Hello Everyone,

We have GE Frame 9E units (open cycle) with MARKVIe control system. Everytime during startup when machine reaches 95% rpm both 88TK-1 & 88TK-2 starts with a gap of 10 seconds. During this 10 seconds two things happen:

1) Sometimes the two flames go out immediately and after few seconds third one too. The unit trips on Loss of flame in between these 10 seconds.

2) Sometimes GT's rpm start decreasing from 2850 to 2760 & then after 10 seconds when the second Fan starts rpm start increasing up to FSNL.

Has anyone experienced these scenarios before? Why machine trips on Loss of Flame during this time?

 

Omair,

When did this problem start?

Did this start happening after a maintenance outage? If so, what was done to the fuel control valve(s) and the control system during the maintenance outage?

Did this start happening after a turbine trip from load? If so, what caused the trip?

Was the Mark VIe supplied with the turbine when the turbine was new, or was it an upgrade from an older Speedtronic turbine control system?

If the Mark VIe was an upgrade to an older Speedtronic turbine control system, how long ago was the upgrade?

Does the unit have DLN (Dry Low NO) or conventional (diffusion flame) combustors?

Is this only happening on one unit?

What fuel is being burned during starting?

Is the unit trying to transfer fuel when this problem occurs?

What alarms are present (Process- and Diagnostic) BEFORE the loss of flame trip?

Basically, the exhaust frame blowers are started at 95% speed, also known as 14HS. And, they are stagger-started because they draw a fair bit of current during starting and if they are both started at the same time quite often the Aux. Transformer protection will trip.

And, at about the same time the Aux. L.O. Pump and the Aux. Hyd. Pump should be shutting down, and on some (older) machines the compressor bleed valves are also told to close at the same time. The IGVs should be already at the minimum operating position (so not fully closed, but not fully open, either).

At approximately 95% speed the unit is starting to transition from acceleration fuel control to speed fuel control (so from FSRACC to FSRN) for the last little bit of speed to FSNL (Full Speed-No Load)--100% TNH (speed), or probably 100.3% TNH (in preparation for synchronization). The operation of the exhaust frame blowers is likely NOT the cause of your problem, but just one of several things which are occurring at the same time.

When the turbine control system annunciates a "Loss of Flame Trip" it's telling the operator(s) that flame was lost in the combustors BEFORE any other condition which should cause fuel to be shut off occurs. The usual reason for a loss of flame trip is that something happens to interrupt the fuel flow-rate to the turbine, such as lack of pressure and/or flow from the fuel supply source or a problem with the fuel stop valve, or some issue with the controls--which would normally be accompanied by alarms, Process and/or Diagnostic. Whether it's gas- or liquid fuel if the pressure flow decreases there is a pressure switch or transmitter which annunciates an alarm to warn the operator. So, understanding what alarms were present before the event, and what alarms were annunciated during/after the event, would be most helpful. As well as understanding some of the history of the unit--was the Mark VIe an upgrade to an older Speedtronic turbine control system; what fuel was being burned during starting; etc.

We need more information; so please provide all the answers to the above questions--all of them, please. There's too much we can't know about the unit(s) at your site, and what has recently transpired to cause this problem (which is presumed to only be happening on one of the units). So, the more information you can provide--the better our response will be. But, at the present time we just don't have enough information.

Looking forward to getting the answers to the above questions so we can provide more help! And, please tell us how your troubleshooting is progressing, also!

 

Thanks Alot ,
Here are your answers :

>When did this problem start?
It started few weeks ago , before this every startup was normal.

>Did this start happening after a maintenance outage? If so,
>what was done to the fuel control valve(s) and the control
>system during the maintenance outage?
The units are newly commissioned & only run for few months. No major maintenance or outage yet.

>Did this start happening after a turbine trip from load? If
>so, what caused the trip?
Turbines trip here a lot due to Boiler issues & pressure from forwarding skids & both normal & backup supply breakers get open. but all other units are not facing problem like this, only 4 units have this issue out of 9.

>Was the Mark VIe supplied with the turbine when the turbine
>was new, or was it an upgrade from an older Speedtronic
>turbine control system?
Yes, MarkVIe was supplied with the package, no upgradation.

>Does the unit have DLN (Dry Low NO) or conventional (diffusion
> flame) combustors?
No DLN , All units with Conventional combustors.

>Is this only happening on one unit?
We have 09 operational Units here. this happens with 4 units everytime & with any other unit sometimes.

>What fuel is being burned during starting?
Start-up on Light Distillate Oil & on 40MW conversion to Crude Oil.

>Is the unit trying to transfer fuel when this problem occurs?
No , Absolutely not.

>What alarms are present (Process- and Diagnostic) BEFORE the
>loss of flame trip?
Everything Normal in Alarm List Before Loss of FlameTrip Alarm. Same alarms appear which come during normal startups of other turbines where there is no issue like that.

Thanks for your timely reply!

 

Omair,

So, the units burn liquid fuel and "only four units out of nine" have this problem--that's 44% of the units. And, the units experience a lot of trips.

The problem has nothing to do with the 88TK exhaust frame blowers; it is occurring at the same time as the exhaust frame blowers are being started--which is 95% speed. The exhaust frame blowers provide cooling air the the turbine exhaust frame (shell) to help reduce expansion which can cause misalignment and vibration. There should be a description of the function of the exhaust frame blowers in the Manuals provided with the units. But, they have nothing to do with firing and can't blow out the flame in the combustors.

Again, a loss of flame trip occurs when the Mark VIe believes everything is fine with the turbine (no loss of L.O. pressure; no excessive vibration; no exhaust over temperature; etc.) and yet flame was lost in three or more combustors that have flame detectors (only four of the fourteen combustors have flame detectors). 99.897% of the time a turbine experiences a loss of flame it's because the fuel supply system is not providing sufficient fuel at the proper pressure and flow to maintain flame in the combustors.

You can use the ToolboxST Trend Recorder to monitor the exhaust temperatures, exhaust temperature spreads, turbine speed, turbine acceleration rate (actual, TNHA, and reference, TNHAR), fuel flow-rate (FQL or FQLM or FQLM1, usually), liquid fuel bypass valve servo current (FAL), and if the liquid fuel bypass valve has LVDTs the liquid fuel bypass valve position. Start the trend just before the unit is started, and then analyze it after a trip. You could also run the Trend Recorder with the exact same signals on one of the units which is NOT experiencing the start-up tripping problem and compare the two trends to look for discrepancies.

Liquid fuel supply systems usually have pressure regulators which can frequently have problems usually caused by air in the regulator sensing lines. Also, air in the supply piping can cause pressure/flow instability problems. Many sites with multiple turbines have groups of turbines supplied from groups of liquid fuel supply pumps (usually called forwarding pumps), and sometimes the control schemes for the forwarding pumps aren't exactly correct in the early stages of plant operation. Especially if multiple units are being started simultaneously and or are being tripped during starting. If the four units experiencing the problem are fed by the same group of supply pumps a problem with the control scheme for those supply pumps could be the problem.

The problem is most likely NOT the Mark VIe--but you haven't told us what alarms (Process AND Diagnostic) are present during starting, so we can't be sure of that. The problem is most likely something in the liquid fuel supply system that is not providing sufficient pressure/flow to maintain flame. This will be evident on the trend if the Mark VIe is trying to close the liquid fuel bypass valve to increase the liquid fuel flow-rate to maintain acceleration rate during starting. The amount of servo current being applied to the liquid fuel bypass servo valve will also be increasing in magnitude to try to increase the liquid fuel flow-rate if the liquid fuel supply is inadequate. The fact that turbine speed sometimes decreases also is indicative of inadequate liquid fuel supply pressure/flow--which is not usually controlled by the Mark VIe.

Please write back to let us know what you discover and how the troubleshooting progresses! Most Mark VIe's have too many alarms during starting (because of poor commissioning) and operators become immune to alarms. But all alarms are important--and if an alarm is a nuisance then it should be investigated and resolved so that when it is annunciated it is really a true alarm condition requiring action to resolve. Just because you experience the same alarms all the time doesn't mean they aren't real problems. If the same alarms always occur during starting--are they really alarms???

Use the Trend Recorder to help gather data to analyze the problem and solve it. I believe you will find the problem is most likely the liquid fuel supply system and not the turbine or Mark VIe. And if you write back for more help--PLEASE provide the list of alarms! Even if you don't believe they are relevant. Include the Diagnostic Alarms--because if the problem IS the Mark VIe there is likely Diagnostic Alarms trying to alert the operate to the problem.

Hope this helps!