Mark V Start Motor and Shaft Turning


Thread Starter


Hi there,

I am rookie about Mark V and GT5 power plant. I've read many things in, but I couldn't see my problem so that I decide to write here. Ratchet is running, clutch is ok, start motor turns. Although oil pressure is ok, start motor does not take over shaft turning. Now mechanical team is draining oil tank in order to see pipe lines condition.

thanks in advance for your recommendations.
How old is the torque converter?

When was the last time the torque converter was refurbished?

Some torque converters have a strainer in the suction; if yours has one, has it been checked and cleaned recently?

Some starting means systems have a "spoiler" solenoid to "control" (limit) speed. If yours has one, is it working pto6?

How hot does the torque converter body get when you try to start the unit and it won't break the turbine shaft away from zero speed,?

Many torque converters have a spring-loaded foot valve in the suction line. If the spring breaks the suction can be blocked, which will prevent sufficient oil from getting into and through the torque converter--and usually the torque converter body gets too hot to touch. The foot valve can usually be removed without draining the L.O. tank.

Please write back to let us know what you find!

thank you so much for your reply. you are right because torque converter was too hot. Mechanical team has changed it. Now turbine turns. I am trying to arrange GCV and SRV valves before IGV.

Thanks again!

Thank you for the feedback! Excessive heat can a sign of too little oil getting into the torque converter, or a worn torque converter.

Could be tough arranging the gas valves; they're pretty heavy and bolted down. And you would have to move a lot of piping to arrange them differently. ;)

Piping system has been installed by mechanical team. I am working for GCV and SRV operation and after IGV and eventually generating power as you already know. This is my first project. I am really working so much, and I have to search so many documents for each step. Now I try to operate GCV and SRV valves. To do that, by using logic forcing window I forced to zero L3GRVFLT fault and L63FGL. After that, L20FG1X forced to 1. I expect that SRV and GCV will move but it did not happen. Could you send me a proper documentation for how can I operate these devices and calibrate their LVDT signals. Next step is IGV and synchronization etc.

If you can send your recommendations to this rookie, I will be grateful sir. my e-mail is [email protected].

Thanks in advance.

Okay, I'm confused--but, then, many would say I'm easily confused. So, Are you adding gas fuel to an existing unit that was previously running on liquid fuel(s)?

Or, is the unit being relocated from some other plant/part of the world to your plant, and is now being commissioned?

To "stroke" or "calibrate" LVDT feedback (because that's the ONLY thing that is being calibrated on any servo-operated device (and if it doesn't have LVDTs, then there's "calibration" that can be done) you need to use the AutoCalibrate feature of Mark V. I believe there are instructions in the Mark V Maintenance Manual, GEH-5980, any maybe also some information in the Mark V Application Manual, GEH-6195. It's not too difficult, but it is very difficult to provide a written, site-specific procedure.

I presume you have set the zero-stroke voltages of all the LVDTs to 0.700 VAC RMS, plus-or-minus 0.020 VAC RMS (so from 0.68 VAC RMS to 0.72 VAC RMS) using a True AC RMS voltmeter. In the case of the GCV and SRV, you need to use a piece of wood or a long piece of pipe (approximately 1 meter long) to gently pry down on the bar that is clamped to the valve stem of each valve (have your prying device as close to the valve stem as possible). Then pry up on the bar that is clamped to the valve stem of each valve (with the prying device as close to the valve stem as possible). You should see a very small gap between the top of the valve actuator rod (which is protruding from the base of the gas valve assembly, below the valve stems) and the valve stems. Then you need to insert feeler gauges into the gap between the valve stem and the actuator rod (the gap should be between 0.030 inches and 0.050 inches)--and LEAVE the feeler gauges in the gap. (You can do this for each valve individually before you try to stroke or calibrate the LVDTs. But you MUST leave the feeler gauges in the gap while calibrating and verifying calibration.)

You need to establish hydraulic system pressure--for some units with an AC motor-driven Aux. Hydraulic Pump that means you have to start and run the Aux. L.O. Pump and then start the Aux. Hyd. Pump. If the unit doesn't have an AC motor-driven Aux. Hyd. Pump, you need to CRANK the unit to get pressure from the Accessory Gear-Driven Main Hydraulic Pump.

Then, for the SRV you will likely need to force L20TV1X and L20FG1X (those are <i>typical</i> names--the ones at your site may be different!) to get Trip Oil pressure to be able to use high-pressure hydraulic oil to open the SRV. (If the unit has a mechanical overspeed bolt on the Accessory Gear, that also needs to be in the RESET position (NOT the tripped position).)

Then you use AutoCalibrate to move the SRV. You choose SRV from the AutoCalibrate menu, and then you would use the MANUAL positioning feature to enter a position (say 50%) and the valve should move to what it thinks the 50% stroke position is. (The actual position at this point is NOT important--it just needs to move to some partially open position.) Then you need to check the polarity of the servo currents being applied to the coils of the servo--this is VERY IMPORTANT. To do this--while the valve is partially open--you disconnect one lead of one of the servo output coils to the SRV (say start with <R> processor's SRV servo output) and the valve should remain open--not exactly at the same position, but it should remain open. Then remove one of the wires of <S>'s SRV servo output, and the value should still remain open--though it will close very slightly, it should still remain open. If it slams shut, the polarity of the current being applied is incorrect, and you need to reverse the wires to the SRV coil being supplied by <T>--because only <T> is supplying servo current to its SRV coil (it's easiest to to at the QTBA of <T> in the Mark V panel).

Once the polarity of <T>'s SRV servo output current has been verified to be correct, then you need to re-connect <S>'s servo output wire and disconnect <T>'s. The valve should remain open and not slam shut. If it slams shut, then the polarity of the current being applied from <S> is incorrect and needs to be changed (by reversing the two wires of <S>'s SRV servo output at the <S> QTBA).

Once the polarity of <S>'s SRV servo output current has been verified to be correct, then you need to reconnect <R>'s servo output wire and disconnect <S>'s The valve should remain open and not slam shut. If it slams shut, then the polarity of the current being applied from <R> is incorrect and needs to be change (by reversing the two wires of <R>'s SRV servo output at the <S> QTBA).

Once the polarity of the currents being independently applied by each processor have been verified (and corrected if necessary), then you can cancel the MANUAL positioning feature of SRV AutoCalibrate and go to the next step.

At this point, you should be able to select AUTO-CALIBRATE and let the Mark V do its thing. It will first send positive servo current to the servo coils to close the valve and when it sees the LVDT feedback voltage not changing for a short period of time it will capture the LVDT feedback voltage (for use later). Then it will send out negative servo current to the SRV coils to make the SRV move to the fully open position and wait for the LVDT feedback voltage to stop changing for a short period of time and it will also capture that LVDT feedback voltage. Then it will send positive current to close the valve again and check to see that when the LVDT feedback voltage stops changing (signifying it has gone fully closed again) the TCQA card will then calculate the 0% and 100% stroke voltages it needs to scale the LVDT feedback.

Then, what you need to do is to manually move the valve again using the MANUAL positioning feature of AutoCalibrate to say a reference of 100% and someone who is watching the valve--but keeping their hands away from the valve stem and LVDT bar!!!--needs to record the reading from the scale ("ruler") mounted on the valve assembly. Then, MANUALly move the valve to a reference position of 75%, and the person watching the scale should see the valve move to 75% of the previously measured and recorded 100% stroke reading. Then move the valve to a reference position of 50%, and the scale should indicate 50% of the previously recorded 100% stroke reading. And do the same for 25% reference position. If all of these look good, the calibration is deemed to be good!

Move the feeler gauges to the GCV and repeat the procedure of verifying servo current from each control processor, and of calibrating and verifying the scaled LVDT feedback.

Hope this helps! Performing servo polarity checks should be done to ensure each coil is getting the proper polarity current applied to it for maximum reliability. And, verifying the accuracy of LVDT calibration is also very important (failing to do so is just wasting time). It's usually not so critical for gas valves, but it is critical for the IGVs (which requires a more detailed procedure for calibration and verification--one which should be using a machinist's protractor for measurement and verification).

When you remove the feeler gauges at some point the LVDT feedback will Ge negative. This is normal and to be expected. It should only be -2% or-3%, and I believe if it is more than-5% there will be a lockout condition, which means either the gap between the actuator rod and the valve stem is too large, or the calibration wasn't done correctly or you pried up too much on the LVDT bar when inserting the feeler gauges.

The reason for the negative position indication is that the designers of the gas valves wanted to be sure the valves would close fully, so they used a valve stem that was not coupled to the valve plug and was free to disengage from the plug when it was closed and the LVDTs are coupled to the valve stem and not the valve plug. So, because the valve stem has to rise up to make contact with the underside of the valve plug in order to open the valve by pushing the plug up there is a small distance (0.030 inches to 0.050 inches) where, when the valve plug is closed the valve stem can drop and result in a negative position. This is normal and to be expected.

Again-2% or -3%, or as much as approximately -4.5% or so is acceptable. If the gap between the actuator rod and the valve stem is too much more than 0.050 inches, that indicates excessive wear somewhere in the valve assembly, and if the gap is much less than 0.030 inches that means the valve was not re-assembled properly and may not fully close to completely shut off the flow of fuel to the turbine.

Hope this helps!!!
Thanks CSA for your explanation.

I am one of your regular follower and works on Frame V, VI and Hitachi machine. I always thinks of what kind of problem a torque converter and ratchet system in GE machine can occur and how to diagnose them?
And what could be the checks for healthy and non healthy ratchet and torque converter.

It will be good if you can share your experience related to ratchet and torque converters problems and their diagnostic.

Thanks in advance

My experience is that it is a VERY good idea to install pressure gauges at various points on both the torque converter AND the hydraulic ratchet system (including a differential pressure gauge across the hydraulic ratchet filter) and record the pressures when the system is in a new and clean condition (such as during commissioning and after replacement or refurbishment of the torque converter). This will give a snapshot of proper operating conditions should suspected problems arise. Site personnel should also be regularly (though not every day or even every week) recording the pressures (and differential pressure) and monitoring them, comparing them to the new and clean pressures and if they begin to see changes in pressures that could portend future problems.

Cleanliness is important to the hydraulic ratchet self-sequencer (VH-41, if I recall correctly). There are lots of very small passages in the mechanism and they can easily become plugged with dirt and varnish. Quite often, because there is usually no differential pressure gauge installed from the factory across the hydraulic ratchet filter the filter gets plugged (choked) and ruptures--releasing most of the dirt and impurities it had trapped into the hydraulic ratchet mechanism. And it rarely gets replaced--again because there's usually no differential pressure gauge or differential pressure switch connected to the turbine control system to generate an alarm on high differential pressure.

That's about it--oil cleanliness and a good set of pressure readings from a new and clean condition to use to compare when there are problems--and a set of gauges to use to gather data when there are problems. Gauges are inexpensive, and can be very important. Every site needs to know what the normal operating pressures and differential pressures are--and they need to have a way of obtaining pressures and differential pressures when problems are suspected.

A read of the torque converter operating and service manual will usually tell you exactly where to install pressure gauges for troubleshooting. But, without a good set of baseline pressures and differential pressures that set of readings won't do anyone a lot of good. Install them early and gather data over time and compare the data to the good conditions--and one can usually predict when a problem is brewing. Or if a problem suddenly arises, it can be spotted pretty easily.

Hope this helps!