Dear ALL,

In our frame-6 GE gas turbine controlled by TMR Mark-V Control system, we got a trip signal on "HIGH EXHAUST TEMPERATURE SPREAD TRIP" Drop # 160.

We have 18 thermocouples. I do not know why this trip although the difference between higher and 3 lowest exhaust temp. is not exceeding 25C. we have 2 TC with up normal reading from long time (-83C) which I think should not enter into calculation and should be excluded by the system.

If now we have these 2 TCs NOs 4 & 18 reading is -83. Do this value exclude from calculations or not and what are the maximum numbers of TCs that can be excluded??

How I can calculate TTXSPL value???

I appreciate if someone can explain and help me in this subject.
Thanks in advance.....

 

Also Please find the exhaust temp. before, during, and after the trip:

TTXD1_1 TTXD1_2 TTXD1_3 TTXD1_4 TTXD1_5 TTXD1_6 TTXD1_7 TTXD1_8 TTXD1_9 TTXD1_10 TTXD1_11 TTXD1_12 TTXD1_13 TTXD1_14 TTXD1_15 TTXD1_16 TTXD1_17 TTXD1_18
____________________________________________________________________________________________________________
533 537 534 -83 556 534 550 544 554 554 546 550 555 545 543 537 539 -83 (at 10:57:49.000) before the trip, O/P Power = 30MW
____________________________________________________________________________________________________________
533 536 535 -83 555 534 550 544 553 554 546 550 555 545 543 536 539 -83 (at 10:57:50.000) during the trip, O/P Power = 0MW
____________________________________________________________________________________________________________
480 455 399 -83 503 499 488 460 469 488 488 504 494 470 461 455 458 491 (at 10:57:51.000) after the trip, O/P Power =0MW
____________________________________________________________________________________________________________

 

So, what makes you "... think [the two failed exhaust T/C readings] should not enter into calculation and should be excluded by the system...."? Is this something you read in a manual, or the way you interpreted the Combustion Monitor algorithm in the CSP?

The Combustion Monitor block, usually TTXSPVn, checks for two "conditions": the magnitude of the spreads (TTXSP1, TTXSP2, and TTXSP3), and the adjacency of the highest and lowest readings. When the magnitude of the spreads exceeds certain parameters <b>and</b> the outlying T/C readings are adjacent to each other, then--and only then--will a trip be initiated.

MANY people confuse the Combustion Monitor algorithm with the Exhaust Temperature Feedback and Protection Algorithm (usually TTXMVn). The latter rejects all T/Cs less than a certain value, and then excludes the highest and the lowest values from the remaining T/C values <i>when calculating the "median" exhaust temperature, TTXM</i> which is not used in the spread determination.

To calculate thee TTXSPL value, you need to use the CSP along with the Control Constant values used by the TTXSPVn algorithm in the CSP in your unit. Note that TTXSPL is a function of axial compressor discharge temperature, so good working CDTAn T/Cs are also very important (which many sites neglect, as well).

By the way, running with two failed T/Cs is just asking for trouble. Most sites jumper (which can be kind of dangerous, also) a good, working exhaust T/C to a failed one. But it shouldn't just be any working exhaust T/C, but one from the either side of the failed T/C selected in order to maintain a more proper profile for the Combustion Monitor.

 

can you please explain in a simplified way to quick understand.
we have same issues at our site, but i was confused to get how TTXSPL was calculated.

 

Without being able to use the figure from the CSP/application code running in the Speedtronic turbine control panel at your site it's very difficult to explain easily.

However, if you can find the signal TTXSPL and where the signal is "written to" you should be able to work backwards from that point.

It should be noted, that I have NEVER found the calculation of TTXSPL to be wrong--except when the axial compressor discharge temperature thermocouples aren't working properly.

In general, TTXSPL is a calculated value, with an upper limt, and I believe most algorithms in use in GE Speedtronic turbine control panels use the axial compressor discharge temperature thermocouple readings prominently in the calculation. A lot of people mistakenly believe the axial compressor discharge T/Cs are not important, but they are--and they are even more important when the turbine is equipped with DLN (Dry Low NOx) combustors.

At any rate, if you would describe the problems you are having we may be able to provide better help. We find, frequently, that when people ask questions like this they are focused on this one signal or aspect as the cause of their problem(s) and we find, if more information about the problem(s) are provided that the real cause is something quite different (not always--but very frequently).

I also find that when people are having exhaust temperature spread problems that they have one or more failed T/Cs and are really trying to understand how may failed T/Cs they can continuously operate with (usually they want to run with two or more failed T/Cs ...).

Finally, it's worth noting that most exhaust T/C spreads are real--that is, they are not caused by that mystical, magical, complicated Speedtronic turbine control system with all those wires and LEDs and alarms (<b>SO MANY</b> alarms). People want to believe that after months or years, or even decades, of trouble-free operation that suddenly and without any indication (that anyone is paying attention to) that the allowable exhaust temperature spread and the Combustion Monitor algorithm has decided to cause problems. According to the Mechanical Department, it simply can't be a mechanical problem--which it almost always is (a mechanical problem).

Sure, it's good to make sure the exhaust thermocouples are installed properly and are wired properly and aren't failing because of high heat and old T/C terminal boards and incorrectly connected T/C wiring. But, in the end, what's usually discovered--after a <b>LOT</b> of wasted effort trying to prove the problem is that darned Speedtronic turbine control panel--that the problem is mechanical, real combustion trouble. Usually by the time that realization is arrived at the turbine has tripped and/or won't run or has suffered catastrophic problems--which, in the end, are (mistakenly) attributed to the Speedtronic because if that darned Speedtronic had been working correctly the damage would not have occurred. (Never mind that operators and Operations Supervisors are under extreme pressure to keep the unit running, and the mechanical people don't want to tell Operations the unit needs to be shut down to check the combustion system.)

Anyway, help us help you by telling us the problem(s) you are experiencing, when the problem(s) started (MOST important!), and what you've done to troubleshoot the problem(s) and what the results of the troubleshooting were. Also, if you can tell us what fuel the unit is operating on, what Process- and Diagnostic Alarms are active on the operator interface we can provide the best assistance in the shortest period of time.

If you have a Mark V Speedtronic turbine control panel, there is a description of how to "read" and interpret the CSP and Big Blocks in the back of the Mark V Application Manual, GEH-6195, in one of the Appendices. (I don't have access to any Mark V documents at this writing, so it might be in the back of the Mark V Maintenance Manual, GEH-5980. But it's in the back of one of them.) It's not too difficult, really; it just takes a little practice.

 

gosp engr,

If by "...we have same issues at our site...." you mean you have two or more failed T/Cs and are asking, essentially, how many failed T/Cs you can run with, the answer is essentially: two (2). When a third T/C goes "low" that's enough to trigger one of the exhaust temperature spread conditions and result in a trip.

The problem is not TTXSPL (the allowable spread), because the Combustion Monitor function doesn't trip solely on the magnitude of difference between the highest and the lowest. (And to be clear, NO exhaust T/Cs are excluded from the Combustion Monitor function--only the median exhaust temperature calculation.) The difference has to exceed some values, determined by a couple of Control Constants, <b>AND</b> the highest and lowest T/Cs have to be adjacent--or as happens when a third exhaust T/C fails low, a trip will occur.

 

Hi CSA,

Can you also explain that if a TC is failed (bad), does it get excluded from spread calculation (TTXD1n & TTXD2n)? I know that if a TC is failed then it gets excluded from TTXM (Average exhaust temperature). but my main query is does the failed TC also gets excluded from spread calculations or does it's value just get replaced by zero value so that it would become the lowest TC? Please clarify.

Thanks,

 

You have 3 Spreads, TTXSP1/2/3. SP1 is difference between highest and lowest TC, SP2 is difference between highest and 2nd lowest and SP3 is difference between highest and 3rd lowest. So the failed TC should not show up in SP2 & 3.

A. Exhaust Thermocouple Trouble Alarm (L30SPTA)
If any thermocouple value causes the largest spread to exceed a constant (usually 5 times the allowable speed) a thermocouple alarm (L30SPTA) is produced. If this alarm persists for four seconds, the alarm will latch and the alarm message "EXHAUST THERMOCOUPLE TROUBLE" will be displayed and remain on until acknowledged and reset.

B. Combustion Trouble Alarm (L30SPA)
A combustion alarm can occur if a thermocouple value causes the largest spread to exceed a constant (usually the allowable spread). If this alarm persists for three seconds, the alarm will latch and the "COMBUSTION TROUBLE" message will be displayed and remain on until it acknowledged and reset.

C. High Exhaust Temp Spread Trip (L30SPT)
A high exhaust temperature spread trip can occur if a combustion trouble exists and the second largest spread exceeds 0.8 times the allowable spread and the first and second lowest thermocouples are adjacent or if an exhaust thermocouple trouble exists and the second largest spread exceeds 0.8 times the allowable spread and the second and third lowest thermocouples are adjacent or if the third spread exceeds 0.8 times the allowable spread or if a combustion trouble exists and a controller failure occurs. If any of these conditions exist for 9 seconds, the trip will latch and "HIGH EXHAUST TEMP SPREAD TRIP" message will be displayed. The turbine will be tripped through the master protective circuit. The alarm and trip signals will be displayed until they are acknowledged and reset.

Hope this explains it. Post back if you need more help

 

you are the best Mr. glenmorangie.

thank you very much

 

following information about 7Fa machine i hope it will be helpful for you...

The purpose of the combustion monitor is to reduce the likelihood of extended damage to the GT if the combustion system deteriorates. The monitor does this by examining the temperature control system exhaust temperature thermocouples and compressor discharge temperature thermocouples.

Three spreads are calculated from the exhaust thermocouple readings:
TTXSP1 (S1)= The difference between the 2nd highest and the
lowest thermocouple reading.
TTXSP2 (S2)= The difference between the 2nd highest and the
2nd lowest thermocouple reading.
TTXSP3 (S3)= The difference between the 2nd highest and the 3rd
lowest thermocouple reading.
The allowable spread is the sum of two values:
a nominal allowable spread and a bias.
TTXSPL-Z1= (0.145 TTXM- 0.08*CTDA + TTKSPL5) deg F
TTXSPL= allowable spread CTDA = compressor discharge temperature
TTXM = exhaust temperature TTKSPL5 = exhaust temp.offset(60 sec)

Nominal allowable spread:
It is the steady state spread limit. It varies, typically, between 50 and 170°F. ( for F class machine)

Bias value:
It is an added to the nominal allowable spread which accounts for the temporary increase in actual spreads that occur during transient operational periods. It is a temperature value which varies b/w 0 - 100F (typically, 100 to 200° F).

When a transient operational condition occurs, such as a rapid change in load, the bias value steps to 100 degrees F.

It remains at that value until 15 seconds (typically, 15 to 60 seconds) after the transient condition ends. Its value then decays exponentially to O F on a time constant of 30 seconds (typically, 30 to 100 seconds).