6B Gas Turbine tripped by Loss of Flame during startup

E

Thread Starter

ebing

Dear Sir,

We have one 6B Gas turbine duel fuel machine which controlled by Mark V.

It tripped around 90% TNH by Loss of Flame for several times with liquid fuel during startup. What we found in trend is that liquid fuel flow FQL1 were dropped to zero within 3-4s, although servo-valve current outputs FAL was changed to negative value from -2.67, -14.10, -10.3, -49.20 then to -67.03mA in order to overcome error between Fuel flow and Command FQROUT=25.83%, but it was not successfully, then tripped by Loss by Flame, all others data are normal before tripped, such as bypass valve command FQROUT=25.83, control oil pressure A63HL_MEAN, Hydraulic oil pressure A63HQ and Liquid Fuel pressure A63HL.

We also replaced a new servo valve and checked the polarity of servo coil one by one, clean the filer of servo valve, and opened the actuator of LFCV to check. We do not found any problem for actuator.

After that, we startup serverl times, the trend and trip log are same as before. There is one things that it is diffcault to confirm what the clutch system working well or not.

In addition, we have startup successfully once or twice before replaced servo-valve, the startup phase is ok, but not stable during transfer to Gas fuel from Liquid fuel or vice versa . The main problem is that liquid fuel flow dropped more than 6% once unintelligibly during increase liquid fuel command or raise once more than 4% unreasonably during decrease liquid fuel command.

Does Servo-Valve has a problem? But we have replace a new one. Or Does clutch system has a problem? Many thanks!
 
Ebing,

Have you replaced with new servo valve or repaired one?

Is there any hunting on LFBV (in your case LFCV)?

Is liquid fuel inlet pressure stable? (confirm no air in the line)

Generally clutch won't be a problem since its working upto 90% speed.

Take care
G.Rajesh
 
Ebing and G.Rajesh,

'Loss of Flame' means: Flame was lost <b>BEFORE</b> any condition was detected that would have tripped the turbine by shutting off fuel (which would ultimately result in a loss of flame--but AFTER a trip condition was detected and annunciated). In other words, 'Loss of Flame' is telling the operator, the operator's supervisor, the technician, and the Plant Manager that flame was lost before the unit was tripped. Flame is lost when there is insufficient fuel flow for the air flow.

Ebing says the fuel supply pressure is normal; the hydraulic pressure is normal, and fuel transfers are unstable--but Ebing does not say if fuel transfers are successful (either going from gas to liquid, or from liquid to gas).

G.Rajesh says the electric fuel pump clutch should not be the problem, but it certainly could be the problem. ANY time liquid fuel is to be flowing the liquid fuel pump clutch should be energized. ALL the time liquid fuel is to be flowing the liquid fuel pump clutch should be energized, and not slipping.

Troubleshooting is a process of elimination. It doesn't seem as if the electric fuel pump clutch has been properly "eliminated" as a possible cause. If the Mark V is putting out a LOT of negative current (negative current increases the flow of liquid fuel) but the flow divider feedback is not increasing or decreases to zero, then it would seem that either there is a problem with the high-pressure liquid fuel pump, the electric fuel pump clutch, the coupling between the liquid fuel pump clutch and the input shaft to the high pressure liquid fuel pump, the relief valve in the line "around" the high pressure liquid fuel pump, or the LFBV (Liquid Fuel Bypass Valve).

Ebing says they used a "new" servo, verified the servo polarity of each individual coil, so it would seem the servo-valve is not the problem (and it usually isn't!).

Another problem could be the high pressure liquid fuel filter, or the low pressure liquid fuel filter(s). If they are dirty and won't allow sufficient flow when flow requirements are increasing during starting, then they could be the problem.

Again, one needs to use the Liquid Fuel System P&ID ("Piping Schematic") and eliminate all of the possible causes one by one until the culprit(s) is(are) found and resolved. Something is causing the liquid fuel flow to be restricted or to stop as flow requirements are increasing.

Is anyone using the manual selector valve at the liquid fuel flow divider output to monitor the high pressure pump discharge pressure during acceleration to see what's happening to it when the flame-outs are occurring?

It could also be possible that there are high spreads during warm-up and acceleration on liquid fuel fuel and that one or more combustors are not "lit" and then as air flow increases flame is lost in one or more of the combustors with flame detectors and that results in loss of flame. What are the exhaust temperature spreads during acceleration and just prior to 90% speed and loss of flame? Are all of the combustors with flame detectors indicating flame during acceleration prior to loss of flame? Or is one or more of them not indicating flame or "flickering" during acceleration just prior to loss of flame?

However, the most likely problem is something with the high pressure liquid fuel pump, the electric liquid fuel pump clutch, the coupling between the liquid fuel pump clutch and the input shaft to the high pressure liquid fuel pump, the relief valve "around" the high pressure liquid fuel pump, or the LFBV. Or some restriction in the line (plugged filters).

There is also one thing we haven't considered: Atomizing Air. Some units have an AC motor-driven Booster Atomizing Air Compressor and some solenoid-operated valves to isolate the compressor from the system at about 95% speed. If there is a problem with the Main Atomizing Air Compressor or the Atomizing Air Bypass Valve (sorry; I can't recall the valve number, but it's on the Atomizing Air P&ID) then the problem could be insufficient atomizing air. But, from the problem description, this doesn't seem likely at all.

<b>From the information provided,</b> it certainly seems like something is causing a loss of fuel flow to the combustors. Again, monitoring liquid fuel pressures using the manual selector valve at the discharge of the flow divider can be very helpful--but it's entirely a manual operator that usually requires a couple of people: one to slowly rotate the valve handle and another to record pressures. It requires an understanding of the various valve positions, and some back-and-forth between valve positions and some very careful positioning and monitoring at about the time of the loss of flame events.

Please write back to let us know how the problem is resolved.
 
Rajesh,

Thanks for you reply!

We try to startup GTG twice or thrice, there were no any Liquid Fuel(FQL1) at all. The situation is getting worse. Later we tested clutch offline. The fuse was broken. Clutch coil grounding with one side was found. We measured resistor of coil is 162 ohm, but one side is 3 ohm to ground. But after replace fuse, the clutch can be work offline by force L20CF1X. Then we startup again, failure again. Another fuse was broken this time. We measured the resistor of coil was not stable, fluctuate between 0 - 224 - 330 ohm. I think the coil is the bad actor. It is that right? We are going to replace the clutch.

Best Regards,
ebing
 
CSA,

Your feedback will be very useful to Ebing to solve their problem.

Ebing also confirmed they have problem with electric motor clutch.

To know more, how clutch will give problem after 90 % speed ALL the time? ("It tripped around 90% TNH by Loss of Flame for several times with liquid fuel during startup").

FYI,
If there is a problem with the Main Atomizing Air Compressor or the Atomizing Air Bypass Valve (sorry; I can't recall the valve number, but it's on the Atomizing Air P&ID) Atomizing Air Bypass Valve name is VA22.

Thank you

G.Rajesh
 
CSA,

Waiting for your reply for the below doubt.

> To know more, how clutch will give problem after 90 % speed ALL the time?
> ("It tripped around 90% TNH by Loss of Flame for several times with liquid fuel during startup").

Thank you
G.Rajesh
 
G.Rajesh,

I said the clutch should not be slipping any time liquid fuel is to be flowing (or, ALL the time liquid fuel is to be flowing the clutch should not be slipping). A clutch is, by definition, a device for slipping and for not slipping. On a GE-design heavy duty gas turbine the electric fuel pump clutch should slip (fully) when liquid fuel is not to be flowing, and should not slip at all any time liquid fuel is to be flowing.

As turbine speed increases, so does fuel flow and pressure--which means the torque required to be transmitted through the clutch will increase as speed (and load) increases. If the clutch slips, it's most likely to do so when the torque transmission requirements are higher.

Electric fuel pump clutches should not slip at all when energized, but they sometimes do. Sometimes the coils fail gradually or as they get warm/hot. This can cause slipping to occur or to increase.

A loose coupling coupling between the clutch and the high pressure liquid fuel pump could also cause the transmission of torque to be less than it should be.

Ebing, whom we've not heard back from about his problem, indicated there are problems with the clutch coil/circuit, including a ground. I would suspect there are also other grounds, and that excessive current is flowing through the clutch fuse causing it to blow. One of the ways excessive current could flow through the fuse is if there is a ground in the clutch coil/circuit and another ground (or grounds) elsewhere in the field wiring and/or devices connected to the 125 VDC battery.

I did not understand you had a question. Why is it deemed "more polite" to say one has a doubt when doubt implies something false has been told? If you want to have more information, or have a direct question (which was not clear) just say, "I have a question," or, "Can you please elaborate or explain what you meant when you said [this, or, that]?"

If you need or want more information or a better explanation, be clear about it, and you'll likely get the response you seek faster.
 
CAS and G.Rajesh,

Thank you for details information. After repaired the coil of clutch, we started our GTG successfully just now. So GTG startup failure was caused by coil failure. Thank you again.

Best Regards,
Ebing
 
ebing,

You are most welcome!

Thanks for the feedback! "Feedback is the most important contribution!"(c) here at control.com. It's what separates good information from great information.
 
Top