LVDT Failures on Splitter Valve Frame-6 Machine


Thread Starter



We face frequent failure of LVDT's on Frame-6 splitter valves. Regulator type 43 (Position control, High selected). Both LVDT deviates and give spiking reading, causing unstable splitter operation.

The failure is repeated in many Frame-6 machines, but all the time on splitter valve. Splitter valves are physically located close to the machine and are subjected to radiant heat form turbine.

Appreciate experience from other on these kind of applications.
Hmmm.... Several things come to mind, the first of which is: Are you using the proper LVDTs rated for the temperature of the environment? Typically, GE uses LVDTs which are rated for high temperature applications, and I have never seen or heard of splitter valves being placed in a turbine compartment. I have seen heated gas fuel which didn't cause any problems with gas valve valve LVDTs, so it would seem there is something unique about the arrangement of the gas valves at your site. (Contrary to (false) popular belief, every GE-design Frame 6 is NOT like every other GE-design Frame 6--especially when it comes to auxiliaries and packaging/arrangement.)

So, that's two things: the temperature rating of the LVDTs, and the placement/location/arrangement of the gas valves.

The next thing is the wiring, and GE and packagers of GE-design heavy duty gas turbines have used four-wire and five-wire LVDTs over the decades that packages have been produced and sold. And I have run across similar problems caused by improper wiring.

AND, then there's the other wiring problem caused by improper selection of barriers/isolators for the application. Also, the barriers/isolators can be intermittently failing, or, as is often the case, they can be improperly installed or improperly connected.

I have never seen or heard of gas valves being placed in turbine compartments. I have seen some dodgey gas valve compartment designs and locations that we're NOT properly ventilated which led to occasional issues such as you are describing.

That's about all I can think of. Please write back to let us know what you discover and how you resolve the issue.
Thanks for your message. As we did in previous cases, we have replaced both LVDT's and Servo. Valve is functional now.

One LVDT had been indicating low (Say around 50%) and the other was reading correct (Say around 80%) but spiking readings. Valve was hunting around 70-100% whereas the Spiking LVDT was indicating 30-90% at times. This is the reason we have doubted LVDT in first place.

We suspect it is because of the radiant heat and the temperature LVDT's are subjected to, although the Sentech LVDT is rated for 176degC and T3 rated.

The removed LVDT;s when tested at workshop reading well. Of course at ambient temperature.

At the risk of sounding like a broken record, it seems there is something atypical about the arrangement of the gas control valves and the gas fuel system on the unit(s) at your site. I have seen some GE-deisgn F-class heavy duty gas turbines that had the gas fuel module placed next to the turbine compartment with a large opening between the two. The gas control valves were in the middle of that opening, and were exposed to high temperatures from the turbine compartment. It was later determined that the gas fuel module ventilation fans were actually drawing hot air from the turbine compartment into the gas fuel module over the gas valves. Combine this with the fact that the gas fuel was heated (performance heating), and the gas valves had a LOT of issues. UNTIL the compartments were insulated from each other (a LOT of work to create the barrier in the opening between the two compartments AND to make openings in the gas fuel module through which outside air could be drawn for cooling. But, that did solve the issues--which were NOT related to LVDT failures, but problems with expansion of valve components due to heat which caused problems with starting when the valves were hot.

If the gas fuel arrangements were designed by GE Belfort, France, it's highly probably, even likely, that there are some problems with the design. GE Belfort are known for changing proven designs just because they can and that has had some ill effects on many systems over the last decade-and-a-half or so.

176 deg C doesn't really seem like a high temperature rating for an LVDT. For decades GE used LVDTs manufactured by Schaevitz, some of which were used on rocket engines and nuclear reactors at temperatures in the range of 1000 deg F (550 Deg C). I think I read or was told that Schaevitz has since been sold or acquired by some other company, but I would think their designs and technology, which were some of the best in the world, were continued after the acquisition.

Yes; from the information provided it would seem that temperature might indeed be a problem. Have you considered trying to build some kind of heat shield around the valve body to prevent heat from affecting the LVDTs? Have you tried working with the packager of the GE-design turbine to find LVDTs with a higher temperature rating? Have you tried working with a sensor manufacturer representative to find LVDTs with a higher temperature rating? Have you tried working with the packager of the turbine and auxiliaries at your site to understand why the turbine compartment heat is adversely affecting the splitter valve LVDTs, and what could be done to remediate the problem?

GE, and the packagers of GE-design heavy duty gas turbines, don't always have the best designs. Many of their designs are the result of cost-reduction efforts, and many of those weren't reviewed for all possible effects before they were put into production. And, this isn't just true of GE designs; it's pretty much true of just about any product which is comprised of components from MANY different manufacturers which all have to work together properly to produce the desired outcome/product. And, it's not unusual for some designs to have to be re-thought and re-designed in order to make them reliable and suitable for the application.

Hope this helps!