our 21 MW gas turbine is controlled by speedotronic (MARk-V) system of GE. Recently generator breaker of our machine tripped due to grid disturbance. After synchronising with grid machine was on free loading. after loading upto 10 MW of grid load its FSR_Status field changed from SPEED DROOP to SHUTDOWN and load on the machine dropped to 6 MW. FSR value field shows 100% and TNR value field shows 107%. After this machine was neither able to add load nor it was rejecting load even after pressing of RAISE/Lower command.So we have to open the breaker at 6 MW load. While on FSNL all parameters were found normal. After some time of observation on FSNL machine was again syncronised and loaded. this time it added the load upto 20 MW.While loading severe hunting was observed in Field current and MVAR of the machine.

Now it has became a regular phenomena for past 4-5 times. whenever this machine breaker trips (or rejects load) due to grid disturbance SHUTDOWN Status appear on synchronising first time and loading halts to 6MW with no further addition or rejection of load even manually.
Machine faces a severe hunting while loading, in field current and MVAR.

Please suggest a solution.

 

[Speedotronic? That's cute! I needed a good belly-laugh this morning; thanks!]

The GE "standard" is that any condition that results in a shutdown is supposed to have a Process Alarm that notifies the operator why a shutdown has been initiated. What Process Alarms are being annunciated when this condition occurs?

What Diagnostic Alarms were present before the event? What Diagnostic Alarms were annunciated during the event?

If there aren't any Process Alarms, then it could be possible that the display for the enumerated data signal was not configured correctly. A shutdown does not usually result in "stoppage" at some load; it usually results in a generator breaker opening and a fired shutdown to COOLDOWN. So, something seems amiss.

There is another class of automatic operation called 'runbacks'. A runback is a reduction in load to some preset value caused by some event. For example, some turbines with redundant exhaust frame blowers will reduce load by 50% on the loss of one exhaust frame blower, and will reduce load to the point that the generator breaker will open and the unit will remain at FSNL (Full Speed-No Load). The automatic reductions in load are called runbacks, and the GE "standard" is that they, too, are supposed to be accompanied by Process Alarm messages to alert the operator what is occurring. So, I wonder if there is some kind of runback that is occurring, but, there should be a Process Alarm to indicate what the problem is.

Next, grid disturbances can play havoc with turbine control systems. If the grid frequency decreases significantly (by say, 2% or more), then the flow through the axial compressor is significantly reduced and therefore the CPD (Compressor Discharge Pressure) is reduced and the amount of fuel that can be burned without exceeding the exhaust temperature limit is reduced. So, the unit can't get to rated power because it's exhaust temperature is high because the axial compressor flow and CPD is low.

The fact that TNR was at 107% would lend credence to the above scenario, that the grid frequency was significantly low and power output was reduced and the Speedotronic (I love that word!) was trying to increase power.

It would be necessary to look at the CSP in your Speedotronic (there's that word again!) to really understand how the unit was programmed, or modified. I've seen some very "creative" sequencing modifications performed at sites where they frequently experience grid frequency problems in a misguided effort to try to make the unit behave in ways it just can't be made to behave. And this was done by some well-intentioned OEM engineers in an effort to try to placate Customers who don't really understand turbines and how they respond to grid frequency disturbances.

Now, as for the MVAr and field current disturbances, we don't know what kind of exciter ("AVR") in use on your unit, nor how it interfaces with the Mark V. Generally, though, with the exception of EX2000 exciters, the interface between the Speedotronic (!) and the exciter regulator is pretty minimal, consisting of discrete outputs for RAISEing and LOWERing generator terminal voltage setpoint from buttons on the operator interface. In some cases, there is VAr/Power Factor control logic in the Speedotronic (!) that drives these two discrete outputs when VAr control or Power Factor control is active.

If the exciter is an EX2000, then there may be a StageLink connection between the Speedotronic (!) and the EX2000 that would be used for getting data from the exciter regulator as well as for sending commands to RAISE or LOWER the generator terminal voltage setpoint.

Again, we don't know what kind of exciter regulator is at your site, but, in general, MVAr fluctuations and field current fluctuations are not usually caused by the Speedotronic (!). But, we don't know what kinds of fluctuations you are experiencing (magnitude and frequency), and you didn't say if VAr or Power Factor control was active at the time, or if it's even possible with your Speedotronic (!).

 

Thanks sir for giving input to progress on diagnosis of this acute problem.

First of all i will like to let you know that no actual tripping of either machine or generator breaker is occuring due to this problem. so the generation of process alarm is not there. secondly, also there was no diagnostic alarm found generated at the instant of this problem. Also i am unable to find any turbine frame blower for our machine. Again severe hunting was particular to the machine and i am not sure about type of regulation. also let me give you brief prelude to this incident;

we have faced some mechanical problem in our machine. For some time we have to switched off the mark-V to carry out some welding process in the machine compartment (it is not clear to me why this is done). while switching ON the mark-V there was sparking in TBQB terminal board of R-core and a diode found burned. we replaced it with new one. Again after fully switching ON the Mark-V we have changed TCDA card in Loc1 & 3 in QD-core as there was various alarms which were unable to reset. after starting of machine it tripped two times due to problem with inverter(now replaced with On-Line UPS) in MCC panel. After this we faced problem with thyristor pack fuse which was blown away. when finally machine was put on load hunting was observed. on the next day on grid disturbance we came across this problem.

is that problem may be due to spark occurred in TBQB or itself new TBQB. while occurrence of this problem GCV was at 97% of its position.

 

I did not understand you to say that any trips (emergency shutdowns) had occurred. I was trying to tell you that any event or condition that results in an automatic shutdown (unloading, breaker opening, and fired shutdown) is supposed to have a Process Alarm message associated with it.

I was also trying to tell you that some units which have exhaust frame blowers (which Frame 5s do not, I realize, but I was just using the exhaust frame blower protection logic as an example) will initiate a runback on loss of one exhaust frame blower, and a runback to FSNL on loss of both exhaust frame blowers.

From the previous post:

>For example, some turbines with redundant >exhaust frame blowers ...

I'm sorry if you understood my example to mean that your unit has exhaust frame blowers; I wished I could have retracted the message as soon as I had clicked on 'Submit' because I was certain you would misinterpret the example to be specific to your unit.

I can't add anything more to this thread. Good luck with your problem(s). I suggest you have someone familiar with Mark V Speedotronic (!) control systems come to your site and provide analysis, comment, and training.

 

I am sorry sir if my reply had hurt you for being specific to the point you raised in your first reply. i didn't mean that. i was just trying to give specific details you raised.

by the way the problem appears again day before yesterday.

Thanks.

regards

 

Consider this scenario which I experienced:

1. GCV was stuck

2. FSR raises so that it can get more fuel for the turbine to support the load

3. FSR reaches to 100%

4. it will be displayed "Shutdown" for the FSR_Control

Note: earlier turbines there were no "GCV not following reference" trip configured.

 

V.Venkatachallam,

Can you please tell us what caused the GCV to be stuck?