Research Article An Architecture for Hybrid Manufacturing...

Research ArticleAn Architecture for Hybrid Manufacturing Combining 3DPrinting and CNC Machining

Marcel Muumlller and Elmar Wings

Institut fur Maschinen und Anlagenbau Constantiaplatz 4 26723 Emden Germany

Correspondence should be addressed to Marcel Muller marcelmuellerhs-emden-leerde

Received 5 July 2016 Accepted 30 August 2016

Academic Editor Fu-Shiung Hsieh

Copyright copy 2016 M Muller and E Wings This is an open access article distributed under the Creative Commons AttributionLicense which permits unrestricted use distribution and reproduction in any medium provided the original work is properlycited

Additive manufacturing is one of the key technologies of the 21st century Additive manufacturing processes are often combinedwith subtractive manufacturing processes to create hybrid manufacturing because it is useful for manufacturing complex partsfor example 3D printed sensor systems Currently several CNC machines are required for hybrid manufacturing one machineis required for additive manufacturing and one is required for subtractive manufacturing Disadvantages of conventional hybridmanufacturing methods are presented Hybrid manufacturing with one CNC machine offers many advantages It enablesmanufacturing of parts with higher accuracy less production time and lower costs Using the example of fused layer modeling(FLM) we present a general approach for the integration of additive manufacturing processes into a numerical control for machinetools The resulting CNC architecture is presented and its functionality is demonstrated Its application is beyond the scope of thispaper

1 Introduction

Modern machine tools are controlled by a computerizednumerical control (CNC) For this reason manufacturingprocesses for example drilling turning and milling arealso referred to as CNC machining when the machine toolis controlled by a CNC The main task of the CNC is tocontrol the relative motion of the tool and the workpieceAll these tools cut away unwanted material to manufacturea part with the desired geometry Therefore these processesare also referred to as subtractive manufacturing Furthersubtractive manufacturing processes are reaming threadingor laser cutting plasma cutting and water cutting Becauseof their commonalities some processes are contained in onemachine For example most milling machines allow a toolchange so that a drill or reamer can be used A millingmachine with an automatic tool changer and an automaticworkpiece changer is called a machining center A fourthaxis offers further processes for example threading andturning The machining center described could carry outmanymanufacturing processesThe combination of different

processes in one machine facilitates more economical manu-facturing because for example the quantity of clamping canbe reduced Furthermore the costs of purchase and upkeepfor one machine are often lower than for two [1 2]

In contrast to the subtractive manufacturing processesadditive manufacturing processes add material to manufac-ture parts with the desired geometries This enables wastereduction because unwanted material will not be addedParticular process steps such as drilling do not have to beapplied for example a hole is formed by not adding materialin the desired position

This paper [3] focuses on energy consumption and indi-cates that accurate assessment and modeling of manufac-turing processes are becoming increasingly important Thecase study presented shows that additive manufacturing hasgreater advantages than conventional manufacturing pro-cesses when the number of parts was small In addition add-itive manufacturing offers new manufacturing strategies andmakes customized solutions possible even for small quanti-ties Furthermore additive manufacturingmakes productionon demand and production on site possible [1 2 4] All

Hindawi Publishing CorporationInternational Journal of Manufacturing EngineeringVolume 2016 Article ID 8609108 12 pageshttpdxdoiorg10115520168609108

2 International Journal of Manufacturing Engineering

these properties indicate that additive manufacturing is akey technology for economical manufacturing Dependingon the additive manufacturing processes postprocessing isrequired for good results The postprocessing can be donemanually or automatically by CNC machining [1 2] Theprocesses of additive manufacturing are completely differentto subtractive manufacturing processes although they havemany commonalities which are outlined later in this paperAdditionally [5 6] suggest that a combination of subtractiveand additive manufacturing processes is recommended

This paper describes a general approach for combiningsubtractive and additivemanufacturing in one CNC architec-ture In this paper we will use the term hybridmanufacturingto describe the combination In Section 2 conventionalmethods for hybrid manufacturing and its disadvantages arediscussed In Section 3 the fused layer modeling (FLM)process and special benefits of combining FLM with CNCmachining are presented Section 4 highlights the objectiveof this paper and Section 5 presents commonalities anddifferences between FLMandCNCmachiningThe approachis presented in Section 6 but its implementation is dividedinto several sections Section 7 describes temperature controlSection 8 describes temperature measurement for tempera-ture control Section 9 describes physical computing for com-munication and Section 10 describes motion control Sec-tion 11 describes the operation principle of the resulting CNCarchitecture for FLM Finally Section 12 concludes the paper

2 State of the Art

A manually organized workpiece transport is flexible butnot effective each machine requires reclamping and its ownsetup which takes time and reduces the achievable accuracy[2] Therefore a production strategy is required

21 Hybrid Manufacturing with a Production Line A pro-duction line with machines which complement each other isillustrated in Figure 1 Machines are labeled with rectanglesand processes are labeled with letters On the one hand theautomatic workpiece transport system of the production linereduces the quantity of clamping and setup increasing theproductivity and achievable accuracy On the other handthe production line takes up much space and flexibility isreducedThere is no possibility ofmanufacturing parts whichneed variants of the process orderThis strategy is optimal formanufacturing identical parts but not individual parts this isa significant disadvantage because one of themain advantagesof additive manufacturing is making customized solutionspossible even for small quantities [1 2 4]Therefore anotherstrategy is required

22 Hybrid Manufacturing with a Machining Center in a Pro-duction Line As previously described a machining centercombines several manufacturing processes in one machine(see Figure 2) This enables several different process stepswith a minimum of clamping to reduce time and costs Forexample a machining center which carries out the processesB C and D can carry out several process orders (BCD CBDCDB ) In addition to the manufacturing processes the

Unmachined part

A B C D E

Machined part

Figure 1 Production line with machines which complement eachother

Unmachined part

A BCD E

Machined part

Figure 2 Production line with machines and a machining center

machining center can also be equipped with a measurementsystem to save a separate measuring machine This enablesthe reduction of required space and can be more economicalbecause fewer machines and buffers are needed betweenthe machines Referring to Figures 1 and 2 the failure ofone machine stops the production of the whole line Incontrast to a production line with complementary machinesthe production with machining centers enables a conceptin which the machines complement and cover each otherAn alternative concept to a production line is a flexiblemanufacturing system (FMS) [2]

23 HybridManufacturing with a Flexible Manufacturing Sys-tem Theprocess steps are distributed to particular machinesor machining centers which are connected by a rotatingworkpiece transport system (see Figure 3) The commonworkpiece transport system offers a minimum of clampingEach machining center offers different andor equivalentprocesses to complement andor cover the othersThe advan-tage of such an FMS is maximum flexibility The failure ofone machine does not stop the entire production becauseanother machine covers the failed process Furthermoredifferent process times do not have to be compensated for bysynchronizing machines if agent-based dynamic schedulingfor flexiblemanufacturing systems arranges an intelligent andflexible process order [2 7 8] Scheduling is a separate issuethat could fill many papers Nevertheless subtractive andadditive manufacturing can be combined with an FMS Insome cases an FMS could be best practice but an FMS needsa great deal of room and is not the least expensive or mosteconomical variant in every case

24 Different Process Times as Bottleneck Themain problemwith the combination of subtractive and additivemanufactur-ing can be the difference in the process times when subtrac-tive manufacturing is only used for postprocessing or shortintermediate steps for example threading a thread reaminga fit or finishing single contoursWith a focus on the differentprocess times the subtractivemachine is uneconomical whenonly one machine for additive manufacturing is availableDepending on the different process times it could also beuneconomical with two or three additive machines whenthe subtractive manufacturing processes are ten times fasterthan the additive Irrespective of this problem such an FMS

International Journal of Manufacturing Engineering 3

CD

AB

A

BCD

E

AE

Unmachined parts

Machined parts

Workpiecetransport

system

Figure 3 Flexible manufacturing system with a rotating workpiecetransport system and machines which complement and cover eachother

Unmachined part

ABCDE

Machined part

Unmachined part

ABCDE

Machined part

Figure 4 Production with machines which cover each other

is not applicable in every company Therefore a large-scaleintegrated machining center which provides both additiveand subtractive manufacturing could be a better application

25 Hybrid Manufacturing with a Large-Scale IntegratedMachining Center A large-scale integratedmachining centerprevents othermachines frombecoming uneconomical whenone machining center fails because a large-scale integratedmachining center produces independently from the othersFurthermore the production rate can be increased using lessspace when large-scale integratedmachining centers are used(see Figure 4)

As a result it is the best application when the additivemachine offers additional subtractive manufacturing proc-esses This is a new kind of machining center which needsa common control providing additive and subtractive man-ufacturing However up to now no control has been foundwhich supports both kinds of processes Therefore processintegration has to be carried out which requires specificprocess knowledge

3 The Fused Layer Modeling Process

This paper focuses on fused layer modeling (FLM) a genericname for the fused depositionmodeling (FDM) developed byS Scott Crump and trademarked (httpwwwtrademarkiacomfdm-74133656html) by Stratasys Inc In the contextof the RepRap project (httpwwwrepraporg) an ongoingproject that made and freely distributed a replicating rapid

Figure 5 Additive part manufactured by FLM with a small nozzlediameter and layer height

prototyper FLM is also called fused filament fabrication(FFF) As described in [1] FLM is an extrusion processin which thermoplastic material (filament) is continuouslysqueezed through a nozzle and deposited on a substrate Thematerialrsquos energy suffices to fuse the substrate after coolingdown a permanent connection is available In contrast toother additive processes the FLM is suitable for both proto-typing and production applications because parts with a highmechanical load capacity can be produced Referring to theRepRap project a wide range of users choose the fused layermodeling [1 9 10]

31 Materials for Fused Layer Modeling Generally plasticsare used with fused layer modeling for example polylactide(PLA) acrylonitrile butadiene styrene (ABS) polycarbonates(PC) or combined plastics for example PC-ABS There arealso types of plastic that can be sterilized by gamma or byethylene oxide for medical technologies In addition withsmall modifications materials other than plastics can beprocessed such as metals and ceramics [1 11ndash13]

32 Fast Hybrid Manufacturing for Higher Accuracy Asdescribed previously additive manufacturing can be donewithout subtractive manufacturing processes for exampledrilling Generally this also applies to fused layer modelingThe postprocessing was cited as a reason for combining bothThis will now be explained with a focus on FLMThe achiev-able accuracy depends on many parameters for example thelayer height and nozzle diameter If the layer height or thenozzle diameter is too large the resulting staircase effect willreduce accuracy and surface quality of the part as well asthe production time Therefore postprocessing is requiredImprovement of surface finish by staircase machining infused deposition modeling is outlined in [14] Generally asmaller nozzle diameter can increase the achievable accu-racy (see Figure 5) but also increases manufacturing timewhereas a larger diameter can reduce manufacturing timebut also reduces the achievable accuracy (see Figure 6(a))An approach will be explained for manufacturing a partwith accurate holes for fit Firstly the part is manufacturedwith undersized holes using a large nozzle diameter (seeFigure 6(a)) Thereafter the holes are rebored This strategyenables an optimal result with accurate holes and minimumproduction time (see Figure 6(b)) This concept is also appli-cable to threads and other fields requiring high accuracy

4 Objective

As described above there are good reasons to add the sub-tractive manufacturing processes to an additive manufactur-ingmachine instead of using an FMSwith differentmachines


(a) Undersized hole (b) Drilled out hole

Figure 6 Additive part manufactured by FLM and subtractivepostprocessed by drilling

Considering the mechanical load occurring with subtractivemanufacturing it is much better to equip a subtractivemanufacturingmachinewith the additivemanufacturing toolinstead of equipping an additivemanufacturingmachinewiththe subtractive manufacturing tool Therefore the objectiveof the development work carried out at the Institut furMaschinen und Anlagenbau was to integrate the process offused layer modeling into a CNC which is designated forsubtractive manufacturing This offers the development ofmachining centers combining both subtractive manufactur-ing and additive manufacturing

5 Commonalities and Differences betweenFLM and CNC Machining

As described above the processes of additive manufactur-ing are completely different than subtractive manufacturingprocesses though they have many commonalities Thesecommonalities and differences are outlined in the followingwith a focus on FLM and CNC machining

51 Commonalities As described above a machining centercould offer many different manufacturing processes For thisreason the CNC provides many process-specific functionsfor a wide range of processes [2]

Both kinds of manufacturing processes (subtractive andadditive) need a precise control to move the tool and theworkpiece on several axes The movement is determined byanNC program code written in G-code whereas the process-specific functions are determined by miscellaneous func-tions (M-codes) The NC program can be generated withcomputer-aided manufacturing (CAM) software from adrawing created using computer-aided design (CAD) soft-ware

Furthermore theCNCneeds sensors to collect data aboutthe tool and the working area In addition actuators forexample cooling aggregates have to be controlled On thesegrounds a common CNC is a good choice

Generally the process of FLM is not restricted to specificmachine kinematics Unusual kinematics is as suitable as triv-ial three-axis kinematics for tool and workpiece positioningThe focus is on simple three-axis kinematics because it isoften used for both kinds of processes

52 Differences The subtractive process needs a clampingdevice to withstand the mechanical load occurring duringmanufacturing Such a clamping device is not intendedfor FLM Experiments at the Institut fur Maschinen undAnlagenbau in Emden and recommendations of the RepRapproject indicate that a heating bed improves the FLMprocess

Therefore a heated clamping device as a combination of bothis needed This could be achieved with a heated vacuumclamp The challenge of engineering such a heated vacuumclamp is to prevent the additive manufactured parts fromabsorbing and deforming while clamping

In addition a special extrusion tool is required In con-trast to milling and threading a further axis is needed insidethe tool for accurately defined motions of the filament forpredefined deposing This tool has to be compatible withexisting tool holder and tool changer systems

Furthermore the process-specific M-codes of additivemanufacturing are completely different from M-codes forsubtractive manufacturing These M-codes have to be inte-grated into the CNC to make additive manufacturing possi-bleThe challenge is tomake aminimumofmodifications andprotect the integrity of existing M-codes

6 Approach

This approach is based on a fully assembled CNC machineor more precisely a three-axis CNC milling machine Theprocess of fused layer modeling is integrated into the existingCNC architecture

61 CNC Architecture for FLM For this approach a CNCwith a customizable architecture is requiredThe CNC has tooffer possibilities for the integration of new hardware addi-tional M-codes and self-developed (sub)programs Further-more a programmable logical controller (PLC) is requiredbecause it is an essential component of the CNC architectureFor this approach an internal PLC is used which is integratedinto the CNC software

Figure 7 shows a CNC architecture for FLM Severalcomponents are integrated into the previously existing CNCarchitecture for enabling FLM All components are inter-connected by a common interface the hardware abstractionlayer The CNC core the PLC and the human machineinterface (HMI) are standard components of a CNC [2]Therefore the previously existing CNC architecture offersthese components although they are modified The modi-fications are outlined later in this paper Furthermore thefully assembled CNC machine offers a motion controllerwith its sensors and actuators for three axes which are usedfor positioning the FLM tool A further axis is required forfilament transport Its integration is outlined later in thispaper But the temperature controller is not a CNC standardcomponent The integration of the temperature controllerwith its sensors and actuators is amajor part of this approachThese hardware components of the temperature control arecontrolled by a software component called microcontrollerinterface (120583CI) The modification of the previously existingCNC components and the integration of the new componentsare outlined in this paper

62 G-Codes and M-Codes As described above the man-ufacturing process is specified with NC programs Thereare three basic standards for NC programs RS-274-D [15](Extended Gerber RS-274X [16]) DIN 66025 [17 18] and


CNC

CNC-core PLC 120583CI HMI

(Sub)programsfor M-codes

Common interface(hardware abstraction layer)

Temperature controller

Sensors Actuators

Motion controller

Sensors Actuators

Figure 7The extended CNC architecture for fused layer modeling

ISO 6983 [19] But no NC program standard definesM-codesfor the fused layer modeling Therefore particular M-codeswhich are defined by the RepRap project (httpwwwrepraporgwikiG-code) (see Table 1) are used in this paper TheseM-codes are accepted by a wide range of CAM softwareprograms (slicers) developers and users The CNC-Corersquosinterpreter needs to know these M-codes and their meaningThe M-codes M104 M140 M141 M109 M190 and M191 aredesigned for setting process-specific temperatures The M-codes M109 M190 and M191 have the additional functionof waiting until the temperature has been reached Thefunctionality can be implemented with (sub)programs EachM-code calls up a (sub)program with well-defined functionsregarding temperature control

7 Temperature Control

A temperature control is required for the extrusion processat the hot-end (nozzle) A heating bed temperature controlcan optimize the process during manufacturing the firstlayers Furthermore a working space temperature controland a cooling fan can optimize the process A distributedmicrocontroller-based closed-loop temperature control canoffer the functionalities of temperature control Thereforethe microcontroller (120583C) controls three systems hot-endtemperature working space temperature and cooling fanspeed A heating bed is not considered because its effect isrestricted to the first layers while a whole working spacetemperature control offers also effects in further layers Theprinciple of the three temperature control systems is similartherefore only the hot-end temperature control is outlined

Table 1 M-codes needed for fused layer modeling (FLM)

Command MeaningM104 P⟨V119886119897119906119890⟩ Set hot-end temperature to ⟨V119886119897119906119890⟩

M106 P⟨V119886119897119906119890⟩ Set speed of hot-end cooling fan to⟨V119886119897119906119890⟩

M107 Set hot-end cooling fan off

M109 P⟨V119886119897119906119890⟩ Set hot-end temperature to ⟨V119886119897119906119890⟩and wait

M140 P⟨V119886119897119906119890⟩ Set heating bed temperature to⟨V119886119897119906119890⟩

M141 P⟨V119886119897119906119890⟩ Set work space temperature to⟨V119886119897119906119890⟩

M190 P⟨V119886119897119906119890⟩ Set heating bed temperature to⟨V119886119897119906119890⟩ and wait

M191 P⟨V119886119897119906119890⟩ Set work space temperature to⟨V119886119897119906119890⟩ and wait

Heatingunit

Sensor

Hot-endminus

+Tset Te 120583C

TRQ0

Q1 Qz

Tact

Tlowastact

Figure 8 Hot-end closed-loop temperature control

71 Hot-End Temperature Control The microcontroller re-ceives the set temperature 119879set from the CNC and con-trols the temperature independently from the CNC witha proportional-plus-integral-plus-derivative (PID) controlleralgorithm Any other control algorithm is also applicable butthe PID controller algorithm is well documented in literatureand easily adaptable to occurring system changes The tem-perature controller needs feedback therefore a temperaturesensor converts the actual hot-end temperature 119879act to thefeedback signal 119879lowastact This can be done for example using athermistor with negative temperature coefficient (NTC) orPT100

72 Hot-End Closed-Loop Control The control loop (seeFigure 8) focuses on hot-end temperature control Duringmanufacturing the cold filament is melted inside the hot-end Therefore a process-related rate of heat flow 1 isrequired to heat the filament 1 and disturbances 119911 causethe error 119879119890 because they are dissipating heat flows The rateof heat flow is defined in (1) where 119876 is the thermal energyof the heat flow The dissipating heat flows are detailed inSection 73

119894 = 119894 =119889119876119894119889119905

(1)

119879119890 is the difference between the set temperature 119879set and theactual temperature 119879act But the extrusion process requires aconstant temperature to manufacture parts with a constant


quality Variations in the filamentrsquos feed rate have to be con-sidered because a lower feed rate requires a lower rate of heatflow 1 Therefore 119879119890 is counterbalanced by the temperaturecontroller The controller output 119879119877119896 is calculated using (2)to eliminate the occurring error 119870P 119870I and 119870D are the PIDparameters which designate the controllerrsquos performance Asdescribed in [20] this algorithm is referred to as the stand(position) algorithm because 119879119877119896 is calculated for every valueof the sampled data period 119879119860 This algorithm is optimal fortemperature control using a large119879119860The signal119879119877 is detailedin Section 74

119879119877119896 = 119870P sdot 119879119890119896 + 119870I sdot 119879119860 sdot119896

sum119894=1

119879119890119894 + 119870D sdot119879119890119896 minus 119879119890119896minus1

119879119860 (2)

From the point of view of the controller 119879Hot-end = constantor more precisely 119879act = 119879set is the objective The heat flowrates have to be balanced for a constant hot-end temperature(see (3)) Therefore the dissipating heat flows are explained

119899

sum119894=0

119894 = 0 (3)

73 Hot-End Heat Dissipation There are two different dissi-pating heat flows On the one hand the process requires aheat flow for heating and melting the filament This heat flowis process-related and volitional It is described by its rate ofheat flow 1 On the other hand disturbances cause furtherdissipating heat flows These are described by a commonrate of heat flow 119911 It is useful to separate 1 and 119911for the process of energy optimization Disturbances are forexample dissipating heat flows which are caused by ther-mal convection thermal radiation and thermal conductionThermal convection is caused by the cooling fan Thermalconduction is caused by the temperature differences betweenfor example the work space and the hot-end as well as thehot-end and a thermal isolator These dissipating heat flowsare summarized in (4) and Figure 9

119911 =119899

sum119894=2

119894 (4)

For practical applications it is not necessary to determineeach effect separately Instead of considering particular 119894an overall 119911 is considered with an efficiency factor 120578119911This is useful because each extrusion tool has individualdisturbances Its efficiency depends on hot-end compositionfor example the size and position of the heating resistor Thesystem is therefore characterized by (5) The parameter 120578119911characterizes the systemwith aminimumof expense because0 is quantifiable bymeasuring the supplied electrical energyand 1 is quantifiable using the filament data and the electedprocess parameters Furthermore the parameter 120578119911 is usefulfor verifications of system optimizations with a minimum ofexpense The greater 120578119911(0 sdot sdot sdot 1) the lower the disturbance

0 =1120578119911 (5)

The fan which is mounted near the hot-end (see Figure 9)cools the deposited filamentThemicrocontroller receives theset velocity Vset from the CNC and controls the fan speed withan open-loop fan control (see Figure 10) V119877 is the amplifiedcontrol signal which controls fan speed 1198853 is the parameterwhich causes 3 as a secondary effect during cooling of thedepositing filament The process of energy optimization anda detailed thermal analysis are beyond the scope of this paperFirstly they are separate issues that could fillmany papers (see[21]) and secondly it is not necessary to determine any rate ofheat flow because the controller needs only the temperaturedifference which is caused by the heat flows [22]

74 Hot-End Heat Supply As described above the PIDalgorithm calculates 119879119877119896 for compensation 119879119877119896 is modulatedby the microcontrollerrsquos pulse-width modulation (PWM)generator The PWM signal is amplified (see Figure 11) bya metal-oxide-semiconductor field-effect transistor (MOS-FET) The amplified control signal is called 119879119877 The heatingunit (see Figure 8) is the actuator It is a heating resistorwhich converts the electrical energy from the amplified signal119879119877 into the thermal energy 1198760 for heating the hot-end Theresulting rate of heat flow is called 0 and counterbalancesthe dissipating heat flows 0 depends on the rate of electricalenergy el It is roughly approximated by (6) where 119880 is theelectrical voltage 119868 is the electrical current and119863 is the dutycycle which is caused by the PWMThe duty cycle is definedin (7) where 119879119875 is the total period of the signal and 119879119867 is thetime during which the signal is active [23]

0 = el = 119875el = 119880 sdot 119868 sdot 119863 (6)

119863 = 119879119867119879119875 (7)

8 Temperature Measurement

As described above a temperature sensor is needed In thispaper an NTC thermistor is used because of its small costs Itis a resistor whose resistance significantly varies with temper-ature Temperature measurement is made according to Fig-ure 12 where 1198800 is the voltage source 1198771 is series resistant119877119879 is the resistant of the NTC and 119880119879 is the voltage acrossthe NTC 119880119879 is the measuring signal The resistant 119877119879 iscalculated according to (8)

119877119879 =119880119879 sdot 11987711198800 minus 119880119879

(8)

The NTCrsquos resistancetemperature characteristic (curve) isrequired to determine the temperature It can be describedby three variants a lookup table the Steinhart-Hart equation(see (9)) [24] or the 119861 (or 120573) parameter equation (see (10))[25]

A lookup table describes the resistancetemperature char-acteristic in steps a specific resistance is assigned to a specifictemperature for example 119877119879 = 550Ω rArr 119879 = 200∘C 119877119879 =500Ω rArr 119879 = 205∘CValues between the steps can be foundby interpolationThe implementation of the lookup table canbe done using an array to store the values A lookup table


Heating resistor SensorFan

Filament

Q1 (process-related)Q3 (by thermal convection)

Q4 (by thermal radiation)

Q2 (by thermal conduction)

Q0

Figure 9 Heat flows which are introduced and dissipated during FLM

Fan Hot-endset

120583C R Z3 Tact

Figure 10 Cooling fan open-loop speed control

Microcontroller board (120583C)

Microcontroller

Reference signals

Serial

Hot-endtemperature

control

Heating bedtemperature

control

Work spacetemperature

control

Fanspeed

control

ADCMeasured

signal

ADCMeasured

signal

ADCMeasured

signal

PWM AControlsignal

Controlsignal

PWM A

Controlsignal

PWM A

Controlsignal

PWM A

Figure 11 Distributed microcontroller-based temperature control

prevents the microcontroller from having to make complexcalculations but it needs a great deal of memory for storingaccurate values

119879minus1 = 119860 + 119861 sdot ln (119877119879) + 119862 sdot ln (119877119879)3 (9)

119879 is the Kelvin temperature and 119877119879 is the NTC resistance 119860119861 and 119862 are the thermistorrsquos constants which may be offeredby the NTC manufacturer

119879 = 119861 sdot 119879119877ln (119877119879119877119877) sdot 119879119877 + 119861

(10)

As previously described 119877119879 is the NTC resistance at thetemperature 119879 in Kelvin 119877119877 is the resistance at the ratedtemperature 119879119877 (eg 29815 K) The 119861 (or 120573) parameter is amaterial-specific constant of the NTC thermistor which maybe offered by the NTC manufacturer

U0

R1

RT UT

Figure 12 Voltage divider for temperature measurement

If no constants are offered or a higher accuracy isrequired the constants can be found by calibration See [26]for further information about thermistor calibration

As previously described119880119879 is themeasuring signal whichrepresents the analog voltage across the NTCThe analog-to-digital converter (ADC) converts the analogmeasuring signalinto a digital value119885 which is available in themicrocontroller(see Figure 11) The digital measuring signal 119885 is calculatedwith (11) Its accuracy depends on the converterrsquos resolutionwhich is limited to 119873 bits The accuracy of the measuringsignal is limited to the converterrsquos resolution because the con-version involves quantization of the input Furthermore theleast significant bit (LSB) voltage 119880LSB has to be consideredIt is determined according to (12) [23]

119885 = 119880119879 minus 119880min119880max minus 119880min

sdot (2119873 minus 1) (11)

119880LSB =119880max minus 119880min2119873 minus 1

(12)

9 Physical Computing

Physical computing describes the interaction of the virtualworld (software system)with the real world (physical system)This requires communication hardware and software with awell-defined interface

91 Communication Hardware Several wired and wirelessoptions exist to connect the CNC with the temperature con-troller Criteria for choosing the best options are for examplecosts data rates distances hardware support environmentalconditions and the requirement for real-time Bluetooth and


Phys

ical

syste

mPh

ysic

al sy

stem

Softw

are s

yste

m

CNC

+

minus

PID

NTC Heating unit

Tact Tset

Tact (Ink)

Tset (Ink)

Te (Ink)

TRk (Ink)

TR

Tact (∘C) Tset (∘C)

T (Ink)rarrT (∘C) T (∘C)rarrT (Ink)

Tlowastact

Figure 13 Physical computing for temperature control

Wireless LAN are two frequently used wireless options butwith a focus on security and reliability a wired option isadvised Common wired interfaces are for example RS-232and (real-time) Ethernet but theCNCcan restrict the optionswhen it does not support all interfaces In this paper a serialcommunication is used with the universal asynchronousreceivertransmitter (UART) and the universal serial bus(USB) physic This enables compatibility with a wide rangeof computer-based CNC

The interaction between the CNC and the temperature-related sensorsactors is shown in Figure 13 The microcon-trollerrsquos functionality (software system) is highlighted Fig-ure 14 shows the steps required to calculate the temperaturefrom increments to degrees Celsius and vice versa

92 Communication Protocol A protocol is required for thecommunication between the CNC and the microcontrollerThe developed protocol supports request-response in plaintext The information is represented in a human-readableformat to enable human parsing and interpretation Thissupports intuitive use of commands for maintenance anddebugging via a command-line interface (CLI) for examplethe terminal emulation PuTTY (httpwwwputtyorg) Theimplemented commands are shown in Table 2 for requestsand Table 3 for responses Commands are separated fromparameters by the equal sign (=) while parameters are

Tact (∘C)Tset (∘C)

Tset (Ink) UT (Ink) = Tact (Ink)

T (∘C)rarrT (K) T (∘C)larrT (K)

see equations (9)and (10)


T (K)rarrRT (Ω) T (K)larrRT (Ω)

see equation (8)RT (Ω)rarrUT (V)

see equation (8)RT (Ω)larrUT (V)

see equation (11)UT (V)rarrUT (Ink)

see equation (11)UT (V)larrUT (Ink)

Figure 14 Calculation order to determine the temperature fromincrements to degrees Celsius and vice versa

CNC 120583C

sht = 200

hts = 200

CNC 120583C

ght

ht = 200

Figure 15 Requests fromCNC and responses frommicrocontroller

separated from each other by a comma An escape sequenceis used to indicate the input end After the input end inputis parsed and interpreted by the microcontroller Unknowncommands cause an error message while correct commandsare executed Figure 15 shows requests and responses betweentheCNCand themicrocontroller On the left side of Figure 15the set temperature is set for the hot-end by the CNC Themicrocontroller confirms the input On the right side ofFigure 15 the actual temperature is requested by the CNCThe microcontroller sends the actual temperature to theCNC Figure 16 shows requests and responses between amaintenance PC and the microcontroller On the left side ofFigure 16 the PID parameters are set for hot-end temperaturecontrol by the PC The microcontroller confirms the inputOn the right side of Figure 16 the actual PID parametersfor the hot-end temperature control are requested by thePC The microcontroller sends the actual PID parametersfor hot-end temperature control to the PC The parser andinterpreter are used by both microcontroller and CNC Astate machine is implemented in the CNC to frequentlycheck temperatures and set new set temperatures whenchanges occur If no human-readable format is required ahexadecimal-based format is an alternative to reduce datasize


PC 120583C PC 120583C

ghpshp =

hps = hp =

KP KI KD

KPKI KD KPKI KD

Figure 16 Requests from PC and responses from microcontroller

Table 2 Protocol for requests to microcontroller

Command Meaningsht = ⟨V119886119897119906119890⟩ Set hot-end temperature to ⟨V119886119897119906119890⟩ght Get hot-end temperature

sbt = ⟨V119886119897119906119890⟩ Set heating bed temperature to⟨V119886119897119906119890⟩

gbt Get heating bed temperature

swt = ⟨V119886119897119906119890⟩ Set work space temperature to⟨V119886119897119906119890⟩

gwt Get work space temperature

shp = 119870P 119870I 119870DSet PID control parameters forhot-end to 119870P 119870I 119870D

ghp Get PID control parameters forhot-end

sbp = 119870P 119870I 119870DSet PID control parameters forheating bed to 119870P 119870I 119870D

gbp Get PID control parameters forheating bed

swp = 119870P 119870I 119870DSet PID control parameters forwork space to 119870P 119870I 119870D

gwp Get PID control parameters forwork space

93 CNC Temperature Control The M-codes M104 M140M141M109M190 andM191 are designed for setting process-specific temperatures These M-codes are not reserved forother processes [18] Both M104 and M109 are used to setthe hot-end temperature M104 sets the temperature andcontinues the process directly regardless of whether thedesired temperature is reached or not This is good for smallchanges during the process but the hot-end needs to beheated before the process starts because a cold hot-end isnot useful Therefore M109 is used M109 pauses the processand continues the process when the desired temperature isreached This is signaled with an enable signal

931 Enable Signal The enable signal is generated bythe CNCrsquos PLC It depends on comparing conditions (seeSection 932) The enable signal could also be generated bythe microcontroller but this has two disadvantages First andforemost further information is transmitted which is redun-dant because all required information is already available inthe CNC Second when comparing conditions are changedrecompilation of the microcontroller program is requiredThe comparing conditions are therefore included in a PLCprogram so that changes can be made without recompila-tion

Table 3 Protocol for responses from microcontroller

Command Meaning

hts = ⟨V119886119897119906119890⟩ Hot-end temperature is set to⟨V119886119897119906119890⟩

ht = ⟨V119886119897119906119890⟩ Hot-end temperature is ⟨V119886119897119906119890⟩

bts = ⟨V119886119897119906119890⟩ Heating bed temperature is set to⟨V119886119897119906119890⟩

bt = ⟨V119886119897119906119890⟩ Heating bed temperature is ⟨V119886119897119906119890⟩

wts = ⟨V119886119897119906119890⟩ Work space temperature is set to⟨V119886119897119906119890⟩

wt = ⟨V119886119897119906119890⟩ Work space temperature is ⟨V119886119897119906119890⟩

hps = 119870P 119870I 119870DPID control parameters for hot-endare set to119870P 119870I 119870D

hp = 119870P 119870I 119870DPID control parameters for hot-endare 119870P 119870I 119870D

bps = 119870P 119870I 119870DPID control parameters for heatingbed are set to 119870P 119870I 119870D

bp = 119870P 119870I 119870DPID control parameters for heatingbed are119870P 119870I 119870D

wps = 119870P 119870I 119870DPID control parameters for workspace are set to 119870P 119870I 119870D

wp = 119870P 119870I 119870DPID control parameters for workspace are 119870P 119870I 119870D

932 Comparing Conditions The enable signal signals truewhen the comparing conditions become true The simplestcondition is shown in

119879set minus 119879act = 0 (13)

Without further conditions small variations are not toler-ated small variations should be tolerated with a range oftolerance because the process can accept small variationsTherefore (14) is a better comparing condition where 119879119879 isthe accepted tolerance

1003816100381610038161003816119879set minus 119879act1003816100381610038161003816 le 119879119879 (14)

The step response 119879act(119905) in Figure 17 shows that the processcan immediately start after 1199051 although the temperature119879act(119905) is overshooting This is not practical because theprocess needs a constant temperature Therefore a furthercondition is used for checking whether the temperatureis settled A Timer-ON-Delay (TON) which is a standardcomponent of a PLC is used for checking whether thetemperature is constant for a specific time 1199053

Currently the CNC notices deviations of the tempera-ture and prohibits manufacturing before the temperature issettled No temperature error management is implementedfor manufacturing however an error management system isrequired to protect parts and themachinewhen the differencebetween the actual temperature and the set temperaturebecomes too large during manufacturing This is importantto guarantee manufacturing with a consistent quality Atemperature error management system will be outlined in afuture paper


Tset

T0

0

Tact(t)

TT

t1 t2 t3 t4

t

T(t)

Figure 17 Step response from temperature controller

10 Motion Control

As described above this approach is based on an existing fullyassembled three-axis CNC milling machine Therefore thispaper does not cover the positioning of the toolTheFLM tool(extruder) by itself needs an extruder axis for extruding thefilament The name of the axis depends on the existing CNCplatform for example E-axis For convenience a rotatorymotor is used Inside the extruder the rotatory motion istransferred into a linear motionThe E-axis is integrated intothe CNC as a linear axis The trajectory of the extruderrsquosnozzle is analyzed in [27]

11 Operation Principle

All required components are described for extending theCNC architecture for FLM which is technically designatedfor subtractive manufacturingThis section describes the lineof action to set the hot-end temperature for manufacturing

(1) TheNCprogram is parsed and theNCprogram codes(G-codes and M-codes) are executed

(2) When the M-code M109 (see Table 1) is reached theinterpreter calls up the appropriate program calledM109 The NC program parsing pause during M109is executed

(3) The program M109 gets the set temperature by com-mand call (M109 P⟨V119886119897119906119890⟩) and commits the value tothe PLC (see Figure 18)

(4) The PLC commits the value to the microcontrollerinterface (120583CI)

(5) The 120583CI commits the value from the PLC to thetemperature controller (120583C) and frequently requeststhe actual temperature from120583CThe120583CI commits theactual temperature from 120583C to the PLC

PLC

Microcontroller interface (120583CI)

M109

Temperature controller (120583C)

Sensors Actuators

Set value Enable

Actual value Set value

Figure 18 Involved components of the CNC architecture forhandling hot-end temperature

(6) The PLC checks the comparing conditions while thetemperature controller controls the temperature

(7) The PLC sets the enable signal when the comparingcondition becomes true and the temperature is set-tled

(8) When the program M109 gets the enable signalM109 is terminated and the NC program parsing iscontinuous

This line of action is similarly applicable for M190 andM191 The programs M104 M140 and M141 are immediatelyterminated after setting the new set temperature The NCprogram parsing is not paused

A software-based CNC with PLC a microcontrollerboard with ATmega328 an EPCOS 100k thermistor anIRFZ44N MOSFET and a heating resistor are used for theimplementation of the CNC architecture presented The lineof action described was followed to set the hot-end temper-ature to 200∘C The actual value of the hot-end temperaturewas logged by the PLC and is shown in Figure 19 to demon-strate the functionality of the CNC architecture

12 Conclusion

Theapproach presented describes the integration of the fusedlayer modeling process into a CNC architecture which is des-ignated for subtractive manufacturing (milling) but withoutreferring to any specific CNC This permits engineering ofhybridmanufacturing centers which offer FLMwith postpro-cessing in one machine The benefits are a minimum quan-tity of clamping more accurate parts and less productiontime Furthermore the integration offers another significantadvantage over previous approaches because only one hybridCNC architecture is needed for offering many processes Thesame CNC can be used for additive manufacturing with sub-tractive manufacturing only additive manufacturing andonly subtractive manufacturing This prevents customersfrom assembling individual CNC systems for FLM and


0

20

40

60

80

100

120

140

160

180

200

220

Time (s)

Tem

pera

ture

(∘C)

0 200 400 600 800 1000

Figure 19 Visualization of hot-end temperature

milling machines Machine customers and machine opera-tors do not need knowledge in different CNC systems becausethe new process (FLM) is integrable in the familiar CNC sys-temThis is economical for developers customers and usersregardless of whether they need hybridmanufacturing or not

The approach presented focuses only on fused layermodeling (FLM) Other additive manufacturing processesneed also postprocessingTherefore the idea of hybridmanu-facturing is also applicable to processes other than fused layermodeling

As described in Section 52 further work is necessaryfor using the CNC architecture for hybrid manufacturingFirstly a clamping device is required which supports addi-tive and subtractive manufacturing Secondly an extrusiontool is required which is compatible with the existing toolchanger of the CNC machine Thirdly an extended CAMsystem is required which supports additive and subtractivemanufacturing at the same time Therefore at the time ofsubmitting this paper the CNC architecture presented wasnot applied to a real hybrid manufacturing process but it hasbeen successfully implemented

Competing Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper

References

[1] A Gebhardt Generative Fertigungsverfahren Additive Manu-facturing und 3D-Drucken fur PrototypingmdashToolingmdashPro-duktion 4 neu bearbeitete und erweiterte Auflage MunchenGermany 2013

[2] H B Kief H A Roschiwal and K Schwarz CNC-Handbuch20152016 Carl Hanser Verlag GmbH amp Co KG MunchenGermany 2015

[3] H-S Yoon J-Y Lee H-S Kim et al ldquoA comparison of energyconsumption in bulk forming subtractive and additive pro-cesses review and case studyrdquo International Journal of Precision

Engineering and Manufacturing-Green Technology vol 1 no 3pp 261ndash279 2014

[4] E Marquardt M Munsch A-K Muller et alHandlungsfelderAdditive Fertigungsverfahren 2016 httpwwwvdideHand-lungsfelderAM

[5] D Espalin D W Muse E MacDonald and R B Wicker ldquo3Dprinting multifunctionality structures with electronicsrdquo TheInternational Journal of Advanced Manufacturing Technologyvol 72 no 5ndash8 pp 963ndash978 2014

[6] V Townsend and R Urbanic ldquoA systems approach to hybriddesign fused deposition modeling and CNC machiningrdquo inGlobal Product Development Proceedings of the 20th CIRPDesign Conference Ecole Centrale de Nantes Nantes France19thndash21st April 2010 A Bernard Ed pp 711ndash720 SpringerBerlin Germany 2011

[7] I BadrAgent-based dynamic scheduling for flexible manufactur-ing systems [PhD thesis] Universitatsbibliothek StuttgartStuttgart Germany 2011 httpelibuni-stuttgartdeopusvoll-texte20115933

[8] L Nie Y Bai X Wang K Liu and C Cai ldquoAn agent-baseddynamic scheduling approach for flexible manufacturing sys-temsrdquo in Proceedings of the IEEE 16th International Conferenceon Computer Supported Cooperative Work in Design (CSCWDrsquo12) pp 59ndash63 May 2012

[9] S S Crump ldquoApparatus and method for creating three-dimen-sional objectsrdquo US Patent 5121329 1992

[10] R Jones P Haufe E Sells et al ldquoRepRapmdashthe replicating rapidprototyperrdquo Robotica vol 29 no 1 pp 177ndash191 2011

[11] S Hong C Sanchez H Du and N Kim ldquoFabrication of 3Dprinted metal structures by use of high-viscosity cu paste anda screw extruderrdquo Journal of Electronic Materials vol 44 no 3pp 836ndash841 2015

[12] S Hwang E Reyes K-S Moon R Rumpf and N KimldquoThermo-mechanical characterization of metalpolymer com-posite filaments and printing parameter study for fused deposi-tion modeling in the 3D printing processrdquo Journal of ElectronicMaterials vol 44 no 3 pp 771ndash777 2014

[13] S Masood and W Song ldquoDevelopment of new metalpolymermaterials for rapid tooling using fused deposition modellingrdquoMaterials amp Design vol 25 no 7 pp 587ndash594 2004

[14] P M Pandey N V Reddy and S G Dhande ldquoImprovementof surface finish by staircase machining in fused depositionmodelingrdquo Journal of Materials Processing Technology vol 132no 1ndash3 pp 323ndash331 2003

[15] ldquoInterchangeable Variable Block Data Format for PositioningContouring and ContouringPositioning Numerically Con-trolled Machinesrdquo Electronic Industries Association StandardEIA-274-D1979 1979

[16] ldquoRS-274X Gerber Format Specificationrdquo Ucamco SpecificationRS-274X 2015

[17] ldquoProgrammaufbau fur numerisch gesteuerte Arbeitsmaschi-nen Allgemeinesrdquo Standard DIN 66025-11983-03 DeutschesInstitut fur Normung eV Berlin Germany 1983

[18] ldquoProgrammaufbau fur numerisch gesteuerte Arbeitsmaschi-nen Wegbedingungen und Zusatzfunktionenrdquo Standard DIN66025-21988-09 Deutsches Institut fur Normung eV BerlinGermany 1988

[19] ldquoAutomation systems and integrationmdashNumerical control ofmachinesmdashProgram format and definitions of address wordsmdashPart 1 Data format for positioning line motion and contouringcontrol systemsrdquo International Organization for Standardiza-tion Standard ISO 6983-12009 2009


[20] S Zacher and M Reuter Regelungstechnik fur Ingenieure Anal-yse Simulation und Entwurf von Regelkreisen 14 korrigierteAuflage Springer Wiesbaden Germany 2014

[21] C Bellehumeur L Li Q Sun and P Gu ldquoModeling of bondformation between polymer filaments in the fused depositionmodeling processrdquo Journal of Manufacturing Processes vol 6no 2 pp 170ndash178 2004

[22] W GellerThermodynamik furMaschinenbauer Grundlagen furdie Praxis Springer Berlin Germany 2015

[23] K Wust Mikroprozessortechnik Grundlagen ArchitekturenSchaltungstechnik und Betrieb vonMikroprozessoren undMikro-controllern 4 verbesserte Auflage Vieweg+Teubner VerlagSpringer Fachmedien Wiesbaden GmbH Wiesbaden Ger-many 2011

[24] J S Steinhart and S R Hart ldquoCalibration curves for thermis-torsrdquo Deep Sea Research and Oceanographic Abstracts vol 15no 4 pp 497ndash503 1968

[25] R Parthier Messtechnik Grundlagen und Anwendungen derelektrischenMesstechnik fur alle technischen Fachrichtungen undWirtschaftsingenieure 4 verbesserte Auflage Friedr Vieweg ampSohnGWV Fachverlage GmbH Wiesbaden Germany 2008

[26] ldquoDAkks-DKD-R-5-1 Kalibrierung vonWiderstandsthermome-ternrdquo Deutsche Akkreditierungsstelle (DAkkS) BraunschweigStandard DKD-R-5-1 2010 httpwwwdakksdecontentkalibrierung-von-widerstandsthermometern

[27] Y-Z Jin J-F Zhang Y Wang and Z-C Zhu ldquoFilament geo-metrical model and nozzle trajectory analysis in the fuseddeposition modeling processrdquo Journal of Zhejiang University-SCIENCE A vol 10 no 3 pp 370ndash376 2009

International Journal of

AerospaceEngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014


Active and Passive Electronic Components

Control Scienceand Engineering

Journal of



RotatingMachinery


Hindawi Publishing Corporation httpwwwhindawicom

Journal ofEngineeringVolume 2014

Submit your manuscripts athttpwwwhindawicom

VLSI Design



Shock and Vibration


Civil EngineeringAdvances in

Acoustics and VibrationAdvances in



Electrical and Computer Engineering

Journal of

Advances inOptoElectronics


Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

SensorsJournal of


Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014


Chemical EngineeringInternational Journal of Antennas and

Propagation




Navigation and Observation



DistributedSensor Networks



these properties indicate that additive manufacturing is akey technology for economical manufacturing Dependingon the additive manufacturing processes postprocessing isrequired for good results The postprocessing can be donemanually or automatically by CNC machining [1 2] Theprocesses of additive manufacturing are completely differentto subtractive manufacturing processes although they havemany commonalities which are outlined later in this paperAdditionally [5 6] suggest that a combination of subtractiveand additive manufacturing processes is recommended

This paper describes a general approach for combiningsubtractive and additivemanufacturing in one CNC architec-ture In this paper we will use the term hybridmanufacturingto describe the combination In Section 2 conventionalmethods for hybrid manufacturing and its disadvantages arediscussed In Section 3 the fused layer modeling (FLM)process and special benefits of combining FLM with CNCmachining are presented Section 4 highlights the objectiveof this paper and Section 5 presents commonalities anddifferences between FLMandCNCmachiningThe approachis presented in Section 6 but its implementation is dividedinto several sections Section 7 describes temperature controlSection 8 describes temperature measurement for tempera-ture control Section 9 describes physical computing for com-munication and Section 10 describes motion control Sec-tion 11 describes the operation principle of the resulting CNCarchitecture for FLM Finally Section 12 concludes the paper

2 State of the Art

A manually organized workpiece transport is flexible butnot effective each machine requires reclamping and its ownsetup which takes time and reduces the achievable accuracy[2] Therefore a production strategy is required

21 Hybrid Manufacturing with a Production Line A pro-duction line with machines which complement each other isillustrated in Figure 1 Machines are labeled with rectanglesand processes are labeled with letters On the one hand theautomatic workpiece transport system of the production linereduces the quantity of clamping and setup increasing theproductivity and achievable accuracy On the other handthe production line takes up much space and flexibility isreducedThere is no possibility ofmanufacturing parts whichneed variants of the process orderThis strategy is optimal formanufacturing identical parts but not individual parts this isa significant disadvantage because one of themain advantagesof additive manufacturing is making customized solutionspossible even for small quantities [1 2 4]Therefore anotherstrategy is required

22 Hybrid Manufacturing with a Machining Center in a Pro-duction Line As previously described a machining centercombines several manufacturing processes in one machine(see Figure 2) This enables several different process stepswith a minimum of clamping to reduce time and costs Forexample a machining center which carries out the processesB C and D can carry out several process orders (BCD CBDCDB ) In addition to the manufacturing processes the

Unmachined part

A B C D E

Machined part

Figure 1 Production line with machines which complement eachother

Unmachined part

A BCD E

Machined part

Figure 2 Production line with machines and a machining center

machining center can also be equipped with a measurementsystem to save a separate measuring machine This enablesthe reduction of required space and can be more economicalbecause fewer machines and buffers are needed betweenthe machines Referring to Figures 1 and 2 the failure ofone machine stops the production of the whole line Incontrast to a production line with complementary machinesthe production with machining centers enables a conceptin which the machines complement and cover each otherAn alternative concept to a production line is a flexiblemanufacturing system (FMS) [2]

23 HybridManufacturing with a Flexible Manufacturing Sys-tem Theprocess steps are distributed to particular machinesor machining centers which are connected by a rotatingworkpiece transport system (see Figure 3) The commonworkpiece transport system offers a minimum of clampingEach machining center offers different andor equivalentprocesses to complement andor cover the othersThe advan-tage of such an FMS is maximum flexibility The failure ofone machine does not stop the entire production becauseanother machine covers the failed process Furthermoredifferent process times do not have to be compensated for bysynchronizing machines if agent-based dynamic schedulingfor flexiblemanufacturing systems arranges an intelligent andflexible process order [2 7 8] Scheduling is a separate issuethat could fill many papers Nevertheless subtractive andadditive manufacturing can be combined with an FMS Insome cases an FMS could be best practice but an FMS needsa great deal of room and is not the least expensive or mosteconomical variant in every case

24 Different Process Times as Bottleneck Themain problemwith the combination of subtractive and additivemanufactur-ing can be the difference in the process times when subtrac-tive manufacturing is only used for postprocessing or shortintermediate steps for example threading a thread reaminga fit or finishing single contoursWith a focus on the differentprocess times the subtractivemachine is uneconomical whenonly one machine for additive manufacturing is availableDepending on the different process times it could also beuneconomical with two or three additive machines whenthe subtractive manufacturing processes are ten times fasterthan the additive Irrespective of this problem such an FMS


CD

AB

A

BCD

E

AE

Unmachined parts

Machined parts

Workpiecetransport

system


Unmachined part

ABCDE

Machined part

Unmachined part

ABCDE

Machined part











4 Objective
















6 Approach






CNC





Sensors Actuators

Motion controller

Sensors Actuators














Heatingunit

Sensor

Hot-endminus

+Tset Te 120583C

TRQ0

Q1 Qz

Tact

Tlowastact




119894 = 119894 =119889119876119894119889119905

(1)




119879119877119896 = 119870P sdot 119879119890119896 + 119870I sdot 119879119860 sdot119896

sum119894=1


119879119860 (2)


119899

sum119894=0

119894 = 0 (3)


119911 =119899

sum119894=2

119894 (4)


0 =1120578119911 (5)



0 = el = 119875el = 119880 sdot 119868 sdot 119863 (6)

119863 = 119879119867119879119875 (7)



119877119879 =119880119879 sdot 11987711198800 minus 119880119879

(8)





Filament




Q0


Fan Hot-endset

120583C R Z3 Tact



Microcontroller

Reference signals

Serial

Hot-endtemperature

control


control


control

Fanspeed

control

ADCMeasured

signal

ADCMeasured

signal

ADCMeasured

signal

PWM AControlsignal

Controlsignal

PWM A

Controlsignal

PWM A

Controlsignal

PWM A





119879 = 119861 sdot 119879119877ln (119877119879119877119877) sdot 119879119877 + 119861

(10)


U0

R1

RT UT





sdot (2119873 minus 1) (11)


(12)





Phys

ical

syste

mPh

ysic

al sy

stem

Softw

are s

yste

m

CNC

+

minus

PID

NTC Heating unit

Tact Tset

Tact (Ink)

Tset (Ink)

Te (Ink)

TRk (Ink)

TR



Tlowastact
















CNC 120583C

sht = 200

hts = 200

CNC 120583C

ght

ht = 200




PC 120583C PC 120583C

ghpshp =

hps = hp =

KP KI KD

KPKI KD KPKI KD

















Command Meaning














119879set minus 119879act = 0 (13)


1003816100381610038161003816119879set minus 119879act1003816100381610038161003816 le 119879119879 (14)




Tset

T0

0

Tact(t)

TT

t1 t2 t3 t4

t

T(t)


10 Motion Control









PLC


M109


Sensors Actuators

Set value Enable








12 Conclusion



0

20

40

60

80

100

120

140

160

180

200

220

Time (s)

Tem

pera

ture

(∘C)

0 200 400 600 800 1000





Competing Interests


References
































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










CD

AB

A

BCD

E

AE

Unmachined parts

Machined parts

Workpiecetransport

system


Unmachined part

ABCDE

Machined part

Unmachined part

ABCDE

Machined part











4 Objective
















6 Approach






CNC





Sensors Actuators

Motion controller

Sensors Actuators














Heatingunit

Sensor

Hot-endminus

+Tset Te 120583C

TRQ0

Q1 Qz

Tact

Tlowastact




119894 = 119894 =119889119876119894119889119905

(1)




119879119877119896 = 119870P sdot 119879119890119896 + 119870I sdot 119879119860 sdot119896

sum119894=1


119879119860 (2)


119899

sum119894=0

119894 = 0 (3)


119911 =119899

sum119894=2

119894 (4)


0 =1120578119911 (5)



0 = el = 119875el = 119880 sdot 119868 sdot 119863 (6)

119863 = 119879119867119879119875 (7)



119877119879 =119880119879 sdot 11987711198800 minus 119880119879

(8)





Filament




Q0


Fan Hot-endset

120583C R Z3 Tact



Microcontroller

Reference signals

Serial

Hot-endtemperature

control


control


control

Fanspeed

control

ADCMeasured

signal

ADCMeasured

signal

ADCMeasured

signal

PWM AControlsignal

Controlsignal

PWM A

Controlsignal

PWM A

Controlsignal

PWM A





119879 = 119861 sdot 119879119877ln (119877119879119877119877) sdot 119879119877 + 119861

(10)


U0

R1

RT UT





sdot (2119873 minus 1) (11)


(12)





Phys

ical

syste

mPh

ysic

al sy

stem

Softw

are s

yste

m

CNC

+

minus

PID

NTC Heating unit

Tact Tset

Tact (Ink)

Tset (Ink)

Te (Ink)

TRk (Ink)

TR



Tlowastact
















CNC 120583C

sht = 200

hts = 200

CNC 120583C

ght

ht = 200




PC 120583C PC 120583C

ghpshp =

hps = hp =

KP KI KD

KPKI KD KPKI KD

















Command Meaning














119879set minus 119879act = 0 (13)


1003816100381610038161003816119879set minus 119879act1003816100381610038161003816 le 119879119879 (14)




Tset

T0

0

Tact(t)

TT

t1 t2 t3 t4

t

T(t)


10 Motion Control









PLC


M109


Sensors Actuators

Set value Enable








12 Conclusion



0

20

40

60

80

100

120

140

160

180

200

220

Time (s)

Tem

pera

ture

(∘C)

0 200 400 600 800 1000





Competing Interests


References
































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation























6 Approach






CNC





Sensors Actuators

Motion controller

Sensors Actuators














Heatingunit

Sensor

Hot-endminus

+Tset Te 120583C

TRQ0

Q1 Qz

Tact

Tlowastact




119894 = 119894 =119889119876119894119889119905

(1)




119879119877119896 = 119870P sdot 119879119890119896 + 119870I sdot 119879119860 sdot119896

sum119894=1


119879119860 (2)


119899

sum119894=0

119894 = 0 (3)


119911 =119899

sum119894=2

119894 (4)


0 =1120578119911 (5)



0 = el = 119875el = 119880 sdot 119868 sdot 119863 (6)

119863 = 119879119867119879119875 (7)



119877119879 =119880119879 sdot 11987711198800 minus 119880119879

(8)





Filament




Q0


Fan Hot-endset

120583C R Z3 Tact



Microcontroller

Reference signals

Serial

Hot-endtemperature

control


control


control

Fanspeed

control

ADCMeasured

signal

ADCMeasured

signal

ADCMeasured

signal

PWM AControlsignal

Controlsignal

PWM A

Controlsignal

PWM A

Controlsignal

PWM A





119879 = 119861 sdot 119879119877ln (119877119879119877119877) sdot 119879119877 + 119861

(10)


U0

R1

RT UT





sdot (2119873 minus 1) (11)


(12)





Phys

ical

syste

mPh

ysic

al sy

stem

Softw

are s

yste

m

CNC

+

minus

PID

NTC Heating unit

Tact Tset

Tact (Ink)

Tset (Ink)

Te (Ink)

TRk (Ink)

TR



Tlowastact
















CNC 120583C

sht = 200

hts = 200

CNC 120583C

ght

ht = 200




PC 120583C PC 120583C

ghpshp =

hps = hp =

KP KI KD

KPKI KD KPKI KD

















Command Meaning














119879set minus 119879act = 0 (13)


1003816100381610038161003816119879set minus 119879act1003816100381610038161003816 le 119879119879 (14)




Tset

T0

0

Tact(t)

TT

t1 t2 t3 t4

t

T(t)


10 Motion Control









PLC


M109


Sensors Actuators

Set value Enable








12 Conclusion



0

20

40

60

80

100

120

140

160

180

200

220

Time (s)

Tem

pera

ture

(∘C)

0 200 400 600 800 1000





Competing Interests


References
































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










CNC





Sensors Actuators

Motion controller

Sensors Actuators














Heatingunit

Sensor

Hot-endminus

+Tset Te 120583C

TRQ0

Q1 Qz

Tact

Tlowastact




119894 = 119894 =119889119876119894119889119905

(1)




119879119877119896 = 119870P sdot 119879119890119896 + 119870I sdot 119879119860 sdot119896

sum119894=1


119879119860 (2)


119899

sum119894=0

119894 = 0 (3)


119911 =119899

sum119894=2

119894 (4)


0 =1120578119911 (5)



0 = el = 119875el = 119880 sdot 119868 sdot 119863 (6)

119863 = 119879119867119879119875 (7)



119877119879 =119880119879 sdot 11987711198800 minus 119880119879

(8)





Filament




Q0


Fan Hot-endset

120583C R Z3 Tact



Microcontroller

Reference signals

Serial

Hot-endtemperature

control


control


control

Fanspeed

control

ADCMeasured

signal

ADCMeasured

signal

ADCMeasured

signal

PWM AControlsignal

Controlsignal

PWM A

Controlsignal

PWM A

Controlsignal

PWM A





119879 = 119861 sdot 119879119877ln (119877119879119877119877) sdot 119879119877 + 119861

(10)


U0

R1

RT UT





sdot (2119873 minus 1) (11)


(12)





Phys

ical

syste

mPh

ysic

al sy

stem

Softw

are s

yste

m

CNC

+

minus

PID

NTC Heating unit

Tact Tset

Tact (Ink)

Tset (Ink)

Te (Ink)

TRk (Ink)

TR



Tlowastact
















CNC 120583C

sht = 200

hts = 200

CNC 120583C

ght

ht = 200




PC 120583C PC 120583C

ghpshp =

hps = hp =

KP KI KD

KPKI KD KPKI KD

















Command Meaning














119879set minus 119879act = 0 (13)


1003816100381610038161003816119879set minus 119879act1003816100381610038161003816 le 119879119879 (14)




Tset

T0

0

Tact(t)

TT

t1 t2 t3 t4

t

T(t)


10 Motion Control









PLC


M109


Sensors Actuators

Set value Enable








12 Conclusion



0

20

40

60

80

100

120

140

160

180

200

220

Time (s)

Tem

pera

ture

(∘C)

0 200 400 600 800 1000





Competing Interests


References
































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation











119879119877119896 = 119870P sdot 119879119890119896 + 119870I sdot 119879119860 sdot119896

sum119894=1


119879119860 (2)


119899

sum119894=0

119894 = 0 (3)


119911 =119899

sum119894=2

119894 (4)


0 =1120578119911 (5)



0 = el = 119875el = 119880 sdot 119868 sdot 119863 (6)

119863 = 119879119867119879119875 (7)



119877119879 =119880119879 sdot 11987711198800 minus 119880119879

(8)





Filament




Q0


Fan Hot-endset

120583C R Z3 Tact



Microcontroller

Reference signals

Serial

Hot-endtemperature

control


control


control

Fanspeed

control

ADCMeasured

signal

ADCMeasured

signal

ADCMeasured

signal

PWM AControlsignal

Controlsignal

PWM A

Controlsignal

PWM A

Controlsignal

PWM A





119879 = 119861 sdot 119879119877ln (119877119879119877119877) sdot 119879119877 + 119861

(10)


U0

R1

RT UT





sdot (2119873 minus 1) (11)


(12)





Phys

ical

syste

mPh

ysic

al sy

stem

Softw

are s

yste

m

CNC

+

minus

PID

NTC Heating unit

Tact Tset

Tact (Ink)

Tset (Ink)

Te (Ink)

TRk (Ink)

TR



Tlowastact
















CNC 120583C

sht = 200

hts = 200

CNC 120583C

ght

ht = 200




PC 120583C PC 120583C

ghpshp =

hps = hp =

KP KI KD

KPKI KD KPKI KD

















Command Meaning














119879set minus 119879act = 0 (13)


1003816100381610038161003816119879set minus 119879act1003816100381610038161003816 le 119879119879 (14)




Tset

T0

0

Tact(t)

TT

t1 t2 t3 t4

t

T(t)


10 Motion Control









PLC


M109


Sensors Actuators

Set value Enable








12 Conclusion



0

20

40

60

80

100

120

140

160

180

200

220

Time (s)

Tem

pera

ture

(∘C)

0 200 400 600 800 1000





Competing Interests


References
































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation











Filament




Q0


Fan Hot-endset

120583C R Z3 Tact



Microcontroller

Reference signals

Serial

Hot-endtemperature

control


control


control

Fanspeed

control

ADCMeasured

signal

ADCMeasured

signal

ADCMeasured

signal

PWM AControlsignal

Controlsignal

PWM A

Controlsignal

PWM A

Controlsignal

PWM A





119879 = 119861 sdot 119879119877ln (119877119879119877119877) sdot 119879119877 + 119861

(10)


U0

R1

RT UT





sdot (2119873 minus 1) (11)


(12)





Phys

ical

syste

mPh

ysic

al sy

stem

Softw

are s

yste

m

CNC

+

minus

PID

NTC Heating unit

Tact Tset

Tact (Ink)

Tset (Ink)

Te (Ink)

TRk (Ink)

TR



Tlowastact
















CNC 120583C

sht = 200

hts = 200

CNC 120583C

ght

ht = 200




PC 120583C PC 120583C

ghpshp =

hps = hp =

KP KI KD

KPKI KD KPKI KD

















Command Meaning














119879set minus 119879act = 0 (13)


1003816100381610038161003816119879set minus 119879act1003816100381610038161003816 le 119879119879 (14)




Tset

T0

0

Tact(t)

TT

t1 t2 t3 t4

t

T(t)


10 Motion Control









PLC


M109


Sensors Actuators

Set value Enable








12 Conclusion



0

20

40

60

80

100

120

140

160

180

200

220

Time (s)

Tem

pera

ture

(∘C)

0 200 400 600 800 1000





Competing Interests


References
































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










Phys

ical

syste

mPh

ysic

al sy

stem

Softw

are s

yste

m

CNC

+

minus

PID

NTC Heating unit

Tact Tset

Tact (Ink)

Tset (Ink)

Te (Ink)

TRk (Ink)

TR



Tlowastact
















CNC 120583C

sht = 200

hts = 200

CNC 120583C

ght

ht = 200




PC 120583C PC 120583C

ghpshp =

hps = hp =

KP KI KD

KPKI KD KPKI KD

















Command Meaning














119879set minus 119879act = 0 (13)


1003816100381610038161003816119879set minus 119879act1003816100381610038161003816 le 119879119879 (14)




Tset

T0

0

Tact(t)

TT

t1 t2 t3 t4

t

T(t)


10 Motion Control









PLC


M109


Sensors Actuators

Set value Enable








12 Conclusion



0

20

40

60

80

100

120

140

160

180

200

220

Time (s)

Tem

pera

ture

(∘C)

0 200 400 600 800 1000





Competing Interests


References
































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










PC 120583C PC 120583C

ghpshp =

hps = hp =

KP KI KD

KPKI KD KPKI KD

















Command Meaning














119879set minus 119879act = 0 (13)


1003816100381610038161003816119879set minus 119879act1003816100381610038161003816 le 119879119879 (14)




Tset

T0

0

Tact(t)

TT

t1 t2 t3 t4

t

T(t)


10 Motion Control









PLC


M109


Sensors Actuators

Set value Enable








12 Conclusion



0

20

40

60

80

100

120

140

160

180

200

220

Time (s)

Tem

pera

ture

(∘C)

0 200 400 600 800 1000





Competing Interests


References
































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










Tset

T0

0

Tact(t)

TT

t1 t2 t3 t4

t

T(t)


10 Motion Control









PLC


M109


Sensors Actuators

Set value Enable








12 Conclusion



0

20

40

60

80

100

120

140

160

180

200

220

Time (s)

Tem

pera

ture

(∘C)

0 200 400 600 800 1000





Competing Interests


References
































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










0

20

40

60

80

100

120

140

160

180

200

220

Time (s)

Tem

pera

ture

(∘C)

0 200 400 600 800 1000





Competing Interests


References
































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation




















RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation









Research Article An Architecture for Hybrid Manufacturing...

Documents

Transcript of Research Article An Architecture for Hybrid Manufacturing...