RTDs and temperature values

A

Thread Starter

ASGHAR

We are often experiencing bad temperature values (less temperature comparing actual) on generator stator and Gen-bearings (GEN cap: 120MW 11KV measuring device: Mark-V). Most of the time problem was solved by cleaning (spring Contact cleaner) and re-tightening the terminals at marshaling panel. Please note that All of these RTDs (PT100 three wire) are having one of their compensation wire grounded at field JB which is not foreseen in general installation. some one have idea on following

1. Purpose of grounding the RTDs compensation wire in field.

2. Can it contribute to the problem stated above?

3. Any Suggestion recommendation to overcome on this problem.

(Note:
a. Marshaling panel same type of terminals are in use for other instrumentation signal without any problem.

b. No any loose wire found during troubleshooting.

c. RTDs shield wires are connected to ground in marshaling panel.
 
The grounding of one of the compensation leads at the generator is <b>VERY</b> common on GE-design heavy duty gas turbine-generators. It is an inexpensive way to reduce the wiring costs by using two-conductor cabling for each RTD signal versus three-conductor cabling.

There should be a least one, possibly two, conductors that are also terminated at the same location as the RTD compensation leads which also run back to the Mark V (through a marshaling panel it sounds like) to the terminal board. Then, there should be jumpers from those two leads to all the other RTD inputs to connect the Mark V compensation lead terminal to ground--the same ground as the RTDs.

The shield drain wires of the two-conductor cables for the RTD signals SHOULD be connected to ground, in only one place, as with all two-conductor instrument cabling with shields.

Have you looked the terminations of the RTDs at the generator end? Are you certain of the connections between the marshaling panel and the Mark V? If you have been spraying contact cleaner and getting good results, then it would seem the terminals may not be very good, or possibly the wires are of poor quality, or there is something wrong with third leg jumpering, which could be done in the marshaling panel or at the Mark V terminal board.

It could even be a problem with the tinning of the ribbon cable connectors in the Mark V between the terminal board and the I/O card in the <C> core.
 
three wire rtd's are for special rtd sensing circuits to compensate for the resistance of the field wiring, not grounding. grounding the third wire is very likely going to cause major problems.

if you have two-wire sensing electronics, you need to wire the rtd's for that configuration without grounding the third wire...
 
Thanks for swift response that I was expected. I had written same to OEM two months back till now not a single word has been replied.

CSA you are right... its good way to save the wiring but what about if three wires and one shield is running from field JB to the Marshalling cabinet and still one compensation wire is grounded (shorted with ground)...? And interestingly this set up run from RTD to Mark-V with no change, means no two-wires at all.

I was thinking to remove these ground links but then a doubt prevents me doing, and that is. "High Voltage (11KV) around the RTD (installed in stator winding) may damage the measuring circuit and this ground is a path for induced voltage to drain out" any idea???

Field JB wiring and termination are perfect (clean terminals, 1.5mm2 wire multi conductor and shining lugs). Marshalling terminals (most suspected for me also) has been examined very deeply (Dismantled in pieces) and found no issue. But sill I have order new of different brand.

Regards
 
Ashgar,

I don't know if I quite understand what you're describing, but if I understand correctly then jumpers need to be installed between the "open" third conductors of each RTD in the Marshaling Cabinet. And at least one or two of the RTDs should have conductors from the generator terminated on them, bringing the ground from the generator JB to the Marshaling Cabinet. Somewhere, the grounded legs need to be jumpered and it sounds like at your site it needs to be done in the Marshaling Cabinet if three-wire, twisted cables are used between the Marshaling Cabinet and the Mark V.

The Mark V can read grounded or ungrounded RTDs (and grounded or ungrounded T/Cs, as well). So, the grounding is not the issue.

There were some early Mark Vs that had problems with generator RTD inputs because the inputs weren't "filtered" properly to reduced the 50Hz influence from being mounted in the generator. So, later versions had different software--at a minimum--to be select either a 50 Hz or 60 Hz filter.

The %-- or 60 Hz filter selection is made in the I/O Configurator. RTDs are usually connected to the <C> core, so you need to look in the I/O Configurator in the <C>. I believe--if the filter selection is available in the software--you should find it in the TBCA card screens.

Other than the above, about the only other thing that could be recommended would be to change the either the CTBA card (where the RTD inputs to the Mark V are usually terminated) or the TBCA card (which converts the analog inputs to digital values in the control signal database).

Please write back to let us know how the troubleshooting fares.
 
Thanks CSA I learned that in past there was problem in measuring temp with RTDs on generators and I also learned that Mark-V has capability to measure grounded and un-grounded RTD/TC. And now I can confirm that our application is grounded RTD and it’s done as it needs to be.

My prime suspect to resolve this issue is the replacement of marshaling terminal block. I am waiting for their delivery (probably it will take some time). If problem re-occur even after replacement of terminals, I'll move to Mark-V hardware as per your advice and will let you know accordingly, meanwhile I’m very thankful for your feedback.
Thanks again…

Regards
 
ashgar,

You are most welcome, and please let us know how this proceeds.

Also, do not forget to check the 50/60 Hz filter setting in the I/O Configurator (if the software in the Mark V supports that filtering capability).
 
Dear CSA,

After replacement of Marshaling terminals and 3 month smooth operation i concluded that problem was the terminals. I also confirm the I/O configurator setting which was found 50Hz similar to grid frequency.

Thanks for guidance.
Regards,
 
Top