Hello,

After more than a half months of the end of the HGPI and during periodic fuel testing, just after the changeover gas/oil at 05 MW, an anomaly occurred at the external interconnection between chamber 7 and 8. Or it was tinged a dark reddish color due to overheating, which forced the operation to stop the machine.

After boroscope and thermographic inspection were carried out to determine possible damage, and have shown mergers crossfire between 2-3 and 7-8, as well as metallic deposits within the liners 7 and 8.

can i get an idea about the origin of this damage?

 

Mirak,

Did the unit experience high exhaust temperature spreads during the time after the HGPI was completed and when the cross-fire tubes were found to be damaged? Because, usually burnt cross-fire tubes are a result of high exhaust temperature spreads. High exhaust temperature spreads result when there is an excess of fuel going into one or more combustors (not usually the case), or when there is a deficiency of fuel going into one or more combustors--and the latter is the most common cause of high exhaust temperature spreads. Other causes include cracked combustion liners and/or transition pieces or badly broken hula seals, allowing excess air to enter the combustor and cool the hot gases flowing into the turbine section. But, plugged fuel nozzles is the usual culprit, and if excessive can lead to burned cross-fire tubes.

Cross-fire tubes "burn" (melt) when there is a flow of hot combustion gases through them when there shouldn't be. When there is no combustion, or poor combustion, in a combustor and the combustor on either side of that combustor has combustion or good combustion the pressure in the can with poor or no combustion has a lower pressure than the can with good/better combustion. This causes hot gases to flow from the can with higher pressure (good/better combustion) into the can with low or poor combustion.

This pressure difference between combustors with flame and those without flame is how the flame propagates around the combustion cans from those with ignitors to those without--through the cross-fire tubes. But once combustion is established in a combustion can the pressures of the two cans virtually equalize and there should be no flow. So, under normal circumstance the only time there is flow through the cross-fire tubes is during starting when flame is propagating around the unit through the cross-fire tubes. After that, there should be no flow through the cross-fire tubes. Especially not at rated speed.

If there is flow through cross-fire tubes, usually both cross-fire tubes in a can with poor or no combustion get burned--especially when the unit is operating a rated speed with hot flames and high combustion gas temperatures. This is because each combustion can has a combustor on either side of it, so the can with lower pressure (the one with poor or no combustion) will have hot gases flowing into it from the combustor on either side through the two cross-fire tubes.

But, you said that only two tubes were burned (I think that's what you said), and if memory serves me correctly the combustion cans you mentioned, 2, 3, 7 & 8, have spark plugs in them (all of them if the unit has DLN combustors). A lot of time what happens is that the clips which hold the cross-fire tubes don't get properly installed and then one end of the cross-fire tube gets damaged.

This isn't a control problem; it's a mechanical problem. And, if the unit was run with high exhaust temperature spreads, then that caused the burning/melting. If the unit did NOT have high exhaust temperature spreads, then something may be amiss with the control system. Each of the exhaust thermocouple inputs to the turbine control system should be checked by connecting a T/C simulator to the wiring at the exhaust JBs (disconnect the exhaust T/C and connect the simulator in its place) to ensure the T/C inputs to the turbine control system are all working fine. Then, when re-connecting the exhaust T/C leads to the T/C extension wire in the exhaust JBs make sure to re-connect them properly. Cold junctions can be caused by improper termination of T/C wiring and T/C extension cable wiring (it usually only affects the reading by 10-40 deg F at 1000 deg F, but that sometimes can be enough to throw things off).

But, there is only flow through the cross-fire tubes when there is either no combustion in a combustor, or the combustion is very poor. Sometimes, a badly cracked combustion liner or transition piece or hula seal can cause poor combustion, but not usually before tripping the unit on excessive exhaust temperature spread.

If you write back for more clarification or help, please tell us which fuel was being burned, and what type of control system is being used on the unit.

Please write back to let us know how you fare and what you find!

 

HI CSA ,

thank you for your quick response

there has never been a problem or spread of high temperature exhaust before or after hgpi. in the historian, I check temperature exhaust and spread TTXSP1, TTXSP2 and TTXSP3. all was normal, and all the spread below the allowable spread. I check in the same time the workstation historical alarms, no alarm is appeared.

we were always crossfire of problem. after each combustion inspection, we find the damaged crossfire but not to the point where the machine is stopped.

which is weird is that when removing fuel nozzle, there were a significant fuel inside.

after we made a mini combustion inspection and we changed crossfire and damaged liners

we found 02 check valve : 03 and 08 combustors pass to 04 bar instead of 07 bar

why alarms are not appeared despite that there were problem in the crossfire?

Are the atomizing air is OK? presence fuel in fuel nozzle during disassembly

Control system is Mark VIe.

Now the change over gas/fuel is hanging waiting to find the origin of the problem.

 

mirak,

Flow through cross-fire tubes is caused by a pressure differential between adjacent combustors. Pressure differentials are usually caused by plugged fuel nozzles (restricting fuel flow), or bad liquid fuel check valves (it would have helped had you told us in the original post this problem was occurring on liquid fuel). You say:

> we were always crossfire of problem. after each combustion
> inspection, we find the damaged crossfire but not to the
> point where the machine is stopped.

I take that to mean that this is not the first time you have found damaged cross-fire tubes--is that correct? Is it always the same cross-fire tubes? Are the flame detectors in combustors 2, 3, 7 & 8? Do you always ensure before bolting the can cover on that the cross-fire tube clips are properly installed and won't vibrate loose? Are you using OEM hardware? Are you using site personnel for the maintenance outages--with supervision by a knowledgeable, experienced person?

> which is weird is that when removing fuel nozzle, there were
> a significant fuel inside.

If there is liquid fuel inside the fuel nozzle when you remove it, well that could be because the unit was shut down or tripped while running liquid fuel and the liquid fuel purge system or one of its components isn't working properly. I would expect to find a little fuel in the liquid fuel portion of the nozzle if the unit was shut down or tripped on liquid fuel. But you aren't telling us where you're finding the significant amount of liquid fuel. Is it in the Atomizing Air (AA) part of the dual fuel nozzle? If so, then it could be dribbling into the AA area because the liquid fuel check valves are working properly and are allowing fuel to dribble out of the nozzle tip and not be burned.

The AA gets into each of the fuel nozzles via hoses/pipes from a common manifold. The AA usually enters the manifold from the bottom, and there is usually a plugged low-point drain which can be opened to see if there is liquid fuel present in the manifold (there shouldn't be).

Also, improper assembly of the liquid fuel portion of the fuel nozzle could allow liquid fuel to leak into the AA cavity.

Are you sure the liquid fuel purge system is working correctly? There are check valves on the purge air system, and there is a three-way valve (VA19-1, if I recall correctly) that ports air to the nozzles or ports the system to a tell-tale drain.

Most incomplete or poor combustion while running on liquid fuel causes excess smoking in the gas turbine exhaust--usually white smoke, or light-colored smoke. Yes; diesel exhaust is usually a light shade of tan or brown, but when diesel is not completely burned it shows up as white smoke.

I can't explain why there isn't an exhaust temperature spread; there could be wiring issues with the thermocouples and/or the thermocouple extension wires which are masking spreads. If all the wiring is correct and you have tested each T/C input to the Mark VIe using a T/C simulator <i>from the JBs at the exhaust</i>, then the problem is probably not poor or incomplete combustion resulting in high spreads and flows through the cross-fire tubes. In this case, the problem is still probably incorrect assembly or hardware of the cross-fire tubes.

<i>Based on the information provided,</i> my best guess is re-assembly of the unit after maintenance outages is not completely correct. Certainly, some of the liquid fuel check valves were found to be opening too soon, which will allow liquid fuel to dribble out of the nozzle during starting and shutdown/trip. The liquid fuel purge system may not be working correctly (failed or failing liquid fuel purge check valves can allow liquid fuel to flow to drain--through the tell-tale leak-off (which should be monitored when running on liquid fuel for evidence of check valve problems).

Have you used the manual selector valve at the discharge of the liquid fuel flow divider to monitor the fuel pressures to the individual combustors during liquid fuel operation? The guideline is that all the pressures to the ten (10) combustors should be within approximately 10% of each other, and any pressure more or less than than 10% is considered to be suspect when there are spread issues.

If the unit doesn't usually run on liquid fuel, the check valves (both liquid fuel and liquid fuel purge) have been known to fail "prematurely" and cause problems when the unit is switched to liquid fuel.

The absolute atomizing air system pressure ratio should be approximately 1.15 to 1.2 times CPD when operating on liquid fuel. So, add atmospheric pressure to CPD and add atmospheric pressure also to the AA pressure, and divide the absolute AA pressure by the absolute CPD pressure to determine what the AA pressure ratio is. (When the unit is running on gas fuel, the absolute AA pressure ratio should decrease to approximately 1.05 to 1.10.)

Lastly, the repeated "burning" of the cross-fire tubes is not good. There is something about the re-assembly procedures or the hardware being used that is suspect. It's not clear if there was actually flow through the cross-fire tubes or if just one end of a cross-fire tube had fallen down into the lower combustor and was subject to high temperatures possible resulting in a small flow through the cross-fire tube.

Proper atomizing air is important to liquid fuel combustion.

When the unit is operating on gas fuel, the AA system should be purging the AA portions of the fuel nozzle, and there should be no liquid fuel entering the AA portion of the fuel nozzle. If there is liquid fuel in the AA cavity of any fuel nozzles, then there's something amiss with the liquid fuel assembly or the liquid fuel check valve of that combustor. Liquid fuel purge air should blow any liquid fuel out of the nozzle and into the combustor above 94% speed (when starting on gas fuel) or after the liquid-to-gas fuel transfer is complete. Liquid fuel purge air should continue flowing all during gas fuel operation, and AA should continue to flow through the AA portions of the fuel nozzles when running on gas fuel (albeit it a reduced pressure ratio).

As was said before, if the pressure differential between adjacent combustors is bad enough, flow will generally occur in both cross fire tubes of the combustor with lower pressure. So, for example, if combustor #3 was having combustion issues (say, plugged liquid fuel nozzle) the cross-fire tubes connecting combustor #2 to combustor #2 and combustor #4 will both be burned, because the pressure in combustor #3 and combustor #4 will be higher (presuming the combustion in #2 and #4 is good) causing hot combustion gases to flow into combustor #3 from combustors #2 & #4, burning the two cross-fire tubes connecting combustor #2 to combustor #3 and combustor #4 to combustor #3.

If you're only finding one cross-fire tube of a combustor burned/melted--and if only one end of the cross-fire tube is burned/melted--then it's likely the re-assembly of the unit was not correct, or the cross-fire tube clip/retainer was not properly inserted or broke, or the wrong hardware (cross-fire tube, cross-fire tube retainers/clips are being used).

Hope this helps!

 

HI CSA,

I want the detailed method of the swirl so that I can determine exactly the origin of the damage in the turbine

Sincerely