What could be the possible reasons of flame toggle and tripping due to loss of flame during load throw-off or when shutdown is initiated and turbine is at FSNL Speed.

For information:
1. Fuel Nozzles are clean and gas operated only.
2. Bleed Valve opening is with Generator breaker opening (not with 95% speed)
3. Spread Pattern is also Hi at FSNL (normal at Base Load).

Regards

 

When did this problem start?

What kind of turbine control system does the turbine have?

Load rejections are always kind of touch-and-go, and require tuning, which can be (and has been) undone by a less-than-accurate GCV LVDT calibration.

And, if you have high spreads (and it would be nice had you told us how high they are), that makes matters even worse.

You say the fuel nozzles are clean, but are they flow-matched and have then been put in locations so the highest- and lowest flows aren't next to each other?

A high spread is not usually a controls problem (though everyone wishes it were).

And, you didn't say how much load is being "thrown off."

The compressor bleed valve opening time should not have an effect of the ability to maintain flame on a load rejection on a properly tuned turbine.

 

Mr. CSA,

We have MARK VI as controller. Load rejections are due to grid problem not operator or instrumentation faults. We normally operate at base ratings. Our Fuel Nozzles are repaired at GE repair centers only and installed as per the movement chart. Our concern was, whether Bleed Valve opening and closing should be with Generator Breaker opening, or, it should see 95% of turbine speed.

We though flame gets disturbed due to bleed valve opening at FSNL and once it happens Hi spread, flame loss etc alarm appears.
Please suggest whether we change the logic for bleed valve opening, or check the fuel nozzles in next opportunity.

Regards

 

You didn't answer the first question: When did this problem start? Has it been persisting since commissioning? Has it just recently started after some maintenance outage? Were the GCV LVDTs recently "re-calibrated"?

Load rejection from Base Load (100+ MW) is hard on a machine; really hard. Lots of thermal stresses, not to mention internal pressure stresses. If you're experiencing multiple load rejections from Base Load, I would suggest you work with the grid personnel to help resolve that problem. When IGVs close "suddenly" and fuel gets cut back suddenly in an attempt to prevent overspeed and yet still maintain flame, that's not a healthy way to be treating a machine.

Opening the compressor bleed valves is going to reduce the air flow coming into the combustors and through the turbine. So, this would seem to be a good thing to my mind. Also, the turbine/compressor speed is likely high shortly after the generator breaker opens so reducing air flow through the unit is something of a good thing (less braking action, but less air for less fuel since the fuel should be cut back to prevent overspeed).

You also haven't said how high the spreads are at FSNL when this occurs. I also wonder if you're initiating a STOP in order to prevent a combustion trouble/spread trip because the spreads are so high. It could well be that flame has been lost in one or more combustors during the load rejection and it's difficult to re-establish flame sometimes. We also don't know if this unit has conventional combustors or DLN combustors.

You could try re-closing the compressor bleed valves at FSNL and see if that helps. But, you don't want to force the bleed valves to stay closed because if the unit trips and they don't open, well, that's a real BAD thing and will lead to lots of really unpleasant things.

Tuning is likely what you really need. And to do that you need to perform some load rejections, and the grid likely won't be very happy about that, either. It's a fine balance between maintaining flame and not tripping on overspeed, but if you have a contractual requirement to re-close the breaker shortly after a breaker open event, then it's likely going to be necessary.

And, if a poor GCV LVDT "calibration" is done subsequently, that can mess up the tuning.

So, to sum up: You haven't answered two very important questions. First, how long has this problem existed? Did these breaker open events from the grid just start recently, or have they been ongoing since commissioning (I'm presuming this is a relatively new unit and that an older control system wasn't retrofitted with a Mark VI.)

And, second, how high are the spreads at FSNL after one of these full load rejections?

I would also like to know if the same spreads occur during a normal fired shutdown of the unit after the breaker opens at approximately FSNL.

I don't recommend making changes to this kind of sequencing (critical) without the packager's approval (GE, BHEL, whomever). Having said that, when you ask the packager for their guidance you should tell them if the unit control system has been retrofitted with a Mark VI and if so, when the compressor bleed valves were opened with the older Speedtronic control system. (GE has a "habit" of making subtle changes during retrofits that don't always result in the desired operation.)

Really, there's a lot we don't know about this unit and it's conditions. It's very difficult to make any sold recommendation, other to contact the turbine packager for assistance before making any sequencing changes to the compressor bleed valve sequencing, especially if the machine has DLN combustors.

Lastly, the whole business of repeated full load rejections is kind of disturbing to me. I've heard of combustion cans collapsing during a full load rejection (on liquid fuel) and believe they should be investigated and resolved in parallel with working to get the turbine to withstand them. Repeated full load rejections are not even good for the generator breaker! Lots of forces involved there, as well, not to mention what's going on inside the turbine and axial compressor.

This isn't just a turbine problem; it's a grid problem; relay settings or something is amiss there. But, to "fix" the turbine is going to require some tuning, likely, and that's hard on the equipment, and the grid, as well.

[Not to mention how difficult it is to trip the breaker manually during testing when the unit is a 100+ MW! That's just scary, for me anyway.]

Best of luck, and let us know how it turns out!

And, please, when responding answer all of the questions, even the ones you don't think are relevant.