hi CSA,

the configuration of regulator1 SRV is:

reg type: 2_LVpsMAX
RegGain: 1.8
RegNulbias 2.67
Dither Amp 2.0
Dither Freq 100.0hz
LVDT1input LVDT1
LVDT2input LVDT2
MnLVDT1_vms 1.033185 1.027448 1.031891
MxLVDT1_vms 3.563585 3.544263 3.559962
MnLVDT2_vms 1.033733 1.032113 1.032384
MxLVDT2_vms 3.475117 3.49347 3.470385
LVDT Margin 5.0
TMR_difflimit 5.0

and for GCV regulator

eg type: 2_LVpsMAX
RegGain: 1.8
RegNulbias 2.67
Dither Amp 2.0
minposvalue 0.0
Maxposvalue 100.0
Dither Freq 100.0hz
LVDT1input LVDT10
LVDT2input LVDT11
MnLVDT1_vms 0.546242 0.543745 0.54634
MxLVDT1_vms 3.354269 3.337259 3.350151
MnLVDT2_vms 0.579185 0.581698 0.576747
MxLVDT2_vms 3.369209 3.384762 3.354698
LVDT Margin 5.0
TMR_difflimit 5.0

so mechanical the valve is open and close when i stoke the valve but for regulator configuration is like this. so no mechanical problem ,

for Control Constant:

FPKGSD: -25% SRV SHUT DOWN COMMAND

FRKCRMN 0% SRV REFERENCE MINIMUM VALUE

FRKCRMX 100% SRV REFERENCE MAXIMUM VALUE

K30FSRSU_T2S:0s Sliding FSR t2 SRV closing early
time

K3GRVA 3s SRV not following alarm time delay

K3GRVAX 3% SRV not following alarm set point

K3GRVT 5S SRV not following trip time delay

K3GRVTX 5% SRV not following trip set point

LK3GRVFB -6.67% SRV VALVE POSITION TROUBLE
SETPOINT

LK3GRVFLTD 2s SRV VALVE TROUBLE ALARM DELAY

LK3GRVO 6.67% SRV VALVE OPEN TROUBLE SETPOINT

LK3GRVSC 30% SRV VALVE SERVO CURRENT TROUBLE
SETPOINT

LK60FSGH 9% HIGH SRV STROKE REFERENCE SETPOINT

SO ALL CONFIGURATION IS LIKE THESE

Can you CSA please help me

now the turbine is running
SRV lvdt= 64.4 valve actuation= -3.07
GCV lvdt= 45.65 valve actuation= -0.26

 

csa,

The indicated position (from the calibrated LVDT feedback) when the turbine is running seems to be reasonable (presuming the turbine is not 100% loaded). The SRV should not be 100% open (because it's trying to control P2 pressure upstream of the GCV) and the GCV should, typically, be at approximately 70-80% for a machine at Base Load if the fuel is as per expected (BTU content; Wobbe index; etc.).

I don't understand the issue--except that you said previously when the turbine "trips" the SRV LVDT feedback goes to -13%. The mnLVDTn_vms values are well above the normal zero-stroke voltage (which should be 0.700 VAC RMS, +/- 0.020 VAC RMS) and the maximum is, well, slightly outside the range of expected maximum voltage--but that's likely because the zero is so high.

And the opposite is true for the GCV mnLVDTn_vms values--they are very low (less than 0.700 VAC RMS, _/- 0.020 VAC RMS), and the maximum seems to be fine (less than 3.500 VAC RMS, +/- 0.020 VAC RMS).

It's also noted you did not provide the minposvalue and maxposvalue for the SRV....

I can't see anything from the information you provided (which did NOT include the type of SRV/GCV) which is unusual--except for the two gains for the two valves being the same. I'm not accustomed to that, but it may be what works at your site.

So, we don't know what kind of SRV/GCV assembly is in use at the site, and we don't know if the position feedback comes from RVDTs or LVDTs. We also don't know if IS barriers are used (I suspect they are because you said this is a 5002D, which is a two-shaft machine, probably driving a compressor so the location would be considered hazardous). Again, there could be some problem with the IS barrier(s)--but you haven't told us if they are present and if they are in good condition.

The -25% reference at shutdown is just to ensure that, if hydraulic pressure were present to the SRV that it would be trying to drive it to -25% position (which it can't do because it shouldn't go less than approximately -5.0%--<b>if the SRV is a combined-type and the gap between the actuator rod and valve stem is 0.030-0.050 inches.</b>

I have seen LVDTs fail, but not like this. Usually, they just go open circuit (very negative feedback) and maybe these are intermittent--but that doesn't make sense if there are two LVDTs because the selection should be the MAXimum of the two and if the LVDT feedback is -13% then at least one of the LVDTs is less than -13% (more negative than -13%) <i>while the other LVDT is also negative (-13%).</i>

This could be the problem--if the LVDT excitation to the two LVDTs is not correct. The LVDT excitation for one of the LVDTs should come from <R> or <S> or <T>, and the excitation from one of the <b>other</b> processors (<S> or <T> if LVDT1s comes from <R>; <R> or <T> if LVDT1s comes from <D>; <R> or <S> if LVDT1s comes from <T>). BOTH LVDTs <b>should NOT</b> be excited by <R> or <S> or <T>. If the LVDT excitation for both LVDTs was coming from the SAME processor's VSVO card (via the TSVO) and there was something wrong with that processor's LVDT excitation (or something was amiss with the TSVO for that LVDT excitation output) then that could explain why the two SRV feedbacks (inputs) are going very negative.

To check this you have to use the Screw terminal information from Toolbox or the Mark VI System Guide for the TSVO and then see where the LVDT excitation is terminated.

That's about all I can think of--<b>based on the information provided.</b>

Sometimes when one asks for help they feel they are being asked questions that are not relevant--but if the information provided is not complete then sometimes irrelevant questions do get asked.

Again, presuming there are two <b>L</b>VDTs on the SRV <b>AND</b> the gas valves are the combined typed (SRV and GCV in a single cast steel assembly) <b>AND</b> the gap is per specification <b>AND</b> the IS Barriers are working correctly <b>AND</b> the LVDT bar is pinned to the valve stem about the only thing that could be wrong (since the regulator is set for max of two) is that the feedback from both LVDTs is being lost for some reason making both of them go negative after a "trip". That could be an excitation problem (difficult to believe that it's happening for two VSVOs at the same time--but then we don't know what other Diagnostic Alarms are present after the turbine "trips" and the LVDT feedbacks are -13%).

The indicated position (on the HMI display) for any 2_LVpsMAX regulator will always be the <b>higher</b> of the two feedbacks. If the higher of the two feedbacks is -13% that means the other LVDT feedback is less than (more negative) that -13%.

When an LVDT fails open, the LVDT output goes, essentially, to zero VAC RMS. Or when the excitation is lost, the LVDT output goes, essentially, to zero VAC RMS. (I say "essentially" because there is usually a little noise on the wiring which makes the output seem like it's not exactly zero--especially when the machine is running.) So, it would seem <b>from the information provided</b> that the output of both LVDTs is going very low (approaching zero VAC RMS) when the turbine "trips". Why this is happening is not clear from the information provided.

Another interesting question to know the answer to is: When you "re-calibrate" the SRV LVDT feedback when it goes very negative (less than -13), by how much does the mnLVDTn_vms and mxLVDTn_vms values change from before the calibration? If they aren't changing by very much, if at all, then it would definitely seem something is causing the LVDT output to go very low--or something is causing the Mark VI not to read the SRV output (feedback) after a turbine "trip" (which a complete list of Diagnostic Alarms when a turbine "trip" occurs would also be very helpful and likely very telling).

I keep saying "trip" because sometimes people use "trip" when they mean stop. Stop and trip are two different things--they are not synonymous (the same). And we also don't know what you mean by "trip."

I've also neglected to ask: When did this problem start? After a maintenance outage? After a VSVO or TSVO card was replaced? After LVDTs and/or IS Barriers in the LVDT circuit(s) was(were) replaced? Has this been going on since commissioning, or did it start recently/suddenly?

It would also be very interesting to know the next time the turbine "trips" and the LVDT feedback goes negative, what the excitation voltages are <b>at the junction box closest to the LVDTs</b>, and what the LVDT output voltages are <b>at the junction box closest to the LVDTs</b>. This would help us to determine if the LVDTs are losing excitation during a turbine "trip" or if the LVDT outputs are going low during a turbine trip. These voltage should also be checked at every junction box between the Mark VI and the LVDTs, including at the TSVO card.

Please write back as you resolve this problem to let us know how it is resolved. If you would also provide all the requested information it would also be very helpful to others reading this post (as well as to me).

 

I wasn't able to pull the documents you are discussing due to some virus protection software that my company has on our computers. So, if this sounds stupid it's because I haven't had access to everything you all are discussing.

With that being said, I remember in the past calculating the null bias current by placing the valves at mid travel and measuring voltage at the MKV terminals. If the null bias current was outside of the tolerance allowed (.267 ma +/- .13 ma), you would adjust the null bias spring on the servo. Has that changed significantly?

It seems like you're masking an issue with the servo if you change the null bias current setting to get the results you're looking for. Am I off base here?

 

Brian Hall,

Adjusting the null bias spring on the servo-valve is just <b>begging</b> for trouble.

The Control Specification has procedures for calculating the null bias current setting--but in my experience it's almost never required to adjust the null bias from 2.67% (per processor in a TMR panel) if all other settings (specifically the regulator gain) and LVDT calibrations are done correctly. And using that procedure usually results in another calculation and yet another calculation. And it's NOT meant for adjusting individual null bias currents--as there's only one null bias spring.

Nearly ALL servo-valve problems--especially these days--can be traced to oil. That's because, and this has been documented, the refiners have changed formulations to improve the lubricity of turbine lube oils--but for those turbines which use the same fluid for high-pressure hydraulics (and there are a LOT of them!) the change has resulted in increased varnishing and particulates which are wreaking havoc with with servo-valves. It's not the servo-valves themselves, or the quality of the servo-valves produced by the manufacturer--it's the oil (formulation and maintenance). And, for some odd reason, F-class turbines seem to have a LOT more problems than most of the rest of the GE fleet of heavy duty gas turbines. And, the IGV servos (located in the Turbine Compartment--with long runs of feed and drain piping to/from the servo-valve/actuator) seem to be most affected on F-class turbines (though units with heated gas fuel also seem to have excessive servo-valve problems).

Servo-valves are frequently and often maligned and blamed as being the cause of a turbine trip or failures to start. People just seem to want to replace a (thousands of dollars) servo at the drop of a hat.

Or "recalibrate" the gas valves or the IGVs--in an attempt to fix what is believed to be a servo problem. (Calibrating only affects LVDT feedback--nothing else. No matter what people believe, or have been told, or surmise, or assume. When someone is (allegedly) calibrating a valve or the IGVs, what they are really doing (and the ONLY thing they are doing) is calibrating LVDT feedback. And almost no one ever checks to see if the LVDTs need calibrating (by comparing LVDT feedback to actual position) before they "calibrate" the gas valves or the IGVs. That's how much confidence people have in GE's AutoCalibrate routine--on the one hand--that they just assume the calibrated LVDT feedback is always 100.00% equal to the actual physical position.

And, yet, when the turbine trips or fails to start, the first thing they want to do is "calibrate" the gas valves or the IGVs. Without ever checking to see if they need calibration. And, when that doesn't resolve the problem the next thing they want to do is replace the servo--and then "re-calibrate" the gas valves or the IGVs.

So, on the one hand these people are 100% confident that AutoCalibrate is always 100.00% accurate (that the results of an AutoCalibrate are 100.00% equal to actual physical position)--but these <b>same</b> people are the first to rush to "calibrate" the gas valves or the IGVs when there's a "problem."

And many of these people are the same people, who before they make an adjustment to a pressure switch or pressure transmitter, will record the as-found conditions--and not make any adjustment if none is necessary. And LVDTs are no different from any other sensor on any process or piece of equipment.

Many people--without verifying that the calibrated LVDT feedback equals the actual physical position--try to use the null bias to make the feedback equal to the reference. What good does it do to make the feedback equal to the reference if the feedback isn't equal to the actual physical position? (Answer: None.)

LVDT calibration rarely drifts. It's just not in the design and construction of the devices that they drift. And, most Speedtronic panel LVDT inputs rarely drift--they just fail (usually because of a short or ground or heat and humidity). So, this constant "calibration" of gas valves and IGVs is pretty wasteful. Especially considering all the steps one must take to use AutoCalibrate (including isolating fuel(s) for safety; forcing logic; starting/stopping pumps). And if one is really going to do the "calibration" correctly it requires setting up a dial indicator or taking dial caliper readings, or using a machinist's protractor on the IGVs--which, these days, requires a Confined Space Entry Permit and hole-watch.

And "calibrating" the gas valves has no effect on the servo, and replacing a servo does not require a subsequent "calibration"--because replacing a servo-valve has nothing to with changing the physical stroke of the device. A servo-valve only controls the flow of oil to the actuator--which does not affect the stop-to-stop travel (stroke) of the device.

The null bias is only to be used to make the LVDT feedback nearly equal to the reference <i>when the LVDT feedback has been verified to be equal to the actual physical position of the device.</i> These people who never verify that the LVDT feedback is equal to the actual physical position and then spend hours and hours trying to make the LVDT feedback exactly equal to the reference are the same people who will rush to "calibrate" the gas valves or the IGVs when the turbine trips or fails to start before checking to see if the LVDTs require calibration. It's really so contradictory and such a waste of time--and just really incomprehensible when one stops and thinks about it.

Again, it's rarely necessary to make an adjustment to the null bias, and it should only be done after the LVDT feedback has been verified to be equal to the actual physical position of the device AND the regulator gains are properly set, and the LVDT feedback (i.e., the actual physical position of the device) is very different from the reference during calibration verification.

But, virtually without exception every time I have been to a site where they adjusted the null bias spring the servo-valve has had to be replaced to get the device to work correctly. Now, I know that's not the case for a lot of sites--but it certainly has been for the sites I've been to.

Lastly, it's important to remember that the null bias specification is a <b>range</b>--that is, it's not a specific value. It's -0.8 mA, +/- 0.4 mA (that's -8%, +/- 4% in Mark V-, Mark VI- and Mark VIe-speak). So, the null bias current for a servo-valve could be anywhere from -0.4 mA (4%) to -1.2 mA (12%) and still be within the acceptable range. It should not require adjustment to the extremes; that means the null bias spring either wasn't adjusted correctly to begin with (which happens with a lot of refurbished servo-valves) or there is truly a problem with the servo-valve (cleanliness; failed/failing o-rings on the spool valve; dirty passages; etc.). And changing the null bias isn't going to improve any one of those conditions--just waste time delaying the inevitable (replacing the servo-valve).

 

HI CSA

1st the problem of SRV is not solved because the turbine is running, when the turbine is stopped i will make troubleshooting and applied your set. My PROBLEM now with other turbine but the same kind 5002D frame 5 2 shaft, so, the SRV work normally but the max opening is 61.5% .

when the turbine is at 0 speed (shutdown) im working on to STROK the valve manual; to send many values for min and max positions when the LVDT is at MAX position 100% on site LVDT blocked at 61.5%. so t think it's mechanical problem !!!!!

but at 0% normal, other position its ok so the problem at 100%

Can you CSA help me and can you give me your experience reference?

 

The hydraulic actuator(s) for the SRV (and GCV) of a combined gas valve assembly (and you STILL haven't told us if the gas valves are combined in a single cast steel assembly or not!!!) are single-acting cylinders (pistons).

As such, there are rings which are supposed to prevent hydraulic oil from leaking past the rings to the other side of the cylinder. If there is any leakage (and some leakage is unavoidable since piston rings, even o-rings) never seal 100%) and oil gets to the other side of the piston what can happen is that if there is no drain for the oil to get out of that other side of the piston then it will not allow the piston to travel to the full open position (oil is not very compressible, so if there is any oil in the side of the piston where it's not supposed to be then eventually it will not allow the piston to move very far).

There is supposed to be a hole in the cylinder through which any leakage past the piston ring can leak out of. Usually this hole, for some strange reason, is threaded--and sometimes people put a plug in the hole. This plug now prevents any leakage from draining out of the side of the piston where it's now wanted, and that causes the actuator to not be able to travel to full open.

Because the actuator rod (piston) rises, the hole which should be unplugged is on the TOP of the actuator, as you are looking at the actuator from the access plate on the side of the base of the combined gas valve assembly. There should be NO plug in that hole. If you find a plug in that hole (on the TOP of the actuator as you are looking at from the side) you should remove it, and that should help with the problem.

That's the usual cause of this problem. It just might be that the valve internals (the plug and/or the valve stem) are just binding and the actuator isn't capable of moving them past 61.5%.

But, really, your post isn't very clear. Are you saying that you are visually monitoring the SRV as you are trying to stroke it and that it does not open past approximately 61.5%? I don't get the part about 100% LVDT and max position. But, yes, the problem is most likely mechanical and NOT controls-related.

If the current to the SRV servo-valve is much less than -3.0% (that's NOT a typographical error-- -3.0%), and if it's more like -70.0% or less (more negative) then the Mark VI is putting out "saturation" current trying to make the valve move to the full open position and something is preventing that from happening--something mechanical.

The servo current's from all three processors should be roughly equal when you are trying to manually stroke the valve to some position, and if each one is -3.0% or less (more negative) then, again, the Mark VI is trying it's hardest to move the device to the reference position--but something is not allowing that to happen.

 

The problem with SRV is mecanical because the min position now is 1.1% and the max position is 63%

so i have strok SRV valve to increase the hydraulic pump from 80 BAR to 89 BAR, to strok the SRV Manual the maX bloked at 63% and when i send -25% the SRV is 1.1%

so when i have ready to start ;

start turbine and after that im triped with theses alarms:

1 L2TV EVT Turbine purge time expired
0 SD_OVRD EVT SHUT DOWN OVERRIDE SIGNAL
0 SD_OVRD EVT SHUT DOWN OVERRIDE SIGNAL
1 L2TVX EVT T1\L2TVX
1 L2TVX EVT T1\L2TVX
1 Q 0286 ALM BASSE PRESS POST ALLUMAGE TRIP
1 Q 0286 ALM BASSE PRESS POST ALLUMAGE TRIP
0 L1X EVT Turb.start sequence initiated - start auxiliaries
0 L4 EVT Master protective signal
0 L4HQ EVT Hydraulic oil auxiliary pump start command
1 L4T EVT Synonimus for trip condition
1 SD_OVRD EVT SHUT DOWN OVERRIDE SIGNAL
0 L2TV EVT Turbine purge time expired
1 TRIP EVT T1\TRIP


please CSA normally it's all events and alarms can you help me ????

 

i forget for SRV is:

the gas valves are combined in a single cast steel assembly are single-acting cylinders (pistons).

 

H A R,

I don't speak French, and the translation programs I used didn't help with understanding the alarms. (Basse is 'farmyard'???) Post is 'after' and 'allumage' is ignition, so something is tripping the turbine after flame has been established--but there's not enough information to be of any help.

The HMI should have a Trip Display, and you should be able to immediately go to the Trip Display after the trip to see what has caused the trip (it's divided--or should be--I'm sure Belfort has had a better idea) into Pre-Ignition, Post-Ignition, Protective Status, etc. So you should be able to see what is tripping the turbine--what's causing the Post-Ignition trip to be energized.

The SRV is charged with dropping the gas fuel supply pressure to something around 1 barg during firing. So, if the gas fuel supply pressure is 20 barg, the SRV will only go to a few percent of stroke to reduce the pressure to about 1 barg. It doesn't have to go to maximum during firing--it <b>SHOULDN'T</b> go to maximum during firing.

If the SRV won't go past 63% when the turbine is running, the turbine might be able to make rated load--because the SRV <b>DOES NOT</b> go to 100% when the turbine is running at rated speed.

The SRV is designed to maintain a pressure that is a function of turbine (HP) shaft speed by the formula:

FPRG = (TNH * FPKGNG) + FPKGNO

You can look up the two Control Constants, FPKGNG and FPKGNO, and then calculate what the FPRG (Fuel Pressure Reference-Gas) SRV reference pressure is based on the speed when the turbine is trying to fire (usually somewhere around 20%), and what it will be at 100%. Again, the SRV <b>does not and should not</b> go to 100% at any time during normal operation (that includes starting and loaded operation). The only time the SRV might go to 100% is if the gas fuel supply pressure is too low, which will cause the SRV to open, possibly to 100%, but at that point the unit will not be able to make rated power output. But that is <b>NOT</b> a normal condition.

If the gas fuel supply pressure is too low during starting, the SRV will go open much more than normal. When that happens, the Speedtronic usually initiates a trip with the Process Alarm "STARTUP FUEL FLOW EXCESSIVE". That may or may not the case--the fuel flow isn't actually being measured or compared to a reference. The Speedtronic looks at the position of the SRV and if it's above some value (usually about 30-35%) stroke the turbine will be tripped, because the implication is that if the SRV is open to or above 30-35% that something is very wrong--either the SRV isn't properly controlling pressure upstream of the GCV, or the pressure sensor(s) between the SRV and GCV aren't working properly, or, there is insufficient gas fuel supply pressure which may cause problems if the pressure suddenly increases.

<b>FROM THE INFORMATION PROVIDED</b> it's not possible to say any more. Something is very wrong. We don't know when this problem started. We don't know how long it's been occurring. It is strongly recommended to have someone knowledgeable come to site to help resolve the problem(s).

 

CSA for alarm:

ALM BASSE PRESS POST ALLUMAGE TRIP its : P2 post ignition low pressure trip logic L86FPG2LT

and for Control Constants
FPKGNG = 3.2875 P_PCT
and FPKGNO = 38.75 PRESSG

FOR EQUQTION : The SRV is designed to maintain a pressure that is a function of turbine (HP) shaft speed by the formula:

FPRG = (TNH * FPKGNG) + FPKGNO
FOR OUR SYSTEM IS
FPRG = (TNH * FPKGNG) - FPKGNO

 

H A R,

As for the formula, if FPKGNO is negative and the formula is (TNH*FPKGNG)+FPKGNO it's the same as if FPKGNO is positive and the formula is (TNH*FPKGNG)-FPKGNO. The Belfort Bunch is at it again!!! AAARRRGGGHHH!!! They are such bloody idjits!!! (Usually, FPKGNO is a negative value; someone in Belfort just thought it should be positive, it seems.)

As for the alarm, it means that the pressure feedback from the intervalve cavity (the area between the SRV and the GCV)--also called the P2 pressure--is less than the constant which energizes L86FPG2FLT (I'm guessing the name of the Control Constant is LK86FPG2FLT, or something similar).

So, it's NOT that the SRV isn't opening past 63.5% during firing, it's that the pressure is too low--which means <b>EITHER</b> that the isolation valve for the P2 pressure transducer(s) (96FG-2A, -2B & -2C (if there are three)) is not open (likely if this problem started after a maintenance outage), OR the pressure transducer(s) are not working, OR the gas fuel supply pressure isn't anywhere near rated or even enough to make the P2 pressure at firing greater than the low-low pressure trip setpoint.

This also means that it's likely that there is an alarm during firing that says something to the effect that "P2 pressure is low" because there should be an alarm to warn the operator before the unit trips (OOPS! This is a Belfort unit--that might not be the case!).

So, there are several things you need to tell us--and if you don't then you need to get someone to site to help (which I strongly recommend anyway).

1) What is the gas fuel supply pressure before you try to start the turbine--when all the manual isolation valves in the gas fuel supply system are open in a ready-to-run configuration? I imagine there is an Auxiliary Stop Valve upstream of the SRV, so, what is the gas fuel supply pressure upstream of the Aux. Stop Valve before you try to start the machine?

2) When you try to start the machine, what does the gas fuel supply pressure do (upstream of the SRV or Aux. Stop Valve)? Does it stay steady? Does it drop by a few tenths of a barg? Does it drop to near zero?

3) When you say the SRV won't go past 63% stroke, is someone physically observing the valve when you are stroking it to see if it goes to full open? There is usually a scale (ruler) on the valve body and a pointer attached to the valve stem that can be used to measure the position. Someone would have to know what 100% stroke is (usually something on the order of 25-35 mm) and then while you are stroking it if the valve doesn't actually go to full open (as indicated by the pointer on the scale) then there is something mechanically wrong with the valve/actuator. It could just be that someone has miscalibrated the LVDT feedback and so it's reporting 25-35 mm as 63% instead of 100%--we don't know. But, tell us: <b>Is someone physically monitoring/measuring valve stroke and that's how you know it's not going past 63% stroke?</b>

The SRV regulator is a pressure loop, not a position loop--even though it has LVDTs, the reference for the SRV is P2 pressure, not position. That means the SRV will go to whatever position it needs to go to in order to make the actual P2 pressure equal to the P2 pressure reference--if that's 12.3% stroke today during firing because the gas fuel supply pressure is 15.6 barg, and 14.5% next week because the gas fuel supply pressure is 15.1 barg, and 9.7% next month because the gas fuel supply pressure is 16.7 barg--it will go to whatever position it needs to go to in order to make the actual P2 pressure equal to the P2 pressure reference. The LVDTs are just there for indication of position, and for stability of the regulator. In other words, the accuracy of the SRV LVDT calibration isn't all that important because the valve is going to go to whatever position is required to make the actual P2 pressure equal to the P2 pressure reference (FPRG).

Now, let's make an example calculation:

Let's presume firing speed is 20% (it may be a little lower or a little higher on your machine--substitute the actual value for your calculation). So from the formula you provided (courtesy of the Belfort Bunch):

FPRG = (20 * 3.2875) - 38.75 = 27.0 (PSIG) = 1.86 barg

So, the P2 pressure (the intervalve pressure; the pressure between the SRV and the GCV) at 20% speed when firing should be 27.0 psig, or 1.86 barg.

At 100% speed, the P2 pressure should be:

FPRG = (100 * 3.2875) - 38.75 = 290.0 psig = 19.99 barg

You did not provide the value of LK86FPG2FLT (or whatever the Belfort Bunch called it); that would be helpful.

Again, <b>from the information provided</b> the trip seems to have occurred because the Mark VI did not see the P2 pressure it thinks should have been present. That's either because the P2 pressure sensors aren't valved in correctly (again--a far too common mistake after a maintenance outage), the pressure sensors aren't calibrated or aren't working correctly, or the gas fuel supply pressure is way too low during starting.

Provide all of the requested information for us to be able to continue helping you. But, it's still recommended someone be brought to site to help with troubleshooting and resolution.

 

H A R,

Also, has someone removed the cover from the side of the base on which the combined gas valve assembly sites to look at the actuators to see if there is a plug on the top of the SRV actuator cylinder?

This can be done while the L.O. Pump is running, but it can be noisy and there will be oil vapours coming off the top of the tank. (I'm presuming the gas valve assembly is mounted on top of the L.O. tank, on the right side of the Accessory Compartment. Some newer units have off-base Gas Valve Compartments, but if they use the combined SRV/GCV then there will likely be an enclosed, rectangular metal box mounted on legs on which the gas valve assembly sits, and there will be a cover (rectangular, with lots of bolts) that can be removed to access the servo-valves, the dump valve, the actuators, and to see the flow from 20TV when it's not energized. I've even seen some sites with the off-base gas valve compartment which installed thick plexiglass (plastic "glass") covers in place of the metal cover--to make troubleshooting easier.

But, again, has anyone looked at the actuator? (This would only be necessary to do if someone has visually observed and confirmed the SRV valve stem does NOT travel the full distance it normally travels. The problem may be an inaccurate LVDT calibration--as difficult as that is to believe.)