GE Model: PG6581(B)
Controls: Speedtronic Mark V <R><S><T>. Can't remember what you call the NOT simplex kind.
Fuel gas skid: Flenco GB DLN1
Part Number: 379A9741.
We're having some trouble holding P2 pressure stable during normal operations. Not sure if turbine generator is causing the swings in the pressure are if the fuel skid is causing the turbine/generator to swing. I'm not saying the DLN skid is the problem, but it's where we're going to start. I've a new field engineer so I thought I'd spend some time working through troubleshooting. He wanted to verify the positioning on the SRV as a first measure. Good deal, smart guy.
We referenced the original control settings. And I quote....
4. SRV Position Calibration:
0% stroke = 0.7 VRMS feedback
100% stroke: (1.25 inch linear travel)
Ok, so no worries. We dig out the old dial indicator and magnet base from back in the log cabin part of the facility and head out to the gas skid. (after blinding off the natural gas of course)
Its a fisher V300 ser B valve. This is a ball valve. Actuated by a Moog contraption (Model 80-113B13CEP)
Question 1. How does one measure linear stroke on a ball valve?
So being confused but determined, we pull out the original papyrus commissioning book, which stated, among other things,
Adjusted LVDT voltages are very close to what I am seeing now.
LVDT 1 LVDT 2 STROKE
REFERENCE RMS VAC RMS VAC Inches
0% 0.696 0.716 0.003
25% 1.190 1.221 0.978
50% 1.691 1.73 1.989
75% 2.188 2.231 3.014
100% 2.684 2.731 3.949
75% 2.194 2.238 2.964
50% 1.698 1.736 1.991
25% 1.195 1.214 0.995
0% 0.695 0.714 0.008
Here are the original regulator 1 settings from the good ol' <Q> card.
Regulator Definition for Servo Output 1 90SR-1
Function type & sub-type: 77
Valid types <00, 2D, 2E, 40, 41, 43, 49, 51, 52, 53, 77, 78, 7A, 7B, 7C>
Suicide enable :- Current Fault: NO LVDT/R fault: NO
Suicide position limits (%):: Low: -5.0 High: 105.0
Current Gain: (0 to 200% rated_cur./%pos.) 1.93
Current Bias: (0 to 100% rated [10,20,40]) 2.7
Zero Stroke (0 to 6.667 Vrms) :- LVDT 1: 0.6999 LVDT 2: 0.6999
100% Stroke (0 to 6.667 Vrms) :- LVDT 1: 2.880 LVDT 2: 2.880
<7> Pos limits (-128% to 128%) :- Low: -25.0 High: 100.0
<7> Integrator convergence gain (0 to 16 psi/%): 0.3
<7> Position reference Gain (0 to 2 %/psi): 0.1
Note: for type 5, enter fuel flow data on Pulse Rate screen.
for type 7, enter time constant data on screen #12.
Any ideas? Anyone?
All the original voltages line up with what we are seeing now. The actuator sounds terrible and I suspect a problem there, but we already knew that.
Yes, its an RMS meter. Yes it's been checked recently.
So, you have a recently calibrated True AC RMS voltmeter and you and your new field service person have verified the linearity of the LVDT feedback over the full stroke, so use the MANUAL FSR function to gag the FSR on load so that you can be reasonably certain the Gas Control Valve(s) (downstream of the SRV) aren't causing the SRV to try to respond to unstable GCV(s). If the GCV(s) are stable when gagged but the SRV isn't, it's more than likely the SRV servo--which you gave very little details about except to say it doesn't sound good. Which should be a clue, if we understand what you didn't really elaborate very well.
So tell your field service person if the LVDT calibration verification is okay, it's not bad LVDT feedback or bad LVDT calibration (and remind him/her that LVDTs rarely drift (that's why they use them on airplanes and rockets)), and if one LVDT is having a non-linear output at some point(s) there would be Diagnostic Alarms to that effect, and during the stroking (verification) check, also.
Have you looked at the SRV servo currents for the three coils when the valve is at a stable position (say, during stroking (verification) to see what they look like? (What are they?)
What is present value of the 'Current Bias' (the null bias) value in the I/O Configurator?
If the valve is stable during stroking (verification), and the servo currents are reasonably balanced and stable, it may not be the servo or the hydraulic actuator. If the unit has multiple P2 pressure transducers are they all providing the same pressure feedback readings (if they aren't, there should be Diagnostic Alarms to indicate a problem with the feedbacks not being relatively equal). But you haven't told anything about Diagnostic Alarms....
Does the SRV seem to be at it's typical position or is it more or less open than usual when the unit is running? If the SRV is much more open than normal it could be a restriction upstream of the SRV causing a flow problem the SRV is responding to.
Or, if there is a pressure regulator upstream of the SRV that could be having problems and negativity reacting with the SRV. But, gagging the FSR should help identify that if it's a source of or contributing to the instability (which you really haven't detailed very well at all...).
When does the P2 pressure instability start? During firing? During acceleration? During Lean-Lean operation? During or after a Premix Transfer during loading?
Are you sure the compressor bleed valves are fully closed?
If the unit has IBH, is the IBH Control Valve stable during P2 pressure instability?
When did this problem start? After a maintenance outage, or a unit trip from load?
What is the hydraulic pressure doing during the SRV instability? If it's also unstable it could be a problem with the hydraulic accumulator and/or the hydraulic pump compensator.
What is the hydraulic filter differential pressure?
As a kind of last resort, you could use the 'Verify Position' and'Verify Current' features to look closer at the servo and/or LVDT health--but the caveat is that you would have to run an AutoCal first before you use either or both of the features. If you're confident in the SRV calibration values in the I/O Configurator (to recover with in case of an issue) then this would be a good option to try to eliminate either the LVDTs or the servo as the cause of the problem. For details on the two 'Verify' features, find the features in either the Maintenance Manual or the Application Manual.
You may not be impressed with the field service person, but everyone has to start learning somewhere. If it's a Field Core person, let Field Core--or whomever the service provider is--know about your dissatisfaction and have them send a replacement at no charge. But we can only help if you provide good information, and while you wrote a lot, there wasn't a lot of value in it. Give us something to work with, and save the attitude for another day and person(s) you want to help you.
You misunderstand, probably my fault. Field Core is not involved at all. I’m just getting our new internal engineer indoctrinated in the ways of the Mark V. I’ve nothing but praise for our GE service engineer. We will most likely bring in Fieldcore to fix it, but I want my internal engineer to spend some time troubleshooting as well.