Gas Turbine Tripping on ACALIB command

A

Thread Starter

Adeel Imtiaz

GAS TURBINE: MS5001 FRAME-V
CONTROL: MARK-V (IDOS)

We recently had few problems in Gas turbine startup regarding GCV improper opening. To resolve this problem (while running on Crank mode with proper hydraulic supply) autocalibration command was executed but gas turbine tripped immediately with out any alarm. Autocalibration files are present in our unit but we have not used it till now.

Plz also send me the exact procedure how to use ACALIB command.
What did I miss in above scenario?

Thanks
Adeel
 
Adeel,

It's unlikely the turbine tripped without any alarm, Process or Diagnostic. We don't know enough about the procedure you used when attempting to use AutoCalibrate.

I believe there is some information in the Mark V Maintenance Manual, GEH-5980, about using AutoCalibrate.

AutoCalibrate involves determining the actual valve stroke as a percentage of effective valve stroke, putting that information in F:\UNIT1/ACALIB.DAT, then running AutoCalibrate to determine the 0- & 100% stroke values to put into the I/O Configurator to download to EEPROM.

Most sites just use full open GCV position as 100% stroke (meaning they think they effective stroke is the same as the actual stroke). This "works" for many units, though it usually results in some unintended problems and some unnecessary manipulations of some Control Constants.

There has been a lot written about AutoCalibrate in previous posts on control.com. I would suggest searching for ACALIB.DAT for info.

If you want more help, you'll need to tell us what kind of Mark V (TMR or SIMPLEX), and what Process- and Diagnostic Alarms were active before the unit was cranked and <b>when the trip occurred</b>.

You'll also need to tell us what you'll be using as the effective stroke of the GCV--the measured stroke as a percentage of the effective stroke from the Control Spec value, or, the actual, full open stroke value as 100% stroke.
 
> We recently had few problems in Gas turbine startup regarding GCV improper opening.

It would also be very helpful and interesting to know what kinds of problems you experienced and how you decided on the course of re-calibrating LVDT feedback as the resolution for the problems.

Thanks!
 
A

Adeel Imtiaz

Dear CSA,

GAS TURBINE: MS5001 FRAME-V
CONTROL: MARK-V (IDOS)

SRV Regulator#77
GCV Regulator#43

During the start-up of gas turbine gas turbine tripped on “Failure to Ignite” alarm. While performing many checks it was decided to manually stroke GCV using USER DEFINED DISPLAY. At that time we did not changed servo regulator type. The gas turbine was operated on CRANK (don’t have Aux lube oil pump) mode to establish hydraulic oil pressure. Only 63FG was forced at that time. As soon as GCV Calibration button was pressed in User Defined Display gas turbine tripped and annunciated following alarms

1) Maintenance- Forcing mode Enabled
2) Gas control Valve Servo Trouble

Calibration of GCV was aborted due to above result and after wards it was found that P2 was 18Kg/cm2 due leakage of SRV. started smoothly when this pressure was released by manual Relief valve.

To investigate this behavior on GCV calibration we are performing similar activity on our MARK-V trainer panel with same specs and software interfaced with SRV GCV assembly (servo + Cylinder+ LVDT on 3kg/cm2 Instrument Air). Gas turbine is brought to CRANK mode by a Frequency Meter 900 RPM. No logic variable is forced. Again as soon as we tried to use both User Defined Display and ACALIB command gas turbine trainer tripped annunciating similar alarm mentioned above (FAG Servo Current Error is latched when investigated with 128% stroke).

Please let us know how to perform GCV/SRV calibration using ACALIB and UDD command. I have read your many related threads but still couldn't get the command going step by step.
 
I don't know what your position is or what your job's responsibilities are, but here's the procedure that needs to be used for the turbine at your site:

Get some training.

I started to write a novella about all the things wrong with the sequence you described, but then I realized it's not necessarily your fault. If you or the people responsible for the operation and maintenance and troubleshooting of the units at your site do not have the proper training and experience to make good decisions about troubleshooting and then use proper procedures to perform the troubleshooting that's the fault of the management and supervision at your site.

Every time (with one exception) I have written a site-specific procedure for LVDT calibration it has been misued, misunderstood and over-ridden by people who had either no concept or incorrect preconceptions or misconceptions about how LVDT calibrations are to be performed. Or even how to just manually "stroke" devices with LVDTs. It <b>ALWAYS</b> comes back to me that the procedure doesn't work. And when I try to understand why it doesn't work I'm always told either that they didn't do this step or that step because someone didn't think it was necessary or that it was incorrectly written, OR that the procedure was obtained from another site (which had different systems and configuration--people just think that there is ONE step-by-step procedure for every turbine and application and that any procedure will work on any turbine or application).

So, I'm not going to open that Pandora's Box on this site.

Suffice it to say that the decision-making that led to the attempt to manually stroke the valve was flawed--severely flawed. If there was 18 kg/cm2 gas fuel pressure upstream of the GCV when the GCV was opened and the spark plugs were energized, flame would have been established--and VERY quickly with a pretty loud <b>BOOM!</b>

Further, the flawed decision-making that led to the attempt to manually stroke, or to "calibrate" the GCV LVDT in response to a failed start attempt is so typical of that used at so many sites. And, it is again indicative of a callous disregard on the part of management and supervision for the training and skillset necessary for operators and technicians to properly and <B>***SAFELY***</b> operate and maintain GE-design heavy duty gas turbines.

I will add one last comment, and that is that if you want to be successful in using either Demand Displays <b>OR</b> AutoCalib, you first need to understand how the Lube Oil-, Hydraulic Oil-, Trip Oil- and Gas Fuel P&IDs (Piping Schematics) all work together to allow high-pressure Hydraulic Oil to get through the appropriate servo-valve to the actuator to allow the servo current signal to make the device move and the LVDT feedback change. Forcing 63FG is not going to work for the GCV (or the SRV, for that matter).

And that is not a simple task--understanding the P&IDs. But, neither is it impossible. It's going to involve a very serious commitment of time and effort, one that will be paid back many-fold over the weeks and months and years ahead. It's going to involve putting aside (but not necessarily ignoring) everything you've been told or read or thought about this concept and working through the drawings and documentation to develop a thorough understanding of what's required for the units at your site to be able to stroke and calibrate the LVDT feedback. And, as you begin to understand the bits and pieces and fit them together in your mind you can recall and review all the things you've been told and heard and thought in light of your new-found understanding, which should help in your quest for truth and understanding.

Best of luck! If you want to use control.com to further your understanding of systems and concepts in order to develop your site-specific procedure for LVDT "calibration" and device-stroking, that's what we're here for. But, there is no one-size-fits-all procedure for calibrating LVDT feedback from servo-operated devices (valves and IGVs) on GE-design heavy duty gas turbines.
 
Anything to do with some form of logic expecting the GCV actual position % to be circa 0% during crank operation, which is then being driven open by the ACALIB function? Mark V tripping Unit on GCV not following position ref?

Just a theory, my experience is with 9FA STAG with MarkV TMR, hence bit of a guestimation!
 
>Please let us know how to perform
>GCV/SRV calibration using ACALIB and UDD
>command.

One more thing: AutoCalibrate and the User Defined Displays (Demand Displays) for servo-operated devices equipped with LVDTs are mutually exclusive. They can be used to perform essentially the same function (stroking the device), but they do so in very different ways.

When you say the unit is tripping on ACALIB, but you are using a Demand Display you are confusing the two. I maintain that when Hydraulic Oil pressure is available at the servo-valve/actuator of a device that simply initiating AutoCalibrate (ACALIB) will <b>NOT</b> trip the turbine. If Hydraulic Oil pressure is not available and you try to use a Demand Display (User Defined Display) to stroke a device, the unit may indeed trip.

Again, it's all about understanding systems, as well as how the Speedtronic sequencing works.
 
A

Adeel Imtiaz

I have now performed stroking of GCV and SRV both through USER DEFINED and ACALIB commands. Trick is simple

1)If hydraulic oil pressure 89 Kg/cm2 is only available in crank mode then take gas turbine to crank mode such that TNH<28%

2) Isolate fuel gas

3) Force

L3GFLT = 0
L60FSGH = 0
L20FG1X = 1

4) Perform stroking using User defined display or Acalib command
One can use view2 command to save the results

5) Master Reset (Otherwise gas turbine might trip)

6) Unforce the variables

The temperature of oil must be monitored keenly as they would tend to rise. However this procedure was figured out on a trainer and afterwards when we got the opportunity the stroking was performed successfully on crank mode. This procedure may vary in other turbines.

Thanks for your help.

Adeel Imtiaz
 
Top