We have a GE 9E gas turbine on site, and the control system uses a GE Mark V HMI. Right now, we’re having an issue with the GCV (Gas Control Valve), and we need to replace it. After replacing the GCV, we need to calibrate the LVDTs (Linear Variable Differential Transformers). Once the automatic calibration is done, the system will generate the zero-position LVDT voltage and full-stroke LVDT voltage.


Normally, we would take these voltage values and enter them into the I/O Configurator, then download the settings to the R/S/T controllers, and finally power them off and restart them. However, we’re concerned that restarting the controllers might cause problems, as we’ve had issues before where the controllers failed to restart properly after downloading.


So, our question is: after the automatic LVDT calibration is complete, can we skip the download and controller restart, and just start the turbine directly? Since the calibration already writes the LVDT voltage values into RAM, the values are already active and being used.


Can we delay the download and controller restart until the next planned shutdown, when a GE engineer is on site to help? They have more experience with this process and can assist if any issues come up.

 

@DGY,

Yes. When AutoCalibrate completes an LVDT feedback calibration it writes the individual LVDT feedback calibration values into each control processor's RAM. You can enter the values from the AutoCalibrate procedure (usually the average of the three values determined by the AutoCalibrate) into the I/O Configurator, validate and save them, close the I/O Configurator and then use the EEPROM Downloader to download the IOCFG partition to EEPROM. Now you have the GCV LVDT values from the AutoCalibrate in EEPROM in case one of more of the control processors gets or has to be rebooted.

The purpose of rebooting after AutoCalibrating and downloading is to write the downloaded values into RAM which is what the processors use when controlling and protecting the machine. The purpose of writing them into EEPROM is to have the "same" values when and if the control processors ever have to be rebooted before another AutoCalibrate procedure is performed. So, downloading the average values of an AutoCalibration procedure into the I/O Configurator, downloading them to EEPROM, and then rebooting the control processors to get the average values into RAM is kind of unnecessary (and it puts average LVDT feedback values into RAM, overwriting the more precise LVDT feedback values determined by AutoCalibrate...!). As you wrote, AutoCalibrate has put the LVDT zero- and 100% stroke LVDT feedback values into RAM when it finishes, and all that's really necessary is to put the average LVDT feedback values into EEPROM without rebooting after downloading.

 

@WTF?
Thank you very much for your reply,These are very useful to me.

 

@DGY,

The Mark* V uses RAM for control and protection functions. EEPROM is just a "safe" place to store the values and programs for control and protection. Each time a processor (control or communication) is started/booted-up as part of the process it goes to EEPROM to get the values and programs it needs and loads them into RAM. The only real exception here is the LVDT zero- and 100% stroke voltage values, which can be written into RAM when a successful LVDT feedback calibration is completed using AutoCalibrate. So, it's not really necessary to reboot the control processors after calibrating LVDT feedback--but it IS necessary to get the closest approximations of the most recent LVDT zero- and 100% stroke voltages written into EEPROM so that if one or more control processors get rebooted or needs to be rebooted that the closest approximations to the last LVDT calibration are "ready and waiting" in EEPROM to be copied to RAM during the boot-up sequence.

 

@WTF?
I have a new question that I’d like to trouble you with.
We’re going to replace the GCV (Gas Control Valve) on site. This GCV was previously removed and sent back to a GE factory for repair, and it has been stored in the warehouse for over ten years. After the replacement, do we need to perform a Cv (flow coefficient) test on the valve? I remember this was required for Siemens gas turbines, but I’m not sure if it’s necessary for the GE 9E model.

 

@DGY,

Most likely not. I presume when you say GCV (Gas Control Valve) you are referring EITHER a single gas control valve OR a combined SRV/GCV assembly (where BOTH the SRV (Stop/Ratio Valve) AND the GCV are in one combined cast steel assembly). I am also presuming the machine has conventional combustors (not DLN-I (Dry Low NOX first generation)). But, it's really not clear from the information you have provided in this post.

One of the beauties of GE-design heavy duty gas turbines has always been that the components are selected and configured so that they don't require any kind of pre-installation testing. One could say the components are, in general, "plug and play."

Having said that, the GE division in Belfort, France, has a particular fondness for changing tested and proven designs and components--just because they can. They have had the design and control philosophy for Frame 9E and 9F gas turbines for nearly 20 years now, and they have changed a lot of auxiliaries at various times over the years because they believe that French designs are much better than American designs. Generally, they seem to have tried to adhere to the "no pre-installation testing" required--BUT, without more information about the machine and its auxiliaries it's very difficult to say.

GE has gone away from the combined SRV/GCV assemblies on new units--primarily because of DLN combustion system requirements. They have also "standardized" (though that word doesn't really seem applicable to GE Belfort designs/configurations) on what's affectionately referred to as IGCV module/skid design--where there are multiple gas control valves serving to control fuel pressure and flow-rates to the various manifolds and nozzles of their heavy duty gas turbines. Though, I haven't seen such a gas fuel module/skid used on conventional combustor-equipped gas turbines nothing is really "standard" any more.

Sorry; I wish I could be more helpful but that's all I can offer based on the information provided.

 

@DGY,

Most likely not. I presume when you say GCV (Gas Control Valve) you are referring EITHER a single gas control valve OR a combined SRV/GCV assembly (where BOTH the SRV (Stop/Ratio Valve) AND the GCV are in one combined cast steel assembly). I am also presuming the machine has conventional combustors (not DLN-I (Dry Low NOX first generation)). But, it's really not clear from the information you have provided in this post.

One of the beauties of GE-design heavy duty gas turbines has always been that the components are selected and configured so that they don't require any kind of pre-installation testing. One could say the components are, in general, "plug and play."

Having said that, the GE division in Belfort, France, has a particular fondness for changing tested and proven designs and components--just because they can. They have had the design and control philosophy for Frame 9E and 9F gas turbines for nearly 20 years now, and they have changed a lot of auxiliaries at various times over the years because they believe that French designs are much better than American designs. Generally, they seem to have tried to adhere to the "no pre-installation testing" required--BUT, without more information about the machine and its auxiliaries it's very difficult to say.

GE has gone away from the combined SRV/GCV assemblies on new units--primarily because of DLN combustion system requirements. They have also "standardized" (though that word doesn't really seem applicable to GE Belfort designs/configurations) on what's affectionately referred to as IGCV module/skid design--where there are multiple gas control valves serving to control fuel pressure and flow-rates to the various manifolds and nozzles of their heavy duty gas turbines. Though, I haven't seen such a gas fuel module/skid used on conventional combustor-equipped gas turbines nothing is really "standard" any more.

Sorry; I wish I could be more helpful but that's all I can offer based on the information provided.

Thank you very much for your patient reply.