PG5371PA frame 5 turbine running parameter fluctuation problem

we are running frame 5 turbine (PG5371PA) in isochronous mode at load 16.96 MW. Turbine connected load is constant as plant load is constant. plant load varies up to 50-200 KW. base load rating is 22.4 MW. All running parameter was stable. At Mark VIe panel, "Back up reference Temp active " alarm appeared yesterday and turbine running parameter starts hunting.

load varies from 16.96 to 15.8 MW. but TNR value was constant 102.8. no change in TNR when speed is varying.
speed hunting from 5123 to 5092. TNH varies from 99.7 to 100.4.
fsr varies from 50 to 42
TTXM varies to 528 to 512
FSRT, FSRACC, FRSN also varies. FSRmin, FSRSD, FSRSU was constant.
P2 pressure is very unstable it varies from 16.4 to 12.4 while FPRG was 14.8. gas fuel supply pressure is stable. Hydraulic oil pressure is stable.
CPD was also changes as IGV changes.
Generator frequency also changes as speed changes.
TTRXP-537, TTRXS- 550.6, TTKI-520 degree before hunting.

As parameter hunting starts
we put IGC Temp control mode put Off, but no change in hunting
we switch GT to droop control from Isochronous mode, but no change in parameter hunting again we switched to isochronous mode.
we check SRV/GCV in field. they are also fluctuating.

Please suggest for checking of scope and reason of the problem.
 
Let me try to take a stab at this and then CSA can correct me where I go wrong. (No doubt that he is smarter than myself) I'm kind of a Mk-V guy so my references are to Mk-V units and tools. I hope you can translate to Mk-VIe.

The alarm "Backup reference temp active" alarm means that TTRXP (primary temp control reference) is higher (and therefore not in control) than TTRXS (secondary temp control reference). Since TTRXP is generated using the value of CPD it is possible that this is caused by rapid changes in the pressure transmitter output. These changes might be a cause of the swings or they might be a result. Couple of questions, 1) do you have triple redundant CPD transmitters or just one? and 2) when these swings happen do you bump into temperature control?

On a Mk-V unit that I have the CSP on, when in Isochronous, the value of TNR is not used for any active control. It's the value of TNR1 that controls speed. Again, in the CSP that I have, the max and min for TNR1 is 100.17% and 99.83%. You are going well above the max (5123 rpm = 100.45%) so I would not believe the speed control is driving this. You can check these Max & Min values on your unit (should be something like TNKRIMX & TNKRIMN).

The P2 pressure being that unstable may be suspicious, but P2 pressure will vary with rapid changes in GCV positions (FSR) so again it may be the cause or the effect. Those are big changes, though.

My initial suggestion would be to run a View2 file looking at P2 pressure, S/SRV command and feedback, CPD pressure, FSR, FSRN, FSRT, DWATT. Run it at a fast rate, like 4 times per second and try to see if the S/SRV command is leading or lagging the P2 pressure and same with CPD. Also, does FSRT take over FSR at any time and is it a leading or lagging item.

I hope this helps. It sounds like you have something maybe with a bad connection or failing transmitter that is giving the unit a kick. Good luck.
Walt
 
Let me try to take a stab at this and then CSA can correct me where I go wrong. (No doubt that he is smarter than myself) I'm kind of a Mk-V guy so my references are to Mk-V units and tools. I hope you can translate to Mk-VIe.

The alarm "Backup reference temp active" alarm means that TTRXP (primary temp control reference) is higher (and therefore not in control) than TTRXS (secondary temp control reference). Since TTRXP is generated using the value of CPD it is possible that this is caused by rapid changes in the pressure transmitter output. These changes might be a cause of the swings or they might be a result. Couple of questions, 1) do you have triple redundant CPD transmitters or just one? and 2) when these swings happen do you bump into temperature control?

On a Mk-V unit that I have the CSP on, when in Isochronous, the value of TNR is not used for any active control. It's the value of TNR1 that controls speed. Again, in the CSP that I have, the max and min for TNR1 is 100.17% and 99.83%. You are going well above the max (5123 rpm = 100.45%) so I would not believe the speed control is driving this. You can check these Max & Min values on your unit (should be something like TNKRIMX & TNKRIMN).

The P2 pressure being that unstable may be suspicious, but P2 pressure will vary with rapid changes in GCV positions (FSR) so again it may be the cause or the effect. Those are big changes, though.

My initial suggestion would be to run a View2 file looking at P2 pressure, S/SRV command and feedback, CPD pressure, FSR, FSRN, FSRT, DWATT. Run it at a fast rate, like 4 times per second and try to see if the S/SRV command is leading or lagging the P2 pressure and same with CPD. Also, does FSRT take over FSR at any time and is it a leading or lagging item.

I hope this helps. It sounds like you have something maybe with a bad connection or failing transmitter that is giving the unit a kick. Good luck.
Walt
Thanks waltrobinson sir for quick response
1) do you have triple redundant CPD transmitters or just one?- we have only single CPD transmitter
2) when these swings happen do you bump into temperature control? No
3) value of TNKRIMX & TNKRIMN are 100.2 & 99.8 respectively
 
The P2 pressure reference is a function of the turbine speed, so if the turbine speed is unstable then the P2 pressure reference is going to be unstable and therefore the SRV position--and probably the GCV position, as well--is (are) going to be unstable. It sure seems like though that the P2 pressure is varying a LOT more than the speed feedback is varying--unless one of the speed pick-ups is varying more than the others (which should be evidenced by one or more Diagnostic Alarms if it's bad enough).

I don't have access to any application code or CSP for the "Back-up Reference Temp Active" alarm at this writing, but if the alarm is in (ALARM) at all times and not dithering (changing quickly from ALARM to NORMAL to ALARM to NORMAL to ALARM to NORMAL to ...) then it should be relatively easy to troubleshoot. What's not known is WHAT EXACTLY is the back-up exhaust temperature control reference based on.?.?.? MW or FSR or ??? And, it should be possible to see what is happening to the primary exhaust temperature control reference--which is based on CPD (or CPR, which is a fancy CPD which uses ambient (barometric) pressure to derive CPR from CPD and varying ambient conditions). It would be helpful to know if the primary or back-up temperature control reference is changing, or both.

I'm a little suspect of the value provided for TTKI_[nn]. I don't think I've ever seen a value of Isotherm that's less than the running value of TTXM.... Something seems off.

Anyway, that's all I can add at this point. To my way of thinking, if the unit is on Isoch Speed Control and speed is NOT stable then either there's some large load in the captive plant which is not stable or something is causing the fuel flow-rate to be unstable. And, if the Mark VIe was bumping into and out of temperature control (primary or back-up) that could cause problems.

It would be interesting to see if using Manual FSR Control (sometimes called FSR GAG) would have any effect on speed/frequency and P2 pressure/SRV stability. Manual FSR Control is a way of reducing FSR (and therefore GCV position) to try to get it to stabilize to try to determine why it might be unstable. Of course, doing this on an a machine running in Isoch is going to cause the frequency (speed) to decrease for the duration of the test. If the GCV doesn't settle down and stabilize it could be a problem with servo-valve (it doesn't sound like it is, and I hate to offer that as a possible problem) OR if the GCV has two LVDTs then it could be a problem with the feedback from one of the LVDTs that is fluctuating more than usual with small changes in position (an indication of a worn LVDT and/or imminent failure).

Hope this helps!
 
The P2 pressure reference is a function of the turbine speed, so if the turbine speed is unstable then the P2 pressure reference is going to be unstable and therefore the SRV position--and probably the GCV position, as well--is (are) going to be unstable. It sure seems like though that the P2 pressure is varying a LOT more than the speed feedback is varying--unless one of the speed pick-ups is varying more than the others (which should be evidenced by one or more Diagnostic Alarms if it's bad enough).

I don't have access to any application code or CSP for the "Back-up Reference Temp Active" alarm at this writing, but if the alarm is in (ALARM) at all times and not dithering (changing quickly from ALARM to NORMAL to ALARM to NORMAL to ALARM to NORMAL to ...) then it should be relatively easy to troubleshoot. What's not known is WHAT EXACTLY is the back-up exhaust temperature control reference based on.?.?.? MW or FSR or ??? And, it should be possible to see what is happening to the primary exhaust temperature control reference--which is based on CPD (or CPR, which is a fancy CPD which uses ambient (barometric) pressure to derive CPR from CPD and varying ambient conditions). It would be helpful to know if the primary or back-up temperature control reference is changing, or both.

I'm a little suspect of the value provided for TTKI_[nn]. I don't think I've ever seen a value of Isotherm that's less than the running value of TTXM.... Something seems off.

Anyway, that's all I can add at this point. To my way of thinking, if the unit is on Isoch Speed Control and speed is NOT stable then either there's some large load in the captive plant which is not stable or something is causing the fuel flow-rate to be unstable. And, if the Mark VIe was bumping into and out of temperature control (primary or back-up) that could cause problems.

It would be interesting to see if using Manual FSR Control (sometimes called FSR GAG) would have any effect on speed/frequency and P2 pressure/SRV stability. Manual FSR Control is a way of reducing FSR (and therefore GCV position) to try to get it to stabilize to try to determine why it might be unstable. Of course, doing this on an a machine running in Isoch is going to cause the frequency (speed) to decrease for the duration of the test. If the GCV doesn't settle down and stabilize it could be a problem with servo-valve (it doesn't sound like it is, and I hate to offer that as a possible problem) OR if the GCV has two LVDTs then it could be a problem with the feedback from one of the LVDTs that is fluctuating more than usual with small changes in position (an indication of a worn LVDT and/or imminent failure).

Hope this helps!
Thanks for quick response
Dear Sir
Back up reference active alarm is based on FSR and Primary Exhaust Temp control is based on CPD. both value was changes as CPD and FSR changed. "Back up temp reference active" alarm changes quickly from alarm to normal and normal to alarm. but when IGV temp control mode put off then this alarm get normal. but turbine speed hunting was same.

The value of isothermal TTKI is 520 which is lower than TTRXP (542 degree at CPD 6.5 kscg) and TTRXS (553 degree at 46% FSR). IGV is maintaining 520 degree C. now IGV opening is 58 degree and TTXM value is 520 degree.

IGV, GCV and SRV have two LVDT each. we check IGV LVDT feedback during problem both was same. we did not check GCV and SRV LVDT feedback.

There are six speed sensor. 03 sensor are for speed monitoring and 03 sensor are for controlling

All running parameter get stable after 23 hours itself. We did not do nothing on field and control system. We still not found the cause for the problem.
 
Hello again,
Looking at a Mk-V CSP, the TTRX value is the minimum of TTRXS, TTRXP & TTKn_I, so I believe the TTXM values we are seeing are OK and from what you are saying above it looks like TTXM is trying to follow the isothermal line.

Another question; when the instability happens is it a consistently repeating cycle or continuously varying? If it is not consistently varying then you might be looking at a failing component. If it is consistent, then it might be a stability issue.
I would suggest, again, reviewing all the constant settings between this unit and the others and note the differences. If a valve stability setting constant has changed or even the mm valve opening vs FRS calibration has changed, the whole loop of SRV / GCV / IGV / TTXM relationships changes and the combination could become unstable.

I have seen liquid systems become unstable due to valve calibration where the valve was only stroking part way after a rebuild. After calibration the relationship of FSR to valve stroke was very off-spec and the valve position during operation became unstable. I know we're talking liquid systems vs gas systems but I believe the principal is the same.

I second CSA's recommendation to try the FSR Gag when the unit goes unstable. This will limit the upper range of GCV travel (limit Max FSR), and doing this can possibly stop the swinging. At that time you might be able to observe one thing that is moving that shouldn't be and therefore find your culprit.

One last question; was there any changes made between the time this unit did not have this issue and now?

Again, good luck.
 
Hello again,
Looking at a Mk-V CSP, the TTRX value is the minimum of TTRXS, TTRXP & TTKn_I, so I believe the TTXM values we are seeing are OK and from what you are saying above it looks like TTXM is trying to follow the isothermal line.

Another question; when the instability happens is it a consistently repeating cycle or continuously varying? If it is not consistently varying then you might be looking at a failing component. If it is consistent, then it might be a stability issue.
I would suggest, again, reviewing all the constant settings between this unit and the others and note the differences. If a valve stability setting constant has changed or even the mm valve opening vs FRS calibration has changed, the whole loop of SRV / GCV / IGV / TTXM relationships changes and the combination could become unstable.

I have seen liquid systems become unstable due to valve calibration where the valve was only stroking part way after a rebuild. After calibration the relationship of FSR to valve stroke was very off-spec and the valve position during operation became unstable. I know we're talking liquid systems vs gas systems but I believe the principal is the same.

I second CSA's recommendation to try the FSR Gag when the unit goes unstable. This will limit the upper range of GCV travel (limit Max FSR), and doing this can possibly stop the swinging. At that time you might be able to observe one thing that is moving that shouldn't be and therefore find your culprit.

One last question; was there any changes made between the time this unit did not have this issue and now?

Again, good luck.
Hello sir
Thanks for reply
1. Sir you are right, TTXM is following isothermal line.
2. Now our GTG is stable. If unit goes unstable we will try FSR gag for observation
3. NO any changes made.
we uprate our GTG in 2017. then some constant were changed by BGTS. GTG was running normal upto now (approx 03 year). Before uprate, FPRG value was 13.2 and after uprate it is 14.8 kscg. There was no issue in last three year.
Thanks
 
Top