152 3BSE041434-600 H



DescriptionName

Digital output unit, 16 outputs200-OB16P

Digital output unit, 2 x 16 outputs200-OB32P

Digital output unit, 8 outputs200-OB8EP

Analog output unit, 4 outputs200-OE4

Analog output unit, 4 outputs200-OF4I

Digital output unit, 8 outputs200-OM8

Digital output unit, 8 outputs200-OW8

S200L I/O and I/O 200C

Table 5.25 shows the different S200L I/Os and Table 5.26 shows I/O 200C.

Table 5.25: S200L I/O

DescriptionName



Analog combined unit, 4 inputs and 2 outputsAX210



Digital combined unit, 10 inputs and 6 outputsDX210

Table 5.26: I/O 200C

DescriptionName

Digital combined unit, 10 inputs and 6 outputs200C-IB10xOB6P

Digital input unit, 16 inputs200C-IB16

Analog combined unit, 4 inputs and 2 outputs200C-IE4xOE2

3BSE041434-600 H 153


Table 5.26: I/O 200C(Continued)

DescriptionName

Analog input unit, 8 inputs200C-IE8

Digital output unit, 16 outputs200C-OB16P

Analog output unit, 4 outputs200C-OE4

Satt Rack I/O

Table 5.27 shows the different Satt Rack I/Os.

Table 5.27: Satt Rack I/Os

DescriptionName

Digital input board with 16 inputsIAPG

Digital input board with 16 inputsIDLD

Digital input board with 32 inputsIDP

Digital input board with 32 inputsIDPG

Digital input board with 32 inputsIDN

Digital input board with 32 inputsIDI

Digital input board with 32 inputsPTC

Digital output board with 16 outputsORG

Digital output board with 16 outputsORGH

Digital output board with 16 outputsOATG

Digital output board with 16 outputsODP2

Digital output board with 16 outputsODPG2

Digital output board with 16 outputsORM

Digital output board with 32 outputsODP.5

Digital output board with 32 outputsODP.8

Digital output board with 32 outputsODPG.8

154 3BSE041434-600 H


Table 5.27: Satt Rack I/Os(Continued)

DescriptionName

Digital output board with 32 outputsODPL.5

Digital output board with 32 outputsODPLD

Digital output board with 32 outputsODN.2

Digital output board with 32 outputsODLD.5

Digital output board with 32 optocoupled outputs, short circuit proofODSG

Analog input board with 8 inputsIBA

Analog input board with 8 inputsIRA

Analog input board with 8 inputsICA

Analog input board with 8 inputsIVA

Analog input board with 8 inputsIVAPOT

Analog output board with 2 outputsOCVA

Analog output board with 4 outputsOCAHG

Analog output board with 4 outputsOCAH

Analog output board with 4 outputsOCAH with handstation

Input pulse analyzer board with 4 inputs, 8 bit countersIPA4

Drives System

There are two types of drives systems, ABB standard and ABB engineered.

ABB Standard Drives

Table 5.28 shows the ABB standard drives.

Table 5.28: ABB Standard Drives

ApplicationName

Standard driveACS400

Crane applicationACS600

3BSE041434-600 H 155


Table 5.28: ABB Standard Drives(Continued)

ApplicationName

Pump and fan applicationACS600

Standard applicationACS600

Crane applicationACS800

Pump and fan applicationACS800



Standard driveDCS400

Standard driveDCS500

ABB Engineered Drives

Table 5.29 shows the ABB engineered drives.

Table 5.29: ABB Engineered Drives

ApplicationName

IGBT supply (ISU) applicationACS600

System applicationACS600

Asynchronous driveACS600AD

Cycle converter driveACS600C

Synchronous driveACS600SD

IGBT supply (ISU) applicationACS800



Standard driveACS1000

System applicationDCS600

156 3BSE041434-600 H


6 Fieldbus

This topic describes the capacity limits and constraints for the application design thatneed to be considered when using standardized fieldbus protocols to connect fielddevices to System 800xA.

Connect Services are supported for the following fieldbus protocols:

• PROFIBUS

• HART, WirelessHART

• FOUNDATION Fieldbus

• IEC 61850

Connect Services allow accessing field device data independent of a controllerapplication, e.g. for Operations or Asset Optimization purposes. The limits that needto be observed for these Connect Services are described in the section for the respectiveprotocol. For constraints regarding deployment of different connect services refer toConnect Service Calculation Rules and Limitations.

6.1 MODBUS

6.1.1 MODBUS RTU Master CommunicationThese below tables show the maximum transmission rate when CI853 acts asMODBUS RTU master at a base controller load of 50%.

Max Transmission RateAC 800M(Total Transactions/second)

50% Load in the Controller PM864A

MBReadMBWrite300 Booleans in Each Telegram

4channels

1channel

4channels

1channel

3BSE041434-600 H 157

6 Fieldbus 6.1 MODBUS

4.82.12.41.01200 baud(8 data bits, 1 stop bit, odd parity)



50% Load in the Controller PM866


4channels

1channel

4channels

1channel




50% Load in the Controller PM891


4channels

1channel

4channels

1channel



158 3BSE041434-600 H


6.1.2 MODBUS TCPThe tables show the maximum transmission rate when CI867 acts as MODBUS TCPmaster or slave.

Table 6.1: MODBUS TCP Performance Data. Reading Dint using one CI867 as Master

AverageTransaction/Slave1

Total Transactions/s(Sum of all Slaves)

MessageLength (in

Dint)

Number of Slaves Connectedand Communicating

149149601task time=100 ms










1. Cyclic read at maximum possible rate is used.

3BSE041434-600 H 159


Table 6.2: MODBUS TCP Performance Data. Reading Boolean using one CI867 as Master



MessageLength (inBoolean)













160 3BSE041434-600 H


Table 6.3: MODBUS TCP Performance.Data. Reading Real using one CI867 as Master



MessageLength (in

Real)







Table 6.4: MODBUS TCP Performance Data. Reading Dint using one CI867 as Slave

AverageTransaction/Master


Number ofData in Dint

Number of MastersConnected andCommunicating







3BSE041434-600 H 161


Table 6.5: MODBUS TCP Performance. Data. Reading Boolean using one CI867 asSlave

AverageTransaction/Master


Number ofData inBoolean

Number of MastersConnected andCommunicating







6.2 PROFIBUSThis topic describes the limits and performance data for PROFIBUS.

6.2.1 Device Management PROFIBUS HART System ExtensionsThis section describes Device Management PROFIBUS HART system extension forAC 800M and AC 800M HI Integrity systems:

162 3BSE041434-600 H

6 Fieldbus 6.2 PROFIBUS

AC 800M System

Load the dependent extensions as in the Table 6.6 before loading PROFIBUS HARTextensions.

Table 6.6: PH System Extensions for AC 800M System

Dependent ExtensionsPH System ExtensionsSl No.

-Fieldbus Builder PROFIBUS/HART1

AC800M ConnectABB HART Device Integration Library– Basics

2

ABB PROFIBUS Device IntegrationLibrary - Basics

3

ABB Asset MonitorEnvironment

ABB HART Device Integration Library- Asset Monitoring

4

ABB PROFIBUS Device IntegrationLibrary - Asset Monitoring

5

AC800M HI Integrity System

Before loading PROFIBUS HART HI extension on safety system, make sure to loadAC 800M Extension followed by AC 800M HI Extension.

Table 6.7: PH System Extensions for AC 800M HI Integrity System

Dependent ExtensionsPH System ExtensionsSl No.

AC800M HI IntegrityABB HART Device Integration Library– Basics -HI

1

ABB Asset MonitorEnvironment

ABB HART Device Integration Library- Asset Monitoring - HI

2

6.2.2 Network Connection through Communication InterfaceCommunication Interface for PROFIBUS DP: CI854A/B

CI854A/B acts as a PROFIBUS Master only.

For PROFIBUS DP following limits and performance figures need to be taken intoaccount:

3BSE041434-600 H 163


Table 6.8: PROFIBUS and CI854A/B Parameters

LimitsParameterCommunication

General Limits

12 (a redundant pair consumestwo of them)

Maximum no. of CI854A/B in anAC 800M (CEX Bus)

126Max. number of nodes per network

124Max. number of slave nodessupported

0Reserved node addresses forCI854A/B (1 for redundant CI)

2 -125Available node addresses forslaves

9,6 kbit/s - 12 Mbit/sSupported Baudrates onPROFIBUS

Configuration Limits

>1,5 Mbit/s: 4000 byte input andoutput data

Max. length of I/O data forCI854A/B

PROFIBUS DPcyclic communication

<= 1,5 Mbit/s: no limitation in CI

Performance Limits

1 ms at 12 Mbit/s with one slavesupporting 1ms

Fastest possible cycle timePROFIBUS DPcyclic communication

10 – 20 ms at 1,5 Mbit/sTypical cycle times

The cycle time on PROFIBUS depends on the baud rate, the sum of I/O data, andthe slave timing parameter and is therefore application-specific.

Further features and constraints to be considered during application design:

• S800 I/O connected to CI840 and/or S900 I/O connected to CI920/CI920Asupports cable redundancy together with slave redundancy.

• S900 (CI920/CI920A) and S800 (CI840 and CI801) support configuration change(changing the parameters) without disrupting the cyclic data communication.

164 3BSE041434-600 H


• If the PROFIBUS master unit, CI854A/B, loses contact with a slave unit, forexample due to a disconnected cable, input values are set according to ISPconfiguration. If the I/O unit does not support ISP, all input values will freeze.

• Reset of PROFIBUS DP master, CI854A/B, and the complete PROFIBUS isdone if one of the following bus parameter settings are changed:

– Node address of CI854A/B,

– baud rate.

– highest station address (HSA).

A change of the other bus parameters does not affect the runningcommunication.

6.2.3 PROFIBUS Connect ServiceFor PROFIBUS the Connect Service is combined with the Connect Service for HART.

Following limits and performance figures need to be taken into account when usingthe PROFIBUS/HART Connect Package of Device Management PROFIBUS/HART:

Table 6.9: PROFIBUS/HART Connect Service Parameters


1Max number of PROFIBUS/HART ConnectServices per Connectivity Server

PROFIBUS/HARTConnectivity Server acycliccommunication 2500Max number of PROFIBUS devices per

PROFIBUS/HART Connect Service

For constraints regarding the deployment of the PROFIBUS/HART Connect Servicetogether with other connect service, refer to Connect Service Calculation Rules andLimitations.


• PROFIBUS/HART Connectivity Server is designed for providing data to assetmanagement applications. It does not support fast update rates as they are expectedfor changing process values of field-devices. Applications requiring fast updaterates need to rely on cyclic communication.

Only Asset Optimization Server can be a client to PROFIBUS/HART ConnectService.

3BSE041434-600 H 165


• Acyclic communication requests through PROFIBUS/HART Connect Serviceare handled sequentially. I.e. a new request is only sent when the previous requesthas been answered from the field device.

• The PROFIBUS/HART Connect Service will run on Connectivity Server. OneConnectivity Server can run only one PROFIBUS/HART Connect Service.

By default, the PROFIBUS/HART Connect Service is configured on the ControlNetwork level. Hence the number of Connectivity Servers required to run thePROFIBUS/HART Connect Services are equal to number of Control Networksin the 800xA System for default configuration.

If any Control Network contains more PROFIBUS/HART devices than supportedby one instance of the PROFIBUS/HART Connect Service, the Service can beconfigured on lower hierarchy levels, for example: Controllers. However, aseparate PROFIBUS/HART Connect Service instance must be configured on allparallel hierarchy levels (for example: Controllers) that PROFIBUS/HART datashall be fetched from. A separate Connectivity Server required for eachPROFIBUS/HART Connect Service configured.

Note that the number of Connectivity Servers may need to be increasedaccordingly as only one PROFIBUS/HART Connect Service is supported perConnectivity Server.

• If HART devices are configured for acyclic communication in the same hierarchylike the PROFIBUS devices, both HART and PROFIBUS devices are handledin the same PROFIBUS/HART Connect Service instance. In that case the sumof HART and PROFIBUS devices must not exceed 2500 per Connect Serviceinstance.

6.3 PROFINET IO

6.3.1 Network Connection through Communication InterfaceCommunication Interface for PROFINET IO: CI871

For PROFINET IO the following limits and performance figures need to be takeninto consideration:

166 3BSE041434-600 H

6 Fieldbus 6.3 PROFINET IO

Table 6.10: PROFINET IO and CI871 Parameters


12Maximum no. of CI871 in an AC 800M(CEX Bus)

General LimitsNote: Only applicationredundancy supported.

126Maximum number of PROFINET IOdevices per CI871

512Maximum supported modules perPROFINET IO device


1 InputNumber of IOCRs (IO CommunicationRelationships) per device

PROFINET IO cyclic RealTime communication 1 Output

1,440 bytesMaximum length of IOCR

4,096 bytesMaximum frame length of Read/Writeservice

PROFINET IO acycliccommunication

Performance Limits

1 frame per ms for bothinput and output directionwith one devicesupporting 1 ms

Fastest possible update timesPROFINET IO cyclic RealTime communication

25 per second with 60bytes data

Maximum number of Read/Writetransactions

PROFINET IO acycliccommunication

The update times on PROFINET IO depends on the number of devices and Ethernetframes that will be communicated and is therefore application-specific.

Examples for typical configurations:

• Example 1: Update times for all devices is configured to 32 ms (default), thenup to 32 devices can be connected to CI871.

• Example 2: Update times for all devices is configured to 8 ms, then up to 8devices can be connected to CI871.

• The information regarding limits and CPU load need to be adjusted with respectto acyclic communication:

3BSE041434-600 H 167

6 Fieldbus 6.3 PROFINET IO

The limitation for the CPU load of CI871 is checked by the system duringdownload. If the system detects that there is a CPU overload, then it is indicatedin the Compilation Summary window and the download is blocked. The CI871may not function properly when there is an overload. The user can check the CPUload before and after download by use of the Web Interface. The limit for the CPUload is 100%. Up to that value the CI871 works stable without any problems orrestrictions.

• Limits of acyclic communication:

CI871 can handle up to 25 transactions/second for a payload of 60 bytes perread/write operation causing a CPU load of about 1% on CI 871 and 0.3% onPM866 for each transaction.

The maximum frame length is defined by 4 KBytes for each service. Longerframes than 60 bytes cause a slightly increased CPU load.

The CPU load for acyclic communication cannot be checked by the systemautomatically during download. CI871 is able to handle up to 10 acyclictransactions/seconds in a stable fashion with high cyclic load in parallel. In casea higher acyclic communication load is required the user needs to take care thatthe resulting CPU load is in a safe area example below 70%.

• Application redundancy is only possible with devices that actively support it,For example: ABB MNSiS.


PROFINET IO connectivity through CI871 supports application redundancy. Thatmeans a second CI871 module needs to be configured separately to communicatethe same data with the voting implemented in the controller application.

6.4 HART, WirelessHART

6.4.1 HART InterfacesHART information is made available to the system by following interfaces:

• S800 I/O cards:

– AI815

– AO815

– AI845

168 3BSE041434-600 H

6 Fieldbus 6.4 HART, WirelessHART

– AO845

– AI895

– AO895

• S900 I/O cards:

– AI930

– AI931

– AO930

These I/O cards allow mapping of the 4 "Secondary HART variables" intothe process data. The maximum number of Secondary HART variables thatcan be communicated by one S900 I/O station is application-dependent (mixof I/O cards) as the total amount of I/O data cannot exceed 216 bytes.

• Supported HART Multiplexers:

(requires HART Multiplexer Connect Option of Device Management HART)

• Pepperl+Fuchs: KFD2-HMM-16

• MTL: MTL4840

• Elcon: Series 2700-F

• Elcon: Series 2700-G

For extending the serial ports of Multiplexers into an Ethernet-based backboneDigi PortServer TS MEI port-extender is supported.

6.4.2 HART Device Management PerformanceFollowing limits and performance figures need to be taken into account when usingthe HART Device Type Managers (DTMs) within Device Management HART:

• Acyclic communication requests from HART DTMs going to the same I/O stationare handled sequentially. that is a new request is queued and only sent when theprevious request has been answered from the field device.

Accordingly, increasing the number of DTMs working on the same I/O stationin parallel or out of the same system node in parallel will slow down the overallresponsiveness. Therefore, following limits shall not be exceeded during parallelwork:

– Maximum 6 DTMs accessing a single I/O station simultaneously.

3BSE041434-600 H 169


– Total 6 DTMs opened simultaneously on an 800xA node for accessing singleor multiple I/O stations.

• Amount of data and communication performance is heavily depending on theDTMs and device's capabilities. Thus the overall communication performancebetween DTM and device may vary depending on device type.

6.4.3 HART Connect ServiceThe HART Connect Service is combined with the Connect Service for PROFIBUS.

Following limits and performance figures need to be taken into account when usingthe PROFIBUS/HART Connect Package of Device Management PROFIBUS/HART:

Table 6.11: PROFIBUS/HART Parameters


1Max number of PROFIBUS/HART ConnectServices per Connectivity Server

PROFIBUS/HART ConnectivityServer acyclic communication

2,500Max number of HART devices perPROFIBUS/HART Connect Service

For constraints regarding the deployment of the PROFIBUS/HART Connect Servicetogether with other Connect Services on the same Connectivity Server, refer to Chapter3.6.


• PROFIBUS/HART Connectivity Server is designed for providing data to assetmanagement applications. It does not support fast update rates as they are expectedfor changing process values of field-devices. Applications requiring fast updaterates need to rely on cyclic communication.

Only Asset Optimization Server can be a client to PROFIBUS/HART ConnectService.

• Acyclic communication requests through PROFIBUS/HART Connect Serviceare handled sequentially. I.e. a new request is only sent when the previous requesthas been answered from the field device.

• By default, the PROFIBUS/HART Connect Service is configured on the ControlNetwork level. If the Control Network contains more HART devices thansupported by one instance of the PROFIBUS/HART Connect Service, the Service

170 3BSE041434-600 H


can be configured on lower hierarchy levels, e.g. Controllers. However, a separatePROFIBUS/HART Connect Service instance must be configured on all parallelhierarchy levels (e.g. Controllers) that HART data shall be fetched from.

Note that the number of Connectivity Servers may need to be increasedaccordingly as only one PROFIBUS/HART Connect Service is supported perConnectivity Server.

• If PROFIBUS devices are configured for acyclic communication in the samehierarchy like the HART devices, both HART and PROFIBUS devices arehandled in the same Connect Service instance. In that case the sum of HARTand PROFIBUS devices must not exceed 2500 per Connect Service instance.

6.4.4 WirelessHARTData from WirelessHART devices is made available to the system by followinginterface:

• Pepperl+Fuchs: WHA-GW-F2D2-0-A*-Z2-ETH

Process data from up to 50 WirelessHART devices is accessible in the controllerthrough a Gateway WHA-GW-F2D2 connected to a Modbus TCP CommunicationInterface CI867. A pre-configured Function Block Library is available for easyintegration of the data into controller applications.

The Device Management PROFIBUS/HART and PROFIBUS/HART Connect Packagecan be deployed on WirelessHART devices as well.

For assessing opportunities for deployment of WirelessHART in specific applications,please contact your local ABB representative.

6.5 FOUNDATION Fieldbus

6.5.1 Network Connection through Communication InterfaceCommunication Interface for FOUNDATION Fieldbus HSE: CI860

For connecting the Controller to FOUNDATION Fieldbus with CI860 followinglimits and performance figures need to be taken into consideration:

3BSE041434-600 H 171

6 Fieldbus 6.5 FOUNDATION Fieldbus

Table 6.12: FOUNDATION Fieldbus CI860 Parameters


General Limits

12 (a redundant pair consumestwo of them)

Maximum no. of CI860 in anAC 800M (CEX Bus)

1(Note: multiple CI’s can beconnected to the same subnet)

Maximum number of HSE Subnetsper CI860

HSE (High SpeedEthernet)


1,000 signals overall out of:Maximum number ofpublish/subscriber signals per CI860

High Speed Ethernet(HSE) cycliccommunication

- Max. 1,000 for analog input- Max. 500 for analog output- Max. 500 for discrete input- Max. 250 for discrete output

300Maximum number of client/serversignals per CI860

High Speed Ethernet(HSE) acycliccommunication (acyclic access to contained FF block

parameters)

150Maximum number of devicessupported for client/server access perCI860

Performance Limit

For performance limits of client/server communication through CI860please refer to section "CI860 Average FF Load Calculation" below.

High Speed EthernetCommunication

CI860 Average FF Load Calculation

To ensure a proper functionality under all conditions the CPU load of the CI860shall not exceed 80% at a maximum. This gives the limit of the Average FF loadof 100% that can be operated by the CI860 during runtime.

Average FF Load = CPU Load * 1.25

The Average FF load is calculated and monitored by Fieldbus Builder FF dependingon the actual configuration.

172 3BSE041434-600 H


The CPU Load is given as Summery of the CPU Load for Publish/Subscribe (P/S)communication, the CPU Load for Client/Server (C/S) communication, and 9% idleload.

CPU Load = CPU_Load_P/S + CPU_Load_C/S + 9% Idle Load

Calculating CPU Load for P/S

CPU_Load_P/S = T*0.105% + N*0.015%

N: Number of configured channels on CI860

T: Number of transfers/sec (publish and subscribe)

Use the following formula for calculating T if number of cycles are used:

T = N1/C1 + N2/C2+.....+ Nn/Cn

N: Number of configured channels on CI860

C: Used Signal Cycle time in [sec] for these channels

The formula accuracy is about +-5%.

Calculating CPU Load for C/S

CPU_Load_C/S = R/C*1.6%

R: Number of client/server requests

C: Cycle time of C/S requests in [sec]

Use the following formula for calculating CPU_Load_C/S if number of cyclesare used:

CPU_Load_C/S = (R1/C1+R2C2+.....Rn/Cn)*1.6%

The formula accuracy is about +-20%.

Verify the CI860 CPU load within Control Builder M, CI860 Hardware ConfigurationEditor in the Connections tab, diagnostic channel IW1.6502 CPU load and IW1.6505Average FF load over a period of time when C/S requests were performed.

3BSE041434-600 H 173


Table 6.13: Average FF Load Calculation Examples

Average FF Load Calculation Examples

60%CPU_Load_P/S500 P/S signals with 1 sec cycletime

+11%CPU_Load_C/S7 C/S signals with 1 sec cycletime

+9%Idle Load

80%CPU Load

(Average FF Load = 100%)

36%CPU_Load_P/S300 P/S signals with 1 sec cycletime

+35%CPU_Load_C/S22 C/S signals with 1 sec cycletime

+9%Idle Load

80%CPU Load

(Average FF Load = 100%)

Further features and constraints to be considered during application design with CI860:

• The CI860 communication interface unit cannot be used in an AC 800M HighIntegrity controller.

• The following FF data types can be communicated:

– Publisher/Subscriber communication:

DS65 mapped to RealIO data type

DS66 can be mapped to either BoolIO or DwordIO data type

– Client/Server communication via CI860:

FFBitStrLen16DS14

FFBitStrLen8DS14

FFDiscreteSTatusDS66

FFFloatDS8

174 3BSE041434-600 H


FFFloatStatusDS65

6.5.2 FOUNDATION Fieldbus Network and Connect ServiceFor the FOUNDATION Fieldbus HSE and H1 network including Linking DeviceLD800HSE and LD800 HSE Ex following limits and performance figures need tobe taken into consideration:

Table 6.14: FF Network Parameters



30Maximum number of LinkingDevices per HSE Subnet

HSE

1,000Maximum number of H1 devicesper HSE Subnet

HSE/H1

4Maximum number of H1 Links perLinking Device

16Maximum number of devices perH1 Link

100Maximum no. ofpublish/subscriber signals in totalbetween one H1 Link and HSE

Performance Limits

40 signals/secMaximum no. ofpublish/subscriber signals/sec perH1 link between H1 and HSE

1sTypical Macro Cycle Time

FOUNDATION Fieldbus recommendations for H1 segment layout is maximum12 devices with up to 4 valves. Refer to the System Engineering Guidelines AG-181issued by the Fieldbus Foundation.

The cycle time on FF H1 depends on the number of device, and the devices' timingparameters and is therefore application-specific. The Macro Cycle Time can befreely configured to adapt to project needs.

3BSE041434-600 H 175


Following limits and performance figures need to be taken into account when usingthe FF Connect Package of Device Management FF:

Table 6.15: FF Connect Service Parameters



1HSE Subnets per FF ConnectService

FF Client/ServerCommunication through FFConnectivity Server 4Maximum number of FF

Connect Services perConnectivity Server

1,000Maximum number of FF H1devices per Connect Service

Performance Limit

50% of total Macrocycle timeRecommended Macrocycle timereserved for acycliccommunication

For constraints regarding the deployment of the FF Connect Service together withother Connect Services on the same Connectivity Server, refer to Connect ServiceCalculation Rules and Limitations.

6.6 IEC 61850

6.6.1 Network Connection through Communication InterfaceCommunication Interface for IEC61850-ED.1: CI868.

CI868 General Configuration Data

CI868 is modeled as an IED with one Access point and supports communicating withother IEDs through IEC 61850 GOOSE as well as IEC 61850 MMS client protocol.

The CI868 IEC 61850 Hardware Library 3.x available in Control Builder, supportsCI868 to communicate over IEC 61850 GOOSE and MMS client protocols.

176 3BSE041434-600 H

6 Fieldbus 6.6 IEC 61850

Figure 6.1 illustrates the data modeling in IEC61850 to explain the terms used in thefollowing sections to define the performance and capacity limits with IEC61850 inSystem 800xA:

Figure 6.1: Logical Node Data and Communication Model

For IEC61850 communication with CI868 following limits and performance figuresneed to be taken into account independent of what communication protocol shouldbe used:

3BSE041434-600 H 177


Table 6.16: IEC 61850 CI868 General Parameters


General Limits

12Maximum no. of CI868 in an AC 800M(CEX Bus) Note: Only application

redundancy supported.

10Maximum number of LDs under MyIED

253Maximum number of LNs per LD

10Maximum number of Data Objects perDataset Analog, Integer or Boolean

type ofData objects.Increasing the number ofData Objects per Datasetwould overload CI868.For more data signals toCI868 from same IED, useadditional Datasets


IEC 61850 connectivity through CI868 supports application redundancy. Thatmeans a second CI868 module needs to be configured separately to communicatethe same data with the voting implemented in the controller application.

The CPU load on CI868 shall be less than 85% for satisfactory performance ofCI868 module with all types of communication.

CI868 Performance for GOOSE Protocol Usage

In addition to the General Configuration Data for CI868 following limits andperformance figures need to be taken into account when using the CI868 for GOOSEprotocol communication only:

The limits with respect to the IEC61850 data model entities need to be consideredalready while engineering the Substation Configuration Description (SCD) filewith the substation engineering tool.

178 3BSE041434-600 H


Table 6.17: Performance Parameters of CI868 with GOOSE Protocol only



80Maximum number of IEDs connectedper CI868

GOOSE communication

150Maximum number of GOOSEDatasets per CI868 Subscribed dataset count

checked by the IEC 61850Wizard during engineering

254Maximum Number of CI868 ReceiveGroup per IED Each CI868 GOOSE Receive

Group constitutes 5Data Objects.Each Data Object constitutes theDataAttributes, for example: stVal /mag.f , q, t

800Maximum number of DataData Objects of types Analog,Integer, or Boolean.

Objects (signals) configured

Performance Limits

160 / secMaximum number of changingGOOSE communicationData Objects of types Analog,Integer, or Boolean.

Data Objects Received

See also CI868 GOOSEPerformance Graph

10 / secMaximum number of changingData Objects of types Analog,Integer, or Boolean.

Data Objects Sent

See also CI868 GOOSEPerformance Graph

3BSE041434-600 H 179


CI868 GOOSE Performance Graph

The GOOSE performance of CI868 module is a function of the following parameters:

CI868 Receive

• Number of Static signals configured in scd-file for CI868 Receive.This is the sum of all Data Objects (DO) signals subscribed to CI868 in allDatasets in all IEDs.

• Number of changing signals out of static signals for CI868 Receive.This is the sum of all Data Objects signals subscribed to CI868 changing at agiven moment across all IEDs.

CI868 Send

• Number of Static Signals configured in scd-file from CI868 to other IEDs.This is the sum of all Data Objects signals in all Datasets from CI868 to otherIEDs.

• Number of changing send signals out of static signals from CI868 to other IEDs.This is the sum of all Data Objects signals changing at a given moment fromCI868.

Figure 6.2 provides the CI868 GOOSE performance for different configurations ofStatic and Changing Receive signals for operating within optimal load of CI868 CPU(85%).

This chart is applicable for CI868 configured for GOOSE communication.

180 3BSE041434-600 H


Figure 6.2: CI865 GOOSE Performance Chart

CI868 Performance for MMS Client Protocol Usage

In addition to the General Configuration Data for CI868 following limits andperformance figures need to be taken into account when using the CI868 for MMSprotocol communication only:

The limits with respect to the IEC61850 data model entities need to be consideredalready while engineering the Substation Configuration Description (SCD) filewith the substation engineering tool.

3BSE041434-600 H 181


Table 6.18: Performance Parameters of CI868 with MMS Client only




MMS communication

150Maximum number of MMS DatasetsSubscribed dataset countchecked by the IEC 61850Wizard during engineering

50Maximum number of LDs under otherIEDs Starting from Position 200 under

other IED inControl Builder hardware tree

1,000Maximum number of Data Objects(signals) configured Data Objects of types Analog,

Integer, or Boolean.

Performance Limits

80 / secMaximum number of changingMMS communicationData Objects of types Analog,Integer, or Boolean.

Data Objects Received

1 / secMaximum number of MMS ControlCommand sent MMS Control Commands sent

from CI868 viaCSWI and XCBR LNs


• High load on CI868 with MMS signals can be avoided by grouping the frequentlychanging signals in the same dataset when creating the SCD file.

For Example: Measurement signals shall be grouped into one dataset, Status signalsshall be grouped into another dataset.

MMS performance of CI868 module functions are as follows:

182 3BSE041434-600 H


CI868 Receive

• Memory: Number of Static Datasets (sent via RCB (Record Control Block)signals configured in SCD file for CI868 Receive.This is the sum of all Data Objects signals subscribed to CI868 in all Datasetsin all IEDs.

• Load: Number of changing DO signals for CI868 Receive.This is the sum of all DO signals subscribed to CI868 changing at a given momentacross all IEDs. The number of MMS signals changing in lesser number ofDatasets consumes less CI868 load as against the same number of signalschanging in more number of datasets.

CI868 Performance for MMS Client and GOOSE Protocol Combined Usage

Following limits and performance figures that are different from the figures given inthe respective sections for GOOSE protocol and MMS protocol need to be taken intoaccount in addition when using the CI868 for parallel GOOSE and MMS protocolcommunication:

Table 6.19: Performance Parameters of CI868 with both GOOSE and MMS Client


Configuration Limit


MMS & GOOSEcommunication

150Maximum number ofSubscribed dataset countchecked by the IEC 61850 Wizardduring engineering

Datasets per CI868 (GOOSE & MMSin total)

3BSE041434-600 H 183


Table 6.19: Performance Parameters of CI868 with both GOOSE and MMS Client(Continued)


50Maximum number of LDs underMMS & GOOSEcommunication(Continued)

Starting from Position 200 underother IED in Control Builderhardware tree

other IEDs

254Maximum Number of CI868 ReceiveGroup per IED Each CI868 GOOSE Receive

Group constitutes 5 Data Objects.Each Data Object constitutes theData Attributes, for example: stVal/ mag.f , q, t

20 Send directionMaximum number of GOOSE DataObjects (signals) configured 20 Receive direction Data Objects

of types Analog, Integer, orBoolean.

600Maximum number of MMS DataObjects (signals) configured Data Objects of types Analog,


Performance Limit

60 / secMaximum number of changing MMSData Objects Received

MMS & GOOSEcommunication Data Objects of types Analog,

Integer, or Boolean

10 / secMaximum number of changingMMS & GOOSEcommunication(Continued)

Data Objects of types Analog,Integer, or Boolean.

Data Objects Sent via GOOSE

20 / secMaximum number of changing DataObjects Received via GOOSE Data Objects of types Analog,


1 / secMaximum number of MMS ControlCommand sent MMS Control Commands sent

from CI868 via CSWI and XCBRLNs

184 3BSE041434-600 H


6.6.2 IEC61850 ConnectFollowing limits and performance figures need to be taken into account when usingthe IEC61850 Connect Package:

Table 6.20: IEC61850 Connect Service Performance Limits



2, 4 if no othet node functions onthe server

IEC 61850 Connect Serviceinstances per Connectivity Server

MMS communicationthrough IEC61850Connectivity Server 16Maximum Number of Subnetworks

per IEC 61850 Connect Serviceinstance

2 (both instances running on thesame Connectivity Server Node)

Maximum Number of IEC 61850Connect Service instances perSubnetwork

80Maximum Number of IEDs perIEC61850 Connect Service instance

80Maximum Number of Data Objects8 Nos. Datasets each containingper IED10 DOs.(1 DO= stVal + q + t).

Performance Limits

3BSE041434-600 H 185


Table 6.20: IEC61850 Connect Service Performance Limits(Continued)


960 / secMaximum Data Objects Change rateMMS communicationthrough IEC61850Connectivity Server

Assumptions:15% of Data Objects arechanging at any given time.80 IEDs x avg. 80 DO per IED x0.15 changing/sec = 960changing DO out of 6400 DOs perConnect Service instance.

5 / secContinuous event flow per IEC61850 Connect Service instance

150 / 100 msecEvent burst capacity per IEC 61850Connect Service instance

100 msecMMS command pass through timefrom 800xA > IEC 61850 ConnectService > IED

For constraints regarding the deployment of the IEC61850 Connect Service togetherwith other Connect Services on the same Connectivity Server, refer to Connect ServiceCalculation Rules and Limitations.

6.7 EtherNet/IP and DeviceNetFor EtherNetIP / DeviceNet configurations with CI873 the following dimensioningguidelines needs to be taken into account.

6.7.1 GeneralThe limitations, with respect to the various devices in general are:

• CI873 can act only as a scanner.It does not accept class 1 and class 3 connections from any other scanner.

• The number of I/O modules that can be connected under Ethernet/IP or DeviceNetdevice adaptor type device is 63.

186 3BSE041434-600 H

6 Fieldbus 6.7 EtherNet/IP and DeviceNet

• The number of configuration parameters supported per EtherNet/IP or DeviceNetdevice is 1000.

• The CI873 supports Listen only connection with EtherNet/IP device, providedthere is already Exclusive owner connection in the device. The CI873 does notsupport Redundant owner connections for EtherNet/IP devices.

• The Read only parameter and monitoring parameters in EDS file are not supportedin this release.

• The tag based Class 1 information should be there in EDS file for communicationwith Allen Bradley PLC where Class 3 tag can be added along with Class 1connection.

• The total number of Input and Output bytes along with channel status bytesshould not exceed more than 80Kb per CI873.

• The Configuration assembly size of 200 is supported per EtherNet/IP or DeviceNetdevice.

• The CI873 supports 20 CIP connections (including Class 1 and Class 3) perEtherNet/IP device. CI873 supports total of 128 connections.

• The CI873 only supports devices which uses EtherNet/IP encapsulation of CIP.

• CI873 does not support PCCC, Modbus encapsulation.

• CI873 supports CH1 Ethernet interface with a speed of 100 Mbps.CH2 is not supported.

• A maximum of 6 non redundant CI873 can be connected to each AC 800Mcontroller.

Performance Data

Typical performance of the CI873 is:

• CI873 can handle a maximum of 10 CIP connections with 10ms RPI.However it can handle a maximum of 128 CIP connections.

• The reaction time of CI873, that is, the time from changed input channel to thetime setting an output channel is less than 100ms at a maximum CI873 CPULoad of 80%, provided the data is sent over a connection operating at an RPI of50ms or less.

3BSE041434-600 H 187


• Data throughput of 1000 CIP I/O packets receive/second and 500 CIP I/O Packetssent/second can be achieved at an optimum load of 85%.Each I/O packet can have data size ranging from 4 to 500 bytes.

• Redundancy Switchover time is 120ms for 10 CIP connections operating at 10msRPI, that is, the time I/O communication stops in primary to the time I/Ocommunication starts in switched primary.

6.7.2 EtherNet/IPThe limitations, with respect to the EtherNet/IP device involved are:

• EtherNet/IP supports three Class 1 connection and three Class 3 tag per AllenBradley Control Logix PLC. The CI873 supports three Class 3 tags with 100mscycle time.

• The data transfer, using the Class 3 connection is slower than the Class 1connection.

• The Class 3 connection is not supported for any EtherNet/IP devices except AllenBradley Control Logix PLC. The CI873 uses tag based Class 3 to write data toit.

• The maximum number of bytes support for Class 1 read tag is 496 and for Class3 write tag is 432.

• 1000 bytes per Class 1 connection is supported,for example O->T: 500 and T->O : 500.

6.7.3 DeviceNetThe limitations, with respect to the LD 800DN linking device (for DeviceNet) are:

• The maximum number of input bytes supported by LD 800DN is 496 bytes. Ifthe total number of input bytes of all DeviceNet slaves configured under thelinking device exceeds 496 bytes, download is stopped.

• The maximum number of output bytes supported by LD 800DN is 492 bytes. Ifthe total number of output bytes of all DeviceNet slaves configured under thelinking device exceeds 500 bytes, download is stopped.

• A maximum of four LD 800DN linking devices can be connected under oneCI873.

• Multiple CI873 cannot listen to same LD 800DN data.

188 3BSE041434-600 H


• The maximum number of DeviceNet connections per device is restricted to 5.

3BSE041434-600 H 189


190 3BSE041434-600 H

7 Applications

This section describes application parameters for Asset Optimization, InformationManagement, Batch Management, and PNSM.

7.1 Application Parameters SummaryTable 7.1 and Table 7.3 details the system configuration for application specificparameters.

Table 7.1: Values

ValuesParameter (Maximum Numbers)

5,000Asset Monitors per AO Service function (PC-0389)

30,000Asset Conditions per AO Service function (PC-0514)

Table 7.2: Connectivity Server History Capabilities 4


10,000 (by default). Max50,000

Event Storage, number of events per event catagory (PC-0515)

3 monthsRetention Period, recommended

10,000Number of data Sources per connectivity server, with redundancyand Parallel enabled (default)

5,000Number of data Sources per connectivity server, with redundancyand Parallel disabled 5

10,000Number of History Logs in a connectivity server,AC 800 Connect, single disk, parallel redundancy enabled, (PC-0509)

4. Refer to the specific tables throughout this section for information on nodes where IM and/or Batch is combinedwith Aspect- and Connectivity services

5. OPC DA items or corresponding for Heritage controller connects (e.g. TTD logs, etc.)

3BSE041434-600 H 191

7 Applications 7.1 Application Parameters Summary

Table 7.2: Connectivity Server History Capabilities 4(Continued)


5,000Number of History Logs in a connectivity serverHeritage Connects, single or dedicated disk, parallel redundancydisabled (PC-0xxx)

1,000Average change rate per second (OPC DA) for connectivity serverlogs

500Average storage rate per second (OPC DA) for connectivity serverlog data written to disk

2,000Burst Change rate (OPC DA) for connectivity server logs

1,000Average number of Trend Requests per minute (OPC DA)

4000Average Trend Retrieval rate per second for historical data (OPCDA)

8,000Trend Retrieval burst rate per second for historical data (OPC HDA)

Table 7.3: History Server Node Parameters


Information Management History

30,000Number of Primary History Logs in an Information Manager Server

2,000Data Transfer rate Values/sec from Connectivity Server History toInformation Manager History Logs (opcHDA)

1,000Average number of trend requests (opcHDA) per min. from anInformation Management Server

4,000Average number of values (opcHDA) Values/sec. for a trend requestfrom Information Management log

8,000Burst Trend Requests (opcHDA) Values/sec. for data from andInformation Management log

30Average Number of Events transfer from Event Server to InformationManager Message Log

4. Refer to the specific tables throughout this section for information on nodes where IM and/or Batch is combinedwith Aspect- and Connectivity services

192 3BSE041434-600 H


Table 7.3: History Server Node Parameters(Continued)


1,000Recommended Maximum History Transfer Rates to InformationManager server from Connectivity (opcHDA) Rate per min.(Staggering can be used to balance and improve performance)

800xA History

12, or 6 PairsNumber of Data Collectors

2(High availability)

Number of History Servers

Note that the disk arrangement is essential when high capacity and performance isrequired. See section CPU, Memory, and Disk Requirements for more details andreferences

Table 7.4: Application Services Parameters


Scheduling Service

200Simultaneous jobs, maximum, if used for reporting, capacity tocontain reports is based on available disk capacity

Calculations

10Number of Services

10Write transactions/sec to AC 800M

100Write transactions/sec to Softpoint objects

2Write transactions/sec to Lab Data Logs

Softpoint Service

2,500Number of Objects

25,000Number of Signals, maximum

3BSE041434-600 H 193


7.2 Scheduling ServiceThe scheduler service can be installed on any number of aspect, application orconnectivity servers as each scheduler functions independently. There is a limit tothe number of simultaneous operations which can occur on each scheduling service.Refer to Table 7.3.

7.3 Softpoint ServiceThe Softpoint Server can have up to 2,500 SoftPoint objects. Each SoftPoint objectcan have up to 100 signals; however, the total number of signals cannot exceed 25,000.Softpoint Server redundancy is also supported. CPU time for each read or writetransaction is one millisecond. The Softpoint server can write 10 events per secondto platform-based Aspect Objects.

7.4 CalculationsTable 7.5 provides Calculations parameters.

Table 7.5: Calculations Parameters

DescriptionParameter

Rate at which input variables are updated by theirrespective OPC data sources.

OPC Base Rate

Range: 100 msecs. to 1 hour.Default: 1,000 msecs. (1 sec.).

Rate at which the Calculations Scheduler scans thelist of cyclically scheduled calculations.Range: 100 msecs. to 1 hour.Default: 500 msecs. (1/2 sec).

Cycle Base Rate

1,000 enabled calculations per Calculation Server.Number of calculations that may be queuedwaiting to be executed

100 calculations/sec., refer to write transaction ratesspecified below to determine capabilities.

Execution Rate

194 3BSE041434-600 H

7 Applications 7.2 Scheduling Service

Table 7.5: Calculations Parameters(Continued)

DescriptionParameter

Refer to Table 7.3 on page 192.Write transactions/sec.

Write transaction/sec. to SoftPoints

Write transaction/sec. to Lab Data Logs

7.5 Information Management

7.5.1 History ServicesThe Information Management History Services (Inform IT Service Provider for oneInformation Management node) can manage up to 50,000 history logs (combinationof primary and secondary).

Table 7.6: Tag Values

ValuesTags

180,000IM History logs per system (PC-0382)

30,000IM history logs per IM server function (PC-0503)

In a system, the maximum number of history logs is according to Table 7.6 on page195. The maximum number of history logs supported is dependent on the configuredsystem including connectivity options, cycle times, and collection rates. Refer to [9],[10], [11], and [12] in Table 2.1 for more detailed information.

7.5.2 Information Manager Server CapacityTable 7.7: Information Management Server Capacity

ApplicationServers SeparateFeature

30,000 1Maximum number of Primary History logs per server

120,000Maximum number of History data points logged per min in I

3BSE041434-600 H 195

7 Applications 7.5 Information Management

Table 7.7: Information Management Server Capacity(Continued)

ApplicationServers SeparateFeature

40Maximum number of alarms/messages logged/sec per IMserver

2,500Maximum number of SoftPoints objects 2

10,000Maximum number of History data points collected from oneconnectivity server

Refer to Table 7.3on page 192

Maximum number of calculations and execution rate

NOTES:Plus additional 20,000 secondary logs (calculated).1.Refer to Softpoint Service and Calculations for additional details regarding Softpoints and Calculations.2.

Licensing of Information Management logs is based on the initial (primary) logfor each signal that is stored in the Information Management server. Secondarylogs are not counted as part of the license count.

7.5.3 Disk and Memory Capacity RequirementsRequirements are shown in Table 7.8, and Table 7.9.

Table 7.8: System Limits and Capabilities - Information Management Server

CapacityCapability

4 Gbytes disk for Batch ManagementPDL Option - Additional disk capacity

256 Mbytes for large applicationsPDL Option - Additional memory

2 Gbytes additional diskAudit Trail Option - Additional Disk capacity

NOTE:Estimated PDL space required in bytes = ab(c(JOB) + 484):

a = number of days to keep PDL data before deleting.b = average number of jobs per day.c = average number of tasks and subtasks per JOB where JOB = (366 + 366d + 366e + 366f).d = average number of resource transactions per task (which transaction causes this).e = average number of history associations per task.f = average number of task variables per task. 366 and 484 are the number of bytes required in various tables per entity

of that type. Estimated PDL message log size in bytes = 3776 * PDL message log capacity.

196 3BSE041434-600 H


Table 7.9: System Sizing Examples - Information Management

RAM(Gbytes)

DiskCapacity(Gbytes)

Number of Disks(SCSI Recommended)System Requirements 1, 2

2152Base - Includes:1,000 TYPE5 History logswith 1 min storage rate,90 day period,one OPC message log w/1,000,000message capacity, small PDL application

3503Up to 10,000 TYPE5 Logs

42004Up to 50,000 TYPE5 Logs

0.54—Large PDL Application

—2—Audit Trail

4.52084Large System - 50,000 logs and all options

NOTES:TYPE5 logs are file based logs with a variable size.1.The maximum space for flat files that IM can consume on a physical or virtual disk is 2 TB no matter how large thedisk is.

2.

7.5.4 Computing PDL Data Disk SpaceWhere:

a = number of days to keep PDL data before deleting.

b = average number of jobs per day.

c = average number of records (tasks) and sub-tasks per job.

A record is created for a job (campaign) another record is created for a batch. Everytime a phase executes a new record is created. This also includes sub-procedures.

d = average number of resource transactions per task.

Resource transactions are created as a result of following BMAs actions, Acquire,Reserve, Select, Release, Unreserved, Deselect. Each time one of those blocks executea new resource transaction is sent to PDL. The database level Acquire is alwaysmatched with Release, Reserve with Unreserved and Select with Deselect. As a resultthere is a single record in PDL for each pair Acquire-Release pair.

3BSE041434-600 H 197


e = average number of history associations per task.

Every time the Data Collection BMA with Start option is used a new historyassociation is created.

Data Collection BMA blocks with the Once option does not create an association.An association links numeric trend data with the production data record.

f = average number of task variables per task.

There are approximately 12 task variables that get created every time a batch startsand ends. Each completed batch will have at least 12 variables associated with thebatch. Every time a bdbput() function is used in an expression, a task variable wouldbe created. All procedure and phase parameters are also recorded as task variables.

g = PDL message log capacity (number of messages), the PDL message log will storeall events where the batch has acquired the equipment, basically a batch id is includedin the message.

Estimated PDL space required in bytes = ab(c(366 + 366d + 366e + 366f) + 484) 366and 484 are the number of bytes required in various tables per entity of that type.

Estimated PDL message log size in bytes = 3776*g. 3776 is approximately the numberof bytes required per message.

Message sizes can vary.

7.5.5 Maximum Number of Entries Per Log (Nominal)Log capacity and number of logs determine table space and disk requirements. Thelimits indicated in Table 7.10 should be sufficient for most applications.

Table 7.10: Entries Per Log

EntriesLog Type

5,000,000TYPE1 - File-based 24-byte



315,000,000 (10 years of data at 1-sec sample rate)TYPE5 - File-based variable size

50,000Oracle-based

12,000,000 per logOPC Message Logs

198 3BSE041434-600 H


Table 7.10: Entries Per Log(Continued)

EntriesLog Type

Limit based on available restored table spaceRestored Logs

5,000 reportsReport Log

7.5.6 History Collection Maximum Sample RateHistory Services can collect from different types of data sources, and limitations onsample rate are imposed based on the type of data source. These limitations aredescribed in Table 7.11.

Table 7.11: Maximum Sample Rates

Rate 1Collection Type

350 - 400 values/secNumeric data supplied manuallyand by data clients

30,000 logs at 120,000 samples/minIM Primary History Log

50,000 logs at 60,000 samples/minConsolidation Node (via TCP/IP)

NOTE:Actual performance is based on the platform where History software runs, and control system topology. Blocking rateand stagger function must be used to balance CPU loading.

1.

7.5.7 Fastest Sample Rate and Timestamp ResolutionTYPE 1, 3, and 5 and Oracle-based logs support microsecond timestamp resolutionand collection at a one-second sample rate. TYPE2 logs can also collect at aone-second rate; however, they support one-second resolution on timestamps.

Some connectivity servers may not support (be able to provide data at) theone-second sample rate. Refer to the applicable documentation for more detailedinformation.

7.5.8 Disk Requirements Per Log EntryDisk space requirements for each History value stored are indicated in Table 7.12.As an example, Disk Space for TYPE1 Log with one minute storage interval, and

3BSE041434-600 H 199


90 day log period would be as follows:

1,440 samples/day * 90 days = 129,600 samples129,600 samples * 24 bytes/sample = 3.1 megabytes per log

Table 7.12: Bytes Required Per Log Entry

Bytes Per Entry 1Log Storage Type

24TYPE1 file-based

4TYPE2 file-based

8TYPE3 file-based

16-32 based on data type (float, double, etc.)TYPE5 file-based

110Oracle-based Numeric

250DCS Message in a message log

3,076OPC Message in a message log(IMMSGLOG/PDLMSGLOG) 2

100Restored log entry

NOTE:It is recommended to leave 5% to 20% extra capacity for expansion.1.OPC message storage is based on the alarm/event generation for a system. Systems typically have an average eventgeneration with occasional bursts of events. The average alarms per second (or hour or day) should be used tocalculate the events per archive interval. The 3076 message size is also the average. Each system generates differenttypes of events and the size can vary with messages being smaller and larger than 3076. Using 3076 as a startingpoint should allow for adequate initial calculations for archive usage.

2.

7.5.9 History Objects Miscellaneous CapacitiesTable 7.13 describes size limitations for composite logs, log sets, and archive groups.

Table 7.13: History Object Miscellaneous Capacities

SizeObject

16Maximum number of Primary and Secondary Logs in a Composite Log

10,000Maximum number of Logs in a Log Set

10,000Maximum number of Logs in an Archive Group

200 3BSE041434-600 H


7.5.10 History ArchiveIM supports several methods to move archive information off the IM system. Theseoptions should be evaluated based on the archive data generated by the configuration.

Disk Archiving

Consider the following for disk archiving:

• Disk archiving requires local disk space to store archive data. Options exist tocreate ISO images, volume backups, or both.

• ISO images can be recorded to DVD or CDs and other ROM media types.

• The volume backups are copied to a local or remote destination. From thislocation, the information can be backed up with any appropriate method to transferthe data to the backup medium.

Cost Considerations for Permanent Storage of Archive Data

The amount of archive data generated will result in a certain cost for backing uparchive data. However, the use of CD/DVD requires a separate step to burn the ISOimage to CD/DVD. If a remote file server is used and the local IT department backsup the data, the information can be backed up with your organizations IT backupprocess. These options should be evaluated based on the configuration requirements,the data collected, convenience, and the cost per archive.

7.5.11 Display Services MDI (Multiscreen Display Interface)Table 7.14: Multiscreen Display Interface

Application Servers SeparateFeature

150Total MDI clients in a system

64Maximum number of graphic display clients (MDI/SDI) perBasic Historian server

150Total Excel Data Access clients in a system

64Maximum number of Excel Data Access clients per BasicHistorian server

200/sec nominalSupported data rate feeding Multiscreen Display Interface1

3BSE041434-600 H 201


Table 7.14: Multiscreen Display Interface(Continued)


Functions:5,460 elements/function call Dialogs:

Maximum size of request returned by Excel Data Access

History - 500 history values/logSQL - 65,536 records

1 value/secMaximum rate for writing to process objects

5 values/secMaximum rate for writing to history logs

150Maximum number of desktop trends

64Maximum number per History server

NOTE:Using LastHistoryValue option will minimize the load while providing near real time data1.

7.5.12 Desktop TrendsTable 7.15: Desktop Trends


150Maximum number of Desktop Trend clients in a system

64Maximum number of Desktop Trend clients per Basic HistorianServer

8 tracesMaximum number of trends on a trend display

50 tagsMaximum number of tags per Tag Explorer Group

50 tagsMaximum number of tags per Ticker file

4Maximum number of concurrent trend controls on one client

15 secs or lessTrend display call-up from local Information Management serverwith eight trends

202 3BSE041434-600 H


7.5.13 ODBC ClientTable 7.16: ODBC Client

Characteristic orValue

Feature

60Maximum number of ODBC-Client Connections in a system

10Maximum number of ODBC-Client Connections per ODBC Historical DataServer

7.5.14 Open Data Access (ODA)- Excel Data Access (EDA)Each ODA server supports up to 20 concurrent client connections. The maximumnumber of values for one query against History is unlimited. However, large queriesmay cause the client application (Crystal Reports or Microsoft Excel) to time out.

7.6 800xA History800xA History comprises of following main node functions.

• 800xA History Server

• 800xA History Embedded Data Collector (DCN)

• 800xA History Archive Server

An 800xA History server can be configured as either a single or redundant Historian,the history is responsible for storage of data from a single System 800xA Installation.An 800xA History server is designed to be installed as part of the control systemoperating environment and is located inside of the firewall of the control room.

7.6.1 800xA History ServerThe capacity of the History server is measured by number of signals stored. A signalis any numeric (Boolean, Integer or Real) data stored in the History Server. A signalcontains attributes like value, time, quality, name, descriptions, range limits,engineering unit etc. A signal can be a raw data from OPC DA source found in System800xA, an aggregate value or result of a calculation.

Table 7.17 provides the capabilities of the 800xA History Server.

3BSE041434-600 H 203

7 Applications 7.6 800xA History

Table 7.17: 800xA History Capacities

CapacityCapability

150,000Maximum Number of History Logs per system

20,000Maximum Number of History points logged per second

400*Maximum Number of Alarm/Messages logged per second

12 (6 Pair)Maximum Number of DCN per History Server

1Maximum Number of 800xA System connected per History Server

NOTES:Events per Second Sustained for 10 Seconds.*.

The maximum number of history logs supported is dependent on the configuredsystem including connectivity options, cycle times, and collection rates.

7.6.2 Embedded Data Collector Node (DCN)The data collector collects the data from System 800xA data sources and forwards itto the History Server. The Embedded Data Collectors, send both events and currentvalue data immediately to the history server, but also provide the capability to bufferthe data for a period of at least 7 days and are utilized as a backup data source fortrending the unlikely event that the network connection or main history servers arenot available.

Table 7.18 provides the capability of the features.

Table 7.18: Capabilities of the Features

CapabilityFeature

18,000Maximum number of History Logs per Embedded Data Collector

5,000Maximum number of History points logged per second

7.6.3 800xA History Archive Server800xA History Archive Server allows the users to archive the data from 800xAHistory. Refer to History Archive section under Information Management.

204 3BSE041434-600 H

7 Applications 7.6 800xA History

800xA History Archive node requires a dedicated Basic History provider. Thismeans the function cannot be installed on nodes with redundant Basic Historyproviders, typically nodes with connectivity node functions.

7.6.4 800xA History Hardware SizingHardware considerations are very important while planning and designing 800xAHistory. The performance of 800xA History depends on the hardware platform. Thehardware specifications are described for different size of history system i.e. smallsize history (< 5k Signals), medium size history (5k-10k Signals), large size history(10k-30k Signals) and enterprise size history (> 30k Signals).

For more information on Hardware sizing considerations for 800xA History, refer to[19] in Table 2.1.

7.6.5 Data Access from 800xA History800xA Operator clients, Smart Client and OPC UA clients can be used to access theOPC DA, UA and HDA data from 800xA History for trending, reporting and analysispurpose. Table 7.19 provides maximum number of OPC DA, UA and HDAsubscriptions.

Table 7.19: Maximum Number of OPC DA, UA and HDA Subscriptions

CapabilityFeature

20,000Concurrent unique OPC DA data points accessed from History Server

10,000Concurrent unique OPC UA data points accessed from History Server

1,000Concurrent unique OPC HDA data points accessed from History Server

Total maximum concurrent client count should not be more than 50.

7.7 Batch ManagementThe following tables provide the tested capacities of a number of Batch Managementfeatures. If the deployed application will exceed one or more of these feature capacities,contact ABB Product Management for further consultation.

3BSE041434-600 H 205

7 Applications 7.7 Batch Management

Using the virtualization host software with 800xA Batch Management server nodesaffects the capacity levels through which the software can be supported. Refer tothe related notes for each of the following tables.

The following table presents the capacities for Batch Management features that arecontroller independent. All times and rates are typical and dependent upon theconnected system configuration and system load.

Table 7.20 details the capacities of Batch Management features.

Table 7.20: Batch Management Capacities (Controller Independent)

SpecificationSystem Level Features

5001Maximum Batch Equipment configured of type Pseudo

500(pseudo) + Maximum batchequipment for specific controller type1,2

Total batch equipment defined in system.Includes real and pseudo units

3+ blocks per second cumulative recipeexecution rate

Procedure Execution (procedure block transitions, alllevels)

Maximum 500 control recipe procedurecompletion events per day (24 Hour)1

Control Recipe procedure execution

401Batch Clients with active batch overview displays

NOTES:Value is reduced by 20% when virtualization host software is used for the 800xA Batch Management server.1.Refer Table 7.21.2.

Table 7.21 shows the Batch Management feature capacities that are dependent uponthe controller deployed.

Table 7.21: Controller Dependent Capacities

FunctionPhase DriverHarmonyDCIMelodyMOD 300AC 800 MFeature

5001,550200150015001,55001,5Maximum BatchEquipment percontroller type

10001,410020016001,410001,410001,4Active Phasesper control type

206 3BSE041434-600 H


Table 7.21: Controller Dependent Capacities(Continued)

FunctionPhase DriverHarmonyDCIMelodyMOD 300AC 800 MFeature

1001501001100110011001Total ActiveBatch Phasesper BatchRecipes

1000/5min1100/5min200/5min1600/5min11000/5min11000/5min1ControllerPhaseTransitions

NA2,350/5min3200/5min1,3200/5min1,3200/5min1,3200/5min1,3Get Batch/DBAFunction

NA2,350/5min3200/5min1,3200/5min1,3200/5min1,3200/5min1,3Put Batch /DBAFunction

NOTE:Value is reduced by 20% when virtualization host software is used for the 800xA Batch Management server. Formaximum performance and capacities, physical server nodes must be used.

1.

The determining factor is the connection to the controller type, OPC update rates, and the use of multiple BatchManager processes.

2.

The subscribe and putm expressions should be used in place of the Get and Put expressions to get this type ofperformance.

3.

This takes into account that a maximum of 500 active phases per connectivity redundant server pair.4.This takes into account that a maximum of 250 equipment are defined per connectivity redundant server pair.5.

7.7.1 Use of Function Phase DriverFunction Phase Driver is a user configurable application which maps batch states,commands and parameters between the batch server and user defined OPC pointsrepresenting the interface to an equipment phase in a PLC or other process controller.This feature is quite useful in process applications that include one or more processequipment skids that are supplied and installed with equipment manufacturer suppliedcontrollers.

The Function Phase Driver is configured in the same manner as other batch expressionsand functions utilizing the same batch variables and commands available in theexpression editor and function wizard. Although an equipment phase programmedusing the function phase driver is represented in the procedure function chart (PFC)as a standard phase type, the recipe manager processes and executes the commandsin a manner similar to that of transitions and other programmed batch expressions.

3BSE041434-600 H 207


7.7.2 Function Phase Driver - Performance ConsiderationsIn many cases, the function phase driver programming limits the ability of the recipemanager to process the starting of multiple phases in parallel. Users need to considerperformance of the recipe and whether parallel processing is going to be applied.Multiple phase starts are processed serially, one after the other. This is different fromthe behavior when specifying the start of two or more phases (using a phase driverof type AC 800M, MOD 300, DCI, or Harmony) in a parallel branch, which willprocess multiple phase start commands in parallel.

Actual performance and time duration for the processing of the phase commandsfrom the batch server to the controller is dependent upon:

• The quantity of OPC points being processed for each phase.

• The configured function phase driver logic for the sequencing and processing ofthe commands and OPC points.

• The configured OPC update rate between the connectivity server and controller.

Users should be aware that there are multiple expressions for handling OPC readsand writes. Proper use of subscribe expression in combination with a get can reduceoverall time to do OPC reads. Use putm instead of using put where multiple OPCwrites are required. Both actions will increase the performance of the function phasedriver.

The Function Phase driver will benefit if multiple recipe manager processes areenabled. This will help balance the load on the batch manager and increase throughput.

7.7.3 Server and Client System SizingThere are several factors to consider when sizing server and client hardware to supportyour batch system and applications. Often users are wondering whether they cancombine applications on a physical host server (physical host) and should they usevirtual host software (VM host). A physical server can support multiple applicationsif the specified server size in Table 7.22 is met for the Batch Management.

7.7.4 Use of a Physical Host Server800xA Batch Management is heavily dependent on physical resources (RAM, CPU,Network, disk space). The use of these resources needs to be taken into account whendetermining the size and type of system. Table 7.22 and Table 7.23 provides

208 3BSE041434-600 H


information for a minimum, and heavily loaded batch system. These are onlyrecommendations.

Minimum system is based on 50% of the capacities listed in Table 7.20 and Table7.21. Specifically Procedure Execution, Batch Equipment, and Controller PhaseTransitions. Heavily Loaded systems would be any system exceeding 50% of thesenumbers. These numbers take into account that these systems would be combiningtwo applications (i.e. Batch Management and Information Management) within thesame server.

Table 7.22: Server System Size

Heavily LoadedMinimumServer

64 bit, 6 core64 bit, 4 coreCPU

32 gigabyte16 gigabyteMemory

150GB150GBDisk Space, see note

1 gigabit1 gigabitNetwork

Note on disk space; Specified space to cover the batch server. The total disk shouldcover all functions running within the server. A good practice is to add another 30%for future updates. RAID 1 is generally faster, but RAID 5 and 6 offer more headroomon the disk with a very minimal speed difference. SSD drives can also be used andare the fastest.

Table 7.23: Client System Size

Heavily LoadedMinimumClient

64 bit, 4 core64 bit, 4 coreCPU

8 gigabyte8 gigabyteMemory

150GB150GBDisk Space, see note

1 gigabit1 gigabitNetwork

Note on disk space; Specified space to cover the batch client. The total disk shouldcover all functions running within the client node. A good practice is to add another30% for future updates. SSD drives are preferred or fastest available.

The performance of any batch system is dependent on the application. Complexityof recipes can play a big role in the performance of the server. Generally ISA 88

3BSE041434-600 H 209


structured recipes perform best. The more levels a recipe has the more complex itgets. Consider items like client counts and the use of dynamic labels. Large activebatch systems require more resources.

For more details on Batch server dimensioning, refer to [29] in Table 2.1.

7.7.5 Use of Virtualization Host SoftwareVirtualization host software (VM hosts) provides a mechanism by which systemresources are shared within the same physical server. Many applications thrive in thistype of environment with little concern of resources, but Batch Management is heavilydependent on the availability of resources (RAM, CPU, network, disk space). If aVM host is to be used, users should take into account the amount of activity withinin the VM environment and ensure that resources are adequate.

The following items should be taken into account when trying to determine whetherVM host is right for the application:

• HowmanyVMswill be hosted by the server? - More VMs means less resourcesto go around. You can increase disk, memory, CPUs and speed up networks, butyou are still using a common pipeline to access these resources, so consider trafficas well as resources on the server.

• How large is your application? - Large complex recipe procedures with severallevels require more resources to support them. Evaluate your recipe structures.Generally good ISA 88 procedure structures perform best.

• How many clients will be active? - Evaluate the active batch clients and theiruse of batch dialogs. Anything over 20 clients is considered a lot of activity.

• How much memory do I need? - The important thing to remember aboutmemory is that you do not want to get into situations where the batch manageris swapping memory resources. The batch manager generally needs certainamount of memory for best performance, see Table 3.21 on page 67. Considerwhat other VMs are being hosted on the physical hardware and the memoryresources they will need. Add these values together and you have your memorysize.

If any of these items are raising concerns a physical server (See system hardwarerecommendations) should be considered. See specifications as listed above.

Table 7.24 details AC 800M RAM usage for the Batch Advanced Templates library.

210 3BSE041434-600 H


Table 7.24: AC 800M RAM Usage for BatchAdvTemplatesLib

Instance(Kbytes)

Type 1

(Kbytes)Module

7.573Unit 2

0.7—Unit Attribute String 3

0.5—Unit Attribute Float 3

0.5—Unit Attribute Integer 3

1278SEM with Server Communication Module 2

0.7—SEM Attribute String 3

0.5—SEM Attribute Float 3

0.5—SEM Attribute Integer 3

8.580SEM Server Ext Comm Module

10.577SEM Client Comm Module

7.554Phase 22

0.7—Phase Parameter String 2

0.5—Phase Parameter Float 2

0.5—Phase Parameter Integer 2

0.1—AE Named Value Item Numeric

0.17—AE Named Value Item String

NOTES:Memory for type is allocated by the first existing instance in the controller, and is reused by all subsequent instanceswithin the same controller.

1.

These estimates are based on modules as delivered in BatchAdvTemplatesLib templates (no user logic).2.The estimates for attribute and parameter modules assume that all ranges and description are configured to themaximum size.

3.

7.8 PC, Network and Software Monitoring (PNSM)The recommended maximum capacity for a single PC, Network and SoftwareMonitoring Connectivity Server is dependent on what node the server is running and

3BSE041434-600 H 211

7 Applications 7.8 PC, Network and Software Monitoring (PNS

the power of the computer running the PNSM Server. Use a dedicated, powerfulserver, such as one with a dual multi core CPU, for larger configurations.

The working recommendations for computers on the System 800xA hardwarecertification list are:

• On its own connectivity server the limit is 150 PNSM assets with no more than50 being SNMP devices.

Refer to [8] in Table 2.1 for more information on SNMP devices.

• On a connectivity server being shared with another product the recommendedlimit is 50 assets with no more than 10 being SNMP devices.

Distribution of the PNSM Server on several nodes is another alternative with severaloptions:

• Spread the PNSM Assets over multiple PNSM Server nodes (multiple PNSMConnectivity Server). This is accomplished by spreading the PNSM Assets insubdirectories of the IT Control structure and then creating and configuringadditional OPC Data Source Definition for each main level of the subdirectories.

• Use Assets from the PNSM Device Library when possible.

7.9 Asset OptimizationAsset Monitoring Engines provide the execution environment for Asset Monitors.The startup time for Asset Monitoring Engines is dependent on the size andconfiguration of System 800xA. Factors affecting startup time include the number ofAsset Monitor conditions and the number of Aspect Server nodes.

The Asset Monitor execution time is determined by the type of Asset Monitor. Asimple Asset Monitor such as Bool Check, Running Time Check, or a field deviceAsset Monitor executes faster than a complex Asset Monitor such as CLAM orHXAM. An Asset Monitor with an active condition and an Auto Fault Report executesslower than the similar Asset Monitor without an Auto Fault Report.

An Asset Monitor detects a change in input data at a rate defined by the OPC GroupUpdate rate. Asset conditions are evaluated at every execution cycle.

For maximum, refer to Table 7.3.

212 3BSE041434-600 H

7 Applications 7.9 Asset Optimization

7.9.1 Control Loop Asset Monitoring (CLAM)System 800xA supports up to 500 control loops for monitoring. CLAM license isscalable from 100 to 500 control loops. The unlicensed version of CLAM only reportsthe overall health of the loop and the final control element, but does not providedetailed diagnostic information.

The licensed version of CLAM provides detailed diagnostic information about thefinal control element and loop performance as listed in the following.

• Final control element diagnostics include the following:

– FCE Action.

– FCE Leakage.

– FCE Size.

– FCE Stiction/Backlash.

– Loop Nonlinearity.

• Loop performance diagnostics include the following:

– Loop Tuning.

– SetPoint Oscillation.

– External Disturbances.

– Data Reliability.

– Harris Index.

– Setpoint Crossing Index.

– Oscillation Index.

– Controller Output Saturation.

– Manual Mode.

– Cascade Tracking.

– Response Speed.

3BSE041434-600 H 213

7 Applications 7.9 Asset Optimization

7.10 800xA Public Address System800xA Public Address System (PAS) comprises of the following main node functions:

• PAS Connectivity.

• PAS Announcement.

7.10.1 PAS ConnectivityTypically there will be one PAS Connectivity Server per 800xA System. PASConnectivity can be a dedicated Application Server or can be combined with anyApplication Server or Connectivity Server in the System. PAS Connectivity is monitorsthe Alarm Lists and adds the announcement messages to the announcement queue.

7.10.2 PAS AnnouncementThe PAS Announcement node can be combined with any 800xA node or can also bea non 800xA Windows server / workstation node. The PAS Announcement picks themessages from announcement queue, perform text to speech conversion and do theactual announcement through the output channel.

For small setups where there is only one PAS Announcement node, optionally thePAS Connectivity and PAS Announcement can combined to a same node.

Table 7.25: PAS Characteristics

SpecificationsCharacteristics

1Maximum number of PAS Connectivity Servers per System

20Maximum number of PAS Announcement nodes per System.This is also equal to the maximum number of output channels allowed.

100Maximum number of pending announcements per System

Typically less than1 sec

Time from an alarm condition active in system until begin of announcement

1,000Maximum number of Alarm list subscriptions (via PASConfiguration Aspect)per System.

10Maximum number of Output Languages per System

214 3BSE041434-600 H

7 Applications 7.10 800xA Public Address System

8 Heritage Control Systems

This topic describes technical data and configuration information for 800xAConnectivity to ABB’s Heritage Control Systems.

This section includes the details for the following Heritage Control Systems:

• 800xA for AC 100

• 800xA for Advant Master with 800xA for Safeguard

• 800xA for DCI (System Six)

• 800xA for MOD 300

• 800xA for Symphony Plus Harmony (includes Symphony Plus)

• 800xA for Melody

• 800xA for Freelance

8.1 800xA for AC 100The connectivity package 800xA for AC 100 supports integration of AdvantController 100 Series controllers via Advant Fieldbus 100 to System 800xA and tothe Function Chart Builder as part of Control Builder A.

800xA for AC 100 does not support connection to Advant Controller 400 Seriescontrollers.

800xA for AC 100 consists of software including AC 100 OPC Server and requiresCI527A for physical connection to the Advant Fieldbus 100 hardware interface.

The 800xA for AC 100 consists of two software parts

• 800xA for AC 100 and

• AC 100 OPC Server

The AC 100 OPC Server runs in the Connectivity Server and requires the AF 100hardware interface CI527A installed in the same node as the AC 100 OPC Server.

3BSE041434-600 H 215

8 Heritage Control Systems 8.1 800xA for AC 100

This software product utilizes the client/server technique, making it possible to runboth client and server on one PC for a small configuration. For larger configurations,one or several servers can be used with several client workplaces.

8.1.1 Prerequisites and RestrictionsOne half length PCI slot is required in each AC 100 Connectivity Server for installationof a CI527A PCI board. If the server is not equipped with a PCI slot, an external PCIExpansion unit must be used.

8.1.2 Maximum System ConfigurationsTable 8.1 specifies maximum supported sizes related to the Connectivity Server rolewhen configured with the AC 100 Connect feature. Numbers in parentheses refer toredundant configuration.

Table 8.1: 800xA System limitations related to AC 100 Connect

ValueDescription

2 (4)Maximum number of Connectivity Servers

10Maximum number of Connectivity Servers per AF100network1

NOTE:1. This is a bus limitation; the maximum number of receivers to which AC 100, and AC 400, controllerscan send events.

Table 8.2 specifies the maximum configurations with AC 100 Connect.

Table 8.2: Connectivity Server limitations related to AC 100 Connect

ValueDescription

15Maximum number of AC100 controllers per ConnectivityServers

10,000Maximum number of tags per Connectivity Server1,2

5,000Maximum number of tags per Connectivity server if the serveralso has other node functions.1,2

NOTE:1. Depending on configuration, such as AF 100 bus load, see [21] in Table 2.1.2. Same limitation applies to both single and redundant pair.

216 3BSE041434-600 H


8.1.3 Technical Data

Supported Networks and Controllers Versions

Various configurations of the Advant Fieldbus 100 network and connections to theAdvant Fieldbus 100 network are supported by 800xA for AC 100.

• Multiple Advant Fieldbus 100 networks.

• Redundant Advant Fieldbus 100 networks.

• Multiple Connectivity Servers on one Advant Fieldbus 100 network.

• Redundant Connectivity Servers on Advant Fieldbus 100 networks.

Advant Controller 100 Series controllers with support for Advant Fieldbus 100 canbe connected to 800xA through the connectivity servers. These controllers types are:

• Advant Controller 70 version 1.0 or later.

• Advant Controller 110 version 2.0 or later.

• Advant Controller 160 version 1.0 or later

Supported Controller Versions

Table 8.3 shows supported Advant Fieldbus 100 options in different controllers. Note,AC 80 not covered.

Table 8.3: Supported options for different controllers

Controller/Controller version

Options AC 160/ ver1.0 or later

AC 110/ver. 2.1 orlater3


AC 70 ver1.0 orlater1

YesYesNoNoExtended DB elements

YesYesYesYesDAT based Objects

YesYesYesYesSDP

YesYesYesYesEvent handling

No(4)No(4)No(4)No4AF100/Optical Bus

3BSE041434-600 H 217


Table 8.3: Supported options for different controllers(Continued)

Controller/Controller version

Options AC 160/ ver1.0 or later



AC 70 ver1.0 orlater1

YesYesYesNo(4)AF100/Coaxial Bus

YesYesYesYesAF100/Twisted Pair Bus

1. Recommended version 1.2/1 or later.2. Recommended version 2.0/4 or later.3. Recommended version 2.1/4 or later.4. Modems to convert between bus types are available.

Specifications

Table 8.4 describes technical data valid for 800xA for AC 100.

Table 8.4: 800xA for AC 100 Technical Data

DataDescription

800xA for AC 100

AIS, AOS, DIS, DOS, DAT_AI, DAT_AO,DAT_DI, DAT_DO, DAT_DAT, MB, MBS, MI,MIL, MR

Predefined ABB process objects types

Process Graphics

From 1 second and upwards. Default updaterate: 3 seconds

Update rates for process graphics

Trend

1 hourTime range default

Advant Fieldbus 100 Capacity

The following topics show some practical limits. If the reason for a limit is a physicalone this fact is mentioned in the text.

218 3BSE041434-600 H


When the system is configured, always check the bus load on the AdvantFieldbus 100.

The capacity of the AC 100 OPC Server is limited by the CI board capacity. Thecapacity of the CI board is indicated in Table 8.5 below.

Table 8.5: Advant Fieldbus 100 Interface Capacity

DataDescription

3,950Total number of DataSet Peripherals

1, 2, 4,... 4,096 msCycle times depending on configuration in AC100 series

As each DataSet Peripheral can reference to up to eight DAT elements, the maximumnumber of DATs is about 32 000.

The Advant Fieldbus 100 runs a data rate of 1.5 MBit/sec. Some of the informationis, however, used for preambles, cyclic redundancy checks and protocol overhead;as a result, the net data rate is somewhat less.

The transfer times for DataSet Peripherals depend on the number of attachedDAT elements (see Table 8.6).

Table 8.6: DataSet Peripheral transfer time

Transfer time (ms)DSP size

0.0880.1080.1560.252

DAT element1

DAT elements2

DAT elements3-4

DAT elements5-8

The Advant Fieldbus 100 bus load has to be calculated when the system is configured.The bus load depends on the DataSet Peripheral configuration and can be calculatedusing the following formula:

This formula provides the bus load as a percentage, where the sum is taken over alldifferent sending DSPs and:

3BSE041434-600 H 219


=number of DSPs (of the same data size and cT)Nbr

=transfer time in ms (from Table 8.6)Ttr

=desired cycle time in ms (1, 2, 4, 8, 16…4096).cT

In order to guarantee that message transfer is possible, at least 25% of the busbandwidth is reserved for message transfer (Event Set and command signals).Up to 70% of the bandwidth may be used freely for DataSet Peripheralcommunication.

Bus Configuration Builder supports automatic calculation of the bus load.

Performance

The performance of the 800xA for AC 100/AC 100 OPC Server depends on the clientapplication program and the Windows system load as well as the load on AF100 andconnected controllers.

Command Response Time

The average for a synchronous OPC write from an OPC client to an object attributein a Controller is about 0.5 seconds. Synchronous means that the time is measuredfrom when the sending starts until the changed value is acknowledged back to theclient.

Event Burst Recovery Time

Typical event burst recovery times, i.e. until the last event was received on 800xABase System

• Burst of 1000 events: 0:55 min



220 3BSE041434-600 H


Data Throughput

The system performance in an open environment is dependent on many factors suchas: PC type, size of memory, other installed software in use, etc.

It is the sum of data subscribed from displays, alarm and event handling, historysubscriptions and the direct OPC access load, that results in the performance for theConnectivity Server. As all data has to pass through the AF100 hardware unit, certainrules need to be applied by the applications and clients that subscribe and write todata located in the controllers. To improve performance in configurations where highamounts of data need to be transferred to and from the controllers, an additionalConnectivity Server can be added to the configuration.

Following table show the maximum subscription throughput via OPC ClientConnection.

Table 8.7: Data Throughput AC 100 Connect

ValueDescription

Connectivity Server performance related to AC 100 Connect

500-5,000Maximum subscription throughput via OPC Client Connection(tags per second)1

10,000Maximum number of concurrent tags per Connectivity Server2,3

5,000Maximum number of concurrent tags per Connectivity serverif the server also has other node functions 2,3

10Continuous event throughput (per second) - Connectivityserver

NOTES:1. Depending on configuration, such as AF 100 bus load, see [21] in Table 2.1.2. Same limitation applies to both single and redundant pair.3. Refer to Table 3.6.

The event throughput depends on the AF100 bus load and the number of eventconsumers (AC 400 controllers or AC 100 OPC Servers) on the AF100 bus.

8.1.4 Configuration800xA for AC 100 is the preferred product to:

• Upgrade and expand current AdvaSoft 1.x and 2.x installations.

3BSE041434-600 H 221


• Connect new 800xA Operator Workplaces to Advant Controller 100 Seriescontrollers.

For general information about node functions and configurations, refer to NodeFunction Deployment.

800xA for AC 100 Connect Configuration

Figure 8.1 shows a workstation based system with separated Client and Server roles.In this example Control Builder A is located in separate node on the client/servernetwork, but it can also be integrated in the Connectivity Server node. This exampleinclude redundant Connectivity Servers and one media redundant Advant Fieldbus100 control network.

Figure 8.1: Configuration Example for 800xA for AC 100

222 3BSE041434-600 H


8.2 800xA for AdvantMaster with 800xA for SafeguardThis section contains technical data and configuration information for 800xA forAdvant Master and 800xA for Safeguard. 800xA for Safeguard is an add on systemextension to 800xA for Advant Master providing additional faceplates and graphicalelements. 800xA for Safeguard is therefore only mentioned explicitly when there aredifferences compared to 800xA for Advant Master.

The system extension 800xA for Advant Master consists of two parts:

1. The server functionality called "Advant Master Connect". This feature is assignedto the Connectivity Server.

2. Predefined Aspect Objects with aspects for process graphics, faceplates andtrends.

To communicate with the Master Bus 300 network, the PU410 Real time Accelerator(RTA) unit is required. This is a 19-inch extension unit directly connected to theserver. Connectivity servers assigned the feature Advant Master Connect can bevirtualized.

8.2.1 Prerequisites and RestrictionsThe following prerequisites and requirements are necessary to support the 800xA forAdvant Master software.

RTA Unit

From 800xA 6.0, the PU515A RTA Board is no longer supported. PU515Ahardware must therefore be upgraded to the PU410 RTA Unit.

The PU410 RTA unit is needed for integration of System 800xA with AC 400 Seriescontrollers via MB300 network. Each Connectivity Server can only be connected toone RTA unit. PU410 RTA unit is connected to the Connectivity Server via Ethernetand to the MasterBus 300 network using RJ-45 connectors.

The PU410 RTA Unit can be placed standalone beside a PC workstation (desktop ortower model) or mounted into a 19-inch cabinet. The software on the RTA unit isautomatically loaded when the 800xA for Advant Master software is started. Figure8.2 shows the PU410 RTA Unit.

3BSE041434-600 H 223

8 Heritage Control Systems 8.2 800xA for Advant Master with 800xA for S

Figure 8.2: PU410 RTA Unit

The PU410 connects over single or dual line via its PC 1 and 2 ports using shieldedRJ-45 connectors for 100 Mbit full duplex communication with the PC (ConnectivityServer). On the control network side, the PU410 connects over single or dual line viaits MB300 1 and 2 ports using shielded RJ-45 connectors for 10 Mbit half duplexcommunication with the MB 300 control network.

System redundancy is based on Connectivity Server redundancy. Only one RTA Unitcan be connected per physical Connectivity Server.

The PU410 RTA Unit supports point-to-point connection to the Connectivity Serverwhich means that the PU410 RTA Unit will require its own dedicated networkconnection and cannot be connected via an Ethernet switch or hub.

The PU410 RTA Unit also supports the Control Builder A engineering tool whenControl Builder A version 1.4/0 or later is used.

History Logs

TTD Logging.

TTD logs is the preferred way to log data in an Advant Master system, to maintain areasonable load throughout the control system. To achieve this, 800xA for AdvantMaster is optimizing the data handling, from the controller through the RTA unit andto the actual logging in System 800xA trends and in IM logs. Refer [9] in Table 2.1.

Direct (OPC) Logging. Direct logs will be coordinated in the connectivity server(when the same objects are subscribed) to avoid unnecessary load on the controllers.Direct logging of data results in a continuous load of the involved connectivityserver(s), the RTA unit(s) and controller(s). This is similar to having a displaycontinuously presenting the same amount of data points.

224 3BSE041434-600 H


Maximum System Configurations

Table 8.8 specifies maximum supported sizes related to the Connectivity Server rolewhen configured with the Advant Master Connect feature. Numbers in parenthesesrefer to redundant configuration.

Table 8.8: 800xA System limitations related to Advant Master Connect

ValueDescription

6 (12)Maximum number of Connectivity Servers

4 (8)Maximum number of Connectivity Servers per MasterBus 300 network

40Maximum number of controllers per MasterBus 300 network

Table 8.9 specifies Connectivity Server limitations related to Advant Master Connect.

Table 8.9: Connectivity Server limitations related to Advant Master Connect

ValueDescription

20Maximum number of controllers per MasterBus 300 network(per Connectivity Server)

15,000Maximum number of tags per Connectivity Server1,2

10,000Maximum number of tags per Connectivity server if the server also has othernode functions.1,2

NOTE:1. Depending of configuration, see [14] in Table 2.1.2. Same limitation applies to both single and redundant pair.

The Advant Master Connect can be combined with other node functions, but fordemanding application it is recommended to run Advant Master Connect in adedicated Connectivity Server where also the RTA unit is located.

Subscription Limitations

There are some controller limitations to consider when using 800xA for AdvantMaster.

Table 8.10 shows the maximum number of event subscribers and (1, 3, and 9-sec)data subscribers per controller node, type, and version.

3BSE041434-600 H 225


Table 8.10: 800xA for Advant Master Data and Event Subscription Limitations

Maximum EventSubscriber Nodes1

Maximum DataSubscriber Nodes1Controller VersionController Type

816Up to 4.0/1MasterPiece 200/1

16164.0/2 and laterMasterPiece 200/1

1616Up to 1.1/latestAdvant Controller 410,Advant Controller 450

32161.2/0 and laterAdvant Controller 410,Advant Controller 450

NOTE:One subscriber node could be, for example, one 800xA for Advant Master Connectivity Server, one Advant Station 500Operator Station, or one MasterView 800/1.

1.

8.2.2 Technical Data

Supported Networks and Controller Versions

Various configurations of the MasterBus 300 network and connections to theMasterBus 300 network are supported by 800xA for Advant Master.

• Multiple MasterBus 300 networks.

• Redundant MasterBus 300 networks.

• Multiple Connectivity Servers on one MasterBus 300 network.

• Redundant Connectivity Servers on MasterBus 300 networks.

Advant Master controllers with support for MasterBus 300 can be connected to 800xAthrough the connectivity servers. These controllers types are:

• Advant Controller 400 Series

• MasterPiece 200/1

• Safeguard 400/3000/9000 Series

Table 8.11 shows the controller versions which are included in the test configurationtogether with System 800xA. For each controller type a version of the latest versionis used.

226 3BSE041434-600 H


Advant Master controller products are maintained according to the published LifeCycle Policy.

Table 8.11: Controller Versions included in 800xA test configuration

Verified with System800xASoftware VersionProduct

VersionController Type

—XMP200 SW*2.0XMP200 SW*3.0

2.0MasterPiece 200/1

—XMP200 SW*4.02.1

—QMP220 SW* 5.03.0

yesQMP220 SW* 6.04.0

—QC01-BAS11*7.01.0Advant Controller 410

—QC01-BAS11*7.01.1

—QC01-BAS11*8.01.2/0

—QC01-BAS11*9.01.2/1-1.2/8

—QC01-BAS11*9.01.3

—QC01-BAS11*10.01.4

yesQC01-BAS11*11.01.5

—QC02-BAS21*7.01.0Advant Controller 450

—QC02-BAS21*7.01.1

—QC02-BAS21*8.01.2/0

—QC02-BAS21*9.01.2/1-1.2/8

—QC02-BAS21*9.01.3

—QC07-BAS41*1.02.0

—QC07-BAS41*2.02.1

—QC07-BAS41*3.02.2

yesQC07-BAS41*4.02.3

Table 8.12 describes the technical data valid for 800xA for Advant Master and 800xASafeguard.

3BSE041434-600 H 227


Table 8.12: 800xA for Advant Master Technical Data

DataDescription

Process Objects

AI, AO, DI, DO, DAT, TEXT, GENBIN, GENCON,GENUSD, GROUP, MANSTN, RATIOSTN, SEQ,MOTCON, VALVECON, PIDCON, PIDCONA,DRICONS, DRICONE, MOTCONI, and GRPALARM

Predefined process object types

FI and GI800xA for Safeguard Predefined processobject types

Process Graphics

From 1 sec and upwards.Three cyclic intervals are preconfigured for efficiency(1, 3, and 9 secs). Default update rate: 9 secs withevent driven update for binary data.

Update rates for process graphics

TTD (Time Tagged Data)

2 secsShortest TTD log sample interval

Direct (OPC) Logging

9 secs (1 and 3 secs possible) 1Recommended update rate when a processobject is used directly as a data source

OPC 800xA Client Connection

9 secs (1 and 3 secs possible) 1Recommended subscription update rates

It is only supported to perform OPC write operationstowards DAT objects. There are no hard limitationson the number of write operations that can beexecuted.

Write operations

NOTES:Not all properties are supported. Refer to [14] in Table 2.1 for more detailed information.1.

228 3BSE041434-600 H


8.2.3 Performance

Data Throughput

The system performance in an open environment is dependent on many factors suchas: PC type, size of memory, other installed software in use, etc.

It is the sum of data subscribed from displays, alarm and event handling, historysubscriptions and the direct OPC access load, that results in the performance for theConnectivity Server. As all data has to pass through the MB 300 RTA unit, certainrules need to be applied by the applications and clients that subscribe and write todata located in the controllers. To improve performance in configurations where highamounts of data need to be transferred to and from the controllers, additionalConnectivity Servers with MB 300 RTA units can be added to the configuration.

It is possible to divide the subscriptions using multiple Connectivity Servers on thesame MB 300 network.

Table 8.13 shows the maximum subscription throughput via OPC Client Connection.

Table 8.13: Data Throughput Advant Master Connect

ValueDescription

Connectivity Server performance related to Advant Master Connect

200-3,500Maximum subscription throughput via OPC Client Connection (tags persecond)

20,000Maximum number of concurrent data points per Connectivity Server1,2,3

15,000Maximum number of tags per Connectivity server if the server also has othernode functions.1,2,3

NOTE:1. Depending of configuration, see [14] in Table 2.1.2. Same limitation applies to both single and redundant pair.3. Refer also to Table 3.6.

Process Data Acquisition

The total RTA CPU load may not exceed 50 percent. For systems with ConnectivityServer redundancy, requiring fail-overs to be free of interference, a maximum RTAload of 40% is recommended. This limit is also recommended when only one of thetwo redundant Connectivity Servers is running. Table 8.14 briefly describes capacity

3BSE041434-600 H 229


examples per Connectivity Server (applies to both single and redundant pair). Theperformance depends on the total load influence on the RTA Unit primarily fromTTD, OPC logging, data for Process Displays and third party OPC clients, but alsofrom Alarms and Events and Status List.

Refer to [17] in Table 2.1 for more details.

Table 8.14: Capacity Examples per Connectivity Server 1

TTD (Time Tagged Data) 2,3

1,400 log variables on 2-sec interval7,000 log variables on 10-sec interval14,000 log variables on 20-sec interval

Examples, each resulting in 20% RTACPU load

Direct OPC Logging, OPC Client Connection, and Data for Process Graphics

9,000 process objects on 9-sec updaterate3,000 process objects on 3-sec updaterate

Examples, each resulting in 20% RTACPU load 4

NOTES:Applies to both single and redundant pair.1.The number of TTD logs must be decreased, if secondary IM logs exist for the TTD logs, where thesample blocking rate is faster than the 20-min. default.

2.

The data source for an Historian log can be either a process object itself, or a TTD log in an AdvantController 400 Series.

3.

For performance and signal error handling reasons, the recommendation is to use TTD logs in theAdvant Controller 400 Series as primary logs. These TTD logs are then used as data sources forlong-term storage in the Historian.Typical mix of data from different process objects.4.

Display Call-up Time

The following information is based on tests in a system using three redundantConnectivity Server pairs, one redundant aspect server (1oo3), one InformationManagement server, five clients, and about 30 controllers.

Call-up times may vary depending on network traffic, controller load, PC hardware,and overall system activity.

230 3BSE041434-600 H


Table 8.15: Display Call-up Times

Display Call-up TimeDisplays

< 2 secs 1Graphic Display with 40 objects (80 OPC items)

< 2.5 secs 1Graphic Display with 120 objects (240 OPC items)

< 3 secs 1Graphic Display with 192 objects (416 OPC items)

< 2 secsTrend Display with 8 traces (TTD or OPC logs)

NOTE:1. Graphic display references are cached after the first call up which makes subsequent display call upfaster. Each display in a system is cached after the first call up which means there is no limitation in thenumber of cached displays. The performance figure reflects a cached display.

8.2.4 ConfigurationSeveral types of system configurations are supported in 800xA for Advant Master.The below configurations are illustrated in this section:

• Single node production system combining all node functions (Aspect Server,Connectivity Server and Operator Workplace) into the same PC node (Figure8.3).

• System configuration combining Aspect and Connectivity Server roles (Figure8.4).

• Large system configuration with node functions for Operator Workplaces, AspectServers and Connectivity Servers separated into different nodes (single orredundant) (Figure 8.5).

• Engineering system configuration for Control Builder A and Online Builder withnode functions for Engineering Workplaces, Aspect Servers and ConnectivityServers separated into different nodes (single or redundant) (Figure 8.6).

For information about maximum supported system sizes, refer to System Size Limits.

This configuration requires a Control Builder A version on the Windows operatingsystem. For more information refer to separate Control Builder A Release Noteand Product Update.

800xA for Advant Master supports multisystem integration. Refer to MultisystemIntegration for more detailed information.

3BSE041434-600 H 231


800xA for Advant Master Configurations

For general information about node functions and configurations refer to NodeFunction Deployment.

SingleWorkstation Systemwith Advant Master Connect. Figure 8.3 show a smallsystem where all node functions have been combined into the same workstation.

Figure 8.3: Single Workstation System with Advant Master Connect

Workstation based systemwith Advant Master Connect. Figure 8.4 show a systemwith separated Client and Server roles, but where Aspect and Connectivity Serversare allocated to client nodes.

232 3BSE041434-600 H


Figure 8.4: Workstation based system with Advant Master Connect

Larger system with separate server nodes and Advant Master Connect. Figure8.5 shows an example of a system with a large MB 300 control network. In largenetworks (>20 controllers) multiple Connectivity Servers, single or redundant mustbe used. In this example the clients are connected via redundant Connectivity Servers.The maximum number of controllers in a large MB 300 control network can be foundin Table 8.8.

3BSE041434-600 H 233


Figure 8.5: Larger system with separate server nodes and Advant Master Connect

Engineering System with Control Builder A

Engineering installations can be made with a single node up to the maximum nodecount. Figure 8.6 shows an example of a large engineering system connected to theMB 300 control network through dedicated Connectivity Servers. The engineeringsystem include installation of Control Builder A 1.4, supporting same Workstationand Server Operating Systems as 800xA Version 6, along with the RTA unit PU410.

For more information refer to [20] in Table 2.1.

234 3BSE041434-600 H


Figure 8.6: Engineering System with Control Builder A

8.3 800xA for DCIThis section contains technical data and configuration information for 800xA forDCI.

The 800xA for DCI server application must not be combined with other control systemserver applications on the same Connectivity Server.

The client part of 800xA for DCI must be installed in all nodes except the DomainServer.

3BSE041434-600 H 235

8 Heritage Control Systems 8.3 800xA for DCI

8.3.1 Technical DataTable 8.16 provide general 800xA for DCI technical data.

Table 8.16: 800xA for DCI Technical Data


80Maximum number of Operator Workplace Clients

4 (single or redundant)Maximum number of Connectivity Servers per system

15,000Maximum number of DCI tags per Connectivity Server

60,000Maximum number of DCI tags per system

The US English version of the operating systems is required even if a translatedversion of System 800xA is used.

For more information on Windows operating system, refer to [7] in Table 2.1.

8.3.2 PerformancePerformance contains display exchange times and server switchover times.

Display Exchange Time

The 800xA for DCI server application collects data from controllers via GDBA, andmakes it available to the OPC clients. Table 8.17 shows the display exchange timeof a process display containing a certain amount and type of display elements.

Table 8.17: Display Exchange Time

Display Exchange TimeGraphic Display Elements

2 secs typical (cached)20 Tags/Graphic164 Total OPC Items subscribed



236 3BSE041434-600 H


Table 8.18 specifications were measured with the product running on a PC with aPentium IV, 2.4 gigahertz processor and 1 gigabyte of RAM. These values will varydepending on the Client Network traffic, number of tags, number of clients/nodes inSystem 800xA and overall activity.

Table 8.18: 800xA for DCI Specifications


25 events/secContinuous alarm throughput

Compatible with ECC, ECC MUX, and standardnon-redundant off-the-shelf Ethernet NICs. An ECCPboard or ECC MUX connection is highlyrecommended.

Connectivity Server node DCInetwork connection

Composer CTK version 6.0 and later. The CTK isrequired on at least one node in a system. It mustbe on a standalone node on the DCI control network.

Engineering Tools

Supported on Server and Client nodesMultiple Processor

Server Switchover Time

The 800xA for DCI server application nodes provide redundant connectivity to theDCI Control network. The connection requires two PC's each with a dedicated ECCP,ECC MUX, or non-redundant Ethernet NIC to the Control Network. Table 8.19specifications were measured with the products running on a PC with a Pentium IV,2.4 gigahertz processor, and 1 gigabyte of RAM and using ECCP card connections.These values will vary depending on the Control Network connection used, ClientNetwork traffic and the number of clients/nodes in System 800xA.

Table 8.19: Server Switchover Time

Redundancy Switchover Time

<5 secsECCP/ECC MUX communications failure

10-20 secs average800xA for DCI connectivity node failure

< 5 secsManual Switchover

10 secs average800xA for DCI Server application failure

3BSE041434-600 H 237


8.3.3 Recommended Hardware ConfigurationWorkstation requirements for 800xA for DCI are intended for new system purchasesor expansions to existing systems. Requirements are for all PCs used within an 800xAfor DCI system.

The ECC MUX interface is used in 800xA for DCI servers when redundant DCUcommunication networks are required. A server class machine running ServerOperating System can be used when a single communication network is being used.

8.3.4 Additional InformationThe following applies to all the 800xA for DCI applications:

• Refer to literature to determine CPU speed, RAM, hard disk capacity, etc. for800xA PC requirements.

• Other ABB applications such as Asset Optimization and Information Managementas well as third party applications may apply additional load on the workstationrequiring additional workstation resources (CPU speed, RAM, hard disk capacity,etc.).

• Conductor NT Server or Client can never be installed on the same workstationalong with 800xA for DCI software.

8.4 800xA for MOD 300This section contains technical data and configuration information for 800xA forMOD 300.

800xA for MOD 300 provides integration of 800xA and the Advant OCS/MOD 300control network (DCN or eDCN). 800xA for MOD 300 consists of two parts:

• One to be run on the workstation.

• One on the RTA unit. The RTA unit is installed on the node where theConnectivity Server is executing. The software on the RTA unit is automaticallydownloaded at start-up.

A RTA unit mounted in a Connectivity Server node provides the physical connectionto the DCN or eDCN and the following controllers:

• AC 460 Series.

• AC 410 Series.

238 3BSE041434-600 H

8 Heritage Control Systems 8.4 800xA for MOD 300

• MOD 300 Controller Subsystem (SC Controllers and Model B).

• MOD 300 Multibus.

8.4.1 Prerequisites and RestrictionsThe following prerequisites and requirements are necessary to support the 800xA forMOD 300 software.

RTA Unit Requirements

The RTA unit is required to connect the Connectivity Server to the Advant OCSControl Network (MOD 300 DCN/eDCN). If you require the RTA unit, you mustinstall the PU412K01 (RTA unit for DCN) or the PU410K02 (RTA unit for eDCN)before you install the PAS System Services software.

The PU410K02 and PU412K01 external Real Time Accelerator (RTA) units areavailable for eDCN and traditional DCN communications respectively. These newcommunication units are mounted remotely from the workstation and connected viastandard Ethernet. This removes the requirements and dependency of the hosthardware.

800xA for MOD 300 System Requirements

800xA for MOD 300 software requires that the Advant OCS with MOD 300 Systembe at the following levels:

• AdvaBuild 3.3/2 P2 or higher for Windows.

• MOD 300 System Version 14.6/x or higher software.

One computer with AdvaBuild for Windows is required by the MOD Importerapplication to populate the Aspect Server with MOD 300 tag data. A GENERICDobject must be configured in AdvaBuild to establish the MOD 300 ConnectivityServer (with RTA unit) as an Advant OCS node in the MOD 300 database.

800xA for MOD 300 Software Requirements

800xA for MOD 300 software requires that either Workstation or Server operatingsystem software and base System 800xA software be installed.


3BSE041434-600 H 239


800xA for MOD 300 client software can be loaded alone, on a Connectivity Server,or on top of an Application Server such as Information Management. The ConnectivityServer can be installed on Server Operating System only.

MOD PAS System Services is installed on the Connectivity Server only.

The Audit Trail option is required in order to record MOD 300 parameter changes toInformation Manager.

800xA for MOD Hardware Requirements

Refer to the computer manufacturers documentation to determine if your computeris working properly when powered up. Basic power up and power down instructionsfor your computer are described in the computer manufacturers documentation. Thecomputer must meet the minimum requirements for the 800xA product.

Multibus Restrictions

The 800xA for MOD 300 software requires that the MOD 300 System be at Version14.6/x software. It is not possible to use the MOD Tag Importer, without the propersystem version and AdvaBuild for Windows.

Performance Considerations when Migrating

It is possible to have both the 800xA for MOD 300 System and your existing MOD300 Multibus or Advant OCS UNIX System connected during the migration period.Users should be aware that they will be adding load on the DCN network and on thesystem controllers if they use combined systems. The MOD 300 Connectivity Serverwill typically add 25% load. Existing heavily loaded systems will have performanceand controller load issues. Users should plan accordingly and take into account theadditional load and minimize any operator or application nodes on the DCN ring.

In general users will see loads decrease as they move to System 800xA for MOD300. Communications to all HMI and application interfaces are done through theConnectivity Server(s). Subscriptions to common points are made once within theConnectivity Server, reducing load on the network.

240 3BSE041434-600 H


Functionality Not Supported in MOD 300

This release of the 800xA for MOD 300 software does not include support for thefollowing functionality.

• MOD 300 Overview display.

• Unit relativity.

• MOD 300 PHL.

• Use MOD 300 Engineering Displays to view TRIO Configuration and PLCConfiguration.

• Disk IO actions – open, close, remove, input, output, filesys.

8.4.2 Server Switchover Time Performance800xA for MOD 300 Connectivity Server nodes provide redundant connectivity tothe MOD 300 Control Network. The connection requires two workstations, eachwith a dedicated RTA Unit (Table 8.20).

Table 8.20: Server Switchover Time Performance

Redundancy Switchover Time

< 5 Sec 1800xA for MOD 300 Connectivity Server node Failure

< 5 Sec 1Manual Switchover

NOTE:1. These will vary based on Control Network loading, Controller loading, and Connectivity Server loading.

8.4.3 Product CapacityTable 8.21: Product Capacity

NumberDescription

Connectivity Servers

6Primary or Backup Connectivity Server Pairs

Tags

3BSE041434-600 H 241


Table 8.21: Product Capacity(Continued)

NumberDescription

6,000Maximum number of tags per Connectivity Server (applies to both single andredundant pair).Include following items when calculating the load on the connectivity server:• CCF continuous and device loops imported for display• TLL objects imported for display• Number of historical logs recorded

NumberAlarm

350Event Burst (events per sec) applied to one Connectivity Server redundant pair.

2,500/secThe maximum number of OPC subscriptions with a combined history/dataConnectivity Server(s) using a fixed rate OPC Server.

NOTE:For design and performance considerations, refer to [15] in Table 2.1.

8.5 800xA for Symphony Plus HarmonyThis section contains technical data and configuration information for 800xA forSymphony Plus Harmony. The Symphony Plus is added here to ensure that 800xAfully supports the Symphony Plus Control, I/O, and Engineering family of products.

800xA for Symphony Plus Harmony consists of two primary components:

• Client software consisting of the operator interface such as faceplates andHarmony diagnostic displays.

• Harmony Connectivity Server which collects data from the Harmony controllers.

242 3BSE041434-600 H

8 Heritage Control Systems 8.5 800xA for Symphony Plus Harmony

8.5.1 Technical DataTable 8.22 through Table 8.24 provide 800xA for Symphony Plus Harmony technicaldata to assist with the deployment solutions.

Table 8.22: 800xA for Symphony Plus Harmony Technical Data


80Maximum number of Operator Workplace Clients

8 (16)Maximum number of Connectivity Servers per system(redundant)

60,000Maximum number of tags per system.

30,000Maximum number of tags per Connectivity Server

30 events/secContinuous alarm throughput

Up to 5,000 events (distributedSOE)Up to 256 events (RIS recorder)1 msec resolution

Sequence of Events

Up to 16Alarm Priorities

15 default configured, no presetuser configurable limit

Engineering Unit Descriptors

15 default configured, no presetuser configurable limit

Logic State Descriptors

Maximum 80 characters permessage, no preset userconfigurable message quantity limit

Text Select Messages/Alarm Comment Strings

800xA for Symphony Plus Harmonymust not be combined with otherconnect servers on the sameConnectivity Server

Remarks

3BSE041434-600 H 243


Table 8.23: 800xA for Symphony Plus Harmony Characteristics


S+ Engineering 1.0 SP1 or laterEngineering Tools

Full support of IET800 and PNI800 interface (64-bit OS)Computer Interface(Ethernet)

Symphony Plus Controllers (HPC800, SPC 700)Harmony ControllersSupported Harmony Bridge Controllers (BRC100, 200, 300 and 400)

Multi-Function Processors (MFP01, 02, 03, 04, 12)Multi-Function Controller (MFC01, 02, 03)Harmony Area Controller (HAC)Analog Master Modules (AMM01, 02, 03)


Table 8.24: Supported Operating Systems

Operating SystemNode Type

Windows server operating systemHarmony Connectivity Servers

Windows client operating system800xA for Symphony Plus Harmony Clients (OperatorWorkplaces)

244 3BSE041434-600 H


8.5.2 Display Exchange Time PerformanceTable 8.25 shows the display exchange time of a process display containing a certainamount and type of display elements.

Table 8.25: Display Exchange Time

Display Exchange TimeGraphic Display Elements

1-2 secs typical (cached)20 Tags/Graphic160 Total OPC Items subscribed

40 Tags/Graphic360 Total OPC Items subscribed

2-3 secs typical (cached)80 Tags/Graphic720 Total OPC Items subscribed

These values will vary depending on the Client Network traffic, number of tags,number of clients/nodes in System 800xA, and overall activity. The tests were verifiedusing one client node loaded to each of the three levels, while ten other client nodeswhere running with level 1 graphic being displayed (20 tags/graphic).

The system includes three redundant pairs of Harmony Connectivity Servers.

Each pair of Connectivity Servers:

• Contained a minimum of 10,000 tags.

• Exception reporting on each pair of Connectivity Servers varied from 75-150XR/second to 1500-1700 XR/second.

• The Event rate was 45-50 events/second for 15-20 seconds.

8.5.3 Server Switchover Time PerformanceThe Harmony Connectivity Server nodes provide redundant connectivity to theHarmony Control network. The connection requires two workstations, each with adedicated INFINET To Ethernet (IET800) communications unit to the ControlNetwork. Table 8.26 shows the server switchover times.

Table 8.26: Server Switchover Time

Switchover TimeRedundancy Switchover

<5 secsIET800 communications failure

3BSE041434-600 H 245


Table 8.26: Server Switchover Time(Continued)

10-20 secs averageHarmony Server node failure

< 5 secsManual Switchover

10 secs averageHarmony Server failure

The values will vary depending on the Control Network used, Client Network traffic,and the number of clients/nodes in System 800xA.

8.5.4 Recommended Hardware ConfigurationTable 8.27 provides general workstation requirements for the 800xA for SymphonyPlus Harmony system.

Workstation requirements for Harmony applications are intended for new systempurchases or expansions to existing systems.

Table 8.27: General Workstation Requirements

RequirementsCharacteristics

Full support of IET800 and PNI800 interface (64-bit OS)Computer Interface(Ethernet)

Ethernet (IEEE ® 802.3) compliant (TCP/IP); best performance achievedwith 100 Mbytes/sec requiring 100BaseT

Network connector

Fixed IP address requiredEthernet addressing

8.5.5 Additional InformationThe following applies to all the 800xA for Symphony Plus Harmony applications:

• Other ABB applications such as Batch Management and Information Managementas well as third party applications may apply additional load on the workstationrequiring additional workstation resources (CPU speed, RAM, hard disk capacity,etc.). Refer to configuration rules for these products at other sections of thisdocument to determine what, if any, additional workstation resources are required.

• The S+ Engineering for Harmony Server should never be installed on any of the800xA nodes, including the Harmony Connectivity Server.

246 3BSE041434-600 H


8.6 800xA for MelodyThis section contains technical data and configuration information for 800xA forMelody.

8.6.1 Sizing DataTable 8.28 shows sizing data to be observed when planning a system:

Table 8.28: 800xA for Melody Sizing Data

RateConnections

≤ 12(Optionally redundant) Connectivity Servers per project

≤ 20PM 875, PM 876, PM 877, or CMC 70 per (optionally redundant)Connectivity Server

≤ 10,000Objects (tags) per Connectivity Server

≤ 90,000No. of Melody tags per Melody system

To determine on site the number of tags relevant for 800xA for AC 870P/Melodysizing, open the Connectivity Server Faceplate and check the number in the Tagsfield.

Sizing Data (Melody)

Table 8.29 shows sizing data that is based in Melody Control. For details refer to theMelody documentation:

Table 8.29: Melody Sizing Data (Continued)

ValueSizing Information

1 msecTime stamp resolution

16Available different Melody message priorities

≤ 4,000Module related message buffer size (raw messages)

≤ 22,500No. of configurable process messages of CCO 30

≤ 60,000No. of configurable process messages of PM 875, PM 876, PM 877,or CMC 70

3BSE041434-600 H 247

8 Heritage Control Systems 8.6 800xA for Melody

8.6.2 Performance800xA for Melody does not limit the System 800xA data transfer rate.

Hardware Prerequisites

Hardware requirements for Melody Connectivity Server or Melody ConfigurationServer are described in Server and Workstation Hardware.

The Melody Connectivity Server is a dedicated Server (running only with MelodyConnectivity Server components).

Supported Operating Systems


Table 8.30: 800xA for AC 870P/Melody Supported Operating Systems

Operating SystemServer

Windows server operating systemMelody Connectivity Server

Melody Configuration Server

Supported Composer Version

Precondition for running 800xA for Melody with S+ Control Melody is S+ Engineering1.0 SP1 (Composer 6.2 SP1).

8.7 800xA for FreelanceThis section contains technical data and configuration information for 800xA forFreelance.

800xA for Freelance is subject to a separate release. Contact ABB TechnicalSupport for more detailed information. Refer to System Updates for prerequisitesand requirements.

248 3BSE041434-600 H

8 Heritage Control Systems 8.7 800xA for Freelance

8.7.1 Sizing DetailsThe performance that can be expected regarding the OPC throughput and upload rateis determined by the Server hardware. The performance of data sources such ascontrollers, field devices, and etc. is important,

Table 8.31: Sizing Details

DetailsSubject

Max. 4 CS objects per800xA for Freelanceproject.

Number of 800F CS objects in a 800xA for Freelance project:• A CS object represents one OPC gateway, or two in case of

redundancy.• Four CS objects correspond to 4(8) OPC gateways.

Max. 3 OPC instancesmay run on a Server.

Number of Freelance OPC gateway instances running on a Connectivitynode:• Install the 800xA for Freelance OPC gateway software on

Connectivity nodes, only.• To ensure the required performance, exclusively use the 800xA

for Freelance OPC gateway software on that PC.

Typ. 10,000 OPC itemsper sec.

800xA tags per CS:• Provided the subscription rate is ≥1 sec.• 10,000 OPC items/s correspond to 2000 tags per sec.• ≥1 OPC gateway instance running on a CS requires sharing of

the specified tag per sec. count.• For ≥ 10,000 items per sec., add additional CSs.

Typ. 1 sec. per tag.Upload rate to the aspect directory.

NOTE:More than one OPC instance on a CS does not increase the specified OPC data throughput. The performance remainsconstant.

For more information on limitations with respect to Freelance System, refer to [5]and [27] in Table 2.1.

3BSE041434-600 H 249

8 Heritage Control Systems 8.7 800xA for Freelance

250 3BSE041434-600 H

9 Process Engineering Tool Integration

This topic contains technical data and configuration information for ProcessEngineering Tool Integration.

Process Engineering Tool Integration is installed on the same node as the 800xAPrimary Aspect Server and may be installed on an Engineering Workplace node withControl Builder M installed. The Web Services and the INtools/SmartPlant ®Instrumentation (SPI) database Import/Export utility are installed on the INtools/SPInode.

9.1 PerformanceTable 9.1 details the performance data of Process Engineering Tool Integrationfeatures.

Table 9.1: Process Engineering Tool Integration Performance Data

Characteristic/ValueFeature

1,000 objects in 45 mins1Create objects in 800xA

400 objects of a 1,000 object database inless than 3 mins1

Update properties of existing 800xA objects

5 secs typical 1SPI Powered by INtools Document call-up in 800xA

< 2 secs3800xA Faceplate call-up during Data Transfer

18 mins for 5,000 objects 1,2Load SPI Data into PETI - Web Service (Database)

10 secs for up to 1,000 objects1Load SPI Data into PETI - File

NOTES:All times and rates are typical and dependent on the connected system configuration and system load. Depends onthe size of the document being generated into PDF.

1.

The web service web site has to be configured to increase the default time-out period.2.Valid for Faceplates delivered with AC800M Connect and AC800M Connect HI.3.

3BSE041434-600 H 251

9 Process Engineering Tool Int 9.1 Performance

252 3BSE041434-600 H

10 Service and Support

10.1 Life Cycle ServicesThis topic contains an overview of the System 800xA life cycle policy. Refer to themanual [26] in Table 2.1 for complete policy description.

10.2 Life Cycle Management ModelThe life cycle management model divides a product's life cycle into four phases:Active, Classic, Limited, and Obsolete.

Each phase has different implications for the end user in terms of services and supportprovided.

All control systems effectively remain in Active phase, with individual componentproducts (hardware and software) transitioned through the lifecycle phases of Active,Classic, Limited, and Obsolete as they are superseded by new technologies andofferings.

10.2.1 Life Cycle Policy

ABBs Life Cycle commitment;

ABBs control systems are designed for continuous evolution. It is ABBs goal to protectour customers' intellectual investment (i.e. application software) beyond the life cyclesof the underlying platform products (i.e. hardware and software).

ABB will not "Remove from Active Sale" any product or "family" of products untilan equivalent replacement to those products is available. Once a product has beenremoved from active sale, ABB will continue to support the product for at least 10years, although exceptions to this may occur if components or technologies neededare no longer available to ABB.

Within this support period ABB will announce a "Last Buy" opportunity at least 12months prior to the end of manufacturing (except in cases where there is a direct

3BSE041434-600 H 253

10 Service and Support 10.1 Life Cycle Services

form, fit and function replacement). It is ABBs intention to provide support for aslong as there is significant customer needs after the Manufacturing End throughfieldservice, repair and by making replacement spares (new or refurbished modules)available.

Life Cycle Phases

The following is a summary overview of the life cycle phases for hardware andsoftware.

Active Phase. The article/version is the base for current sales. Price list is active,product is supported.

During the Active phase and with an active Automation Sentinel subscription at theMaintain Level or above, the software version will receive ABB’s normal productmaintenance including enhancements and corrections, testing of Microsoft securityupdates and third party software updates. The software is available for new systempurchases and expansions.

Classic Phase. Support is available. Use www.abb.com/partsonline to order systemexpansions.

During the Classic phase and with an active Automation Sentinel subscription at theMaintain Level or above, the software version will receive corrections of businesscritical issues, testing and support of Microsoft security updates, and maintenancesupport for the specific version of the operating system and other associated thirdparty software applications. The software is available for expansions of existingsystems.

Limited Phase. Support is possible on a best effort basis and might be subject to acharge. An active Automation Sentinel subscription at the Maintain Level or abovewill provide access to all available software updates, technical corrections, productdocumentation, etc. at myABB/My Control System. The software is no longer availablefor sale.

Obsolete. Support not longer available through the product organization. Servicemay still provide support on a best effort basis. An active Automation Sentinelsubscription at the Maintain Level or above will provide access to all available softwareupdates, technical corrections, product documentation, etc. at myABB/My ControlSystem.

254 3BSE041434-600 H

10 Service and Support 10.2 Life Cycle Management Model

PC Hardware Qualification Testing

The commercial life of a PC is normally less than one year. New PC models areverified against the currently released system version as soon as they becomecommercially available. New system versions are regression tested against the specifiednew, as well as older hardware. At some point a new system version will not besupported on older PC models. In order to provide replacement hardware newer PCsare verified against older system versions. However this support may be limited tocertain PC variants, and to one vendor only.

The PC hardware qualification testing is an exclusive service for the AutomationSentinel subscribers. A report documenting all the relevant findings is published onregular basis at myABB/My Control System.

System Software

System 800xA software versions are supported up to a period of 7 years from thedate of release. A valid Automation Sentinel subscription at the Maintain Level orabove, is required as a pre-condition to receive any control system and softwaresupport that is not safety related. A system software version is actively sold from avalid price list until it is replaced with a successor version. There is a six monthoverlap before the price list for the previous software version is withdrawn from theavailable price books. A system software version stays Active for at least a period of3 years.

Extensions of Existing Installation

Previous major system versions can be ordered for example at expansions of a plantwhere upgrade at the same time is not desired. Price lists and configurator (Wizards)for classic, or even limited, versions exist but are not generally available. A requesthas to be sent to the regional sales organization, which will quote the requested systemversion (limited versions only if feasible).

Special terms and conditions (no warranty) apply in these cases. Prices will increasewith the age of the system version.

Normally business critical errors are found early in the plant life cycle, but can showalso if for example the mode of operation changes. Major rebuilds and extensions ofplants may make latent errors to show up. Customers should be aware that in thosesituations an alternative is to upgrade the plant.

3BSE041434-600 H 255


Revisions

Revisions of a software version are provided in the form of system revisions. A validwarranty or a valid Automation Sentinel subscription at the Maintain Level or aboveis required as a pre-condition to receive any software revisions. System revisions arecharacterized by:

• Contains error corrections only, typically for several products with 800xA.

• No functional additions. Existing functions do not change.

• No new (higher) requirements on PC hardware.

• Applications built on previous revision are directly compatible after installationof the revision (binary compatibility).

• A revision is possible to install online, node by node, in a running system, withoutdisturbance to the controlled process.

Software upgrades from one System 800xA software version to the latest availablesoftware version is only possible under a valid Automation Sentinel at the MaintainPlus Level or above. For customers with 800xA 5.1 or 5.0 versions, AutomationSentinel Maintain Plus is required. For customers with versions earlier than 5.0 (3.0,3.1, 4.0, 4.1) Automation Sentinel Maintain and Evolve is required as of April 1st2016.

Out of band corrections to an individual product can be released as a product specificrollup. A valid warranty or valid Automation Sentinel subscription at the Maintainlevel or above is required to receive any such software rollup's. Safety relatedcorrections can be based on case-by-case decision, provided generally to all userswithout cost. The existing license file will work after installation of the revision. Noupdate is required.

Temporary Corrections, Certified Corrections

A temporary correction (TC) is intended for immediate correction of critical problems.Normally a temporary correction is customer specific and made available only to oneor a few customers where a problem is identified, and where an immediate action isrequired. A valid warranty or a valid active Automation Sentinel subscription at theMaintain Level or above is required to receive any such software corrections.

A certified correction (CC) is intended for immediate correction of problems in safetycertified products. There are two types of certified corrections, safety critical, andnon-safety critical.

256 3BSE041434-600 H


A safety critical CC is always made available to all end users regardless of theirAutomation Sentinel agreement status, and is maintained as all other softwaredeliverables listed in the safety annex.

A non-safety critical CC can be selectively delivered to end customers based on thecriticality and business impact in their installations.

10.2.2 System 800xA Software SupportAutomation Sentinel is a life cycle support program aimed at providing services forthe maintenance, continuous enhancement, and evolution of the ABB installed baseof control systems. Each and every ABB control system must be under AutomationSentinel to make use of the multi-tiered services and deliverable offered under theprogram to meet the differing customer life cycle service needs. Automation Sentineloffers four subscription options (see Figure 10.1) that can be briefly described asfollows:

Figure 10.1: Automation Sentinel - Program Options and Main Deliverable

Maintain Basic

The Maintain Basic subscription level provides a set of life cycle support servicesand deliverable that do NOT include any software maintenance or software updateof any type. This subscription level includes L3/L4 expert product support and isprimarily designed for those customers who are focused on regularly maintaining the

3BSE041434-600 H 257


IT security status of their control system as well as getting web access to timely andrelevant control system information through at myABB/My Control System.

Maintain

The Maintain level includes all the life cycle services included under the MaintainBasic level along with software maintenance of the latest system software version.This level is designed for those customers who are focused on maximizing theavailability of their system while minimizing the introduction of changes into thecontrol system environment.

Maintain Plus

The Maintain Plus includes all the life cycle services included under the Maintainlevel along with access to new software license versions and enhancements. Withinthis level, customers may upgrade to new software products within the same systemproduct family (does not include evolution from Heritage Systems to 800xA orSymphony Plus). This level provides increased flexibility to those customers whowant to maintain their system, keep it updated, and enhanced with new functionality.

Maintain & Evolve

This includes all the life cycle services available included under Maintain Plus withthe addition of the possibility to evolve the existing control system to ABB state-of-theart System 800xA. This level is designed to provide the flexibility to meet the needsof those customers who are on the leading edge of productivity and performanceimprovements and/or are in the process of step-wise evolution of their existing controlsystem.

Refer Automation Sentinel Product Guide (3BSE047996) for more information.

10.2.3 Life Cycle Parts Services

Policy:

ABB offers a comprehensive range of spare parts with short lead times, which helpsminimize downtime if a failure occurs. In addition, preventive maintenance kits make

258 3BSE041434-600 H


it possible to plan maintenance in advance. This maximizes availability and increasesmaintenance performance.

The Life Cycle Parts Service portfolio meets customers' needs to minimize costs andmaximize the value of their investments in ABB equipment.

Whether you need to repair a broken part or purchase a spare part, our servicesachieve cost-efficient maintenance.

ABB is highly qualified and provides a wide range of technical knowledge and support.ABB ensure that customers receive the best possible return on their assets throughoutthe entire product life cycle.

Spare Parts Service

Choose this service when you wish to purchase a new spare part shipped within aday.

Our high quality genuine and certified ABB spare parts increases reliability leadingto longer lifetime of your equipment.

With our large stock of ABB spare parts for all Open Control Products, we offer quickhandling and shipping of your spare parts needs through our 24 hours web-basedordering process. The use of optimal means of transport results in cost effective andquick arrival of your spare parts.

Emergency Parts Service

Choose this service when quickest possible delivery is the most important factor.Parts service availability is critical to on-going operations, not only during businesshours. ABB maintains a complete stock of certified ABB parts to ensure availability.

For emergency cases, our personnel and partners are available 24 hours to provideimmediate response to your emergency parts request and our global logistics networkensures the quickest possible delivery.

Refurbished Parts Service

Choose this service when you want to purchase a spare in the most cost-effectiveway, or during later Life Cycle phases when new parts are no longer manufactured.

The Refurbished Parts Service is a cost-effective alternative to purchasing new sparepart modules. Parts sold with this service are 'like-new' parts that have been recovered

3BSE041434-600 H 259


and updated by ABB. What's more, they have been tested to meet the originalequipment specifications and the current component standards.

Parts Repair Service

Choose this service when you wish to have your part repaired and returned. Our repairnetwork provides repairs that meet original equipment specifications. Our world-classturnaround time for repairs is typically less than two weeks. We include a repair reportin every return shipment.

Parts Exchange Service

Choose this service when you wish to purchase a fully functional spare in return foryour broken unit.

Parts Exchange Service keeps lead-times to a minimum while offering a cost efficientalternative to a spare part. Parts under this service can be refurbished or new dependingon availability. A repair report for the broken 'sent-in' unit can be supplied if it isrepaired. Request this report when ordering the exchange part.

Parts Test Service

Choose this service when you want assurance that the spares you have used or storedfor a longer period are still operational.

Parts Test Service determines the status of parts, for example: if they are functionaland meet the original manufacturing specifications, with a 3-week turnaround timeor less.

Units that passes the test are re-sealed with factory labels to provide the bestperformance when placed in your system. ABB Parts Repair Service is available forany non-functioning units. A Test Report is always included with the unit.

Parts StepUp Service

Choose this service when you want to replace your current system hardware withmore powerful and up-to-date units.

Your system hardware must be in control of your process at all times, regardless ofsystem load. To assure this, it is now possible to Step-Up your hardware, by replacingyour hardware units with a more powerful ones, that also extends the expected lifetime of the system.

260 3BSE041434-600 H


Parts Root Cause Analysis Service

Choose this service when you want to know valuable knowledge and understandingthrough analyzing the root cause of a hardware failure.

With the only purpose of providing valuable knowledge and understanding of thatcaused a hardware failure, our hardware design experts analyze your failed hardware,with access to laboratory and original design and test documentation. After the analysishas been performed, an extensive report is created and issued.

Preventive Maintenance Kits

Choose this program when you want to secure continuing operation and know thecalculated maintenance cost.

Preventive maintenance is the most efficient maintenance program because it ensuresthat equipment is serviced before a breakdown actually occurs. Preventive maintenanceis carried out during planned productions shutdowns and the required parts andresources should be reserved in advance.

Our pre-specified, easy-to-order kits consist of genuine ABB spare parts that arenecessary for a specific scheduled maintenance. These cost-efficient kits have a lowerprice compared to the price of individual spare parts and are delivered to a lead-time,unlike normal spare parts.

Inventory Access™ Program

Choose this program when you want to have spare parts located at or near yourpremises.

Our Parts Inventory Management program, inventory Access™ (iAP), is a costeffective alternative to purchasing parts inventory. It provides a customized spareparts inventory at or near your location yet owned and maintained by ABB. We takefull responsibility for obsolescence, depreciation, administration and inventoryownership costs until the parts are put into service.

A fixed monthly fee gives you the security of on site parts availability, while reducingthe initial capital expenditures for a spare parts inventory. Actual purchase does nottake place until the part is taken out from the Inventory Access part stock.

This is also the time at which the warranty period starts.

3BSE041434-600 H 261


262 3BSE041434-600 H

11 Ordering and Delivery

This topic describes the ordering and delivery of System 800xA.

11.1 Ordering Procedure

11.1.1 Ordering ToolsWhen ordering an 800xA System, use the appropriate and valid price books andmatching version of Wizard 800xA. Orders for 800xA shall under normalcircumstances be carried out using Business Online (BOL).

The following rules apply:

• Use BOL and optionally the Wizard 800xA to order System 800xA.

• Use the Wizard 800xA and BOL to order license expansions.

The tools and price books are found in the Lotus Notes "BU IIT & OCS Pricingdatabase". They are also available through the Control System home page. Informationregarding BOL is found on Operations Centers home page.

11.1.2 Sales Configurator Wizard 800xAThe Sales configurator Wizard 800xA contains all configuration rules and prices forSystem 800xA. The Wizard is used to generate a valid configuration and a correctprice. The Wizard also supports quotation and ordering.

By using the data created in the configuration process for quotation and ordering, youobtain a single point of data entry, minimizing the risk of divergence in quoted andordered data.

General Wizard 800xA Capabilities

• Create a correct configuration with correct purchase and sales prices.

3BSE041434-600 H 263

11 Ordering and Delivery 11.1 Ordering Procedure

Help to create an optimized configuration. In the Wizard, the user enters basicdata regarding the System 800xA (number of I/O channels, which controllers toinclude etc.). Based on this, the Wizard will suggest a configuration as optimizedas possible. This configuration may be changed in a later stage.

• Templates for quotations to support the bid process.

The Wizard contains extensive possibilities to generate different reports. Forexample you can generate different versions of Bill of Material, OverviewDrawing for the 800xA configuration, Load Calculations for AC 800M etc.

These reports may be used directly or the information can be copied as a basefor the quotation.

• Single point of data entry.

Data is entered once and used for configuration, quotation and orderingminimizing the risk of distortion of data.

Include any Product in a Configuration

Although the configurator contains configuration rules and prices for System 800xA,it can still be used to select any product with a price list in Wizard format or productswithout price lists.

Example Project:. A project will contain System 800xA, Harmony Controllers andworkstations from HP/Dell.

The Wizard will help to generate the 800xA configuration. However, by using the"Manual Selection" feature in the Wizard, the Harmony controllers are also included.And with Manual selection of "Ad-hoc" items the HP/Dell workstations are alsoincluded.

11.1.3 Business OnlineBusiness Online (BOL) is a global ABB ordering tool available for internal andexternal customers 24/7 365 days a year. Business Online has connectivity with theconfiguration tool Wizard and license handling tool Software Factory to make yourordering easier. In Business Online you will also find your order status, current stock,

264 3BSE041434-600 H


pricing for your business, detailed information about products, such as images andproduct data and many more interesting features.

You can find Business Online on online.abb.com or as a service on www.abb.comBusiness Online (BOL) is the ordering tool used for purchasing of System 800xA.

11.1.4 Order Box and Wizard Order ModuleWizard order module is used to enter order and delivery information needed by theFeeder Factory. This information may also be entered in the Wizard before the datais exported to the order module. The order module creates the order message (EDIfile) to be mailed to Feeder Factory order-box. The mail address and other order boxrelated information is found in the header of the project in the order module.

Use the Wizard order module (which is one of the Wizard Utilities) only if the productsare not available in Business Online.

All spare parts are ordered through Business Online through the assigned local ServiceCenter.

11.1.5 Ordering and Configuration ProcedureA System 800xA order normally consists of:

• 800xA Software licenses (Licenses for Tags, Operator licenses, Engineeringlicenses, History Logs, Batch Management etc.)

• 800xA Hardware (AC 800M controllers, S800 I/O modules, cables, power supplyetc.)

• Media (Solid state drive, DVD/CD)

Available Sales tools (Wizard 800xA and Wizard Order module) help the user to setup a correct 800xA configuration, meaning a configuration.

Up to date instructions on how to order new systems, expansions of existing systemsand spare parts are available at Operations Center web site under the section "Howto order".

3BSE041434-600 H 265


11.2 Special ServicesA number of special services are available for the ABB sales unit. These can beordered from the price list available for each of the Control Systems.

Information about these services such as samples, download forms etc. is availableat Operations Center web site.

11.2.1 License ChangeTransfer of license options or change of licensed options can be ordered from pricelist (3BSE041220) Certificates and Services for Order and Delivery.

Use the form Control System 800xA License change (3BSE058690) available atDelivery Center web site and attach with the order. In general are no changes of activelicenses allowed. All requests for changes in active licenses will be treated individually.

Options which are base for calculation of Automation Sentinel subscription cost canonly be moved in case the two systems have same expiry date or if the recipientsystem has a shorter validity than the "sender". New audit sheets are required.

11.2.2 Pre-installed PCSystem 800xA Pre-Installed is a service from Delivery Center in Västerås offeringhardware with latest Industrial IT System 800xA and Third party softwarepre-installed.

Some of the benefits with this service are: Time saving, Efficient, Reliable andReady-to-use.

More information about this service is available at Delivery Center web site.

11.2.3 Certificate of QualityThe purpose of this certificate is, to certify that the units are new, unused and that allunits successfully have passed all the manufacturer inspections and tests in accordancewith Control Technologies standard requirements.

Certificate of Origin - (Chambers of Commerce Certification)

The certificate of origin is sometimes required by the end customer due to importregulations, for example when it comes to identifying the origin of goods in subjectto anti-dumping regulations or licensing and monitoring arrangements.

266 3BSE041434-600 H

11 Ordering and Delivery 11.2 Special Services

Certificate of Origin - (ABB Certification)

This is the same as the as the previously described Certificate of Origin but signedby ABB AB.

11.2.4 Suppliers DeclarationThis declaration is used when it is beneficiary that the goods is of EU origin and isdelivered within the EU. (The EU is a customs union; therefore there are no suchthings as imports or exports as far as products transported within the Community areconcerned. Products can be moved within the EU countries without the need forcustoms checks or clearance).

11.2.5 Packing CertificateThe packing certificate is to certify that the wooden packaging material has beentreated against pests and is labelled according to the IPPC standard. This is for examplesometimes required for deliveries to China.

Special Packing

All special packing preferences besides normal Control Technologies packing isavailable as price list items or price upon request.

Among the non-standard packing options are wooden packing and waterproof packing.

Special Goods Marking

All special goods marking preferences besides normal Open Control Systems goodsmarking is available as price list items or price upon request. Prices are per packingunit.

3BSE041434-600 H 267

11 Ordering and Delivery 11.2 Special Services

268 3BSE041434-600 H

12 Warranty

12.1 Extended Warranty - HardwareExtended warranty time for hardware products / articles can be ordered for 1, 2 or 3years. Extended warranty time for hardware products can be ordered from the pricebook 3BSE049907; Extended Warranty Time for

• 800xA

– AC 800M

– S800 I/O

– S900 I/O

– Fieldbus components

Included product classes (type of products) are:

• Power supplies

• Central units

• Communication Modules

• I/O modules

Extended Warranty time is not available for:

• Batteries

• Panels

Extended Warranty can be ordered using the Wizard 800xA. A special function inthe Project Wizard automatically adds the function to the order. Separate quotationdocuments are also available.

Usually the local ABB units are responsible for handling warranty claims and needapproval from Open Control Systems. The local ABB units can now cut out the riskexpense from the project cost calculation and assign the risk to Open Control Systems.

3BSE041434-600 H 269

12 Warranty 12.1 Extended Warranty - Hardware

This will reduce the cost of poor quality (CoPQ) for goodwill spending after theregular warranty period.

Always add the price for Extended Warranty Time for the HW into the quotation asan option (even if not previously discussed). If not in current quotation(s) or discussedbefore, bring up the possibility to extend the warranty time with the end customerbefore placing the purchase order for HW.

The order for extended warranty time must be included in the HW purchase and thereis one price list article per extended warranty time period in addition to the regular18 month hardware warranty.

• 12 months for 3% price increase



The price increase is calculated on the standard transfer price (no discounts). Formore details regarding Value Based Selling, how to sell, how to purchase, orderconfiguration-, quotation-, ordering- and warranty time period validity check toolssee the presentation 3BSE061483; Extended Warranty Launch in ABB Library.

270 3BSE041434-600 H

12 Warranty 12.1 Extended Warranty - Hardware

13 Revision History

The revision index of this User Manual is not related to the 800xA 6.0 SystemRevision.

The following table lists the revision history of this User Manual.

DateDescriptionRevisionIndex

December 2014Published for 800xA 6.0 releaseA

October 2015Published for 800xA 6.0.1 releaseB

October 2015Published for 800xA 6.0.1 releaseC

April 2016Published for 800xA 6.0.2 releaseD

September 2016Published for 800xA 6.0.3 releaseE

not publishedF

November 2017Published on 800xA media for 800xA 6.0.3.1 releaseG

November 2017Published on ABB Library and myABB/My ControlSystem portal for 800xA 6.0.3.1 Release

H

3BSE041434-600 H 271

13 Revision History

13.1 Updates in Revision HDescription of updateUpdates Section/sub-section

Updated the compatibility of Windows 10Enterprise 2015 LTSB with 6.0.3.1

Section Supported Operating Systems

13.2 Updates in Revision GDescription of updateUpdates Section/sub-section

Updated with respect to new operatingsystems supported

Section Supported Operating Systems

Update regarding server hardwareSection Server and Workstation HardwareClarification regarding non-validatedhardware

Added support for highly distributed plantsusing Multiple Aspect Services

Section Example topology with MultipleAspect Servers

Added support for Auxiliary clientsSection Auxiliary Clients

Support for Microsoft Hyper-V includedSection Virtualization Host Software

Maximum Batch Equipment for Melodycontrollers updated

Table Controller Dependent Capacities

Parameters for Information Managerupdated. Notes on disk configuration andits relation to performance added

Section Application Parameters Summary

272 3BSE041434-600 H

13 Revision History 13.1 Updates in Revision H

IndexNumerics

800xA for Advant Master800xA for DCI, 236Configurations, 232

800xA for DCI800xA for Melody, 247Composer CTK, 237

800xA for FreelanceSizing details, 249

800xA for Melody800xA for MOD 300, 242Composer, 248

800xA for Melody softwareOperating systems, 248

800xA for MOD 300, 238AdvaBuild 3.3/2 P2, 239Harmony, 243

800xA for Symphony Plus HarmonyBatch Management, 246

AABB Engineered Drives, 112, 144ABB Standard Drives, 112, 144AC 800M, 129AC 800M Connect

OPC Server, 127performance, 127

AdvaBuild 3.3/2 P2, 239800xa for MOD 300, 239DCN, 238

Advant Controller 400 Series with Mastersoftware

MasterPiece 200/1, 227Advant Master Controllers

Advant Controller 400 Series with Mastersoftware, 227

Alternative dimensioning methodOPC subscriptions, 41

ArchivingClients, 203

Asset optimization, 212CLAM, 213

BBatch management, 205

Capacities, 204, 206Batch Management

800xA for Symphony Plus Harmony, 246BC810, 133BC820, 134Bus Bandwidth, 220Bus Load, 219

CCapacities

Function phase driver, 207Capacity, 219Caution

Information, 13Caution icon, 13CI853, 90, 135CI854, 112, 135, 163CI854A, 90, 136CI854B, 90, 136CI855, 90, 126, 136CI856, 90, 115, 136CI857, 90, 126, 137CI858, 90, 116, 137CI860, 90, 137

3BSE041434-600 H 273

Index

CI862, 90, 137CI865, 90, 137CI867, 90, 137CI868, 90, 137, 176CI869, 90, 138CI871, 90, 138, 166CI872, 90CI873, 90, 138, 186CLAM, 213

PNSM, 211Clients

Desktop trends, 202COMLI, 90Compatibility, 217Composer

800xA for Melody, 248Hardware configurations, 246Hardware prerequisites, 248

Composer CTK, 237800xA for DCI, 237ECC MUX, 237

Conductor NT, 238Configurations

Graphics, 228Control Builder M, 251Control loop asset monitoring, 213

DData Rate, 219DataSet Peripheral (DSP), 219DCN, 238

Hardware requirements, 240Desktop trends

Disk and memory, 196Direct OPC logging, 224Disk and memory

History collection, 199Document conventions, 13

EECCMUX, 237

800xA for DCI, 237ECC MUX

Graphics, 236ECCP, 237

800xA for DCI, 237eDCN, 238Engineering tools

AdvaBuild, 239Engineering workplace, 251

GGraphics

800xA for Advant Master, 228Objects, 236

HHardware configurations

800xA for Symphony Plus Harmony, 246Server switchover time, 245

Hardware prerequisites800xA for Melody, 248Objects, 247

Hardware requirements800xA for MOD 300, 240Objects, 242

HarmonyObjects, 243PETI, 251

History collectionLog capacity, 198

IIcons, 13

Caution, 13Information

Tip, 13Information icon, 13

274 3BSE041434-600 H

Index

Information managementArchiving, 201

INtools, 251INtools/SmartPlant Instrumentation

Load data, 251INtools document call-up, 251

LLoad data

Objects, 251PETI, 251

Log capacityPDL, 197

MMI, 44MODBUS RTU, 90MODBUS TCP

Performance, 178Modulebus

scan cycle time, 108Multisystem integration, 44

Supported connects, 46

OObjects

800xA for DCI, 236PU514/PU514A, 239RTA board, 238

OPC Serverperformance, 127

Operating systems800xA for Melody software, 248

OverviewSystem 800xA, 15

PPAS, 240PDL, 197

Server capacity, 195–196performance

AC 800M Connect, 127Performance

DataSet Peripheral transfer time, 219PETI, 251

INtools/SmartPlant Instrumentation, 251PLC connect, 48

Performance, 50PM851, 88, 129PM851A, 88, 129PM856, 89, 129PM856A, 89, 129PM858, 89, 130PM860, 89, 130PM860A, 89, 130PM861, 89, 130PM861A, 89, 130PM862, 89, 131PM864, 89, 131PM864A, 89, 131PM865, 89, 132PM866, 89, 132PM866A, 89, 132PM867, 89, 133PM891, 89, 133PNSM, 211Process engineering tool integration, 251PU514/PU514A, 239

800xA for Melody, 239Sizing data, 247

RRTA board

800xA for MOD 300, 238Software, 239

SS100 I/O, 144, 150

3BSE041434-600 H 275

Index

S200 I/O, 144, 152S200L I/O, 144, 153S800 I/O, 144–145S900 I/O, 144, 147S3964R, 90Scan cycle

ABB Engineered Drives, 112scan cycle time

Modulebus, 108scanning ABB Engineered Drives, 112

scanningModulebus, 108

SerialLib, 90Server, 245Server capacity

Sizing examples, 197Server switch over time

800xA for Symphony Plus Harmony, 245Sizing data

800xA for Melody, 247Sizing data (Melody), 247

SM810, 89, 134SM811, 89, 134SM812, 89, 135SmartPlant Instrumentation, 251Standard Drives, 112System 800xA, 15

Overview, 15

TTip

Warning, 13Tip icon, 13TTD logging, 224

WWarning icon, 13

276 3BSE041434-600 H

Index

800xA 6.0 System Guide Technical Data and Configuration 2010. 7. 5. · System800xA SystemGuide...

Documents

Transcript of 800xA 6.0 System Guide Technical Data and Configuration 2010. 7. 5. · System800xA SystemGuide...