Experiencing high exhaust temperature spread during load ramp up between 18 MW and 21 MW. High differential temperature for wheel space 1st stage aft 1 & 2.
We had been facing high exhaust temperature spread on our Hitachi H-25 gas turbine since August; therefore, HGPI was undertaken in October. However, high exhaust temperature spread observed again on machine start up.
A comprehensive presentation with pictorial observations and start up trends can be accessed from the dropbox link:
Summary is as follows.
Major observations during HGPI were:
- Deposition of molten cross-fire tube on 1st stage nozzle
- Splatter deposition on 1st stage bucket
- Extensive TBC / Parent loss on 2nd stage buckets (leading edge and trailing edge)
- No observations found on fuel gas lines, fuel nozzles, liners and transition pieces.
- Cross-fire tubes and retainers replaced
- Transition pieces cleaned
- 1st stage nozzles cleaned and cooling holes de-choked
- 9 segments of 2nd stage nozzles (from upper half) repaired
Observations on S/U:
- Combustion Alarm (High exhaust temperature spread) observed from 18.5 MW to 20 MW. After 21.5 MW, first exhaust spread is generally lower than allowable spread; however sometimes exceeds the allowable spread.
- Wheel space temperature alarm (High 1st Stage AFT wheel Space temperature) appeared above 18.5 MW. Delta between right and left thermocouple is about 85 Celsius
1) What is the possible reason for high temperature spread between 18 & 21 MW? What can be done to rectify it?
2) Why is the first exhaust spread fluctuating at high loads?
3) What could be possible reasons for large difference in wheel space 1st stage aft # 1 & 2 and what are the consequential damages?
4) Does anyone has a swirl chart for H-25? If yes, please share.
5) Do you think this is related to combustion issue? For your information, All fuel nozzles installed during box up are clean with no choking. New cross-fire tubes and retainers installed. No damage on liners or transition pieces. Fuel gas composition is okay too.
6) How is the allowable temperature spread calculated for Hitachi Turbine (H-25) with Mark VI control system?
Looking forward to hearing from you guys soon.
I have read your problem statement and you explained it well. Before i go for detail analysis, tell me one thing have you guys replaced the "cross fore tube metal seals"?
Thank you, Abu bakar, for a response.
>I have read your problem statement and you explained it
>well. Before i go for detail analysis, tell me one thing
>have you guys replaced the "cross fore tube metal seals"?
6 of 20 cross fire metal seals were found worn out. Unfortunately, we did not have any in spare hence subject part was not replaced.
I am unable to view the files in the link; my anti-malware software is complaining LOUDLY there is some problem with the file(s). So, sorry if some of my questions could have been answered by reviewing the link, but I'm unwilling to bypass the warning and this particular anti-malware won't let me quarantine files to look at them.
Is the spread high (more than approximately 35 deg C) before and after the 18-21 MW load range, or does it just spike during the 18-21 MW load range?
Was the wheelspace temperature differential high before the spread problem started?
Is the hottest wheelspace reading much more than expected, or is the coldest wheelspace temperature reading much lower than expected? In other words, is the differential because of a higher than normal reading or a lower than normal reading?
It's probably not likely the wheelspace differential is related to the cause of the spread problem, but that's only my personal opinion. The most likely cause of a high wheelspace temperature differential is improper T/C insertion, and/or improper 2nd stage nozzle assembly.
The problem of metal deposition on the first stage nozzles/buckets suggests one of two things. Either there is a serious combustion problem causing flow through the cross-fire tubes resulting in melting of the cross-fire tube material, or there is a material problem (wrong or inadequate cross-fire tube material was used by the supplier of either the tubes or the retainers/clips). Neither is a good condition.
Combustion troubles, if severe enough, result in flow of hot combustion gases from adjacent combustors into the combustor which has a lower pressure due to poor combustion. On units with non-DLN combustors it's very rare (except on liquid fuel operation) for there to be excessive fuel flowing into one or more combustors--but it can happen, and for gas fuel operation the cause is the opposite of "choking"--it's excessive fuel flow caused by improperly matched fuel nozzles or leaking fuel nozzle components because of poor assembly techniques. For liquid fuel operation the most common cause of a higher fuel flow to one or mor combustors is a broken flow divider shaft or gear mechanism.
The other possibility is that there is an instrumentation problem with either the exhaust thermocouples or the T/C input card. Are the problem exhaust T/Cs and the problem wheelspace T/C connected to the same T/C input card? Are there any Diagnostic Alarms (please list all Diagnostic Alarms, even if it's felt some or all might be irrelevant). If the unit exhausts into an HRSG, it's possible there is a problem in the exhaust duct transition piece causing a flow issue resulting in a temperature differential which is not related to a true combustion problem--but the physical damage suggests otherwise (unless it's a "simple" materials issue). It's also possible, but unlikely the exhaust T/C radiation shield(s) might be bent or blocked by some material (like insulation which has blown out from the flexible seals or from behind exhaust duct plates and is blocking flow through the radiation shields); but again the physical damage suggests otherwise.
However, cross-fire tube metal deposition strongly suggests a true combustion problem, as does the 'Combustion Trouble' Process Alarm. Or, a materials problem. Loss of TBC (Thermal Barrier Coating) suggests higher than normal hot combustion gas temperatures impinging on the area where the TBC was lost, or insufficient air flow for cooling, or FOD (Foreign Object Damage--loose materials impinging on the component causing the TBC to crack and break off).
Was the metal deposition primarily downstream of one or two combustors, or was it mostly uniform around the nozzles? If it was primarily on one or two nozzle segments that would point to the combustor(s) or cross-fire tubes experiencing the problem. If it was uniform then that would suggest, to me, a materials problem.
What does Hitachi say about the problem?
It's very likely the allowable exhaust temperature spread is calculated using the same block/algorithm as GE-design heavy duty gas turbines use--which means it relies heavily on properly reading/feedback from the compressor discharge temperature thermocouples. Are they (the compressor discharge T/Cs) working correctly? It should be possible to right-click on the Spread Monitor block in Toolbox and select 'Block Help' to bring up a written description of the block's functions. Sometimes it's helpful; sometimes it's not. But it's always worth trying.
Again, my apologies if some or most of my questions were "covered" in the link.
But it certainly sounds like there is a true combustion problem, and/or a materials problem. Please write back to let us know what you find.
Thank you for the detailed response and apologies for the malware issue. The files shared were a power point presentation and an excel spreadsheet. I have converted them into a pdf and uploaded on one-drive. Sharing the link again. Hope it will work this time.
Find, below, a point wise response to your queries.
1) 1st actual spread has always been greater than 35 deg at all loads
2) The wheel space temperature differential was not this high before the HGPI. There was a differential of 12 degrees which is now running at ~ 70-80 deg
3) Hottest wheel space reading is much more than expected i.e. 1st stage aft # 1 is ~ 530 deg versus 1st stage aft # 2 at 445 deg
4) I understand there shouldn't be any flow of hot gases through cross-fire tubes during normal operation. However, we have observed marks of local hot spots on our cross-fire tubes during inspection. What do you think could be the reason for this? Could worn out cross fire tube seal plates be a reason? When you mention combustion issue, what does it exactly entail? For the record, all fuel nozzles installed during box up are clean with no choking. New cross-fire tubes and retainers installed. No damage on liners or transition pieces. Fuel gas composition is okay too. IGV / GCV filters inspected and found okay as well.
5) We are using natural gas as fuel. The fuel nozzles are verified for correct matching in every box up and all fuel lines are checked for choking as well. No major observation identified in this area
6) The unit discharges into an HRSG. Duct inspection was carried out and no abnormality found
7) T/Cs radiation shields are open - no clogging found
8) The metal deposition was generally uniform throughout the nozzle. Pictures of the nozzles can be found in the one drive link.
9) Since we boxed up the machine with partial repairs 2nd stage nozzle, Hitachi is of the opinion we need to change both 1st stage shroud (shrouds observed to be worn out in last HGPI) and 2nd stage nozzles. OEM also recommends checking the fuel supply system / path and nozzles for choking or contaminants.
Additionally, you mentioned "improper 2nd stage nozzle assembly". What does it specifically entail that could cause high wheel space temperature?
About the diagnostic alarms, will need to check from the manual/instrument engineer. Will get back to you as soon as I have the required info.
Thanks for the valuable opinion. Really appreciate your help.
use following formulae:
i derived it myself when Hitachi gave its own data. they definitely will not share the swirl angle chart
y = -0.0603x2 - 2.8594x + 150.49
y is angle and x being load in MW
identify the nozzle with same; check for check-valve. we replaced all HSD check valve to ensure no passing and cleaned our nozzles. we were able to get rid of our problems.
how many H-25 do you have?