RLNG Pressure Controllers of Gas turbines

S

Thread Starter

Seeker1988

We have five 30MW Frame6 Gas Turbines. One of the Gas Turbine's exhaust temperature is higher even at low loads (more than 590 degC at 21MW). Its exhaust air temperatures are higher; it takes more fuel (400kg/hour more as compared to other GTs at the same load of 21MW); and the RLNG pressure controllers have unusual opening (6833B valve is 65% open and 6833A is 40% open).

What could be the root cause of the problem?
Should we focus on the controller? Even if the 6833B controller is stuck at 65% opening, still the opening is high for 21 MW. What do you people suggest?
 
Seeker1988,

What Process Alarms and Diagnostic Alarms are active on the affected unit?

When did this problem start--after a maintenance outage, or after a trip from load?

I can kind of determine what RLNG is (some flavour of Liquified Natural Gas?). But 6833A and 6833B are completely new valve designations for me. Are you referring to a Stop-Ratio Valve and a Gas Control Valve assembly--two gas fuel control valves in a single cast assembly?

What kind of turbine control system does the affected turbine use? GE Speedtronic Mark IV, Mark V, Mark VI, or Mark VIe?

If the power output of the generator is lower than normal and the fuel flow to the turbine is higher than normal and the exhaust temperatures are higher than normal that suggests a very dirty axial compressor and IGVs, and/or dirty inlet air filters, or a possible problem with one or both of the compressor bleed valves not being fully closed.

But if you can explain the nomenclature in terms that we can understand what you're referring to, we might be able to offer more or better assistance.

Please write back with clarification and let us know how long it's been since the affected unit has had a thorough off-line compressor walking, what the differential pressure across the inlet air filters is (and what the average differential pressure is across the inlet filters of the other five units is), and what the positions of the two compressor bleed valves are of the affected unitb(by visually observing the compressor bleed valve positions).

Looking forward to hearing back from you, Seeker1988!
 
S
Sorry for not pointing out the details.

I am referring to the off-base RLNG (regassified LNG) pressure control valves. The Two pneumatic control valves (6833A and 6833B) operate in cascade mode. When the common controller is given any command/set point, first 6833B (30%) opens and when it opens fully 6833A (70%) opens to accommodate the increased flow.

In other GTs 6833B is open like 35% and 6833A is closed. But this GT 6833B is 65% open and 6833A is 30% open.

The last water wash was done on April 2016.
Air filter dp is not high.

We have MKV controller; frame 6 30MW GT. There are no alarms related to this. But normally TTXM of GT reduces when loading is reduced and at around 20MW it is around 530decC. But for the concerned GT, it is as high as 590degC. This is no critical condition as TTXM still is following TTRX. But the exhaust TC readings are still high at 20MW when compared to other GTs. I suppose this is due to opening of both the off base control valves. Fuel consumption is higher in this GT.

PS: I don't know if we can attach any picture with posts here else I would have attached one related to this.

<b>Moderator's Note:</b> You cannot attach pictures to messages on Control.com. I suggest you use a download site like dropbox.
 
Seeker1988,

What is the difference in load <b>and</b> exhaust temperature between the affected unit and one of the units which is operating more efficiently?

Are any of the units being operated in Combined Cycle mode (that is, IGV Exhaust Temperature Control ON)? If so, which ones if not all of them?

What are the exhaust temperature spreads of the affected unit (TTXSP1, TTXSP2, TTXSP3)?

Do the unit exhaust in HRSGs? Do the units have a by-pass stack, and if so, does the affected unit run better in by-pass (Simple Cycle) mode than in Combined Cycle mode through the HRSG?

Are there any exhaust duct pressure sensors (transmitter(s)) for the exhaust duct? If so, how do the reading(s) from the affected unit differ from those of the more efficiently running units? Could there be some flow restriction in the exhaust? (It's like trying to exhale through several thicknesses of cloth if there is a high restriction of exhaust flow--it reduces the efficiency of the gas turbine.)

Again, have you checked the compressor bleed valves? What is the CPD of the affected unit versus one of the units which is running more efficiently (at the same load, say 20 MW)?

What is the difference in IGV angles at the same load, say 20 MW, between the affected unit and one of the unit is running more efficiently?

What is the inlet filter pressure drop?

Do the unit have conventional (diffusion flame) combustors, or DLN-I combustors?

I don't personally have any experience with a re-gassifier skid and controls. I have seen three different skids--all provided by different sources (one by GE; two by the EPC, via low bid purchasing). None of them seemed to work all that well, but they were out of my scope; all I had to do was prove the problem with pressure/flow stability was not being caused by the gas turbine control system.

How long since the affected unit had a maintenance outage (CI; HGPI; Major)? What was the last maintenance outage (CI; HGPI; Major)?

What Process- and Diagnostic Alarms are being annunciated by the Mark V of the affected unit when the unit is running? (This is important--even if some think the alarms may not be relevant.)

What is the P2 pressure of the affected unit versus the P2 pressure(s) of the more efficiently running units?

Are all the unit paralleled to the same grid/load? Is the frequency stable?

It seems like I'm asking for a lot of data--but, based on the information provided it's difficult to say what might be the problem. There are a lot of possibilities--hence the number of questions and the amount of data requested. But, without knowing more all that can really be said is that reduced output with increased fuel flow, or increased fuel flow for the same output, usually means there's something amiss mechanically with the unit. Either the nozzles/turbine buckets are very worn, there's a loss of air flow through the machine (leaking compressor bleed valves; dirty compressor; increased exhaust duct back pressure). My best guess is that the increased RLNG valve opening is the result of decreased performance/efficiency of the gas turbine, and/or its axial compressor, or the exhaust. It's likely not something amiss with the gas skid--it's supplying the fuel necessary for the unit to maintain load.
 
S
At 23.5 MW, exhaust temperature is 590degC for affected GT and 551 of unaffected GT. Spread is 24/23/22 and 22/21/19 rspt. Exhaust duct pressure is 160 mmWC and 220mmWC; IGV opening is 63DGA and 57DGA; CPD is 7.4 and 7.6 rspt; P2 pressure is 17.7 in both the cases; both the GTs have DLN type combusters; Filter pressure drop are 1.6 and 1.7 inches of WC.

Last CI of affected unit was done in Jan 2014; HGPI in July 2015 and MI in July 2012.

No diagnostic alarms are existing. Some alarms were there, but they had normalized.

The unit exhaust goes to HRSG. But the unit are never run in Simple cycle mode. So, I cannot compare the performances in the two modes. All the units run in parallel and frequency is normal.

The matter came to light when the unit had high exhaust temperature at low load. Due to load constraints all units were running at low loads.

PS: I was on a long leave and so there was quite delay in the reply.
 
Seeker1988,

Interesting the differences in the exhaust duct back pressure, and the differences in IGV angles, and the difference in CPD values. I would expect with a higher IGV angle (more air flow) the CPD would be higher and the exhaust duct back pressure would be slightly higher--not lower by 60 mmWC. At this point I have to suspect instrumentation--not the exhaust T/Cs, but the CPD transmitter(s) and the exhaust duct back pressure transmitter(s). The accuracy of the IGV LVDT calibration may also be suspect.

UNLESS the axial compressor is dirty. There's off-line water washing, and then there's off-line water washing. The difference is usually the effectiveness of the wash. If the unit wasn't properly cleaned, and/or wasn't properly rinsed (which is usually the case--because people usually use too much detergent), then the compressor can foul faster than normal. And, even if a good rinse is done sometimes the compressor requires a second wash, perhaps with a longer rinse. When performing water washes it's always important to visually inspect the IGVs and first couple of stages of the axial compressor. Many sites also use a borescope to check later stages of the compressor for cleanliness.

I RARELY suggest checking the calibration of instrumentation, but something sure seems amiss with this set of data.

What is TNR and TNH for the two machines?

What is TTRX and TTXM for the two machines?

What is CSRGV_TEMP and TTRXGV for the two machines?
 
S
>I RARELY suggest checking the calibration of instrumentation, but something sure seems amiss with this set of data.
>
>What is TNR and TNH for the two machines?
>
>What is TTRX and TTXM for the two machines?
>
>What is CSRGV_TEMP and TTRXGV for the two machines?

Hello CSA,
For the power load of 23.5 MW, the TTRX of the affected GT was 590 C and TTXM also 590 C; the values for other machine are 586C and 551C.
For other parameters you asked for, I don't have data at 23.5 MW but at present, when the affected GT is running at 28.7 MW, TNR is 102.7%, TNH is 100%; CSRGV_temp is 82.9DGA and TTRXGV is 569 C.

For healthy one (running at 29.7MW) these values are 102.6%, 100%, 75.2 DGA and 566 respectively. TNR of the affected machine is higher even at lower load!

The offline water wash of the affected machine was done 7 months ago while the healthy one had underwent WW 4 months ago. Would this add to the observed difference?
 
Top