Dear Sirs,

we are running a simple cycle GT-Fr9 on crude oil. one of the units started to trip by "high exhaust spread", our mechanical department used "SWIRL" program to identify affecting cans. they replaced 3 fuel check valves. we ran the unit again but same problem happened then they replaced 4 check valves, and we start the unit again but the same trip happened.

After each start-up, SP1 started to increase firstly at high load (95MW). then after like 90 minutes the same value of SP1 recorded but at lower load (85MW). by counting operational hours the occurrence of the same value of SP1 keeps to happened but at lower loads up to (60) MW.

QUESTIONS:

1- why SP1 value started to be high at high load (our case 95MW/sp1=52) then the same 52 happened at 85MW then after hours the same 52 happened at 60MW.

2- what are the possible causes of high spread?

3- is the value of SP1 that matters? or is it the difference between SP1 and allowable spread that matters?

 

Maythem,

Swirl charts are not always accurate; they are slightly better than guessing and only when used by knowledgeable people. They are designed for particular groups of machines with particular compressors and combustors and exhaust diffusers. They are not (contrary to popular belief) one-size-fits-all for any Frame-size GE-design heavy duty gas turbine.

Can annular combustor gas turbines must have equal amounts of fuel flowing into each combustor's fuel nozzles in order to have a uniform hot gas temperature entering the turbine section. (This presumes all of the combustion hardware is intact and not cracked or leaking.) Liquid fueled units have a flow divider which takes the output of the liquid fuel control valve (usually a bypass valve) and divides the fuel equally to all the combustors, which have similarly-sized fuel nozzles. This should result in similar fuel flow-rates and combustion in each combustor, which should result in the same hot gas temperatures entering the turbine section, which should result in uniform exhaust temperatures as the gases leave the turbine section.

There is a manually-operated selector valve (sometimes called a "switch"...) at the liquid fuel flow divider. When the same amount of fuel is flowing into each combustor the fuel pressure to each combustor will be very nearly identical. Using the selector valve to monitor liquid fuel pressures is the BEST way to determine which combustor or which combustor's components (liquid fuel check valve; purge air check valve (if a dual fuel (gas/liquid) machine); fuel nozzle) is experiencing some issue which results in a higher or lower fuel pressure. Failed check valves (that don't maintain the proper cracking pressure) usually result in lower than normal liquid fuel pressure. Plugged check valve or plugged fuel nozzles usually result in higher than normal fuel pressures. Leaking liquid fuel purge air check valves (if so equipped) also usually result in lower liquid fuel pressures.

If the unit has a liquid fuel purge system, a leaky or failed purge check valve will result in a discharge out of the system's Tell-Tale Leakoff (see the appropriate P&ID for the liquid fuel and/or liquid fuel purge system). It's a fast way to identify if one or more liquid fuel purge check valves are leaking, and the manual selector valve will tell you which purge check valve is leaking (a lower pressure than the others).

Higher fuel pressures to a particular combustor or combustors usually mean there is some restriction to fuel flow in the fuel nozzle(s) or the liquid fuel/fuel purge components. Lower fuel pressures indicate a problem with a liquid fuel check valve or a liquid fuel purge check valve. But, liquid fuel pressure differentials will almost always change when there is a problem with the liquid fuel system/nozzles.

Gas fuel fired machines don't have a manual selector valve and gauge to check fuel pressures to each combustor, and so finding problems resulting in spreads can be more difficult.

If a combustion liner or transition piece is cracked and excess compressor discharge air is flowing into the hot gas path the fuel pressures will be nearly equal for all the combustors, but the exhaust temperature spread will increase--a cold spot will develop in the exhaust temperature profile. This is where a swirl chart can be helpful--but, again, only if it's for the particular machine at the site. Differences in axial compressors and exhaust diffusers and combustion hardware can all have a large impact on swirl angles.

So, those are the most common causes of exhaust temperature spreads for liquid fuel-fired machines. Spreads will generally get worse over time if the problem isn't resolved. Crude oil is usually not very well filtered and can have a lot of different constituents which can start to collect in small passages/orifices (such as in check valves), and can even begin to plug fuel nozzle orifices. But, using the manual selector valve/gauge on the liquid fuel flow divider can help to quickly identify the particular combustor which is experiencing problems.

The issue with exhaust temperature spreads is two-fold. When the same amount of fuel is being burned in each combustor the pressures in each combustor are nearly identical, which means there is no flow through the cross-fire tubes which interconnect each combustor to it's adjacent combustors. When the amount of fuel flowing into one combustor is reduced the pressure in that combustor is reduced and that can cause flows into that combustor through the cross-fire tubes of the adjacent combustors--which can damage the cross-fire tubes and also cause damage to the combustion liners of the three combustors. This usually requires a very high pressure differential--but at sites which "bypass" spread alarms can eventually see the cross-fire tubes glowing red (which is REALLY bad--and dangerous).

The other issue is that the rotating turbine buckets pass each combustor 50 times per second. If the gases from a particular combustor are much hotter or much colder than the combustors on either side of it the turbine buckets are experiencing very fast heating/cooling cycles on every revolution--which is NOT good for the turbine buckets if allowed to continue for long periods of time, and will eventually lead to damaged buckets and even catastrophic failure of the turbine section.

Hope this helps! One has to remember that it's important to keep the same amount of fuel flowing and combusting in each combustor to keep the hot gas temperature of each combustor as close to the others as possible. And that's to protect the turbine buckets from experiencing high thermal stresses on every revolution (and on a Frame 9E that's 50 times per second!). Temperature differentials can be (and are usually) caused by uneven fuel flows into a combustor, but can also be caused by cracked or broken combustion liners and/or transition pieces. But, for liquid fuel-fired machines use of the manual selector valve/gauge can help to pinpoint the problem combustor(s) and even to eliminate liquid fuel issues as the cause of high exhaust temperature spreads.

By the way--there is a 'Search' field cleverly hidden at the far right of the Menu bar at the top of every control.com webpage. And, this particular topic--exhaust temperature spreads on liquid fuel-fired machines--has been covered many times in the past on control.com.

Please write back to let us know what is found and how the spread problem is resolved.

 

A little further from CSA's reply

1) One of the main causes of high spreads are faulty fuel nozzles, usually the liquid tip being clogged or distorted. When the machine is started the spread will increase as the load increases as the uneven flow from the defective nozzle has more and more. This can become more and more apparent the longer the machine runs and will show up at lower loads as the temperature differential caused by the faulty nozzle increases. Using the "swirl" effect to pinpoint the faulty combustion cans is very inconclusive, in my own opinion, pretty near worthless. As CSA has pointed out, the true test is checking individual nozzle pressures.

2) For me, and from experience, the No. 1 cause of spreads is faulty fuel nozzles, probably followed by check valves (which are easy to check by looking at tell tale flows).

3) We are talking about a control system here, SPL is setpoint and SP1 is the actual. Like any control system, when the actual exceeds the setpoint, you have a problem. If you look closely at your system you find taht have an SP2 and possibly an SP3. Below I have given you an extract from the Combustion System Monitor document which gives an explanation of trips are calculated.

a. SPREAD #1 (S1): The difference between the highest and the lowest thermocouple reading

b. SPREAD #2 (S2): The difference between the highest and the 2nd lowest thermocouple reading

c. SPREAD #3 (S3): The difference between the highest and the 3rd lowest thermocouple

High Exhaust Temperature Spread Trip
(L30SPT)

A high exhaust temperature spread trip can occur if:
1. "COMBUSTION TROUBLE" alarm exists, the second largest spread exceeds a constant (usually 0.8 times the allowable spread), and the lowest and second lowest outputs are from adjacent thermocouples

2. "EXHAUST THERMOCOUPLE TROUBLE" alarm exists, the second largest spread exceeds a constant (usually 0.8 times the allowable spread), and the second and third lowest outputs are from adjacent thermocouples

3. the third largest spread exceeds a constant (usually the allowable spread) for a period of five minutes

If any of the trip conditions exist for 9 seconds, the trip will latch and "HIGH EXHAUST TEMPERATURE SPREAD TRIP" message will be displayed.

Thanks again to CSA's super explanation

 

Thanks to glenmorangie's great explanation of combustion monitoring!

 

Dear Sirs,

Thank you both for all the efforts spent in explaining;kindly note the below:

1- the fuel pressure after flow divider is almost identical.

2- our control department engineer reduced the FSRMAX constant to be 62% instead of 100% for the purpose of load limitation to prevent high exhaust spread trip to happen at high load; it is a spin around the solution.

3- our mechanical department opened one can (i think it is can no 6) and as they said there is nothing abnormal

4- all check valves have been replaced by new ones

5- 3 nozzles have been changed also

6- the problem still persist but now the load is limited to 85MW ; SP1=36; SPL= 66.

i will keep you posted with the updates when happened.

regards,

 

Changing only 3 fuel nozzles is probably the worst thing you could have done, Don't you understand that fuel nozzles are a flow tested set, not individual nozzles.

How did you know which nozzles to change? I think I said already, the first thing you need to do is remove all the fuel nozzles and replace them with a set of at least repaired and flow tested nozzles.(You can repair and flow check the removed nozzles and keep them as a set) As you have changed the check valves and you reckon all the nozzle pressures are OK, if you still get a spread, it is not the fuel system.

Below is another extract fro a document on the combustion monitoring system

COMBUSTION TROUBLE MODES TO BE CONSIDERED
A. Combustor
1. Failed Liner (Cracked or Burned)
2. Failed Transition Piece (Cracked or Burned)
3. Collapsed Liner
4. Hot Crossfire Tubes

B. Fuel System
1. Break in Liquid Fuel Line
2. Break in Gas Fuel Line
3. Plugged Check Valve
4. Check Valve Stuck Open/Closed
5. Liquid Fuel in Gas Manifold
6. Stuck Flow Divider
7. Failed Fuel Pump

C. Fuel Nozzle
1. Plugged Fuel Nozzle (Liquid or Gas)
2. Unscrewed Fuel Nozzle
3. Fuel Nozzle Erosion
4. Red Hot Fuel Nozzle

D. Atomizing Air System
1. Break in Atomizing Air Line
2. Faulty Purge System
3. Atomizing Air Compressor Failure
4. Plugged Atomizing Air Passage at Manifold or Nozzle

E. Pressure Vessel Integrity
1. Cracked Combustor Casing
2. Blown Gasket
3. Damaged Crossfire Tube Piping
4. Cracked or Blown Sight Port
5. Leakage at Flame Detector or Spark Plug

F. First Stage Nozzle
1. Burned Out First Stage Nozzle
2. Plugged First Stage Nozzle

As you can see, a lot of things can cause a high spread, it is a process of elimination.

Can you give us some history to help us evaluate. Did the spread increase suddenly or progressively? When was this machine last overhauled, particularly the combustion system. Has anything else changed (Fuel,fuel supplier,fuel storage)?

I'm also not so happy about reducing FSRMAX either, it could have other implications (recovery after sudden load increases for example), why don't you just put a MW limit on the machines operation?

Good luck and give us some answers

 

Can you also tell us what is Spread 2 TTXSP2 when TTXSP1 is 36 Deg. also can you give us the complete Exhaust Temp. profile, all thermocouples with positions.

 

>Temperature differentials can be (and are usually)
>caused by uneven fuel flows into a combustor, but can also
>be caused by cracked or broken combustion liners and/or
>transition pieces. But, for liquid fuel-fired machines use
>of the manual selector valve/gauge can help to pinpoint the
>problem combustor(s) and even to <b>eliminate liquid fuel
>issues as the cause of high exhaust temperature spreads</b>.

If the liquid fuel pressures for the fourteen combustors are within a range of less than 10% of each other when the spread problem exists then it's pretty likely the problem isn't related to the liquid fuel system--it's a combustion liner or transition piece or side seal or some other issue like a problem in the exhaust diffuser/compartment, etc.

As glenmorangie said--the usual cause is a flow nozzle issue--but it's NOT always the fuel system. And if you've eliminated the fuel system components by use of the manual selector valve/gauge then you need to move on to the other possible causes.

And if the problem is getting worse and occurring at lower and lower loads then the likelihood of a serious catastrophic failure of the turbine section is increasing every time the machine is run and gagging the FSR is not the answer to protect the machine.

Use a borescope to look at the liners/transition pieces when you pull nozzles--even if you have to pull all the nozzles. While this is happening, go into the exhaust and inspect the diffuser and exhaust T/C radiation shields for damage. It's entirely possible to look for problems in multiple areas simultaneously. As glenmorangie says--troubleshooting is often a process of elimination. It seems you've eliminated the fuel system--start looking at other possible causes. And realize that it may not be long if the unit is left running before it comes crashing down and the cause will become apparent and cost even more in parts and lost generation to repair.

And please write back to let us know how you progress and how the problem is resolved!

 

Dear Sirs,

i work in shifts, yesterday i was on my day shift. on my days off the control dept once again reduced FSRMAX to 52% due to exhaust spread started to increase even when they set FSR to 62% so this is the case now. Higher authorities need the unit to keep running even with part load.

CSA, what is the T/C shield you had mentioned? could you explain it for me?

we are running on droop mode with a very unstable grid frequency, that is why even if we limited the load in operator screen by preselect load the MW will eventually increased when the frequecy drops.

below is the requested readings:
Fuel: crude oil
FSR: limited to 52%
MW: 69.4
SPL: 50.3
SP1: 31.2
SP2: 24.3
SP3: 21.2
TTXM: 420.4

TTXD's in order:
429, 426, 426, 421, 420, 417, 414, 426, 420, 421, 422, 421, 422, 421, 415, 414, 405, 398, 408, 423, 420, 424, 429, 428

4 months ago the unit was overhauled (hot path); the spread problem started to happened 20 days ago with no significant difference in parameters except the type of Vanaduim inhibitor which had been changed, it has been changed at 05/05/2016; the unit started to trip due to spresd at 06/05/2016; however other units also uses the same MGO material weeks before the said unit but with no significant impact to their spreads readings.

the problem is still persist, your kind help is highly appreciated.

thanks,
Maythem

 

Maythem,

The problem at your site is a cold spot. The highest exhaust T/Cs are 429 deg C; the lowest is 398 deg C; the "median" exhaust temperature (420.4 deg C) is closer to the highest values, and there are a couple of low values on either side of the coldest T/C value (which I deem to be located at position 18 from the data you provided).

Does the location of the cold spot move with changes in load--or does it always seem to remain in the 17/18/19 region?

Because if the location of the cold spot doesn't change with load then it could be an "instrumentation issue." The exhaust T/Cs are inserted axially into tubes called "radiation shields" in the aft wall of the turbine exhaust. These tubes have a cover over then end of the tube and generally have two openings, one on either side of the tube 180 degrees across from each other, and oriented so the two openings are in a radial line out from the axis of the machine.

When properly inserted into the radiation shield, the tip of the exhaust will be a few mm from the cover at the tip of the radiation shield tube--not touching the cover at the tip of the radiation shield tube. All of them should be uniformly located in about the same position--with the tip of the exhaust T/C (the tip is usually tapered to a dull point) visible in the openings of the radiation shield. The exhaust T/C tips should be aligned with the axis of the radiation shield tube and not bent to one side or the other. (People working in the exhaust of the unit tend to like to bend the T/C tips with their fingers....) But, again, the tips should not be touching metal and should be uniformly visible in the openings at the end of the radiation shields.

It's possible for insulation from the flexible seals or behind the metal plates of the exhaust plenum to get stuck in one or more of the radiation shield openings (they are small oval openings on either side of the radiation shield, probably about 2-2.g cm in diameter). This can prevent the flow of exhaust gases through the radiation shield opening--and can cause a cold spot, one which doesn't change location with changes in air flow or load.

It's also possible for a leaking compressor bleed valve (which flow into the exhaust compartment) to blow cold air in the direction of one or more exhaust T/Cs causing a cold spot--one that doesn't change location with changes in air flow or load.

It's also possible for cracked or damaged exhaust diffusers to cause uneven exhaust gas flows which can cause a cold spot--one that doesn't change location with changes in air flow or load.

I've heard of--but not actually witnessed--exhaust T/C failures that led to cold spot, ones that don't change location with changes in air flow or load.

AA (Atomizing Air) is also very important to combustion of liquid fuel. BUT, the discharge of the Main AA Compressor flows into a ring manifold around the axial compressor and then into the AA passages of the fourteen fuel nozzles. This type of manifold design generally insures even AA flows into each combustor. I have seen balls of duct tape and weld slag and rocks and even cigarette butts get lodged in the internal AA passages of fuel nozzles. But, this is not common--not impossible, but not common.

Improper operation/adjustment of the AA Pre-cooler Temperature Regulating Valve can cause AA compressor inefficiencies and even failures over time--but this doesn't generally result in spread problems.

So, exhaust temperature spreads are not just about the magnitude of the temperature differentials, when trying to troubleshoot spread issues the location, or change of location, of cold (or hot spot, if that's the case) needs to be considered, also. If the cold (or hot) spot doesn't move with changes in air flow and load, then it's probably not a fuel or combustion hardware issue.

You've said the fuel nozzle pressures are all pretty uniform--and that's good because it basically eliminates liquid fuel system components (liquid fuel check valves; leaks; leaking purge check valves; fuel nozzle blockages, etc.). And, if the location of the cold spot is "stationary"--meaning it's always in the 17/18/19 region of the exhaust that pretty much eliminates that "new" combustion hardware. That leaves T/C installation and exhaust compartment problems.

Two final questions: First, it's common for controls tech's to "temporarily" jumper a working exhaust T/C to a failed exhaust T/C. Has this happened at your site and are any exhaust T/Cs presently jumpered ("temporarily")?

Second, which version of Speedtronic turbine control system is in use at your site?

Please write back to let us know what you find--and to provide the answers to the questions above! (Does the location of the cold spot move with load? Are any exhaust T/Cs "temporarily" jumpered"? Which turbine control system is in use on the turbine at your site?)

 

I'm surprised that you don't see any difference in fuel nozzle pressures, the three temperatures 405,398,408 seem to show where the problem is, if those three temperatures could be normalized your spread would drop to about 10 degrees. I would still recommend a change of the full set of fuel nozzles which have been flow checked if that is possible, it could be a quick and cheap fix. Looking at the spread pattern I would guess that it is only one nozzle that is causing the problem. It is hard to analyze if the change to vanadium inhibitor could be causing the problem but, if you have one bad nozzle specifically on this machine, it is maybe producing more impact from the change.

If you don't do the nozzle change, you are really heading for a Combustion Inspection and that is time and money. As CSA has said, if you don't get this fixed quickly, you are heading for some combustion system damage fairly soon.

 

Maythem,

A true cold spot is caused by either a reduced fuel flow-rate into a combustor (plugged fuel nozzle; non-functioning liquid fuel check valve; leaking purge air check valve; etc.), <b>OR</b> excessive air flowing into a combustor (cracked liner or transition piece; holes in liner or transition piece; bad side seals; etc.). That's for a combustion problem.

As was said above, a "fake" cold spot (one that doesn't move with load) is an "instrumentation" issue, either a failed or failing T/C, an improperly inserted T/C, something blocking the flow through the radiation shield openings, or some problem with exhaust gas flow, or a leaking compressor bleed valve blowing air in the direction of one or more exhaust T/Cs.

Please keep us informed!

 

Dear Sirs,
thank you for all the efforts spent in explaining the spread possible causes and remedies.

the unit controller is MK VIe; i asked our control dept. and they said that all thermocouples are expressing themselves on the screen and they did not use one for two channels.

i am not so sure about the cold spot moving, i guess it is kinda fixed when they reduced FSRMAX; below is a data dated May-08-2016 for your reference(it was before they replaced 3 nozzles but after changing some check valves [maybe 7 of them])

Fuel: crude oil
FSR: original set 100%
MW: 69.1
SPL: 48.6
SP1: 34.5
SP2: 29.9
SP3: 24.4
TTXM: 405
TTXD's in order:
409.4, 407.9, 409.5, 408.6, 410.3, 408.8, 405.2, 412.7, 409.7, 411.6, 414, 410.2, 388.2, 404, 412.4, 410, 411.2, 408.1, 408, 409.4, 393.6, 389.8, 382.8, 378.2.

thanking you as always,
Maythem

 

Maythem,

I haven't asked about Water Injection (for NOx emissions reduction).... I assume it's not being used, but that's not a good thing to do. Does the unit use Water Injection for NOx emissions reduction, and if so, is it running during these periods of high exhaust temperature spreads?

And, based on the data you just provided, the location of the cold spot is moving--it was located at position 13 (more or less).

You say the unit has been tripping on high exhaust temperature spread; exactly what Process Alarms were annunciated for say, 30 seconds before the trip? As glenmorangie has said, the unit won't (generally) trip strictly on the difference between TTXSPL and TTXSPn, it also looks at the adjacency of the hottest and coldest T/Cs AND the difference between TTXSPL and TTXSPn.

Another assumption I've been making is that the unit has conventional combustors--not DLN-I combustors. (I've never seen a unit which burns crude oil use DLN-I combustors (for gas fuel), but then I haven't seen every GE-design heavy duty gas turbine--and the Belfort bunch have a way of doing unorthodox things just because they (think) they can). So, what kind of combustion system does the unit have--coventional (single fuel nozzle body per combustor, perhaps for gas fuel as well), or multiple nozzles per combustor with various combustion modes when operating on gas fuel?

These data snapshots you are taking--are they taken when the grid frequency (and turbine-generator speed) are near rated and relatively stable, or during grid frequency excursions?

Does the unit exhaust to atmosphere, or through an HRSG (Heat Recovery Steam Generator--a boiler)?

Can you ask your controls department to tell you which version of the Combustion Monitor algorithm is in use in the Mark VIe? It should have a "code" like TTXSPV4, or TTXSPV5, or something similar.

Again, continuing to run the unit and having to keep decreasing FSRMAX (in effect, gagging FSR) probably means there is something pretty seriously wrong with the combustion hardware--and getting worse. And, it would seem the cold spot moves with load, which is, again, indicative of some combustion problem.

I agree with glenmorangie--if the site has a full set of nozzles (from which the three which were swapped out were removed) it would be a very good idea to swap out all the remaining nozzles. And, then to closely examine the nozzles which were removed for signs of plugging/obstruction--both the liquid fuel passages, as well as the AA (Atomizing Air) passages. And, if we're talking about conventional combustors, it would be extremely advisable to use a borescope to inspect the combustion liners and transition pieces when the fuel nozzles have been removed for any signs of damage (cracking; holes; etc.). It's not always easy to see, for example, the "hula seals" of the end of the combustion liner where it fits into the round opening of the transition piece, but it should be checked as best as possible.

Again--the fact that FSRMAX has to be continually reduced because the spread seems to be increasing (if I understand what's happening correctly) is not a good omen. And, the cold spot appears to be moving--so it's either a fuel system problem (which it doesn't really sound like, but we haven't seen the pressure readings from the manual selector valve/gauge at the liquid fuel flow divider when the grid frequency was stable) or a combustion hardware problem. It could be a problem in the exhaust, but based on the information provide it doesn't seem so.

It could be a failing or intermittent couple of exhaust T/Cs causing the problem; not likely, but possible. Also, wiring could be an issue in the T/C circuit; again, not very likely, but possible. But, knowing what Process Alarms are active and which ones are being annunciated during high spread operation just before the trips would be VERY helpful. Also, if it's possible, could someone have a look at the STATUS LEDs on the various Mark VIe I/O Packs and tell us which ones have yellow LEDs for STATUS (an indication of Diagnostic Alarms)?

Lastly, is the Mark VIe a TMR control system, or a DUAL redundant control system, or a SIMPLEX control system?

 

Dear CSA,

when i came to night shift two days ago the designated unit was off for off-line washing, day shift finished washing, Mechanical dept. replaced fuel nozzles for chambers 7 and 8.

we had the chance to start the unit, it was ok, i guess the problem have been solved, now FSRMAX is 100% , L=92 MW, SP1=30, SPL=61.

FYI, it is a TMR controller, TTXSPV4 algorithm.

thanks,
Maythem

 

Maythem,

Thank you for the feedback!!!

Thanks, also, to glenmorangie for his assistance and input!

 

CSA
you have said ; if the location of the cold spot doesn't change with load then it could be an "instrumentation issue
could you explain it please? why it could be

 

Process12,

This is copied from one of the responses to this thread above:

”As was said above, a "fake" cold spot (one that doesn't move with load) is an "instrumentation" issue, either a failed or failing T/C, an improperly inserted T/C, something blocking the flow through the radiation shield openings, or some problem with exhaust gas flow, or a leaking compressor bleed valve blowing air in the direction of one or more exhaust T/Cs.”

A T/C is part of the instrumentation on the unit connected to the turbine control system. So, if there is a problem with one or more T/C's, or something that is preventing exhaust gas flow across the T/C tip, or even something blowing colder air across the T/C (such as a leaking compressor bleed valve) that would mean the instrument (a T/C in this case) is the reason for the exhaust temperature abnormality. (Admittedly a blockage preventing exhaust gases from flowing across the T/C tip or a leaking compressor bleed valve blowing cold air on the T/C isn't a problem with the T/C--but they are causing the T/C reading to be abnormal, so the root cause (a radiation shield problem or a leaking compressor bleed valve) is affecting the instrument (T/C) reading.

If the T/C is failing, or is blocked, or has a source of cooler air blowing across it then the temperature reading from that T/C will not change with load, or it will not change much. Under normal circumstances as load changes and IGVs move a cold spot will move, say from the 8:00 o'clock position to the 12:00 o'clock position. So, the cold spot will change (move) from one T/C to an adjacent T/C. If there is a problem with a particular T/C or its ability to read exhaust gas temperature then the reading from that T/C will always be abnormal even as load changes and IGVs move. So. If exhaust T/C 11 was always reading lower than the other exhaust T/C's as load changes and the IGVs moved that would be a ”fake” cold spot because it's not the result of a combustion problem or hot gas path hardware problem.

A ”true” combustion, problem--say a plugged fuel nozzle in combustor #7--will result in a cold spot that moves when load and IGVs move. That means if the coldest T/C reading at 5 MW was at exhaust T/C #5, and as load increased the coldest exhaust reading ”moved” to exhaust T/C #6 then to exhaust T/C #7 then to exhaust T/C #8--that would indicate the problem was combustion related. But if the value of exhaust T/C #5 is always the lowest value regardless of load and IGV angle, then that's more an indication of an instrumentation problem than a combustion problem.

Hope this helps!