Frame 6B High Exhaust spreads

We have a dual fuel Frame 6B (standard combustor) with Mark VIe operating on gas in combined cycle.

The unit has low spreads at low power outputs but spread increases at higher power outputs and high spreads even occur when operating in simple cycle on IGV temperature control mode even on part load (with HRSG out of service whilst running IGV mode to simulate how unit will operate during part load operation with HRSG on with diverter damper open).

We removed the fuel nozzles and found some sticky residue on the nozzles, all nozzles were then removed and cleaned and were re-installed. The unit ran with a failed atomising air compressor for a few weeks before it was replaced. After cleaning and re-installing the FNs the issue still persists. Unit runs the latest and greatest combustion hardware (advanced extendor) and has done 22000 hours (still 10,000 hours from the next MI). Spreads appear to rotate with the swirl so it appears to be a combustion issue. Any advice if we can continue to run another 10000 hours on this machine, ignoring the spread alarms. The machine doesn't seem to be tripping on high spread presently.

Another issue was that unit was test run and the three conditions we have in our MkVIe logic (as per an old MKV revision of the control spec ) for trip triggered by combustion monitor were fulfilled. i.e.,
>0.8 x SPL
Adjacency of the SP1 and SP2

But the unit did NOT trip. The above criteria met quite a few times after we took the raised load from 20MW in 2MW increments with IGV temperature control mode ON.

After we looked at the MKVIe control logic block for combustion monitor , it seems to be using the TTXM but the logic is not clear and help pages are pretty scarce in helpful information. Is MKVIe using TTXM-TTXD(min) to calculate SP1 (etc.) instead of TTXD(max) - TTXD(min)?

Interesting moniker....

Anyway, I can't really add too much to this thread--because GE is changing things. Whether it's for the good or not remains to be seen, but they are changing things. And, anything to do with GE, Belfort, France, is most subject to change. And, GE-design Frame 6B units have been given over to the engineers in Belfort.

There are two problems here as I see it. One, the lack of documentation--no "Item Help" files made available when new algorithmic functions are released. Why? Well, because there is no standard in GE that says there has to be any documentation released with new functions.

The second problem, as I see it, is that GE likes to believe that the methods and schemes in the functions are intellectual property and are therefore not made available to the "public." They want to sell services--that's where they really make their money. Not on new unit sales; not on parts; but on services. Too bad they don't put more effort into making their services better. But, that's another story and another set of problems that GE doesn't want to face or recognize.

So, it's incorrect to say that the three conditions for a combustion trouble trip in the Mark VIe (based on what was in the Mark V that the Mark VIe replaced) did not result in a trip. The scheme appears to have been changed, and that happens when someone purchases a control system upgrade from GE. One expects the logic and sequencing to be the same or relatively the same--but, remember what the GE salesperson said: All the new TILs and improvements would be included free of charge with the Mark VIe. Just try asking what changes were made between the Mark V and the Mark VIe.... You will get nothing. Except, the statement, "Functionally, it's exactly the same."

So, to prove it is or it isn't is not on GE--it's on the purchaser/user. That's right--the purchaser/user. GE isn't going to give you a list of all of the changes--that wasn't in the contract. (Of course "the latest and greatest" probably wasn't in the contract either, but that's what was sold even if that wasn't what was in the contract, verbatim.)

You haven't used Trender to provide us with any data. None. Zero. Zilch. Nada. Niente. Nothing. So, how can we possibly help? We dont' know what the spreads are, how they change in magnitude or location. Nothing. Not even what fuel(s) the unit has been operated on.

That's for the controls.

As for the question about extendor hardware--that's between you and the seller. A failed MAIN Atomizing Air Compressor, or a failed BOOSTER (Starting) Atomizing Air Compressor? Did you know that air from the Main AA Compressor is used for cooling and purging of the liquid fuel passages when running on gas fuel? And, without it, that hot combustion gases can get back into the liquid fuel lines--all the way back to the liquid fuel flow divider and even to, and in some cases, past the Liquid Fuel Stop Valve? Now, those gases usually cool as they get further away from the turbine compartment, but when then flow into liquid fuel lines they push liquid fuel out of other liquid fuel lines, and quite often right back into fuel nozzles/combustors. And that's one way sticky residue can build up on fuel nozzles. Another is leaking check valves. Another is carry-over from a failed atomizing air compressor (you didn't say how it failed). Another is oil and other contaminants entrained in the gas fuel.

There's lots of things that can cause sticky residue, and that sticky residue can usually be traced back to its source. Have you opened the gas fuel manifold piping? The atomizing air manifold piping? The purge air manifold piping? Because if it's in the fuel nozzles, it's probably in the piping somewhere, too.

The sticky residue can also be sent out for analysis.

I would suggest you speak, soon if you haven't already, with the supplier of the extendor hardware and tell them--and give them Trender data--about the perceived problem which is not causing the unit to trip. (I have heard that some new combustion monitor algorithms are initiating a unit shutdown, to prevent the thermal stresses inherent in a trip from causing damage to hot gas path parts.) And ask them what they think should be done, and how they are going to warrant the parts of they fail based on how the unit is being protected with the Mark VIe?

That's about all I can add. Data. Actionable data. That's what's required. That's easily obtainable using Trender. And, then talk with the suppliers of the control system and the extendor hardware and figure out if the unit is properly protected. GE should at least be able to tell you that: Is the unit being properly protected with the new combustion monitor algorithm in the Mark VIe?