Wheelspace Temperatures in Gas Turbine

K

Thread Starter

KRManish

HI all,

We have 2 GE Frame 9E gas turbines at our facility. The design base load is 115 MW & ambient temperature of 28 degree Celsius and a Mark 5 control system. We have a 17 stage axial compressor, 14 canister type combustion chamber & a 3 stage turbine. Design turbine inlet temperature is 1124 degree Celsius. The first 2 stages of turbine buckets has internal air passages for cooling while the third stage does not.

Recently, in one of the gas turbines, both the third stage wheel space temperatures i.e. forward outer & aft outer have been registering differential temperatures of more than 30 degree Celsius 402\434 degrees & 284\310 degrees respectively. The wheelspace differential alarm is already persisting in Mark 5 MMI as one of the second stage forward wheelspace thermocouples in unavailable.

Is this wheelspace temperature differential likely to cause any problem? If so, at what temperature range should we think of a shut down? The manual mentions only absolute wheelspace temperature values as indicative of problems, which are well within limits. Also, no rising trend has been observed for the absolute wheelspace temperature values.

Also, some time ago we had an issue with 3rd stage wheelspace temperature rising for the same gas turbine. On shutting down the unit & on inspection, cracks had been observed in the exhaust diffuser struts. The same had been welded & the unit has been running fine. The newly procured exhaust diffuser from GE has already arrived at site & we are planning to replace the same also soon. My concern is whether these wheelspace differential values are also tell-tale of any problems in the exhaust casing. The wheelspace thermocouples' installation happened way back in 2011 but this phenomenon is a much more recent one [I don't remember when exactly it started].
 
KRManish,

These kinds of issues boil down to trending--what are the differentials doing over time? Are they increasing quickly, or slowly? Since there are two T/Cs in each wheelspace, are both increasing or is only one of them increasing? If only one of the T/Cs is increasing, is it in the area where the crack was repaired?

Have you noticed any increase in #3 bearing vibration?

Does the unit have Bently-Nevada proximity vibration sensors--and if so, do the sensors on the #3 bearing indicate any increase in vibration (when the seismic vibration pick-ups might not)?

Any increase in bearing metal temperature in the #3 bearing, or the bearing drain temperature?

Does the unit have an axial position indicator (usually a B-N sensor)? If so, has the axial position changed recently?

This has been covered (ad nauseum) many times before on control.com--wheelspace temperatures attempt to measure the air temperature (when properly inserted--and this is the crux of the measurement method here) in the space between the rotating turbine wheel/disc (that the rotating turbine buckets are mounted on) and the stationary turbine diaphragm (nozzle segment). GE-design Frame 9E heavy duty gas turbines rotate at 3000 RPM--so there's a LOT of "mixing" of the air in the space (which is just a couple of mm, axially, on each side of the rotating turbine wheel/disc). Unless there's a crack or some other similar type of problem which is allowing hot or cold air to impinge on the T/C tip it's virtually impossible for there to be a difference in temperature on the two sides of the machine. Again, it's rotating at 3000 RPM, and presuming the two T/Cs in each wheelspace are inserted similarly (which is also virtually impossible to achieve) a differential between the two T/Cs is very common.

It's what happens over time to these readings--and while it's uncommon for one of the two T/Cs in a wheelspace to increase or decrease more than the other, in the situation where there is cracking of the turbine casing/exhaust frame it has been known to happen.

Your post isn't clear about what each of the T/Cs in each of the wheelspaces is doing. Are they both increasing, at a similar rate/differential, or is one of them increasing (or decreasing) more than the other? The latter would be an indication of some kind of possibly serious problem.

Also the rate of change over time is an important indicator of problems.

If you're going to be shutting down soon to replace the exhaust frame and/or the turbine casing, unless the "differential" is increasing quickly over time it's probably not going to be too much of a problem--especially if this machine has been repaired in the past. This is a very difficult thing to diagnose from afar, but based on the information provided, I would say unless the "differentials" are increasing quickly over time it's not a serious problem.

Having said that, excessive heat is not kind to turbine wheels/discs--especially over time. And it's usually excessive heat that causes cracking, along with torsional forces. You will be removing the turbine shaft and so it will be possible to inspect the turbine wheels/discs for signs of any damage.

Hope this helps!
 
CSA,

I will check up on the exact values of wheelspace temperature differential trends & let you know as soon as possible. I have been observing it for the past 3 weeks now & there has been no rising trend [Before that, the unit was under reserve shutdown for more than 8 months & I don't recall the exact values but I am positive that the differential temperatures were within 10 degree Celsius of each other]. None of the third stage wheelspace temperature values have increased, only they are not matching.

There has not been any increase in the bearing#3 seismic vibration.
Yes, the unit has 2 proximity vibration sensors for bearing#3. One of them have indicated a slight increase in vibration, but not a large [from about 25 microns to 55 microns]. The second proximity vibration sensor is showing same value.

No, there has not been any increase in bearing#3 metal temperature or bearing drain temperature.

Yes, the unit has axial position indicator. Again, no discernible change.

Again, no relative change in wheelspace temperature values over time.
But the cracks are what we are worried about. Actually, the shutdown plan is not yet finalised as the grid people at the load dispatch centre have not yet agreed due to high demand & expected long duration of outage [45-60 days]. But, if the machine is at risk, it would be a totally different matter of course.
 
Top