Steam Turbine Problems and Corrosion Monitoring

5/25/2018 Steam Turbine Problems and Corrosion Monitoring

1/11

1

STEAM TURBINE PROBLEMS AND

THEIR FIELD MONITORING

Otakar Jonas, P.E., Ph.D.and

Joyce M. Mancini

Jonas, Inc.1113 Faun Road, Wilmington, DE 19803 USAPhone: (302) 478-1375, Fax: (302) 478-8173

Email: [email protected], Web: www.mindspring.com/~jonasinc

ABSTRACT

This paper describes typical turbine problems and the monitoring techniques developed and usedfor the determination of corrosive conditions, corrosive characteristics of materials, exfoliatedoxides causing solid particle erosion, and deposit buildup in steam turbines and other steamhandling equipment (utility and industrial). Available instruments include the Steam TurbineDeposit Collector/Simulator, Converging-Diverging Nozzle for LP Turbines, Converging Nozzlefor HP Turbines, Drying Probe for Wet Steam Stages, Boiler Carry-over Monitors, EarlyCondensate Samplers, Particle Flow Monitor for Exfoliated Oxides, Steam Blow, and WaterDroplets, Water Induction Monitors, and Corrosimeter for Erosion-Corrosion Monitoring. Inaddition, U-bend specimens can be installed inside turbines to determine susceptibility ofdifferent materials to pitting and stress corrosion cracking and fracture mechanics specimens can

be used to determine stress corrosion and corrosion fatigue crack growth rate and crackincubation times.

TURBINE PROBLEMS(1 to 10)

Typical turbine problems include pitting corrosion, corrosion fatigue, stress corrosion cracking,deposit buildup, solid particle erosion, and water droplet erosion of the wet stages. Turbinecomponent failures and loss of efficiency and MWs due to deposits are very costly; the cost oflost production being about ten times higher than the cost of repairs. Figure 1 shows typicalturbine problem locations.


2/11

2

Figure 1. Typical turbine with locations of deposition, corrosion, and related problems

It has been demonstrated by experiment and field experience that up to 15% of MW generatingcapacity loss can be caused by erosion and buildup of blade deposits. In addition, solid particleerosion caused by exfoliation leads to expensive maintenance and blade replacement on top ofthe lost output.

There are thousands of steam cycles worldwide where turbine performance has been impaireddue to deterioration of blade surface finish and/or buildup of turbine deposits. These problemsand opportunities for improvement have been recognized for over 50 years, but today=scompetitive business environment is now providing strong incentives for improvement. Withreplacement power typically at over $100/MWh and costing as much as $7,000 per MWh, thesavings can be very high. For a unit with a copper deposition problem resulting in a 30 MW lossof capacity and a difference between the sale price and the cost of generation of $30 per MWh,the yearly loss is over $5 million.

The current trends of increasing turbine inspection and cleaning intervals to ten years and

extending the warranty period to up to twenty years requires the best achievable control ofturbine deposits, erosion, and corrosion. The root causes of these problems include:

Operation with a high concentration of impurities in turbine steam. It is estimated thatabout 40% of U.S. utility and 60% of industrial turbines operate with high concentrationsof impurities in steam.Marginal design of blades and disks (high steady and vibratory stresses, not consideringimpurity concentration and all appropriate corrosion properties, improper materials)Poor, improper, or inadequate monitoring of steam chemistry. Sodium and cation


3/11

3

conductivity should be monitored as a minimum and should be the routine way to preventdeposition of salts and hydroxides which can lead to corrosion and efficiency and MW loss(11, 12). For troubleshooting and commissioning, additional parameters should bemonitored.

DEPOSIT BUILDUP

The impact of deposits on turbine performance depends on their thickness, their location (steampressure), and the resulting surface roughness. Deposits will change the basic profile of thenozzle partitions resulting in losses caused by changes in energy distribution and aerodynamicprofiles, as well as, by surface roughness effects. These changes can result in large megawattand efficiency losses in both fossil and nuclear turbines in addition to the cost of extra fuel,emissions, turbine washing, boiler chemical cleaning, and maintenance. Table 1 summarizes theprocesses by which deposits are formed.

Table 1. Processes that cause Turbine Deposits

Precipitation from superheated steam and deposition of mineral particles and aciddropletsEvaporation of moisture (liquid films) on surfaces above saturation temperature andretention of the mineral residueDeposition of metal oxides formed elsewhere in the steam cycle (mechanical boilercarry-over, exfoliation in superheater and reheater, etc.)Adsorption of gases and dissolved impurities in superheated and wet steam on oxidizedsurfaces

CORROSIVE ENVIRONMENTS

The steam turbine environment consists of superheated and wet steam with low ppb levels ofdissolved and suspended impurities, deposits which have formed by precipitation and plating-outin the superheated steam sections and drying of moisture on hot surfaces, and of moisture filmand droplets containing concentrated low volatility chemicals. Each of these chemicalenvironments has a different effect on turbine materials and performance. In addition to thereactive chemicals (salts, acids, and hydroxides) that can cause various forms of corrosion, solidimpurities, prevalently magnetite exfoliated from superheaters, reheaters, and steam pipes, cancause solid particle erosion and chemically interact with the reactive chemicals.

Turbine corrosion and impurity behavior zones are shown, together with turbine steamexpansion lines, in the Mollier diagram in Figure 2. The current understanding of the physical-chemical processes occurring in low pressure turbines is summarized in Figure 3. In thesuperheated region, impurities dissolved in steam may precipitate and deposit because theirsolubility in steam decreases as the steam expands. In the Asalt zone@, salts and acids can bepresent as highly concentrated corrosive solutions. Sodium hydroxide can bethermodynamically stable as a concentrated solution anywhere in the superheated steam region.In the wet steam region, impurities distribute between the gas and moisture which, at low pH andhigh velocity, can cause erosion-corrosion. Some surfaces which are surrounded by wet steam


4/11

4

may be heated by heat transfer or flow stagnation which can result in evaporation of moistureand concentration of impurities on the surface.

Figure 2. Mollier diagram for a drum boiler cycle with regions of impurity concentration

and corrosion


5/11

5

Figure 3. Cross section of an LP turbine with the locations of the physical-chemical

processes

NEW MONITORING DEVICES

The field monitoring equipment described in this paper can be used to diagnose and preventmany common problems (13). Table 2 summarizes the devices available and Figure 4 shows the

installation locations for the equipment used in LP turbines. In addition, there are also monitorsavailable to detect vibration, blade and rotor cracking, steam leaks, air inleakage, rotor position,and wear of bearings.


6/11

6

Table 2. Summary of Devices available for Deposit, Corrosion, and Erosion Monitoring

DEVICE APPLICATION MONITORING RESULTS

Steam Turbine DepositCollector/Simulator

Simple troubleshooting tool forcollection and analysis of HP, IP

and LP turbine deposits withoutturbine disassembly

Deposit composition,morphology, and rate of

deposition vs. operation

Converging-DivergingNozzle for LP Turbines

Simulates steam expansion infossil and nuclear low pressureturbines along the blade path

Types of impurities depositingon LP turbine blades andcorrosiveness of theenvironment

Converging Nozzle forHP Turbines

Simulates HP turbine deposition;copper deposits, MW loss, solidparticle erosion

Types of impurities depositingon HP turbine blades

Drying Probe for Wet

Steam Stages

Simulates moisture drying on hot

surfaces in LP turbines; related topitting, SCC and corrosionfatigue

Deposits of low volatility

impurities in LP turbines arecollected

Boiler Carry-overMonitors

Direct monitoring of % moisture Mechanical carry-over

Early CondensateSamplers

Corrosion and flow-acceleratedcorrosion in LP turbines

Chemistry of water dropletsformed in the final stages of theLP turbine

Particle Flow Monitorfor Exfoliated Oxides,

Steam Blow and WaterDroplets

Solid particle and water dropleterosion; also used to monitor

effectiveness of steam blow

Number and size distribution ofoxide particles in superheated

and reheated steam and waterdroplets in wet steam stages

Water InductionMonitors

Liquid water in steam pipes Alarms, % of liquid water insteam pipes

Corrosimeter forErosion-Corrosion(Flow-AcceleratedCorrosion) Monitoring

Erosion-corrosion vs. waterchemistry and operation - pipewall thinning

Thinning rate for materials ofconcern in the specificsuspected areas of piping

U-Bend Specimens Corrosion, stress corrosioncracking

Detect general corrosion,pitting, and stress corrosion

cracking conditionsFracture MechanicsSpecimens

Stress corrosion cracking andcorrosion fatigue

Stress corrosion crack growthrate and crack incubation times


7/11

7

Figure 4. Installation locations of equipment available for LP turbines

Deposits

Several diagnostic tools are available for determining deposition rates and deposit compositionand morphology in turbines. These include:

Steam Turbine Deposit Collector/Simulator- This filter-like device simulates theturbine conditions governing deposition from superheated steam and can be installedanywhere in piping or a turbine (Figure 5).

Figure 5. Steam Turbine Deposit Collector/Simulator installation

Converging-Diverging Nozzle for LP Turbines - Simulates steam expansion along theblade path in fossil and nuclear low pressure turbines to determine the amounts and typesof impurities depositing on LP turbine blades (Figure 6).


8/11

8

Figure 6. Converging-Diverging Nozzle for LP turbines

Converging Nozzle for HP Turbines - Simulates steam expansion in the HP turbinecontrol stages to determine the amounts and types of impurities depositing on HP turbineblades.

The deposits collected with each of these devices can be analyzed using optical and SEMmicroscopy, energy dispersive x-ray spectroscopy, x-ray diffraction, and wet chemistry methods.The results of the deposit analysis can be used to predict deposition and corrosion inside theturbine, and to assess boiler carry-over and the effects of steam and water chemistry upsets.

Wet Steam Environments

Several instruments have been developed which can determine the corrosivity of the wet steam

environment. One such instrument is the Drying Probe for Wet Steam Stageswhich simulatesthe evaporation of moisture and the concentration of low volatility impurities on hot surfaces.The droplets carrying dissolved impurities impact the Probe, evaporate, and leave behind the

impurities as deposits on the Probe surface. By analyzing the deposits, the corrosiveness of theenvironment can be determined.

Two new Boiler Carry-over Monitorshave recently been developed which can be used formonitoring of high boiler carry-over and the % moisture in process and heating steam and thesteam used for oil recovery. One monitor is based on the absorption of gamma radiation and thesecond on on-line calorimetry combined with isokinetic sampling.


9/11

9

Early Condensate Samplersare used to collect samples of steam moisture in turbines. Thecondensate is collected and withdrawn for analysis while saturated steam is vented into theturbine space. Two types of samplers are available: internal and external.

Exfoliation and Solid Particle Erosion

Solid particle erosion in steam turbines is mainly due to exfoliated oxides from the superheater,reheater, and steam piping. The primary effects of solid particle erosion include decreased unitefficiency, increased maintenance costs for steam path inspections and repair/replacement costs,lengthened outages, more frequent inspections for ensuring the mechanical integrity ofcomponents, and increased probability of forced outages as a result of a failure of an erodedsteam path component.

A Particle Flow Monitor for Exfoliated Oxides and Water Droplets is used to determinenumber and size distributions of oxide particles or water droplets in steam. The Monitor probe isinserted into the pipeline and a computer records each particle or droplet as it impacts the probe.

The number of particles/droplets, average mass of particles/droplets, and the mass of eachindividual particle/droplet can all be measured. In addition, the particle/droplet size can becalculated. The data obtained from this monitoring is used to determine the operating conditionswhen erosion is highest and to indicate when chemical cleaning of the superheater or reheatermay be necessary.

Water Induction Monitors - Water induction and accumulation in steam piping and waterreentrainment into the turbine can lead to water hammer and other damage. Three types ofinstruments can be used to quickly detect this condition: gamma gauge, water droplet monitor,and fast response thermocouples.

Corrosion and Erosion

A Corrosimeter for Erosion-Corrosion Monitoringcan be installed in wet steam piping andinside turbines to continuously monitor the thinning rate and the effects of operation and steamchemistry. Several types of this instrument are commercially available. They are based on themeasurement of electrical resistance of a wire loop exposed to the flow and environment.

Properly placed U-bend corrosion specimens(near the saturation line) give a good indication ofturbine corrosion. They can detect general corrosion, pitting, and stress corrosion crackingconditions. Under constant environmental conditions, pitting of turbine materials is proportionalto time

1/3.

At some point in time, depending on stress, corrosion fatigue and/or stress corrosion cracks canstart propagating from the pits, leading to blade and other component failures. Typically, utilityturbine blades may crack from pits about 10 mil deep, but cracking from pits as small as 3 milshas been observed; mostly in LP L-1 blades. Figure 7 shows the installation location of the U-bend in a turbine and Figure 8 is an example of an exposed U-bend.


10/11

10

Figure 7. Location of the U-bend specimens installed in a turbine

Figure 8. Carbon steel U-bend showing general corrosion, pits, and flaky deposits on the

outside surface

Fracture mechanic specimenssuch as 1T-WOL wedge loaded specimens have been used to

determine stress corrosion crack growth rate and crack incubation times. Hydraulically loadedspecimens in autoclaves using turbine steam have been used to determine corrosion fatiguecharacteristics of materials in turbine environments.

CONCLUSIONS AND RECOMMENDATIONS

1. The diagnostic devices described in this paper have all been tested and used introubleshooting of steam systems and/or R&D and have performed as expected.

2. These devices are providing new information on transport and behavior of steam impuritiesincluding:

relationships between bulk steam chemistry and deposition

the rate of deposition and form of precipitates of steam impurities in LP turbinesthe rate of deposition and form of copper in HP turbinestransport, concentration, and size distribution of exfoliated oxideshideout and release of contaminants from superheaters and reheatersdistribution of moisture film on LP turbine bladescomposition of early condensatedrying of moisture on hot surfacesinteraction of oxides with other impurities


11/11

11

number, size distribution, and chemical composition of potential nucleation seeds3. In troubleshooting, the following turbine problems have been investigated with the use of

the new devices:corrosion in the phase transition zone of LP turbinesexcessive deposits leading to loss of capacity and efficiency and corrosion

hideout of corrodents in superheater and reheater after contamination of a boiler byseawatererosion-corrosion in two phase regionsexfoliation and solid particle erosioneffectiveness of chemical cleaning of superheaters and reheaterseffectiveness and monitoring of steam blow

REFERENCES

1. O. Jonas. ASteam Turbine Corrosion.@Materials Performance, February 1985.2. Characterization of Operational Environment for Steam Turbine-Blading Alloys. EPRI.

Palo Alto, CA. 1982. CS-2932.3. A. Whitehead and G.F. Wolfe. ASteam Purity for Industrial Turbines.@Presented at the 8

th

ASME Industrial Power Conference, Houston, TX. October 1980.4. Steam, Chemistry, and Corrosion in the Phase Transition Zone of Steam Turbines. EPRI,

Palo Alta, CA. February 1999. TR-108184.5. T. McCloskey, B. Dooley, and W. McNaughton. Turbine Steam Path Damage: Theory and

Practice. EPRI, Palo Alto, CA. 1999.6. O. Jonas and B. Dooley. ASteam Chemistry and Its Effects on Turbine Deposits and

Corrosion.@Proceedings of 57th

International Water Conference. October 1996.7. O. Jonas, et al. ACopper Deposition and MW Loss: Problem and Its Solutions.@

Proceedings of 57th

International Water Conference. October 1996.

8. O. Jonas. AErosion-Corrosion of PWR Feedwater Piping - Survey of Experience, Design,Water Chemistry, and Materials.@Report for U.S. Nuclear Regulatory Commission,NUREG/CR-5149. ANL-88-23. March 1988.

9. O. Jonas. ANew Continuous In-Line Method for Monitoring of Solid Particles and Resultsof Field Testing.@Proceedings of Solid-Particle Erosion of Utility Steam Turbines: 1985Workshop. EPRI. Palo Alto, CA. August 1986. CS-4683.

10. K.C. Cotton. Evaluating and Improving Steam Turbine Performance. Cotton Fact Inc.Rexford, New York. 1993.

11. Interim Consensus Guidelines on Fossil Plant Cycle Chemistry. EPRI, Palo Alto, CA.June 1986. CS-4629. and other EPRI guidelines.

12. O. Jonas. ADiagnostic Monitoring - an Overview.@Power. January 1992.

13. O. Jonas. AOn-line Diagnosis of Turbine Deposits and First Condensate.@Proceedings of55

thInternational Water Conference. October 1994.

Steam Turbine Problems and Corrosion Monitoring

Documents

Transcript of Steam Turbine Problems and Corrosion Monitoring