We have two Temp loops A & B for Drive End Bearing drain temperature. and C & D for non drive end bearing drain temp for Generator side, which is driven GT Frame 5 of GE.

What is the correct logic to be applied for Trip?
could anybody suggest.

regards,
[email protected]

 

Sundar2015,

Why do you want to trip the unit on high bearing drain temperatures (this is oil temperature leaving the bearing, right?)? Have you, or someone at your site, had problems with high bearing drain temperatures causing a bearing failure so it's felt it's necessary to trip the turbine on high bearing drain temperature?

Most sites are trying to reduce the number of trips to reduce the thermal stresses on the turbine. Many former trips are being converted to automatic shutdowns, or even runbacks (a runback is when the turbine control system automatically unloads the unit to some low load until the operator corrects the problem or decides to shut the unit down).

Is it felt there is not enough time for the operator to take action on a high bearing drain temperature to save the bearings? Is there no alarm on high bearing drain temperature? Could you consider a runback or a trip instead of a trip? Does the insurance company or owner require a trip on high bearing drain temperature?

If you definitely want to trip the turbine (or even initiate a runback or shutdown), it is suggested to take the average of the two readings for a single bearing as the bearing drain temperature, and then alarm on a high temperature and initiate the desired automatic action on a high-high temperature. You should also consider an alarm on a failed signal (either high or low) or if the temperature differential between the two readings of a single bearing exceeds some value--just to make sure the operators are conscious of a problem with the readings and have notified the Instrument Department so they can repaired/replaced as soon as possible.

 

Sundar2015... with the equipment you have, I suggest 2oo4!

If you want a copy of my JSSS Paper, "Probabilistic Risk Assessment of Equipment Safety Systems", contact me at:

[email protected]

Regards,
Phil Corso

 

Rarely, but once in awhile I agree with CSA!

In this case his suggestion to provide a pre-trip, or an off-normal alarm is right-on!

BTW, 2oo4 logic is std on the nuke-side of Nuke plants! My paper discusses the Pro's and Con's on its implementation.

Phil

 

Rarely do I ever agree with Phil Corso, and this is no exception. How common is it for both bearings of a generator to experience oil flow problems simultaneously? I venture almost never. So, hence my recommendation to alarm on the average of each bearing's temperature sensors, and if both bearings do have problems then the unit is still protected.

Unless failed sensors can be "rejected" or accounted for somehow (depending on how the control system deals with a failed sensor--some go high, others go low) there is a risk with 2-out-of-4 (2004) of nuisance problems.

 

CSA, you are correct again! But the proper answer is in my JSSS paper.

In case your curiosity is piqued JSSS means "Journal of System Safety Society."

Phil