We start the Frame 6 Machine HAVING Mark-VI after CI. and Machine was tripped on L86GCVT when Machine changeover in IFO.

We replaced the GCV Plug during CI. and calibrated LVDT after the Job. and LVDT feedback was found acceptable range during Calibration.
Machine was started on NG Fuel and thereafter change over on Liquid Fuel, FSR2 came to zero and Machine change to Liquid Fuel. mean time L86GCVA and L86GCVT,and L3GCVFB alarm came and Machine got tripped.
LVDT feedback of Both LVDT was found around -8% while FSR2 zero.
Actually machine got tripped due to the L86GCVT but I want to know why this happen when we have calibrated the LVDT properly.

Any one please help me to find our the root cause of the malfunctioning...

Please mail me your valuable suggestion if any one face this type of problem after changing the GCV Plug.

 

Is the GCV part of a combined gas valve assembly, where the SRV and GCV are cast into a single assembly, each with it's own actuator and plug and set of LVDTs? The SRV and GCV are usually "side-by-side" in this combined gas valve assembly. (The original manufacturer of this combined gas valve assembly was Young & Franklin in the state of New York in the USA. This combined gas valve assembly was used for many Frame 5s and Frame 6Bs over several decades. A different combined gas valve assembly was used for many years on Frame 7E/EA units and Frame 9E units for several decades as well, but the SRV and GCV were "on top" of each other, not side-by-side.)

Also, when the unit is at rest or running on liquid fuel, what is is the LVDT feedback value from the GCV?

Also, most of these alarms are the result of one or more conditions (an "OR" gate driving a single alarm text message). Did you have a chance to determine which specific condition caused the alarm messages to be annunciated?

 

Since there was no response from the original poster about the GCV in use, the "drawing" and overly long description (sorry, ProcessValue) below presumes that the gas valve assembly being referred to is part of the combined SRV/GCV gas valve assembly typically provided with GE-design heavy duty Frame 5 & -5 gas turbines.

Please study the drawing carefully; it's not easy to draw in basic, generic ASCII text, but I'm not posting anything in any other site. (They tend to disappear after a while and then people are always asking to re-post.)

It's very important to note that the LVDTs are attached to the GCV valve stem, not to the GCV plug. So unless the top of the GCV valve stem is in contact with the bottom of the GCV plug the position indicated by the LVDTs is NOT the position of the GCV plug, and it's really the position of the GCV plug that is important.

And the the GCV valve stem is in contact with the GCV plug whenever hydraulic pressure is applied to the GCV actuator to force the actuator rod up to pust the GCV valve stem up to push the GCV plug open, such as when gas fuel is flowing during normal circumstances. However, during LVDT feedback calibration, the <b>ONLY</b> way to ensure that the GCV valve stem (with the LVDT cores attached to it) is in contact with the GCV plug is to insert and maintain feeler gauges during the calibration and verification process.

If feeler gauges were used to achieve and maintain the zero stroke position during calibration of the LVDT feedback (see below), then when the feeler gauges are removed from the "gap", if the GCV valve stem falls down when the valve is closed to contact the GCV actuator rod, the "gap" now exists between the bottom of the GCV valve plug and the top of the GCV valve stem. If the GCV LVDT feedback was calibrated properly, because the LVDTs are attached to the valve stem ***THEY WILL INDICATE A NEGATIVE GCV VALVE POSITION!!!***

This is normal and to be expected because the zero stroke position is when the the GCV vavle stem is pushed up to contact the bottom of the GCV valve plug and the LVDTs are attached to the GCV valve stem. But, when top of the GCV valve stem is NOT in contact with the bottom of the GCV valve plug, the valve is still closed (at zero stroke!), but the LVDTs will indicate a negative stroke, because the valve stem is NOT at the zero stroke position (presuming the feeler gauges were used to achieve and maintain the zero stroke position during calibration of the LVDT feedback).

A common occurrence is for the GCV to be calibrated with the GCV valve stem fully down and not in contact with the bottom of the GCV valve plug as the zero stroke position. This is actually a negative position, and can also lead to Diagnostic Alarms when trying to use AutoCalibrate to calibrate the LVDT feedback. It will also cause the first 0.030-0.050 inches of movement to be completely ineffective (presuming the gap is within specification), because the GCV plug won't be rising until the actuator rod pushes the GCV valve stem up to contact the GCV plug. This can cause problems with firing and acceleration.

When gas fuel is shut off (either because of an emergency trip or during a normal fired shutdown) the GCV valve stem may be forced all the way down by the force of the GCV closing spring shutting the valve very quickly (and forcing the actuator rod fully down, as well, in the absence of hydraulic pressure). This would also cause a negative position to be displayed, and is to be expected if feeler gauges were used to achieve and maintain the zero stroke position during calibration of the LVDT feedback. This is because the GCV LVDTs are attached to the GCV valve stem, and when the valve stem is not in contact with the GCV valve plug they will indicate a negative position if they were calibrated at the true zero stroke position.

In the above case, if the gap between the actuator rod and the GCV valve stem is more than the upper limit (greater than 0.050 inches), then another problem can occur. Most Speedtronic turbine control systems check to see if the LVDT feedback is more than -5.00% stroke and if it is will initiate a turbine trip condition. So, if the GCV valve stem falls "excessively" (more than approximately 0.050 inches), then this problem is going to occur (the detection of a "position" less than -5.00%). A gap larger than 0.050 inches isn't in and of itself a bad thing, but it might indicate the GCV isn't fully closing (something is preventing the GCV plug from being fully seated on the GCV valve seat). It might also indicate excessive wear of either the top or bottom of the GCV valve stem, or the bottom of the GCV valve plug, or the GCV valve seat. And in the worst case, it might also cause the GCV not to be able to admit the maximum amount of fuel possible (though this would be an extreme worst case, and probably would only occur if the valve had been "short-stroked" by the turbine packager and the gap was very large).

My best guess at what happened here is that calibration was done with the GCV valve stem in the true zero stroke position (most likely unintentionally), but, the gap between the GCV valve stem and the actuator rod is more than 0.050 inches, which means that when the top of the GCV valve stem is NOT in contact with the bottom of the GCV plug that the indicated position is much less than the Control Constant looking for feedback failure (I believe that's LK3GCVFB in many Speedtronic panels; it may be a different name, but it's usually set at -5.00%.)

The originator said they had replaced the GCV plug, but probably didn't check or correct the gap between the GCV valve stem and the actuator rod, and when the unit is tripped and the GCV valve stem and actuator rod get driven down by the GCV closing spring, then the LVDTs are indicating less than -5.00% (-8% was mentioned, I believe).

The <b>proper</b> thing to do would be to measure and correct the gap. Or to calibrate using the feeler gauges. But, they likely won't shut down to do that....

One more (unlikely) possibility is that the roll pin that is used to ensure the LVDT bar stays in the proper position and doesn't rise or fall during GCV operation didn't get re-installed properly or has sheared off, allowing the LVDT bar to shift downwards from the position it was in during calibration/verification. But, the originator said the LVDT calibration was verified, so this isn't likely (if the verification was done with feeler gauges and a dial indicator or -calipers to measure the actual physical position).

<pre> **NOT SHOWN** in this drawing is a very large,
powerful GCV closing spring above the GCV plug
that pushes the GCV plug down to close the GCV.

The GCV plug rises ("lifts" off the GCV valve
seat) in order to allow gas fuel to flow to
the gas manifold and nozzles.

Hydraulic pressure against the underside
of the actuator piston forces the actuator rod
up to contact the GCV valve stem which is pushed up
to contact the bottom of the GCV plug to open
the GCV against the force of the closing spring.

The GCV valve stem is **NOT** fixed (attached)
to the GCV plug, and must be pushed up
to contact the bottom of the GCV plug to
open the GCV against the force of the closing spring.

Also, the GCV valve stem is **NOT** fixed (attached)
to the actuator rod. The GCV valve stem is
free to move up and down and can be moved
up and down slightly when the GCV is closed and there
is no hydraulic pressure in the actuator.

The gap between the top of GCV actuator rod and the bottom of
GCV valve stem, and the gap between the top of the GCV
valve stem and the bottom of the GCV plug is
<b>exaggerated</b> in the drawing below in an attempt to show that the
actuator rod is <b>not</b> connected to the valve stem *and* that
the valve stem is <b>not</b> connected to the valve plug.

The GE specification for the total gap is 0.040 inches,
+/-0.010 inches, or 0.030-0.050 inches.
When the valve is assembled and installed, the only way
to measure the gap is by ensuring the actuator rod is down as
far as it can travel, and the GCV valve stem is up as far as
it can travel, and then to measure the tap between the actuator
rod and the GCV valve stem.

Gas Control Valve
_____________________
| |
Gas Fuel In--> GCV Plug |
|___ ___________ ___|
GCV Valve Seat---|-->\\ | | //<--|----GCV Valve Seat
|____\\_||-||_//____|
| |*| |
| |*| --> Gas Fuel Out
|--------|*|--------|
|________|*|________|
|*|<------------------GCV Valve Stem
__|*|_________|____
|______________|____|<----LVDT Bar (Clamped & <b>pinned</b> to GCV Valve Stem)
|*| |<------LVDT Core (Rises and falls inside the LVDT
|*| ||| Armature as the GCV Valve Stem rises
|*| ||| and falls)
|_| |||
___ |||
| | | |<---LVDT Armature ( Bolted to stationary base plate)
____________________| |________|_|_______________
____________________| |__________________________ <---Stationary Base Plate
| | | |
| | | |
| | | |
| | | |<---GCV Actuator Assembly (bolted to stationary base plate)
| | | |
| | |<---|-----GCV Actuator Rod
| | | |
|-----------|<--GCV Actuator Piston
|_ ________|
||
_____||
Hydraulic Oil<--->______|
From Servo-valve
(Bi-directional flow)

For simplicity, only one LVDT is shown; most GE-design heavy
gas turbine control valves have two LVDTs.
</pre>
Note that the LVDT core(s) moves up and down with the GCV valve stem (the LVDTarmature is stationary; it's bolted to the stationary base).

When there is no hydraulic pressure in the system through the GCV servo to the GCV actuator, one can force the GCV actuator rod down until it touches the bottom of the actuator (usually by prying lightly against the top of the LVDT bar). Then one can lightly pry up against the bottom of LVDT bar until the top of the GCV valve stem touches the bottom of the GCV valve plug. A gap between the GCV actuator rod and the GCV valve stem should be visible, and the specifications say the gap should 0.030-0.050 inches. <b>***THIS IS THE ZERO-STROKE POSITION OF THE GCV!!!***</b>

When calibrating the GCV LVDTs, one is <b>calibrating the LVDT feedback so that it accurately reflects the actual GCV plug position</b>, this because the GCV valve stem (to which the LVDTs are attached) is not physically attached to the GCV plug. So, the GCV valve stem (to which the LVDTs are attached) must be put into a position during calibration (and verification, either during the calibration, or any time after the calibration!) that makes it move along with the GCV plug. And, in this case the only way to do that is to insert feeler gauges between the top of the actuator rod and the bottom of the GCV valve stem, so that the LVDTs will only rise and fall as the GCV plug is driven up and down. And, so that the GCV valve stem will reliably return to the true zero stroke position during any calibration and/or verification.

To make sure the GCV valve stem rises and falls with the GCV plug one must insert feeler gauges between the top of the GCV actuator rod and the bottom of the GCV valve stem to force the GCV valve stem up against the bottom of the GCV valve plug--which is the true zero stroke position. Those feeler gauges must be inserted in the gap and remain in the "gap" during the LVDT calibration procedure and when verifying the accuracy of the LVDT calibration after an LVDT calibration, or when checking the accuracy of LVDT calibration in order to ensure that the GCV valve stem does not "drop" below the zero stroke position.

There is a seal at the bottom of the GCV assembly that the GCV valve stem passes through. Depending on the age and condition of the seal, the GCV valve stem may be "held" in place by the pressure of the seal spring. In other words, the seal spring can "bind" the GCV valve stem and prevent it from being easily moved up and down when the GCV is closed and there is no hydraulic pressure in the actuator. (There's also a seal on the actuator rod where it passes through the stationary base plate that can "bind" the actuator rod and not let it fall down fully, which is why it's necessary to use some light force to push the actuator rod down fully when measuring the gap and/or installing the feeler gauges.)

 

As BKM seldom comes to say or appreciate others efforts which are for his benefit, i am going to so that for him.

excellent "overly long " explanation CSA.

 

Wow. I'm kinda shocked, ProcessValue. Thank you. I just hope BKMEENA, aka B K M, aka Brijendra Kumar Meena, will find something useful, presuming his machine has this type of combined gas valve assembly.

And you're right about his penchant for not providing any feedback. It's about time we stop providing any responses to his questions, though some of them are very difficult to understand and no responded probably for that very reason.

By the way, the above is exactly true for the SRV in the combined, side-by-side gas valve assembly as well. The gap should be exactly the same, and it's measured the same, and the feeler gauges must be used during any LVDT calibration and calibration verification/check.

I don't recall exactly, but for the larger GE-design Frame 7B, -7E and 7EA, and Frame 9E machines with the combined vertical gas valve assembly one of the two gas valves also has this gap between the actuator rod and the valve stem, and the other valve does not. I don't recall if that's the SRV or the GCV, but I believe it's the valve on top of the assembly that does not have the gap. And, the gap specification is exactly the same, and the feeler gauges must be used for LVDT calibration and calibration verification/check.

This is one reason why writing a procedure for calibrating gas valve LVDTs is so problematic; it's very difficult to tell someone to insert feeler gauges during the procedure if they don't understand why the feeler gauges are necessary. Especially if they've never seen anyone use feeler gauges during the calibration. And, they're not necessary for all gas control valves. So, a 'one size fits all' description can be very misleading and bad documentation is worse than no documentation. Because it causes people not to use or search out any documentation, and leads to all manner of myths, wives' tales, and legends--most of which are completely unfounded and which lose something in the telling or get more grandiose in the telling.

And people think that LVDT calibration issues are caused by the Speedtronic panel, when mostly it's because they don't understand the devices which the LVDTs are attached to and how they work and what they're measuring. They think AutoCalibrate knows everything about the device to which the LVDTs are attached; and it can't, and it doesn't. It's funny what a name can mislead people to believe. And this one has mislead many people.

This is true of a lot of "issues" mistakenly attributed to Speedtronic panels: a lack of understanding of the field devices and systems the Speedtronic interfaces with. And, it causes people to do some very crazy things with the Speedtronic panel in misguided attempts to fix what is a non-problem, or worse, to spend hours, and days even, trying to troubleshoot a problem that isn't really a problem. It's a perception issue, mostly, because they think or they are told or they have come to believe that something should work or not work as they think it should or should not work, or as they have been told it should or should not work. And, again, bad documentation is worse than no documentation, as in the case of this information which is missing from almost any GE documentation, including the Control Specification. This isn't even taught to GE field engineers/TAs.

B K M said the GCV plug was replaced during an outage, and I have to wonder if the drawings for the gas control valve were used during the refurbishment. And, as is the case with many I&C technicians I encounter these days they don't believe they need to understand the field devices and systems the control panels they work on interface with. They know all about how to change logic or sequencing or make pretty displays, but they don't know much, if anything, about the process or the field devices or instruments the control system uses to monitor, control and protect the equipment the process (in this case, the process is generating torque to produce electric power). But they know little about anything outside the control system or even the PC used to interface with the control system.

B K M just wants someone to tell him to change the Control Constant that senses loss of LVDT feedback to prevent the L86GCVA and L86GCVT alarm and trip, and nothing will ever be done to fix the real problem. Or even to understand the real problem.

 

>L86GCVA and L86GCVT

I haven't read this in detail so just a quick comment.

I have seen this trip on dual fuel units when changing from liquid to gas fuel. Does this unit have trip oil. I've had software were the valve reference would increase as soon as l20fg1x would energize, but it would take 2 seconds for the trip oil pressure switches, 63HG1/2/3/L to pickup. The solution from GE was to include the pressure switch status in series with the valve enable permissives, L3PM1 and L3PM1. Where's the same software on 9E GT's that don't have trip oil, Woodward gas valves with an integrated shutdown solenoid, transferred without problem.

Don't look exclusively for a valve calibration/setup problem

 

When you do read the original post, completely, you will see that the problems began after the GCV plug was replaced during a maintenance outage. And, the problem occurs after the fuel transfer from gas to liquid is complete, when the GCV and SRV should be closed. Further, when the GCV is "slammed" shut by the action of the GCV closing spring when hydraulic pressure is removed from the actuator, the LVDT feedback indicates less than -8.0% position, which is less ("more negative") than the typical -5.0% loss of feedback setpoiont. So, the time it takes to establish Trip Oil pressure can also be ruled out.

And, many of the smaller Frame size machines don't require Trip Oil pressure to "stroke" (actuate) the GCV, only the SRV requires Trip Oil Pressure. And this post is about the GCV, not the SRV.

It's likely not a calibration issue. It's most likely a problem with an excessive gap between the actuator rod and the GCV valve stem which was not properly set when the GCV plug was replaced. And this results in a larger than normal, less than expected, negative position on what may be a GCV with properly calibrated LVDT feedback. The originator indicates the LVDT feedback was checked and found to be calibrated "properly".

The intent of the overly long write-up was to show that there is an intentional gap between the actuator rod and the GCV valve stem, that the LVDT feedback from a properly calibrated set of LVDTs <b>should</b> indicate a slight negative position because of the gap, and that an excessive gap would result in a less than expected ("more negative") LVDT position and feedback.

Which is what the originator is reporting.

The other intent of the overly long write-up was to point out to readers of the post that the true zero-stroke LVDT position is when the top of the GCV valve stem is fully in contact with the bottom of the GCV plug. And, the only consistent, reliable way to achieve that zero stroke position during LVDT calibration and calibration verification is to insert feeler gauges into the gap and leave them there during the calibration and verification.

 

CSA,

We have recently disassembled our gcv srv (GE Frame 9e) for overhaul. I noticed that the srv plug has holes on it. Why are the holes there? does that mean that whenever srv is shut at default position, gas can still go through it and raise the p2 pressure? Thanks for ur relentless guidance.

 

I have never seen holes such as you described. The SRV is primarily a stop valve, so it should not pass or leak.

 

Holes are served as pilot, so when SRV or GCV opens up empty space is filled slowing to come to the ambient pressure instead of giving a jerk.

Correct me CSA or PV or anybody please, if I am wrong?

Thanks

 

Sardar9,

Are you guessing?

 

When we disassembled the plug, we could see the spring in it. We can just conclude that when there is pressure, the plate will hold the spring down and therefore close the opening by a lil.

You are supposedly right. It is a pilot valve which is a small valve inside the plug assembly of srv. It seems like it controls the high pressure of gas inlet by controlling the amount of gas entering p2 cavity.

 

Pilot in SRV ??!!

well as far as i know there is no pilot. The gas stop ratio valve (VSR - as per PID's) the default position is close. Ie with no hydraulic oil supply, it will be in closed condition. but the question piqued my interest and i asked my rotary engineer about it. he explained to me the following.

the SRV is not a ordinary control valve with a stub type metal seat. it has a skirted arrangement. the holes that you see in the plug is the skirt guide. they are used in places when there is a high pressure flow control is needed. it supposedly reduces the eddys and turbulent flow and maintains a laminar flow. he showed me a control valve with skirted arrangement and it became clear. just search "skirt guided control valve" or "skirted plug" in google and you will get the necessary info.

I have never actually seen a disassembled SRv. i should check it out the next time, but normally even during M&I i have not seen them disassembled. did you have any problem with the valve ??

 

Hi processvalue,

Thanks for your feedback. I was search hi and low for information on gcv srv but to no avail. Your information just answered my question. Thanks again. We recently changed our GCV SRV (GE Frame 9E)due to gas leaking at the stem of the srv during GT start up. After we have changed our GCV SRV, we have got another problem with the spare unit. During shut down, the srv always get stuck at 8% open. We had to isolate the discharge after the gas skid filter and vent out the remnant gas in the line for the srv to close completely.

This prompted us to disassemble the original unit for overhaul and replace the spare whenever there is another outage.

 

Sardar9 is right. SRV has internal piloting mechanism that is required to control low pressures and flows while turbine is at firing and warming. Pilot valve then wide opens after the main valve plug is lifted off its seat.

Regards

 

Piloting mechanisms and skirted control valves are not the same devices.

GE, and it's packagers, have used a variety of different types of valves as Stop-Ratio Valves (SRVs) over the years, including the combined SRV/GCV assemblies (of which there were two types: one used primarily for Frame 3s, Frame 5s, and Frame 6B units; and one for Frame 7B/E/EA and Frame 9E units), rotary cam vee-ball valves, and right-angle plug-type valves (most recently ones manufactured by Woodward). There were other valves manufactured by Fisher used as well for some units over the years.

So, unless one qualifies the specific type of SRV being referred to it's inappropriate to refer to all SRVs as having the same construction and similar characteristics which can cause confusion.

 

Hi,

Such piloting mechanism exists in the combined SRV/GCV made by Y&F. It is an inner mechanism on SRV only and not mentioned in any document submitted by GE or Y&F. I was not aware until I saw the service bulletin submitted by GE Oil & Gas, which was about the maintenance philosophy applies to SRV and GCV.

Regards

 

TOral,

Can you please post the bulletin/document number?

Thank you.

 

Hi CSA,

Submitted By: GE Oil & Gas
Service Bulletin No: NIC 04.34.SL-Rev.00
Dated: 23rd December 2004
Subject: Gas Control and Speed Ratio Valves Maintenance Philosophy.

Regards