Unit trips at PMX to LL xfer during almost every shutdown

Our 9E DLN1 unit trips at PMX to LL xfer during almost every shutdown. We have CTXF thermocouples installed in crossfire tubes and a logic that trips the unit if one of them senses temperature higher than CTDA + 32 DegC for about a minute, or if one of them senses overranged temperature - above value MKV able to read and goes to negative, - for 5 seconds. All trip come from the latest. Temperatures start rising the same second with reignition signal and every time between different combustors. Lately we started experience also a problem at transfer from LL to PMX state and switching IBH ON helps, but we noticed it happen mostly after a run on liquid fuel, even though we run in LL mode for 30 min on gas to purge liquid fuel leftovers.
Multiple checks were carried out - all gas valves, purge valves, tightness tests, check valves, tubing connections, borescopes... All good.
Two thing must have been added before I'm switching to "waiting for an answer" mode: 1) We have a very high temperature in the turbine room, about 130-150 DegC, and we've found turbine casing leaking ( In other identical unit we have exactly the same problem but never a problem with DLN transfers ). 2) The unit underwent a Major two years ago and a whole new set of combustors, liners, nozzles was installed. Since then this problem started to appear, at first very rare so no one have connected this fact to the Major Overhaul, and got worse with a time.
After having performed same checks with minimum results ( sometimes we do find stuck check valves ) we strongly think that this unstalled hardware is the cause of a problem, although mechanical engeneering department insists that ist exactly the same hardware as before.

Will appreciate any thoughts and suggestions written.
 
Jolek,

I understand "...All trips come from the latest...." to mean that all the trips which are occurring when the unit is being unloaded and during the Premix Steady-State to Lean-Lean combustion mode transfer are the direct result of the logic added to the Mark V associated with the crossfire tube thermocouples. Please confirm this is the correct understanding.

Do you know why and when thermocouples were added to the cross-fire tubes and the tripping logic was added to the Mark V? Was there some kind of serious problem which was attributed to a combustion problem (cracked or collapsed combustion liners, for example; or fuel nozzle tip damage (if it was fuel nozzle tip damage can you tell us if it was primary or secondary nozzles, or both?))?

Let's talk about what happens when the unit transitions from Premix Steady Stage combustion mode to Lean-Lean combustion mode. I don't have any DLN-I software to look at, but if I recall correctly when the Combustion Reference Temperature (TTRF, or TTRF1 for many units) drops below the specified temperature the first thing that happens is the ignitors ("spark plugs") are energized. The ignitors are located in the Primary Combustion zone where the fuel has been burning in a very lean fuel/air mixture (it's not a diffusion flame, but there is combustion and the fuel is burning--just at a lower temperature than if there was diffusion flame). Then, I believe, when the turbine control system detects diffusion flame in one of the combustors with flame detectors the fuel split is changed to approximately 50-50 (meaning approximately 50% of the fuel flow is directed to the primary combustion zone and approximately 50% of the fuel is directed to the secondary combustion zone).

The ignitors--two of them (they are redundant; only one has to be working)--can only ignite the fuel/air mixture into a diffusion flame in the Primary combustion zone of the combustors where the ignitors are located. The fuel/air mixtures of the remaining Primary combustion zones without ignitors are ignited into diffusion flame through the cross-fire tubes, just like during initial firing of the the gas turbine when starting. The only time there should be any flow of air or hot gases to speak of through the cross-fire tubes is when there is a difference of pressure between the Primary combustion zones of two combustors on either side of a cross-fire tube. And, when there is diffusion flame in the Primary combustion zone in one combustor and no diffusion flame in the Primary combustion zone in the other combustor connected by a cross-fire tube there is a sufficient difference in pressure to cause hot gases from the combustor with diffusion flame to flow through the cross-fire tube into the combustor without diffusion flame--which then ignites the fuel/air mixture in the Primary combustion zone of the combustor into a diffusion flame. (Combustors with diffusion flame have slightly higher pressures than those without diffusion flame.) This is how diffusion flame in all the other Primary combustion zones of all the combustors without ignitors gets started--through the cross-fire tubes.

Normally, this should only take a couple of seconds, at best, to occur--the propagation of flame through the cross-fire tubes from the Primary combustion zones of combustors with ignitors to the Primary combustion zones of combustors without ignitors. So, it should be clear, that for a brief period of time the temperature of the gases passing through the cross-fire tubes will be higher than "normal." (The cross-fire tubes are normally "cooled" by compressor discharge air which flows past (around) the cross-fire tubes, which is at compressor discharge temperature--which is MUCH lower than diffusion flame temperature. And, under normal conditions there is also no fuel/air flowing through the cross-fire tubes, which is another reason the cross-fire tubes should be just slightly higher than compressor discharge temperature (they will be slightly higher simply because of the radiant heat from the Primary combustion zones; even Premix flame is hotter than compressor discharge temperature--and we all know how hot the combustors are from the outside when the unit is running in Premix Steady State combustion mode!).

If I were you, I would be looking at the temperatures of all of the crossfire tube thermocouples when the unit is running in Premix Steady State. I would also be trending them as the unit is unloaded down to just before the unit transitions to Lean-Lean combustion mode. I would also be trending the combustion reference temperature and fuel split percentages. It's entirely possible that what is happening is that something is igniting diffusion flame in one or more combustors BEFORE it should be ignited.

I would also REALLY LIKE to see how the thermocouples are mounted on the crossfire tubes--and how the thermocouple extension wires are supported to keep them from touching other hot metal surfaces as they are run out to junction boxes. It's entirely possible that the leaks you are describing are impinging on the thermocouple extension wires and causing problems.

USUALLY when there are problems transitioning from Premix to Lean-Lean it's because both of the ignitors are not functioning properly, and flame is NOT being ignited within approximately one minute of when it should be during the transition.

It would also really help to be able to analyze the logic to see when it's being enabled and when it's not enabled, as well as just to see it.

MANY years ago I heard of some units which were being tested for DLN while running on liquid fuel (the holy grail of DLN!), and if I recall correctly the crossfire tubes were instrumented with thermocouples because of issues the units were having during liquid fuel operation. But, that liquid fuel DLN testing was stopped (it wasn't very reliable) and it's possible the instrumentation and associated logic was left in place and causing problems--because of improper installation or hot air impingement on the extension wires.

Hope this helps! Looking forward to hearing back from you and how this problem is resolved!
 
Jolek,

I want to add information about the ignitors being unable to ignite flame in the Primary combustion zone during a Premix-to-Lean-Lean combustion mode transfer. The high-energy ignitors do not work indefinitely; they will eventually need to be replaced, and when they are replaced the new ignitors will have to be properly inserted at the proper depth when setting the compression fitting which secures the ignitors to the tube on the combustor.

If the ignitors are inserted too deeply into the Primary combustion zone the tips will be damaged by the heat (my personal belief is that it's diffusion flame in Primary and Lean-Lean (and Extended Lean-Lean) that causes the most heat damage to the ignitor tips; Premix flame has a lower temperature). So, if the unit is operated in Primary and/or Lean-Lean for extended periods of time, or even Extended Lean-Lean (which is NOT recommended) then it's very likely the ignitor tips, especially if they inserted too deeply into the Primary combustion zone, will be damaged prematurely.

So, usually, when the unit fails to ignite diffusion flame in the Primary combustion zone during a Premix-to-Lean-Lean transfer it's usually because both (because there are two--and it only requires one ignitor to be working properly) ignitors are not capable of producing enough spark to ignite the fuel in the Primary combustion zone when the unit is operating in Premix steady state at very high fuel flows and air flows at rated speed. Yes; the ignitors in this situation are capable of igniting the fuel during starting--there isn't much air flow at that speed and time. But, when there are very high air flows the spark just gets "blown out" when the ignitor tips are damaged or failing or have failed. Again, this situation means both ignitors have failed or are failing, or one has failed and the other is beginning to fail.

Insertion depth is very important. Other threads on control.com have discussed how to check and measure insertion depth; it's not hard but it's not intuitive, either (unless you're a very experienced mechanic or controls person who has done similar things in the past or who has very good critical thinking skills).

And, ignitors do fail over time. Yes; there is a Process Alarm which indicates an ignitor exciter channel failure--but it's useless for detecting failed ignitors/ignitor tips. (Actually; that Process Alarm is pretty useless in most cases, as it's usually improperly configured during commissioning or afterwards and most sites just ignore the alarm anyway.)

Hope this helps! Looking forward to getting good data, and perhaps some clear photos or screen captures of the crossfire tube T/C logic. It would also be helpful if you could provide the exact Process Alarm text message that is annunciated when the unit trips (or all of them if there are two or more, which there are often are), along with the times of the alarms.
 
CSA,
Thanks for a quick, as usual, responce!
Sorry for confusion, I supposed to have written "latter" instead of "latest", and still probably I'm not very clear. I meant that unit trips from second logical option - 5 sec of very high CTXF temperature, above 1100 DegC. To emphasize again - crossfire temperature starts rising at the same second command L2TVX to reignition is given. It can be any thermocouple out of 14, with no pattern , not only those surrounding chambers equipped with ignitors. So, problem is not been unable to ignite primary zone but flame not spreading to all chambers after reignition. And, both ignitors are sparking good, we check them every outage and know exactly how they must be installed.
To your questions:
All CTXF readings are stable, normal and close to each other prior to the moment of reignition. Also TTRF1 and Splitter Valve behave normal. Maybe I could provide a screenshot of a trend later.
All 4 our units are equipped with crossfire thermocouples and same logic for about 25 years now, so its not some kind of test. Thermocouples are 7-8 meters long metal coated, running to JBs at both sides of turbine base.
Please, see attached alarm window ( Finally its possible ! ) :
1585587197782.png

And here is a part of block ( contains all 14 CTXFs) where L30CTXF_NEG ( alarm 423) trips a turbine after 5 sec delay:
1585587797673.png

I can provide any additional info in order to keep you thinking ;) , because you are good at it.
Thanks.
 
Jolek,

Thank you; it's clearer now. Almost....

You wrote:

" ... if one of them senses temperature higher than CTDA + 32 DegC for about a minute, or if one of them senses overranged temperature - above value MKV able to read and goes to negative, - for 5 seconds."

You also wrote:

" ... I meant that unit trips from second logical option - 5 sec of very high CTXF temperature, above 1100 Deg C."

Because, most T/C inputs on Mark V's I have worked on can read inputs higher than 1100 deg F. So, I don't understand this "over-range" thing at all. Is the value of CTXF_NEG 1100 deg F? Or, is it a calculated value?

And, let's step back and look at this logically. 1100 deg F is quite often the maximum allowable (Isothermal) exhaust temperature (TTKn_I, where "n" is a value from 0 to 7). And THAT is the temperature of the hot combustion gases AFTER they have passed through the turbine section. TTRF or TTRF1 is an approximation of the value of the hot combustion gases exiting the first stage turbine nozzles--and that's AFTER the Premix combustion gases mix with the Secondary combustion zone hot gases (which includes a small diffusion flame). So, the temperature of the hot gases in the Primary combustion zone are very likely closer to 2000 deg F, which would definitely cause the crossfire tube temperatures to increase, and increase pretty quickly. Now, it could be that the cross-fire tube metal shouldn't be exposed to temperatures more than 1100 deg F, but saying that's "over-ranging" the Mark* is ... not true.

High crossfire tube temperatures is going to indicate there is flow between combustors because of a pressure difference between the combustors. The hotter gases from the combustors with diffusion flame are going to flow through the crossfire tubes into the combustors with Premix flame--and that's supposed to happen because that's how the diffusion flame is ignited in the Primary combustion zones of combustors without ignitors. Normally, it happens very quickly (the crossfiring of Primary combustion zones), and at some point the Mark* (after it sees flame in at least one or two (I can't recall exactly) of the combustors with Primary flame sensors) then tells the Gas Fuel Splitter valve to move to approximately 50% position (so approximately 50% of the total fuel flow to the turbine goes to the fuel nozzles in the Primary combustion zone and the remainder goes to the Secondary fuel nozzles and the Secondary combustion zone.

So, what is happening to the exhaust T/C readings when the crossfire tube values are so high? Because flow through the crossfire tubes for more than a few seconds is indicating there is diffusion flame in the Primary combustion zone of some combustors and no diffusion flame in the others--and THAT should be causing the exhaust gas temperatures to change, resulting in high exhaust temperature spreads and an uneven exhaust temperature profile.

I'm wondering if the Mark* isn't cutting back the fuel to the Primary combustion zone too quickly and that's making it harder for diffusion flame to ignite in the Primary combustion zones of some of the combustors.

Some more "things" from the partial Alarm List you provided.... It looks like the unit uses IBH (Inlet Bleed Heat) in order to stay in Premix Steady State combustion mode longer than would be possible without IBH. That means that the fuel/air mixture in the Primary combustion zone is VERY lean--there is a LOT of air relative to the amount of fuel (because, one, the IGVs are closed most likely below 57 DGA, and the IBH control valve is open and recirculating some compressor discharge air back the axial compressor inlet. AND, when the Mark* detects combustion trouble it is then immediately turning off IBH--which also probably means that the IGVs are moving more open and heaven knows what else is happening as the IBH control valve is also being closed. There's just a LOT going on--and it's happening at a time when the unit is trying to transition combustion modes.

When IBH is NOT used, the unit can only go down to about 80% of rated load before it has to switch to Lean-Lean combustion mode from Premix Steady State. This is because to reduce load fuel is being reduced, and the IGVs are pretty far open and the unit is at rated speed and the amount of air entering the combustor (specifically the Primary combustion zones of the combustors) is high relative to the amount of fuel. So, to prevent losing Premix flame the unit is switched from Premix Steady State to Lean-Lean. With IBH, the IGVs are closed to reduce the amount of air entering the Primary combustion zone thereby enriching the fuel/air mixture slightly--but to close the IGVs it's necessary to recirculate compressor discharge air from the combustion wrapper to the axial compressor inlet, which also serves to reduce the amount of air entering the Primary combustion zone also.

So, now if during this transition the IBH control valve is quickly closed AND the IGVs are also probably moving more than they would otherwise be moving AND the Mark* is trying to change the fuel split which further leans out the Primary combustion zone fuel/air mixture (from approx. 80% to approx. 50%), well, you can see--there's gonna be problems.

I think the crossfire temperature logic is kind of ... unusual. (Well; not for GE-Belfort--they are notorious for grossly over-complicating things. Their way is always better than GE-Greenville's/Schenectady's ways of doing things. ALWAYS.) That's one thing. You have other issues--DLN Trouble (which can be one of several things); probably high exhaust spread trouble (there's probably a COMBUSTION TROUBLE alarm that's not shown in the photo). And that's driving the IBH Control Valve to close quickly and that's indicative of probably quickly moving IGVs and fuel splits and other things.

Let this be a lesson to everyone reading this thread: Process Alarms are important. Once more, for emphasis: PROCESS ALARMS ARE IMPORTANT! It's important to list all alarms (Process and Diagnostic) when trying to explain or understand an event. And, this is why trends of values are important.

Jolek, if you have Historical Data of a good shutdown, can you share it? Because there are probably some glaring differences, or even subtle ones, that could help explain what's going on here. I would be very tempted to bet a LOT of money that on good shutowns IBH isn't being switched off, and that exhaust temperature spreads aren't increasing very quickly. We need to see actionable data to be able to understand what's happening and when, and for both good shutdowns and these troublesome shutdowns.

Enough thinking?
 
Thank you, CSA.
A little bit more info and corrections.
First of all, tripping CTXF temperature with sort 5 sec delay is 1100 Deg Celsius, not Fahrenheit. You just misread it. So, reading of temperature higher than 1100 Deg C goes negative ( - 1133 DegC ). In order to MKV not consider this negative value as simple thermocouple fault ( - 83 Deg C when Open Loop), the logic monitors deeper negative values.
In the alarm window you can see there are no process alarms long before DLN transfer starts . Unit was running with IBH system ON but the valve actually did not open since Base Load, so the alarm just says that IBH system goes OFF but no valve movement happened. I've checked trends, IBH closed and IGVs did not move towards open to compensate. Also, no Exhaust Spreads before, during and after trip. At the moment of trip:
TTXSP1 = 24 Deg C
DWATT1 = 92.5MW
TTRF1 = 1087 Deg C
IGV pos = 62.4 DegA
GCV pos = 59 %
SPLT vv pos = 75%
AFPEP stable.
I've noticed that TTRF1 at Base Load of 107MW was a bit low, at my opinion - 1107 DegC. One day before the trip it was a successful DLN transfer and shutdown with TTRF1 = 1130 DegC at Base Load of 110MW.
Another fact needs attention - almost every trip happens after unit runs on liquid fuel. We run weekly on fuel oil up to Base, then down to 20-30MW for fuel changeover and than 30 min at LL on Gas fuel.
Thats all I can add today.
Thanks.
 
Jolek,

This is one of the things I find most frustrating about troubleshooting on control.com--the pertinent information comes in dribs and drabs, if at all (a lot of time when we ask questions we get nothing in response--radio silence; the deer-in-the-headlights look). And, I have learned to deal with the non-responses; hopefully enough food for thought was given that either the problem was solved, or something was learned by the original poster or those who read the threads later.

But, it's this dribbling in of information that just really rubs me the wrong way (actually, I don't like being rubbed any way, any where). It's like going to the bathroom to leave a poop, being satisfied with the results, and then having to return again in a few minutes to go again. It stinks (going in the second time, that is). If you find this analogy of pooping offensive, I find the dribbling of information just as offensive.

My participation in this farce is over. It seems pretty clear: There is some kind of agenda here--and the information being supplied is being done in a way to force the consideration of a particular, preferred explanation. I have much better things to do than to be "lead" in a direction, with breadcrumbs dropped at intervals along a trail like a fairy-tale story--just not to a good ending.

Hire someone who is knowledgeable and who you will work with and supply the requested information to in a timely fashion and you will most likely be surprised with the results. (Especially if you have a preferred explanation.)

Stay healthy in this crazy time--everyone!
 
Not to get in the middle of things, but I thought I would add just a bit.
Having any thermocouples on crossfire tubes is VERY UNUSUAL. I have never seen it and I have worked on many Frame 9E DLN-1 units. I don't know where they came from and would question their need. I am assuming you do not have one of those strange liquid fuel DLN machines that CSA was referring to.

So a question, do you have a "standard" DLN-1 combustion liner installed in your units? The only reason I could think of for having TC's in the crossfire tubes would be to prevent burnout of the tubes, but I do not believe that is a 9E issue on DLN-1 units. Have you seen burning in your crossfire tubes anytime over the years? I have worked on 9E duel fuel machines from EGT and know the software pretty well, and have not seen instrumentation on any crossfire tubes.

If I may suggest, you have sister units that do not have this tripping issue. Try running a View2 of those temperatures during a transition from Premix to Lean-Lean and see what they do. I would suspect they would increase as the primary zone re-ignited but they should then drop back as gas flow through the crossfire tubes goes to zero. Compare the "good" unit with the one having an issue. Also verify software rungs and constant values to see they have not changed.
Good luck.
 
CSA,

I've been really surprized by your last post on this thread. Even shocked. For several years I follow your answers and explanations to many colleagues and find them very helpful and contributing, sometimes reread them and even save some interesting points. And I REALLY appreciate your experience, work and time you spend.
And I TRULY don't understand your reaction. At the very first post I've tried to bring up as much useful information as I could at the time. After every your responce I've tried to get info and data you mentioned and asked for as quickly as possible. Not every day I have a possibility to get to this Unit and gather data. We have 8 other turbines frrom different manufacturers including 2 Combined Cycles that keep us busy. So "dribs&drabs, farce, agenda, silence..." are just totally NOT true. We have checked A LOT, tried a LOT and, while still having a problem, I've decided to refer to a wider community of experienced colleagues, for different angle of view. I believe my poor ability to express myself in English is a cause.Or maybe its a crazy time talking. Anyway, CSA, thanks a lot for your time and attempt to help!
 
Waltrobinson,

As for me, I've been sure that all machines of this type in the world are equipped with Crossfire tube thermocouples! Those units we have originally were liqiud oil fuelled only. Than, about 10 years ago, we had an upgrade to dual fuel done by Ansaldo-Thomassen. I don't know what "standard" combustion liner is. I agree that TC's main function is to prevent tubes burnout and no, we never had tubes seriously damaged at my 20 years period of time on site ( maybe thanks to this odd logic ?). And we have not had a lot of trips from this protection. TC's do provide a lot of help in investigating combustion problems. In "healthy" units during re-ignition of Primary zone there is a slight and short increase of temperatures, but flame propagation lasts about a second and then they drop back to normal and equal. Thats what not happening in this "unhealthy" unit recently. Will have software checked to see any differences.
Thanks for your attention.
 
Hi Jolek,
Thanks for your reply. I can assure you that not all "E" class DLN-1 GE type turbines have crossfire tube TC's. I have never seen crossfire TC's and I've worked on a lot of "E" class units on different sites.

Also, in my experience burned crossfire tubes are not a classic "E" class DLN-1 issue, so I'm not sure why you have the TC's. They may be solving a problem that doesn't exist.

Keeping in mind what I said above about other units and lack of crossfire tube TC's, I would consider temporally changing the trip function to an alarm function and record (with a trend) what happens on the next transfer Premix to Lean-Lean. I know I am recommending defeating a trip and therefore it must by your call to do that. With the vast majority of the fleet not having these TC's, I can not see an issue with a test with the trip disabled but that is your call. (Defeating trips is always a delicate proposition.)
Whether the trip is defeated or not, a high speed trend during that mode transfer might be helpful anyway.

Lastly, I'm assuming the location of these TC's has been double checked and verified. Are the TC's installed on the crossfire tube itself? There is no way the TC could have found its way through a cooling hole to the inside of the crossfire tube? Just wondering.

Walt
 
waltrobinson,

"A picture is worth a thousand words," as they say.

We asked for pictures of the cross-fire tube T/Cs, but didn't receive any. (More evidence of an agenda for this question/issue.)

The last drab (or was it a drib--I've lost track at this point) of information suggested this ONLY happens after the unit has run on liquid fuel.... It's pretty inconceivable how any liquid fuel could get trapped in the cross-fire tubes and live to tell about it (it would be vaporized and the residue would appear as carbon deposits on the surface(s) in the cross-fire tubes. There have been reports of carbon deposits on fuel nozzle tips getting so hot they can ignite premix fuel/air mixtures when running in Pre-Mix Steady State--resulting in unexplained primary zone re-ignitions (unexplained until the carbon deposits were discovered, that is). This has happened many times I'm aware of, with the carbon deposits resulting from dribbling liquid fuel nozzles, lubricating oil entrained in the gas fuel supply, lubricating oil in the atomizing air lines (from leaking Main AA Compressor seals), etc.

I wonder if it's possible that the site has some kind of hot deposits in the cross-fire tubes that are igniting the premix fuel/air mixture into a diffusion flame as it's passing through the cross-fire tubes during propagation when trying to switch from Premix to Lean-Lean. Kind of far-fetched, but it would seem something is going on as a result of liquid fuel operation. It could even be that the liquid fuel nozzle purge sequence isn't correct, resulting in purging at the wrong time (not likely, but I have seen something similar on a GE Belfort Frame 6b once before).

I would be a relatively simple matter (well, not that simple, but not that difficult, either) to open a combustor end-cap (primary fuel nozzles) as they are hinged, I believe, and use a borescope to look at the two cross-fire tubes to see what might be happening--if there's any evidence of problems. Yes; the mechanical department has to do some manual labour, and there are some consumables lost in the work (Flexitallic gaskets mostly), but if it's going to be a long time before the next maintenance outage it might be worth a look in at least one combustor. (It might even result in finding a problem which might cause more severe issues before the next maintenance outage).

Methinks the control system is trying to be blamed for this issue, though it's hard to see how--except it is a Mark V, and it does have LOTS of wires and LED bargraphs which flash and LCC Displays which display all kinds of cryptic information--so it MUST be the cause of the problem(s). Right?

Using the VIEW tools as you suggested during good and bad transfers would help. But, sometimes a look-see is best for all concerned.
 
CSA,
I agree with everything you say. I also have seen flashbacks just after operation on liquid fuel. At that site, after a short time operating on gas fuel the flashbacks would stop occurring, until the next time running on liquid. But this doesn't seem to be a flashback issue but just an occurrence during primary re-ignition.

It is suspicious that a unit that never did this before starts right after mechanical changes. Occam’s razor? "Simplest answer is usually the correct one".

I also agree that a picture of the TC would be very helpful for me to understand better, as I've never seen one.
Also in agreement with CSA, opening a combustion can end cover is, I believe, justified. Try opening one adjacent to a crossfire tube that had caused a trip and then you may be able to get a view of what the issue really is.
Again, good luck.
 
Walt,
Here are pics of Crossfire TC.
1586193661287.png

1586193704160.png

An HGPI is scheduled for this unit at July, so it's difficult to organize endcover opening, especially these coronavirus days when amount of people is down to minimum.

CSA,
Still hot on me...
If you would have read carefully my very first message you would have noticed "... but we noticed it happen mostly after a run on liquid fuel..." . I never said ONLY, though.
 
Hi Jolek,
Thanks for the pictures. That is something I've never seen before. I'm assuming the TC goes through the outer piping and is either in the air space outside the crossfire tube or it contacts the crossfire tube to read the crossfire tube wall temperature. Do you happen to know which it is?

I touched base with a couple of old friends with tons of field experience (mostly on GE manufactured units but also on EGT units and others of GE design) and neither had ever heard of an "E" class unit with TC's on their crossfire tubes. I've done a search on my hard drive and I believe I have software for at least a dozen "E" class units, 7's and 9'. I was looking for the front end of the logic you showed, L30CTXF and found nothing. So all the units that I have software for do not have that logic.

I'm kind of thinking two things. First, you might want to talk to whomever does your overall service and ask them why that logic is there since GE never seems to use it on their 7EA's or 9E's. You might be able to totally remove both the hardware and software and have one less thing to give trouble.

The second thought just goes back to why is this unit giving trouble now when it never had in the past. Could there be a mechanical issue now? What has changed? Don't know.

I think we are back to what CSA recommended, which is the mechanical inspection as the next step, understanding that it may not happen for a while. (I wonder what you could see with a flexible borescope going in the ignitor hole? Might just be able to get around to going into a crossfire tube although I'm not sure you could see the TC. Just a thought as it would be easier, faster and use less people.)
 
Walt,
All 14 TCs go through both outer and inner tubes and thermocouple tip located at the center of the inner tube. We assemble/dissassemble/change these TCs A LOT - every CI, HGPI, Major, chamber repair or inspection,...and we have 4 units equipped with these TCs, so you can imagine how much work and time we had spent on this issue. Not to mention that they about 7-8 meters long and must be fixed between chambers and pipes very carefully and not to be loose.
Agreed with you and CSA about what can and should be checked and done. Investigation continues and its progress dictated and limited by grid demand. This unit is been started and stopped every day, sometimes twice a day.
Thanks.
 
waltrobinson,

I agree--I only know of ONE site, in a country on the coast of the Mediterranean Ocean, which had thermocouples installed in the cross-fire tubes. And that was for Dry Low NOx testing on liquid fuel (distillate; diesel). I would NOT be surprised if after that effort was abandoned the T/Cs were not removed, nor would I be surprised to learn that GE Belfort (the former EGT) had used them on some units around the world. (Again, they are notorious--in my estimation--for needlessly over-complicating any process they can, simply because they can and because their way is better than anyone else's, particularly Schenectady/Greenville GE's.)

If this is happening on one of four unit there MUST be something different about the installation or wiring or something. It sure doesn't seem like simply liquid fuel operation for a limited period of time should be causing this problem. And, again, with the dribs and drabs, now we are learning that the unit in question starts and stops, sometimes twice a day. And, according to early reports this happens "... almost every shutdown. ..." and " ... it happen mostly after a run on liquid fuel ... " Does this mean they ALWAYS switch to liquid fuel before every shutdown--sometimes twice a day? (I'm really confused here. I hope I'm not the only one...!)

I think Jolek should re-write his original post to include all the pertinent information that has been dribbed and drabbed into the thread. Do the trips occur mostly after liquid fuel runs--" ... even though we run in LL mode for 30 min on gas to purge liquid fuel leftovers. ..."? How often does the unit get run on liquid fuel--after every run on gas fuel? Is the unit run at L-L for 30 minutes, then loaded back up to Premix Steady State (to what load--Base Load? or ???), and then unloaded to shutdown, when the trips mostly occur, except when they happen mostly after a run on liquid fuel? (Again, I'm really confused here.)

How many times is the unit run on liquid fuel in one week, or one month? What, exactly, is the shutdown procedure after a liquid fuel run? How long is the unit run on liquid fuel? When is the unit run on liquid fuel, after a run at load on gas fuel?

How often does the unit trip after a shutdown when the unit was ONLY run on gas fuel? Or, if it's easier, how often does the unit trip during a shutdown when it was ONLY run on gas fuel?

This just doesn't add up or make sense. And, neither does the one unit out of four thing make sense, either. Something else, a Frame 9E dual fuel unit with a Mark V and enough extra T/C inputs for 14 cross-fire tube T/Cs? That's pushing the limits of a Mark V's I/O capability, even with extra analog I/O cards (TCQB and TCCB)? And if the T/Cs are connected to <C>, then it's odd that they would be used for tripping purposes (well, not for GE Belfort, but that's another issue altogether).

I hope I'm not the only one that's confused here.
 
As I already wrote in post #6 of this thread, we have a routine : once a week we start on gas, switch to liquid at FSNL or 20-25MW, go up to Base for about half an hour, go down to 20-30MW for changeover to gas, run on L-L mode for half an hour and then up to Base. Shutdown on gas. Once a week means One time at One day of One week , four times in a month units run on liquid fuel.
Someone simply confused himself. Or, more likely, I unable to express myself clearly, which happens to me often in any language.
It is impossible to predict all questions that can arise and to write down all pieces of information in one first post, and then someone keeps calling it "dribs and drabs".
There are no exact answers, only based on statistics. Not always unit trips during shutdown (on gas) after it ran on liquid fuel before it was switched to gas. Sometimes not. Lastly we switch to Extended LL prior to shutdown to be more stable. And yes, one can say that unit trips always after run on liquid the same day or a day before.
About something else. Application Manual GEH-6195D, Appendix, Signal Flow Diags, D-11,12,13 explains how and where these TCs connected. <S> core, TBQA 1-28, jumpered correspondingly to 31-58 and 61-88 to make a voting capability.
Thanks and stay healthy - a lot of people need you.
 
Based on the information provided my personal opinion is: This problem is not related to liquid fuel operation. It's virtually impossible to imagine how it could be.

T/Cs on the cross-fire tubes are pretty rare.

Jumpering T/Cs on the TBQA of <S> was not done for voting purposes; it was done to prevent Diagnostic Alarms because the inputs being used are voted inputs, and without the jumpers there would be permanent Diagnostic Alarms because there was nothing connected to the <S> and <T> inputs.

There is--or was--something unusual with the combustion system of these machines that made the use of cross-fire tube T/Cs necessary. Presuming that ALL the cross-fire T/Cs have been checked using a T/C simulator from the JBs in or outside of the turbine compartment to confirm there is no wiring problem, and presuming that the proper T/C Type was used on the most recent outage/replacement corresponding to the selected T/C type in the I/O Configurator for the cross-fire tube T/Cs, there is something about the combustion system that caused the designers to install these T/Cs and configure the protection the way it was done.

You are going to have to work with the packager of the turbine, or GE, to understand why the cross-fire tube T/Cs are necessary--and probably to understand why what is happening is happening.

My apologies for upsetting you; I have been party to too many recent issues where information was withheld to try to shift the troubleshooting to a particular desired outcome. This thread seemed to be headed in the same direction based on my recent experiences.

Based on what you say you have been told about the wide-spread use of cross-fire T/Cs, something certainly seems off. I saw something like this 20 years ago when a development project involving a Customer's unit was being dropped and the cost of converting back to standard combustion hardware was too expensive so it was "decided" (by both the Customer and the OEM) that the unit would continue to use the development hardware. The people involved with that decision moved on (in both the Customer's organization and the OEM's) and the reasons for having the special hardware were lost with the change in personnel, and there was all kinds of speculation about reasoning and requirements, and a big mistake was made which resulted in a crash of the unit which took several months to get the unit running again. And several years of lawyers working on pointing the smoking finger of blame. I don't know how that ever worked out, but I have a big suspicion that standard hardware was installed in the machine during the rebuild. At who's expense I don't know.

Hopefully you will write back to let us know what you find out, if you find out anything. A visual inspection is probably going to tell you something--maybe a lot. But I believe there is a lot you still don't know--about how those T/Cs came to be installed on the cross-fire tubes.

Best of luck!

And you stay healthy as well! These are unusual times, and what we don't know is really, really scary.
 
Top