We have a GT frame 6 engine with Mrk V control system, baseload at 30 MW load. we plan to do houseload testing with load rejection at 75% baseload. at the time of load rejection, the 52G will be released and the turbine FSNL with a load of 0.2 MW. the governor mode will automatically change from droop mode to isocronus mode to avoid reverse power. This is because the turbine load drops below the reverse spinning load. during the isocronus mode with a load of 0.2 MW (FSNL mode), the turbine will have difficulty controlling the flow rate resulting in an unstable turbine frequency, but the isocronus mode will still maintain the turbine frequency at 49.9 Hz to 50.1 Hz.
My question is, how long is the safe limit to operate the turbine with a load of 0.2 MW? What parameters should be considered to ensure that the turbine is operating normally?

We kindly ask for your advice and input in order to keep the unit in good working order.

 

@Abdul majid bakti,

If 52G is open, how is the turbine going to produce any power? Is this possibly a black-start machine?

What fuel will be burned when the machine is performing this load rejection test?

I doubt the machine will maintain system frequency throughout the test between 49.9 Hz and 50.1Hz. It might settle to something around 50.15 Hz (which is 100.3% TNR in frequency and the normal FSNL speed when waiting for synchronization).

UNLESS there WILL be some kind of load on the machine when 52G opens it's not necessary for the machine to switch from Droop Speed Control to Isochronous Speed Control. (When a GE-design Frame 6B heavy duty gas turbine is started and accelerated to FSNL in preparation for synchronization it operates in Droop Speed Control, and depending on several factors it (should be) automatically switched to Isochronous Speed Control when 52G closes. When I was commissioning machines and we were obligated to perform a load rejection test and maintain flame in the combustor AND supply a small load (a couple of MW or so, not 0.2 MW as in your description) the machine would automatically switch from Droop Speed Control to Isochronous Speed Control--but 52G would stay closed.

Spinning reserve is not the absolute minimum sustainable load for a GE-design heavy duty gas turbine, and some machines (usually larger machines) even have a difficult time holding stable load (but not frequency--because that's controlled by the grid) at spinning reserve. Why? Because the fuel control valve is usually at about it's minimum controlling position at machine rated speed (frequency). So, frequency will be stable but load sometimes oscillates by a MW or more around the spinning reserve setpoint.

Operation between FSNL and spinning reserve isn't particularly hard on the machine. Again, it's more about holding a stable load--when loaded (when 52G is closed), and that's more about the fuel control valve(s) not anything else. And some of the possible load instability is about the tuning of load control variables, as much as it can be about the "small" opening of the fuel control valve(s).

The major reason for performing a load rejection test is to prove the machine won't flame out when the fuel is cut back so much when 52G is opened. And sometimes that requires multiple tests and some adjustment of the minimum FSR variables to achieve. Most utilities want to make sure the machine can re-synchronize within a couple of minutes after a breaker-opening event (NOT a turbine trip!) and if the machine flames out when fuel is cut back so quickly it can't be re-synchronized until it coasts down after the trip and is re-started.

This is something MOST people misunderstand about a load rejection test. Uneducated and inexperienced people will say the purpose of a load rejection test is to "prove the control system." And that's a very, Very, VERY unspecific purpose and is a recipe for failure on so many levels. If there's no contractual requirement for the machine to resynch very quickly after a load rejection (breaker-opening event...) there's little point in performing a load rejection test. AND understanding the reason(s) for performing a load rejection test are every bit as important as the test itself because if the reasons for performing the test aren't understood by everyone then the interpretation of the test results will be in doubt and questioned.

Best of luck. Unless you want to hold 0.2 MW for an hour or so, and do so with very little fluctuation in load or frequency (especially if not synchronized to a grid with other machines) the machine will be fine.

 

@Abdul majid bakti,

When you do any kind of load rejection test on a large piece of rotating equipment, especially from high load (I count 75% of Baseload rating as high load)--even if the generator breaker (52G) remains closed--the end results are going to be, first, an overspeed/overfrequency. There is a LOT of fuel between the gas control valve(s) and the fuel nozzles and combustion chambers which has to be burned (it has nowhere else to go). Any load is like a brake; the more load the bigger the braking action. Suddenly losing that load is like suddenly releasing the brake--and the machine WILL overspeed, hopefully NOT to the point of tripping on overspeed, but there will be a very quick increase in frequency as that fuel is dissipated (by burning it in the combustors and passing the hot combustion gases through the turbine section). And that will cause the frequency of the generator to increase (because generator rotor speed and frequency are directly related). IF the generator breaker (52G) is closed when this happens it's very likely the over-frequency relay/function may trip the generator breaker, or trip the turbine (by completely shutting off the fuel flow to the combustors).

The turbine control system (a Mark* V in this case) is going to immediately sense the speed increase and will reduce fuel flow to the absolute minimum calculated to not result in loss of flame in the combustors. The axial compressor is also a brake on the machine rotor speed (and generator rotor speed) and the second thing that's going to happen in a load rejection test is that the machine will underspeed--not by much but it will certainly experience a slight underspeed because until that happens the turbine control system is going to hold the fuel flow at minimum until it detects an underspeed and starts to increase the fuel again. (This is IF the machine didn't lose flame when the fuel was reduced to the calculated minimum to maintain flame in the combustors!) And, again, the "other" brake--the axial compressor--is trying to slow the machine speed by using energy from the turbine section to spin the axial compressor. And if there's little load on the generator (if 52G is closed) or no load on the generator (if 52G is open) under-frequency relay (and the underspeed function of the turbine control system) might also trip the generator breaker or the machine.

I also want to stress that the minimum fuel calculated to maintain flame at rated speed is just that: a calculated value. Sometimes it's the right value; other times it's not and the machine will lose flame (remember the machine is going to overspeed which means the air flow through the axial compressor is going to increase, even if the IGVs are a minimum operating angle which is what will extinguish the flame in the combustors). If the machine has been in service for a long time, it's likely the fuel sources have changed since the original commissioning and that can--and does--have an effect on that minimum fuel required to maintain flame.

There's a LOT going on during a load rejection--and if the machine has DLN (Dry Low NOx) combustors, there's even MORE going on with the fuel control valve(s) (especially if them machine is burning gaseous fuel).

I can't stress enough--the reason for a load rejection test MUST BE understood and agreed to by all parties. And, a load rejection test doesn't "prove" the turbine control system operation. The most common reason for performing a load rejection test of a heavy duty gas turbine-generator is to prove the machine will maintain flame when load is suddenly lost AND the machine will settle quickly to FSNL (Full Speed-No Load) so it can quickly be synchronized again to provide power to the load/grid. I can't really think of any other reason to perform a load rejection test on a heavy duty gas turbine-generator other than to demonstrate (prove) that--per contractual requirements!--the machine will maintain flame and settle at FSNL after the load rejection and can be resynchronized quickly without waiting for the machine to coast down to or near zero speed to be restarted and resynchronized. There are SO many things which can affect how the machine performs during the load rejection test and if everyone doesn't understand what happens and why then it's guaranteed there will be an argument about whether the machine passed or failed the load rejection test. And how to proceed.

I suppose a machine could be tested to see if it would maintain flame and speed so it could be quickly resynchronized if the grid operator or captive load/plant wanted to know if that was possible and even necessary for emergency operation/restoration. But, again--the purpose of the test and what constitutes pass or fail must all be understood by all parties (many of whom may not really understand heavy duty gas turbine-generator operation and need to be educated PRIOR to the test).

I'm not saying the initial test will or won't be successful--I am saying that without clear expectations on everyone's part determining success of failure can be extremely difficult, as would be resolving the issue to everyone's satisfaction. Setting expectations is a key component of a load rejection test of a heavy duty gas turbine-generator.

 

I apologize for replying to your post and thank you for your helpful input.
Yes, our machine is designed to operate black start with the initial load of the turbine being the auxiliary power turbine with a total load of 0.2 MW.
I think, your explanation about fuel control really answers our question. we have done load rejection test and the turbine successfully hoseload with 0.1 MW load. but the engine failed to synchronize again.

 

@Abdul majid bakti,

Thank you for the clarifications; things are a little clearer now. Although I'm not exactly clear what an "auxiliary power turbine" is; in my experience, there's a axial compressor and turbine coupled together to make one (single) shaft. That is coupled to an Accessory Gear Box which mechanically drives the Main L.O. Pump, the Main Hydraulic Pump, and the High-Pressure Liquid Fuel Pump, and is coupled to the Starting Means to transfer torque from the motor through the Torque Converter. The other end of the axial compressor/turbine shaft is coupled to a Reduction, or Load, Gear, which is coupled to the drive end of the three-phase synchronous generator which converts the torque of the generator rotor to amperes to provide power to a load or loads through the 52G generator breaker. Some Black Start machines have an auxiliary transformer (generator terminal voltage to something like 380- or 440 VAC) to provide electrical power to loads such as Cooling Water Pumps and Cooling Water Fans and a few other AC motor-driven auxiliaries. But, an auxiliary power turbine? I haven't seen every GE-design Frame 6B heavy duty gas turbine ever put into service, but that would be a first to my knowledge.

Anyway, without being able to see the CSP (Control Sequence Program) running in the Mark* V turbine control panel it's not possible to say for certain but many heavy duty gas turbine-generators with Mark* V turbine control systems required the machine to be in a START sequence to be automatically synchronized after a generator breaker open event (such as a load rejection test). This required selecting a START from the GE Mark* V HMI (or <I>) to re-initiate a START sequence (while running at FSNL) so that the machine would automatically adjust speed and voltage and re-synchronize again. A manual synchronization would have likely been possible (depending on how the 52G generator breaker was opened during the load rejection test), but even though it's possible to manually "select" an Auto Synch after 52G was opened (without a turbine trip) the START sequence permissive is not satisfied.

Again, it's impossible to say for certain that this was the cause of the failure to re-synchronize but many machines had this sequencing for Auto Synchronization. (It should have been possible to see the permissives on the Synchronizing Display--the little squares which all have to be green to permit synchronization; any red square indicates the corresponding permissive(s) is(are) not satisfied.)

There are other possible permissives to allow re-synchronization which would require review of the CSP to identify; some sites had some rather unusual requests/requirements during commissioning which forced commissioning personnel to make sequence changes and add inputs to satisfy.

The good news is it sounds as if the fuel control parameters are good for the conditions at your site to prevent flame-out on loss of load (and to prevent excessive over- and underspeed conditions).