We have Speedtronic Mark IV+ control system for controlling GE Frame 9 Gas turbine machine.

When I was stroking the Liquid fuel bypass valve after shutdown, I found that valve position is not going as per forcing voltage. for example if I give (+5 vdc) the valve is going only to 45 % closing (it should be 50%). When I am giving 8 vdc (full stroke) it is only going to 82% (it should be 100%). Both physical indication plate (which is not linked at all to LVDT) and LVDT reading on HMI are showing the same.

If I am giving more voltage also (around 9 vdc) the valve is closing more, at around 9.5 vdc the valve will close 100%.

My question is: I have to calibrate the servovalve? Or it should be replaced?

And if I am going to calibrate this one, what is procedure for that?

Regards,

 

First of all, one doesn't calibrate the servo-valve, or the LFBV, or whatever device the servo-valve is being used to position. One only calibrates the feedback from the LVDTs installed on the device.

It's been a considerably long time since I've worked on a Mark IV, but as I recall the "forcing voltage" you referred to was not directly proportional to position. I don't recall if the Analog Forcing display (for the servo-outputs only) allowed one to put in a reference position or not. If so, and you put in a reference of 75%, what position (on the indicator on the valve) does the valve go to? What is the displayed position on the CRT from the LVDT feedback when the reference position is 75%?

If the actual position of the valve plug and the displayed position feedback from the LVDT agree, then the LVDT feedback is calibrated properly.

Although most people refer to the act of calibrating LVDT feedback from the LFBV (or GCV, or IGVs, or SRV, or whatever device) as "calibrating the LFBV", they aren't calibrating the servo-valve or the LFBV. They are only calibrating the LVDT feedback from the LFBV. Not the servo. Not the LFBV. Just the LVDT feedback from the LVDTs mounted on the LFBV (or GCV, or IGVs, or SRV, or whatever device).

As for the servo-valve, if this is a TMR Mark IV+ then you should be able to see the three individual servo current output values from the three processors, as well as the null bias values. Presuming the three null bias values are correct and equal (as they should be; "correct" being 0.267 ma per processor if I recall correctly), then when the valve is at some mid-stroke position the three servo currents should be relatively equal and the algebraic sum should add up to be 0.8 mA, plus or minus 0.4 mA. [NOTE: The Mark IV "inverted" the displayed values of servo current, just to keep things interesting for everyone!]

If the three servo currents being applied to the servo coils have the proper polarity (remember the servo-valves used on GE-design heavy duty gas turbines are <b>bi-polar</b> devices) and the three servo currents are relatively equal and their algebraic sum is within the above specification, then there is likely no problem with the servo-valve. (Again, this is presuming the actual valve position and the displayed valve position from the LVDT feedback are very close.)

I just can't recall if there was a way to put in a position reference in the Analog Forcing display of the Mark IV. Sorry; it's been too long.

Even if there isn't a way to manually set a position reference, if the actual position and the displayed position are nearly equal <b>AND</b> the three servo currents are relatively equal and within the specified range, and the null bias currents are within specification and equal, then there's likely nothing wrong with the servo-valve.

Since the servo-valves are bipolar devices, at any mid-stroke position the observed (measured) servo output current from the Mark IV should be approximately equal to the specified range. If the actual current were proportional to 5 VDC or 8 VDC or 9.5 VDC, then the valve would always be the fully closed position. Output current and reference voltage are not proportional.

 

Thank you very much CSA

GREAT job.

G.Rajesh