Flame scanner delay for GT Frame 9E

Our GT frame 9E set is having an intermittent problem with our flame scanners 4 & 5. The scanners can only detect flame at 40% TNH and above. Since our GT is a peaking unit, this problem at times causes panic for the operators.

Maintenance crew has already cleaned the scanners a few times but the problem still persists.

Any pointers? Thanks.

If the unit is also experiencing high exhaust temperature differentials during starting which diminish (but may not go away completely) shortly after flame is detected by the last two flame detectors, then it may may be a matter of adjusting the firing or warm-up fuel flows. If the unit isn't establishing flame during light-off, or fuel gets cut back too much during warm-up, there may not be flame in the two cans with detectors which are indicating flame much later than the other two. Do the #4 & #5 flame detectors initially indicate flame, and then "go out" before coming on about 40%?

But the biggest indicator is the exhaust temperature spread; even during start-up and acceleration it shouldn't be really high. If it is, then not enough fuel is being admitted during firing or warm-up.

All of the above presume this condition has persisted for some time, and hasn't just started after a maintenance outage which included LVDT calibrations. There have been units which have had cross-fire tube problems after a maintenance outage which have similar problems; if this problem just started after a maintenance outage involving the cross-fire tubes this may be a possibility, also. If the fuel has changed at all recently, this may also require an adjustment of firing and warm-up FSRs.

The flame scanner #4 & #5 do not "go out" after there is flame but the scanner tends to detect the flame "late" after 40% speed.

We did test and calibrated our GCV and SRV during our forced outage last year. We also drained our lube oil during the outage last year. We also noticed on the same unit the speed TNH tends to slowly increasing once the speed reaches 95%. It takes longer time to reach 100% compared to the other unit.

Are you kidding? Do you want help with your problem or not? WHAT ABOUT THE EXHAUST TEMPERATURE SPREAD BEFORE AND AFTER FLAME IS DETECTED IN CANS 4 & -5?

So, someone "calibrated" the SRV and GCV. Did you know you don't calibrate the SRV or GCV (or IGVs)? You calibrate the LVDT feedback from these devices. Was the accuracy of the LVDT feedback checked before the calibrations were done? (For that matter, how was the accuracy of the calibration checked after the procedure?)

It's been discussed before here (use the 'Search' feature of control.com; it's *very* good): Before anyone calibrates a temperature switch or a pressure transmitter they always perform as "as-found" check, and if the device doesn't require adjustment, that's it!

But for some completely unexplainable, and no logical, reasons whatsoever, people calibrate the SRV or the GCV or the IGVs without ever checking to see if the current calibration is correct or not. Trying to understand this is like trying to understand women (if you're a man) or men (if you're a woman--well, to be honest, men are much easier to understand).

I'm going out on a limb here (but it's a short distance, and the limb is about half a meter in diameter so there's no danger) and going to say that the calibration of the GCV LVDTs has changed sufficiently to affect the light-off and warm0up fuel flow-rates.

Do you have the pre-calibration values? (That's another thing, most times when an instrument is calibrated, people record the settings before and after, *but* when they calibrate LVDTs, they rarely keep records.... What's up with that?)

If you have the pre-calibration values for the GCV LVDTs, try changing back to them and see what happens.

And let's concentrate on one problem, answer the questions, and then we'll try more problems.

Also, how about telling us a little about the control system on your Frame 9E?