VC3 Valve Hunting and Machine Tripping

We are having trouble during the start-up of our gas turbine (MS9001E) on liquid fuel. Whenever turbine reaches upto around 2900 ~ 2950 RPM, the VC3 valve starts hunting and within 5 seconds the machine gets trip over LOSS OF FLAME signal.

We have replaced VC3 servo valve, its hydraulic cylinder, checked and verified fuel oil accumulator's bladder pressure, fuel oil pump (PF1) working, Flow divider (88FD) discharge pressure. We are now also replacing VC3 valve, but we are still not sure that the problem will be solved or will remain stand still.

We are very sorry to hear about your troubles.

The infamous 'LOSS OF FLAME TRIP' Process Alarm means flame has been lost in the majority of combustors with flame detectors for no known reason--in other words, there was not a trip condition or shutdown condition that would have resulted in a loss of flame prior to the actual loss of flame. The unit just "ran out" of fuel. Now, that could be because the fuel supply is not sufficient, or the control valve (liquid fuel bypass valve) is oscillating causing the flow to the combustors to drop so fast flame is being lost in several combustors.

You didn't mention which type of turbine control system the unit has; if we know we can probably make some recommendations for troubleshooting (such as using Trend Recorder or Trender if the unit has a Mark VI or Mark VIe; or using one of the VIEW tools if the unit has a Mark V).

When you replaced the servo, did you perform a servo current polarity check for each of the three coils, independently? If so, what were the results?

Some newer GE-design Frame 9# units do not have LVDTs on the liquid fuel bypass valve (VC3). Does your unit have them? If so, have you tried manually stroking the valve slowly and monitoring the LVDTs with the Calibration function of the Mark* to see if there is any stability at a certain point in the valve movement? (That would indicate a problem with a worn actuator or possibly a worn valve, but you have replaced the actuator, so it shouldn't be that.)'

If there is a "last chance" filter in the hydraulic supply to the liquid fuel control valve actuator, have you replaced it?

What Process Alarms are being annunciated when this problem occurs?

What Diagnostic Alarms are present during starting and acceleration and prior to this event?

When did this problem start? After a trip from load?

After a maintenance outage?

In other words, what might have changed just prior to this problem starting?

The speed you mention is when the unit is typically switching from acceleration control (if the unit has a Mark V or later turbine control system) to Droop Speed Control.

Are you certain the liquid fuel supply piping is free of air--including the filter vessels (low- and high pressure)? When was the last time the filter supply piping was bled of air--including the liquid fuel filter vessels?

Is the liquid fuel supply pressure stable prior to this event occurring? (I imagine the liquid fuel supply pressure starts oscillating when the liquid fuel bypass valve, VC3, I believe, starts oscillating. Please confirm.)

When running on liquid fuel, the Mark* liquid fuel reference is a flow-rate (signal name FQR, usually), and it is compared to the feedback from the liquid fuel flow divider speed pick-ups (signal name FQ or FQL, usually). These are two of the signals which should be monitored during starting and at the time of the problem to see what is occurring, as well as turbine shaft speed (TNH). You should also be looking at FSR1, FSR2 and FSR to make sure they are stable and not changing.

I have seen choked (plugged) liquid fuel forwarding strainers cause a problem only above a certain fuel flow-rate; below that flow-rate they don't cause too much problem but the liquid fuel bypass valve is usually much more closed than usual prior to the start of the instability. (The liquid fuel bypass valves CLOSES to send more liquid fuel to the unit.)

What is the hydraulic system pressure prior to the start of the VC3 oscillations? It should be very stable, and even when VC3 starts oscillating it should still be pretty stable. If not, then one should check the hydraulic accumulator(s) for proper valve positions and a proper charge (usually approximately one-half (50%) of rated system hydraulic pressure).

About the only other thing which might be causing problems (other than air in the liquid fuel supply piping and/or filter vessels; unstable liquid fuel supply pressure; choked liquid fuel strainers; unstable hydraulic pressure or non-working hydraulic accumulators; incorrect servo polarity on one servo coil; problems with the liquid fuel flow divider speed pick-ups, or excessively worn liquid fuel flow divider bearings) might be a problem with the liquid fuel pump clutch or coupling between the clutch output shaft and the high-pressure liquid fuel pump input shaft (loose coupling set-screws). I have seen some liquid fuel pump clutches that start slipping above a certain flow-rate, but that's not too common.

There are LOTS of components in the liquid fuel system, and all of them have to work properly. MANY of them are NOT directly, or even indirectly, controlled by the Mark*.

Again, the liquid fuel reference is a flow-rate reference, and that flow-rate reference should be close to or transferring between acceleration control and Droop Speed Control. If there are problems with speed pick-ups not being stable, that could cause an issue--but they would be accompanied by Process- and Diagnostic Alarms. The flow-rate reference is compared to the liquid fuel flow divider speed pick-up feedback connected to the Mark* (which is scaled into #/sec, or kg/sec). If there's a problem with the liquid fuel flow divider, or the liquid fuel flow divider speed pick-ups, or the feedback from the speed pick-ups (the inputs to the Mark*), that could cause a problem--but it's hard to understand why the unit could start and accelerate to near FSNL (Full Speed-No Load) before encountering fuel control valve instability.

Is the unit capable of starting and running on natural gas fuel, and if so, have you tried starting on gas fuel and trying to transfer to liquid fuel at FSNL? If so, what were the results?

But, we really need to know what alarms (Process AND Diagnostic) are active during the start-up and what alarms are annunciated at the time of the event. We can tell you a LOT more about what might be happening if you can tell us about the alarms--ALL of them, please. (Even the ones which might not seem relevant.)

Looking forward to hearing back from you!
This issue was live at our plant site from 8 years. The problem was kept solving after taking different measures everytime, sometimes after replacing the Fuel oil bypass valve (VC3), and sometimes after replacing VC3 servo-valve and its inlet filter. Although its filter was found clean every time, but in-spite of that after trying start-up twice or thrice the machine started anyway.

This time we did everything one by one.
— We charged the N2 pressure in the fuel oil system accumulators' bladders. (Hunting persisted)

—We replaced the main hydraulic oil filter. (Hunting persisted)

—Again performed liquid fuel priming. (Hunting persisted)

—Replaced the VC3 servo valve inlet filter. (Hunting persisted)

—Replaced the VC3 servo valve. (Hunting persisted)

—We increased the fuel oil pressure through PCV-355 (up-to 5.3 barg). Hunting persisted.

We looked for everything. Checked any hydraulic oil leakage, any fuel oil leakage, opened the vents from fuel oil HP filter and then closed. But problem could not be solved.

Finally, when we increased the tension of VC3 actuator spring. The machine started very smoothly and synchronised with the grid. So the conclusion we made that if we increase the tension of VC3 actuator spring from its free position, the hunting will be
eradicated. We increased the tension up-to 3mm from spring free position. You must try 3 mm first then if problem is not solved then I recommend to increase the tension up-to 5mm.