Mark-V trip on a non clear reason

H

Thread Starter

hanywepco

We have frame-V GE Gas Turbine controlled by Mark-V control system , it was running normal under fuel gas and suddenly tripped ..but unfortunately the trip reason is not clear at all.

The trip event file (as collected from the HMI) can be downloaded from the following link : http://www.speedyshare.com/files/26433256/Trip-Events.txt

The strange thing in this event file is that post the trip the Dwatt is still showing 15.2MW while L4 already became 0 ,

It is showing in the event list also “GAS FUEL HYDRAULIC TRIP PRESSURE LOW” but no problem found in the field..

Can any one please advise...

Thanks in advance.
 
There is insufficient information to determine what may have happened. The alarm list does not indicate what was the cause of the trip because the dithering alarm filled the queue.... The alarm printout must be consulted to see what condition tripped the unit and what alarms might have been annunciated immediately prior to the trip.

I have seen some MW transducers erroneously indicate power flow but never for three seconds. The worst offenders were the 0-1 mA transducers, but even some of the 4-20 mA transducers have been problematic at times. Even when the fuel stop valve closes there is still fuel under pressure in the line(s) between the stop valve and the fuel nozzles which takes some time to dissipate, so as the flame detectors indicate, there was still flame for a short period after L4 dropped out.

As can be seen from the POST trip history, the generator breaker was still closed for two seconds, waiting for reverse power detection. Something which many people forget, or don't realize, is that when the fuel stop valve is shut on a trip the generator breaker isn't opened until reverse power is detected. In other words, when a trip condition is annunciated, the Speedtronic doesn't open the generator breaker immediately on the trip, the breaker is opened on reverse power. This is done to use the grid as a "brake" to dissipate the energy from the fuel that must be dissipated when the fuel stop valve is opened (see above). If the generator breaker is opened at the same time the fuel stop valve is closed, then the energy of the fuel trapped between the stop valve and the fuel nozzles will be dissipated as speed, as in overspeed, not necessarily an overspeed trip but a speed increase above rated.

Later Mark V panels included the ability to detect a reverse power condition and trip the breaker (very early Mark V panels did not have this feature). There should also be an external reverse power relay (device number 32) which monitors for reverse power and would also trip the breaker even if the Mark V has the reverse power feature. The external relay has something the Mark V doesn't have because it usually has the ability to detect rate of change of power (the inverse time function) as well as magnitude of reverse power. The inverse time function can trip the breaker faster than the Mark V in many instances.

Most Customers never bother to reset the flag on the external reverse power relay, so when it actually does trip the breaker the flag is already "dropped" and this makes it very difficult to determine if the external relay tripped the breaker or if the Mark V tripped the breaker (presuming the Mark V has the reverse power feature). [Resetting any and all external relay flags prior to a unit start should be part of every pre-start routine at every site, but sadly it's not.]

One last thing which is puzzling is the P2 pressure had suddenly dropped 1 sec before the trip, going from 10.8 to 7.91 barg while FSR remained unchanged. This would indicate either a problem with the gas fuel supply pressure or the SRV was closing or had closed prior to the trip.

If you're going to rely on the Trip History Display for troubleshooting unit trips, you need to print it immediately after a unit trip, as it's decelerating before it unit even reaches zero speed. And, operators need learn how to use the Alarm Lockout feature to prevent nuisance, dithering alarms from filling the Trip History Display alarm queue.

If you have the alarm printout from the trip event, please tell us every alarm (and event, if so configured) that was existing and was annunciated or cleared for a period of at least two minutes prior to the trip. <b>EVERY</b> alarm (and event), <b>including</b> Diagnostic Alarms. Whether you think it was relevant or not.
 
thank you for your always support and help. I have uploaded also the trip log file as per your request.

You can download it on the following link:
http://www.speedyshare.com/files/26499743/gt2_LOG.txt

As per my previous post and from the trip events file, the GT Trip has occurred at about 09:54:38.000 on 14/1/2011. We have replaced on 19/1/2011 the TCEA2 card by a new one as we noticed the error leds on it during GT start-up are not showing the same as the other 2 TCEA cards. Then again on 20/1/2011,we replaced also TCEA1 card although there is no alarm from it on Mark-5 but we noticed also the error leds on it are not showing during startUp the same as it showing for the other TCEA2&3 cards.

but the trip reason is not clear??
 
The trip reason is very clear:
<pre>14-JAN-2011 09:54:37.375 T2 1 Q 0121 ALM GAS FUEL HYDRAULIC TRIP PRESSURE LOW
14-JAN-2011 09:54:37.437 T2 0 L4 EVT MASTER PROTECTIVE SIGNAL
</pre>
The unit tripped because the gas fuel trip oil pressure was low. Plain and simple. Something caused the trip oil pressure to go below the switch setpoint (there are sometimes three of these gas fuel trip oil pressure switches; we can't know how many there are at your site without looking at the CSP or you posting that information).

Sometimes the pressure switches fail intermittently (not often, but it has been known to happen). Also, sometimes the trip solenoid (20FG-1, in this case) fails intermittently (not often, but it has been known to happen). 20FG-1 allows trip oil system pressure to flow to the gas fuel system dump valve (VH5-1, I believe; refer to the Gas Fuel Piping Schematic (P&ID) for details). Sometimes VH5-1 fails intermittently (not often, but it has been known to happen).

You need to examine the Trip Oil Piping Schematic (P&ID) for the turbine, and review all the components. Sometimes, there is an overspeed bolt which can dump trip oil pressure. Sometimes, 20TV-1 needs to be energized to establish and maintain trip oil pressure. If either of these devices are necessary for trip oil pressure and they fail or pass sufficiently to allow a drop in pressure below the pressure switch setpoint then there's going to be a trip.

And, there's always the wiring. Loose crimps; loose terminal screws; corroded connections; loose cables; these can all cause it to appear as if the trip oil system pressure is low.

There are also valves on the Access. Gauge Cabinet for the pressure switches which must be in the proper positions. There are usually block valves (spring-loaded to open; push to close) and bleed valves (manual twist to open/close). If these aren't in the proper position they can cause problems. I've also seen air in the tubing (it's usually 1/8-inch stainless steel tubing and is rarely properly bled of air) cause intermittent problems, no not usually if it's been running for some time. If this occurred after a maintenance outage when the switches were "calibrated", then it could be the problem, but unlikely.

My money is on the wiring. Something is not right with the wiring or terminations. Sometimes tough to find, especially if the unit uses plug-connected cables ("J-cables") which can be prone to corrosion in some environments (I'd say your site is one which would be prone to such conditions if J-cables are used). If you haven't already done so, you should be checking every termination in all of the trip oil pressure switch loops, at the Mark V, any intermediate JB, and at the switch(es). If there are J-cables and you're not sure if they are corrosion- and moisture-free, then now's as good a time as any to plan to check them during the next shutdown. You'll need to power down the Mark V and all unit-related circuits in order to be able to safely unplug the J-cables, check them, and re-connect them (PROPERLY! Most sites don't plug them back in fully which is a recipe for disaster!), and then power the control system back up.

The last thing we haven't talked about is grounds. Is the 125 VDC battery ground alarm "in" on the panel? If so, what is the incoming DC voltage (each leg, with respect to ground)? (You can safely measure this with a voltmeter while the unit is running!)

If you know you have a "nuisance" alarm (Process or Diagnostic) and you know you can't do anything about it for a while, learn to use the Alarm LOCK and -UNLOCK feature of the Proces- and Diagnostic Alarm windows. It can prevent a lot of wasted alarm log file space, printer paper, and make troubleshooting alarms much easier.

Finally, you will find the LEDs on the Mark V cards will all blink in unison and with the same LEDs lit if there are NO Diagnostic Alarms. You will find that when there are Diagnostic Alarms associated with a card in a processor that the LEDs will blink out of sequence with the other cards which do not have Diagnostic Alarms. So, "non-conforming" LED blinking is associated with Diagnostic Alarms, and vice versa, Diagnostic Alarms are associated with "non-conforming" LED blinking.

But, learn to use the Alarm LOCK and -UNLOCK feature. It's a valuable "tool". One just needs to remember to UNLOCK an alarm when the alarm has been resolved. (Actually, one needs to unlock the alarm to see if it has been resolved so it can be reset!)

If the Alarm LOCK and -UNLOCK buttons don't work, then you need to get GE or the GE Mark V HMI supplier back to correct the problem!
 
P

Process Value

Mark V trip on unclear reason

the reason for trip is clear <pre>
14-JAN-2011 09:54:37.375 T2 1 Q 0121 ALM GAS FUEL HYDRAULIC TRIP PRESSURE LOW
14-JAN-2011 09:54:37.437 T2 0 L4 EVT MASTER PROTECTIVE SIGNAL</pre>
your trip oil pressure has gone down which has caused the mark v to initiate a trip.

as there are no accompanying alarms about the lube oil system so the problem is isolated to the trip oil circuit.

you can eliminate the trip oil pressure switches , the fuel gas pressure has reduced from 10.8 to 7.9 prior to the trip , this means that the hydraulic stop valve had partially closed due to reduction in the trip oil pressure. from this fact i think you can eliminate faulty wiring also i guess , as the system responded as it should have.

now what are all the possible reasons

1. trip oil filter choke - this is a remote possibility , but plant operators may not have given sufficient care to look at this. the normal trip oil pressure low activation is about 0.56 kg. to become this low it would have to have been somwehe near the trip value. the filters choke really fast after a particular DP. like in mark VI versions , mark v do not normally have trip oil pressure transmitters and alarm signals. check this possibility.

2. trip oil drain solenoid - a passing or a corroded trip oil drain solenoid will also a trip oil pressure low. i do not know if the trip oil pressure suddenly fell or slowly. but check and see it the trip oil drain solenoid is holding oil is it passing.

3. intermittent signal to the trip oil drain solenoid - this is also a possibility , intermittent trips are very hard to pinpoint and very difficult to solve. but as csa has recommended check the wiring for trip solenoid and the voltage levels in the terminal board to which trip oil drain solenoid is connected.
 
Top