Mark V Gas Turbine Trip on high exhaust temp

H

Thread Starter

HANY SALEH

We have frame 6 GE Gas Turbine, suddenly while we are in normal opeation it was tripped on exhaust temp. high alarm although the exhaust temp is showing normal reading during this time. We replaced one thermocouple that gives the alarm to kill the doubt even we know that one thermocouple will not make trip. After that we started the turbine and everything is normal.

Can anyone give advice about what we can do and what the cause of this trip as there is no indication. Could the scantime of the I station cause it not to detect a fast trip and the exhuast temp. really increased?

Thanks for all. My mail: hanywepco @ yahoo. com
 
SIMPLEX or TMR Mark V?

Did you use the Trip Log (Trip History) display/printout to determine the exhaust temperature was normal--or was this just an operator who happened to see at the exact instant of the exhaust over-temperature trip that the exhaust temperature was normal?

The Trip Log would generally have had the information on each individual exhaust T/C reading for each second of the ten seconds immediately prior to the trip.

What kind of fuel was being burned? Could there have been some kind of fuel pressure fluctuation (particularly if it was liquid fuel) during the event?

What were the exhaust temperature spreads prior to the trip? (Again, generally that information would have also been present in the Trip Log.)

It is possible that the <I> couldn't display an extremely sudden (and unlikely) exhaust overtemperature, but it would have most likely been caught on the Trip Log because the trip log is stored in the Mark V (on <C>, I believe). And it's not a single exhaust T/C exceeding the trip level which will cause a trip, it's the average exhaust temperature exceeding the trip level which will cause a trip (if it's programmed properly, anyway; one has to presume some things!) which means that more than one would have had to go high.

And, if you look at the development of the TTXM signal I think you should see that the highest and the lowest T/C readings are tossed out just to prevent against a single T/C causing a problem like this (going high or going low!).

So, unless you already had a failed T/C (high or low) *and* you had another go bad, something else happened.

What was the load immediately prior to the trip--did it suddenly spike high or low? (Again, this would likely have been caught on the Trip Log.)

Were there any Diagnostic Alarms present before the event? If so, what were they?

If you aren't familiar with the Trip Log, read the Mark V User's Manual, GEH-5979, section on the Trip Log. And make sure everyone on site reads it and knows how to use it. And then, *every* time there is a trip view the Trip Log *before* anything else is done (except silencing the audible Alarm Horn) and print the Trip Log for review and analysis. Do this for every trip, even if you already are sure what caused the trip as it gets people in the habit of printing the Trip Log on every trip. Although it doesn't have everything in it, and it's resolution sometimes leaves something to be desired, it's better than nothing. It can be a very useful tool at times.
 
H
Thank you for your kind reply. We checked the trip Log but I think it was deleted or acknowledged, only the Post trip alarms exist.

I have one question: can one thermocouple of the exhaust cause trip for the GT if it has a problem?

hany saleh
 
Poor maintenance of lubricating oil (it does require more than just changing the filters when the differential pressure gets high) is the usual cause of servo-valve problems.

Lubricating oil conditioning and maintenance is becoming a "science" (as technology proliferates) and many vendors can offer all manner of services, some for free or a very nominal charge in an effort to sell their products.

But, one simply can't fill a tank with lubricating oil, change the filters when they get dirty, add some now and then, and expect to have trouble-free operation of hydraulic systems for decades. Just doesn't happen that way. And servo-valves get the blame!

Most plant maintenance people are absolutely astounded when, after twenty years, they drain the lubricating oil tank and clean it before refilling it. They just can't believe the accumulations of dirt and rust they find in the bottoms of the tanks--even when they practice good maintenance and housekeeping practices on the outside of the tanks and around the accessories and turbines.

Yeah; it could be the servo. But consider the operating conditions the servo has to endure before just continually replacing the servo or saying, "We must have gotten a bad batch of servos!"

As has been said elsewhere on this site and many times as well, servos are one of the most misunderstood--and most often improperly blamed--components on a GE-design heavy-duty gas turbine.
 
B

Bob Johnston

1) One single exhaust thermocouple should never cause this to happen.

2) This highlights the importance of operating the Trip Log correctly. Try to get into the habit of immediately printing the Trip Log Report. I know the "Update Saved Log" buttons, etc. are a bit confusing but, read the manual, get to understand how to use the log. Without Trip History you would pretty well need a crystal ball to try to analyze the cause of the trip.

3) If a sticky servo-valve caused a High Exhaust Temp. trip on a MKV machine, it would be the first I have ever seen.
 
Every time the unit is re-started, the contents of the Trip Log (in the Mark V <C> RAM) are discarded and a new Trip Log is started. (The logic signal L4 is the trigger for the Trip Log.) That's why it's *so* important to print the Trip Log after *every* trip, almost immediately (after silencing the audible alarm horn) to have the operator print the Trip Log.

As for the exhaust T/C question: Under *normal* circumstances, a single failed exhaust T/C (whether it fails high or low) should be rejected by the Combustion Monitor to prevent a trip. *Normal* to many operators (and technicians) means the unit is running, regardless of the Diagnostic- or Process Alarms annunciated at the time. Of course, *normal* really means all exhaust T/Cs reading properly, no unresolved and no neglected (because of not attempting to understand the meaning/source of the condition) Diagnostic Alarms. Just because the turbine is running, doesn't mean everything is okay.
 
Top