GE 9FA Gas Turbine Fuel Flow High


Thread Starter


In 9FA gas turbine, are there any possibilities to initial excessive gas flow? What are the reason behind this? I know about 9FA, here gas fuel parameters are controlled strictly and very correctly by FSR algorithm. And also by controlling all VGC valve and also VSR valve, fuel flow is controlled accurately. I guess, if any malfunction occurs in control logic or control algorithm, valve controlling will be failed, which will occurs the excessive flow. But without this reason, what are the others factors, which may affect the fuel flow rate? like, calorific value, Gas supply pressure, Gas supply temperature ? and if VGC and SRV will fail to control amount of P2 Pressure(flow), What problems can be happened on GT ? And if really these things happen as a good operator what should do?

Another problem,In our 9FA unit, when we increase load up to base load, when it crosses around 54% load, mode change happens from piloted premixed to premixed (It is about DLN 2.0+ nozzle), I assume, that as our IBH (Bleed heating) works properly, so it is very normal. But yesterday, while we were increasing load, when we set load 110MW , suddenly mode change happens, at 93MW ( piloted premixed to premixed), within 1 minuet, it again changed to (Premixed to piloted premixed), It is very harmful for GT to frequently mode changeover. I don't know, why this happened. If anybody knows, please kindly explain. Also let me know, what factors are related to mode changeover, I mean, what condition should be maintained to successfully mode changeover? As far I know, P2 pressure, CA_CRT value, Fuel gas supply temperature. Any others factor?

As I am not so much familiar with 9FA DLN 2.0+, please help me by answering my curiosities .

Thanx a lot.



Any time I've built and tested Trent 900's, Trent 1000's, Trent XWB's, and BR725 turbofans or even V8's, straight 6's, or 4 bangers (cylinders), if the beast doesn't work as expected it's almost always an issue with a sensor.

In your case it could be as simple as a messed up P2T2 (just the P or the T could be the issue).

If you know the 9FA, (and I'll admit I don't but a turbine is a turbine, is a turbine, etc.) you know suck, squeeze, bang, blow. You also know a normal operating machine from one that is not normal.

In order to help you out I have a few questions for you. And please know I'm not questioning your ability or knowledge.

1) Has it performed properly in the past?
2) Referring to question 1, is this a first time run?

The golden rule of "it ran fine before" is:

A) Someone changed something - [Then identify what was changed (who cares about "who" right now)]

B) Something (sensor?) isn't working as expected - [calibration/routine maintenance solves this one normally] <- I said normally, not always :)

Okay, after my long winded "Sensors." rant...

That's what I would at least start with, SENSORS.

Go to your roots and go with your gut.

Don't forget the basics of troubleshooting. You already nailed the best one :) Swallow your pride and ask for help. :D

Let me know if this helps you or if you need more information.




thanks for your interests. This problem is new in our unit. We run our GT before 4 months ago. We calibrating all instruments associating with this system. Now, we are trying to find out any other reasons, which are responsible for this type of problems.

Have a nice day. Thank you again.

impossible it is actual gas fuel flow as you just said "gas fuel parameters are controlled strictly and very correctly by FSR algorithm" to ensure the correct A/F ratio. And i guess that there was no change either on IGV opening nor on Exhaust temp. so it is simply Error of the measurement of Gas fuel ultrasonic flow meter in the entrance of gas fuel system supplying your plant (just the signal going to HMI).

your next guess: What problems can be happened on GT? And if really these things happen as a good operator what should do?

Well, of course there are protections that will never allow such catastrophic thing to happen (actual excessive gas flow). you have Exhaust spread protection then you have High Exhaust protection and if both don't happen there is isothermal Ehaust protection which immediately trip the unit as Exhaust Temp reaches 671 degrees c.

Your problem with high initial gas fuel flow-rate is not clear--because you didn't say precisely when it occurred. And, really, based on the information provided it has nothing to do with DLN 2.0+--most GE-design heavy duty gas turbines all start relatively the same regardless of the combustion system.

Since you have a Mark VIe it's entirely possible it came with something called MBC: Model-Based Control. If so, then many, many things about control have changed and we may not be able to answer your questions with any certainty in this regard.

Before MBC, gas fuel flow-rate was never controlled. Yes--you read that right. It surprises most people--but these gas turbine control systems aren't nearly as sophisticated as many automobiles are (though with MBC they're getting closer to even some low-cost automobiles). And, many people don't believe it when they are told this--but it's true. Gas control valves are/were chosen by the OEM such that position was proportional to fuel flow-rate--but fuel flow-rate was neither the reference, nor the feedback, only position. In other words, by controlling the pressure upstream of the gas control valves (using the Stop/Ratio Valve) flow through the valves were proportional to valve position (stroke). Firing FSR (it's presumed you by 'initial FSR' you mean Firing FSR--the FSR used to establish flame during starting) is a valve position that's proportional to the amount of fuel calculated to be necessary to establish flame. So, the gas control valve(s) were commanded to a position calculated to be sufficient to establish flame--but the actual flow-rate was not measured and compared to any reference and adjusted accordingly. It's strictly a position reference based on the calculated fuel flow-rate as a function of valve position (stroke).

(Again in the past) After firing, fuel was cut back to a slightly lower flow-rate called Warm-up FSR to allow the unit to warm up without a high temperature spike which is not good for the hot gas path parts during a unit start. Again, actual fuel flow-rate is not compared to any fuel flow-rate reference--it's all done by knowing the flow-rates through the valves at various positions (strokes).

After warm-up, FSR was increased slowly. For F-class gas turbines (including the FA units), the starting means (the generator which has been converted to a motor for starting the unit) provides the torque necessary to accelerate the unit according to a programmed rate, and fuel is increased mostly just to prevent flame-out during this time.

>I guess, if any malfunction occurs in control
>logic or control algorithm, valve controlling will be
>failed, which will occurs the excessive flow.

This isn't the case. There is (was) likely an alarm "STARTUP FUEL FLOW EXCESSIVE" which was a totally erroneous message. The position of the Stop/Ratio Valve was monitored during firing and if the SRV opened more than a programmed amount the <b>presumption</b> was that the fuel flow-rate was too high--but it was NOT because the actual fuel flow-rate exceeded some value.

Recently, the OEM has implemented some tests to make sure the gas valves can control pressure(s)--much to the dismay of many sites, because if the valves aren't maintained they will start to pass (leak) and this will cause trips, and there have been many posts on about this "logic" and the reasoning behind it.

If fuel flow-rate is ever truly uncontrolled what will happen is that the gas turbine exhaust temperature will exceed the trip temperature and the turbine will trip on exhaust overtemperature. That's always the uncontrolled fuel flow "back-up" protection--exhaust overtemperature.

To your second scenario, again, before MBC, combustion modes were changed based on TTRF (or TTRF1)--Combustion Reference Temperature, <b>not</b> load or FSR. This is a calculated temperature and varies with load (and other operating conditions--see the unit manuals). Because of this if the Pre-Selected Load Control setpoint causes the TTRF/TTRF1 to be very near to a switching temperature the Speedtronic will toggle back and forth, which isn't the Speedtronic's fault--it's the operator's fault for not knowing what the switching temperatures are and monitoring them when choosing a load setpoint. (A lot of time the Speedtronic will automatically raise load during a combustion mode switch--which will cause the actual load to be above the Pre-Selected Load Control setpoint when the mode switch has finished, which will cause Pre-Selected Load Control to reduce load, which will cause another mode switch, and so on, in a vicious circle of up and down load and combustion mode changes.)

And the Speedtronic is wrongly blamed for this >>perceived<< "problem." It's operators that cause this by not knowing and not understanding the unit and how it operates and how it should be operated. (I'm always amazed at how operators who have been "operating" a machine for many years never really know how it operates, even by just observing how it loads and unloads as they change load. It's really telling, actually, about how much training or experience owners/supervisors require or give their operators. Sorry--but, these are multi-million US dollar machines that in some cases are critical to the electricity supply in some parts of the world, and operators don't even make casual observations during the course of years of operating the turbines. One of the most common refrains I hear from operators is, "It's never done <b>THAT</b> before!" And, it's always done that. I'm surprised that operations supervisors and their management don't require operators, particularly new operators, to record--manually--unit operations so that when there's a question, or a comment such as the one above, there is operating data to review to see how the unit normally operated under similar circumstances in the past. Amazing, and sad at the same time. Again; sorry, but I've just come from such a site where a lot of <b>perceived</b> problems were incorrectly being attributed to the Speedtronic with the above quote being repeatedly uttered by operators, and even their supervisors.)

Actually, Pre-Selected Load Control is a really poor way to operate a GE-design heavy duty gas turbine--especially in places where the grid frequency is not stable; but I have NEVER been able to convince anyone of that. EVER. And, I'm not going to even try now. Or later. Just keep operating the machine the way it's being operated, but learn what the combustion mode switching temperatures are and the loads at which they generally occur and ensure the desired load setpoint isn't within about 10 deg C of a switching temperature and this won't occur again.

Again, the issue you tried to describe is not related to DLN 2.0+; it's just the way DLN in general (before MBC) works and what can happen when operators aren't knowledgeable about DLN switching temperatures and their supervisors insist on using Pre-Selected Load Control to operate GE-design heavy duty gas turbines.

The other thing to know about DLN combustion systems is that they're not really designed to be operated in anything other "premix" mode, where the NOx emissions are lowest. And, choosing a Pre-Selected Load Control setpoint that is at or near the temperature at which the units goes into "premix" mode will cause the kinds of problems you tried to describe. Unless emissions aren't really a problem, or the problem you are having occurs during starting and loading to premix mode, any Pre-Selected Load Control setpoint should be above the combustion reference temperature at which the unit switches into premix mode. And, then there won't be "problems" like this.

Hope this helps! (Sorry for the diatribe above. Guess I'm getting too old to be a consultant/"trouble"shooter. It's getting tougher to be away from home getting the brunt of complaints that have nothing to do with the control system. And it's happening more and more these days; people just presume too many things about the operation of the units that are entrusted to their care.)


hello CSA,

Many many thanks for your detail explanation. Now it is clear to me. Actually I read again again what you have written and try to understand all of lines. Because you may know, that English is not my native language.

By the way, It's my badluck, because, we are people working here on 9FA machine, but management did not arrange for us the training from GE perfectly. we got only 3 training from GE. Mark vIe trouble shooting training, GE 9FA operation training, and GE 9FA maintenance training. All of the training were very short time (5-7 Days). So, u understand now, its how tough for me to understand everything !!! Actually I post here for learning something. Basically I should learn these things at the training. As I was new in 9FA, so, without working, How I ask a question to trainer? Our management should extend all of those training or they should especially arrange some other GE training. But now, we have nothing to do. they will never do that.

Have a good day