DLN1 facing secondary flame out during premix to lean lean


Thread Starter


When DLN-1 Frame6B unit transfers from premix to lean lean mode during unit shutdown (when unit is un-loading), we observe that secondary flame out is happening for a moment (say 2 seconds, at least two secondary flame detectors show no flame). I need to know is it normal or secondary should not flame out?

Can you please elaborate the process that what should be normal shutdown process during this mode transfer?


The process is that when TTRF1 drops below the Premix-to-Lean-Lean transfer point drop-out temperature the ignitors ("spark plugs") are energized for a short period of time, and the gas fuel splitter valve moves from its Premix fuel split position (usually approximately 80% primary) to its Lean-Lean fuel split position (usually about 50% primary). The ignitors are energized to ensure that diffusion flame is quickly established in the primary combustion zones (and propagates around the unit through the cross-fire tubes of the combustors).

I've been thinking hard about what could cause a momentary loss of flame in the secondary combustion zone, and I can't really think of a likely cause. The fuel in the secondary combustion zone INCREASES from approximately 20% of fuel flow to approximately 50% of fuel flow--which should ensure diffusion flame is stable.

Even if the secondary flame detector lenses had some humidity/moisture condensation on them, the diffusion flame intensity should INCREASE when the fuel flow to the secondary combustion zone increases during/after the Premix-to-Lean-Lean transfer.

I really can't imagine a scenario where the secondary diffusion flame intensity would decrease, or be lost, during the Premix-to-Lean-Lean transfer. It's quite a quandary because it's counter to what should happen if fuel to the zone increases.

What happens to the exhaust temperature spreads during this loss of flame event?

What happens to the unit load during this loss of flame event?

If there is a large increase in the exhaust temperature spread, I would imagine there would be a 'Combustion Trouble' alarm, which would clear before a trip was initiated if flame was re-established quick enough (usually there's only about 9 seconds delay between Combustion Trouble and a high exhaust temperature spread trip--if the adjacency permissives are met).

Also, if there is a large increase in exhaust temperature spread (because of an actual loss of flame--and I'm NOT saying the loss of flame is not real!) I would expect that the load would drop due to the lower hot gas temperatures entering the first stage turbine nozzles (meaning less work can be extracted during expansion of the hot gases through the turbine section).

And, I would expect that the spreads and the load swing(s) would stabilize once flame was re-established.

It is also helpful to know when this problem started--is it something that started after the last maintenance outage?

How long since the last maintenance outage?

What are the exhaust temperature spreads in Premix Steady State?

What are the exhaust temperature spreads in Lean-Lean?

Does the unit experience primary zone re-ignitions when operating in Premix Steady State (that is, does it intermittently transfer from Premix to Extended Lean-Lean)?

I suppose if the gas fuel splitter valve moved extremely suddenly and increased the gas fuel flow to the secondary nozzles extremely suddenly that could result in a short "blow-out" of the diffusion flame in the secondary combustion zone. But, it's not clear why this would suddenly start happening, if the unit had been running without similar problems before. It's certainly unusual, that's for sure.

I don't think it's really a problem--unless the spreads increase during the loss of flame event, AND the load changes. The Speedtronic will sense the <i>drop in load</i> and increase the fuel automatically, which could be another cause for the load swing(s). It's kind of difficult to analyze without some high-speed data to determine the sequence of events and what happens first.

Hope this helps!

Please write back with more information, as well as to let us know what you find.
Thanks CSA, Response is quite useful.

FYI, last major outage was 3 months before.
There are no spreads, no flame detectors issue and no noticeable load swings.

Will let you know if we conclude something.



We tuned the re-ignition timer from 3 seconds to 1 seconds and now tested with atleast 20 starts and there is no flame out now. Before it was flame out 4oo5 times.

> Thanks CSA, Response is quite useful. Will let you know if we conclude something.

Thanks for the feedback, ..., but, I have never heard of "tuning a re-ignition timer."

And there's more--the purpose of re-energizing the ignitors (spark plugs) in the primary combustion zone is to re-establish diffusion flame in the primary combustion zone (which propagates around the combustors through the cross-fire tubes). So, how does reducing the re-ignition timer from 3 seconds to 1 second help improve the flame stability in the secondary combustion zone? Is it causing the fuel splitter to start increasing the fuel to the secondary nozzles two seconds sooner than otherwise would have happened?

And, the question that should always be asked if the unit had been running properly with the original value: What has changed to make it necessary to "tune the re-ignition timer?"
It is observed that by the time the unit re-ignites primary (during shutdown) the "Secondary Load Recovery Mode (SLRM)" (83FXS3) is getting initiated and if SLRM is initiated it increases splitter setpoint thereby increasing primary fuel flow and potentially blows out secondary flame. Note that SLRM occurs when TTRF < 1925 which is considered the secondary zone flame out TTRF level.

Normal Re-ignition (2TVXR) occurs when TTRF < 1975 F for 3 seconds (On-Delay) but by the time 3 seconds timer is done the TTRF reaches to SLRM level which is 1925 F. So reduction of this timer to 1 second allowed ignition of primary zone well before TTRF reaches the SLRM trigger level.

I want to contribute to your discussion by sharing my experience regarding the subject.
Recently, I was facing a problem with Primary re-ignition failure when transferring from PMX to LL, the problem was solved by increasing the split by 1% (more gas was introduced in the primary zone during PMX SS). and when we went for re-ignition the primary re-ignited normally.
However, when we increased the split by 2% just to scan a wider split range. and during re-ignition the primary re-ignited and in the same time we faced a loss of flame in secondary for less than 1 second. I want to highlight that the lean lean fuel split will not be active until the primary flame is detected and the LL mode is enabled in other words during re-ignition only spark plug are being energized.
I suspect a sudden change of pressure inside the combustion chamber by re-igniting the primary with a higher split (more gas into primary) is the cause of the phenomena but I still cannot rely on any evidence for that.
I would be pleased if we exchange ideas and brainstorm together.
Note : initial split was 77%