Hi,

Please who can help me about the specifications of the limits (alarm and trip) for the temperatures of wheelspace for a GT GE FRAME 3?

After upgrade Mark II to Mark VI, do have I to modify the limits or not?

The GT is for crude oil pumping. Thanks.

 

You didn't say if the problem with the wheelspace temperatures was high differentials or higher than previously experienced temperatures. Remember, some Mk IIs had cold junction compensation calibrations which had to be performed, and which could drift....

IF ALL THAT WAS DONE was a control system retrofit from Mk II to Mk VI, then, no; you should not have to modify the limits.

IF, HOWEVER, there were some performance modifications done to the unit, and/or a Hot Gas Path- or Major Inspection was done at the same time, it's likely that either the wheelspace temp's changed because of cooling air flow changes due to the performance modifications and the Control Constants weren't appropriately modified, OR the wheelspace T/Cs weren't properely re-installed.

After servo-valves, wheelspace T/Cs are the most misunderstood monitoring element on heavy-duty gas turbines. The key words in the previous sentence are: monitoring element. That's all wheelspace temperatures are used for--monitoring. Yes, there are alarms when the temperatures are "excessive" or when the differential is high between the T/Cs in the same wheelspace. But, the unit is (generally) not tripped on high wheelspace temperatures or high differential temperatures.

Further, it is very common for wheelspace T/Cs NOT to be inserted properly after a maintenance outage. If they aren't positioned properly (the tip isn't fully inserted or it's touching metal other than the shallow thermowell) it's very common for them to indicate an erroneous WHEELSPACE temperature--rather, they're measuring metal temperature, metal other than the thermowell or sometimes they're not even measuring an temperatures. It's virtually impossible to determine if wheelspace T/Cs are installed correctly, because one can't see the thermowells where they're installed.

If you have two wheelspace T/Cs in each wheelspace, and one of them differs by, say, 100 or 150 deg F, from the other--it physically can't be. The turbine rotor is turning at several thousand RPM for a Frame 3 mechanical drive turbine, churning the air in the wheelspace as it rotates. It's virtually physically impossible for there to be that much difference in the WHEELSPACE temperature from one side of the machine to the other. BUT, if one T/C isn't inserted properly and isn't measuring the same temperature as the other, this can lead to high differentials.

The BIG thing about wheelspace T/Cs is that they need to be monitored and TRENDED. If the temperature in a wheelspace is actually increasing, both T/Cs will increase at a similar rate. If it's increasing quickly or increasing steadily without leveling off to levels not previously experience then there's probably a real mechanical problem. Similarly, if the actual temperature is decreasing quickly or continously, there's probably a real mechanical problem.

Another thing which can happen during a maintenance outage is that orifices can be forgotten to be installed, or the wrong orifice can be installed. There are usually metering orifices in the Cooling and Sealing Air piping which are used in part to control the air flow to the wheelspaces. Sometimes if performance upgrades are done, the new orifices are missed and the old ones are re-installed.

You should have the wheelspace temperature alarm limits defined in the Control Specification for the Mk VI. If you feel they are not correct, you should contact the control panel supplier for clarification/assistance.

markvguy

 

Our GE GT after CI also have the same problem:

Tem. at 2nd stage wheel space are A & B (our model GE is M5100)
Before CI, A is higher than B, but in control limit (diff. between A& B < spd_lim and A < alarm H).

After CI, B is higher than A, and sometime out control limit (diff. between A& B > spd_lim and A > alarm H if generator <16 M).

Please help!
Thanks.

 

Were the wheelspace T/Cs removed during the CI, and then re-installed after the CI prior to the start-up?

 

Yes, Mr CSA. The wheelspace T/Cs removed during the CI, and then re-installed after the CI prior to the start-up.

 

GT,

First of all, during a CI, nothing that affects the wheelspace cooling (or heating) is changed.

Second of all, it is physically IMPOSSIBLE for the temperature of any wheelspace to be so radically different on opposite sides of the unit. The shaft of a Frame 5 is spinning at approximately 5,100 rpm, which is mixing the air in the wheelspace (which touches the metal in the wheelspace). There is no way the temperature of one side of a wheelspace can be so different from the temperature of the other side of the same wheelspace.

Third of all, there are two thermocouples in each wheelspace for redundancy; that's why the average of the two is used for calculating the temperature that is compared to the alarm setpoint.

Fourth of all, improper insertion (including re-insertion) is the cause of almost ALL wheelspace temperature problems. Period. Full stop. End of discussion. <b>Especially after maintenance outages.</b>

Fifth of all, as was said earlier in this thread: wheelspace temperatures are to be TRENDED. If you have a problem like this after wheelspace T/Cs are removed and re-inserted when there's little or no chance that anything else could have caused such a change in wheelspace temperature, the problem is IMPROPER RE-INSERTION. So, since the most likely cause of the problem is IMPROPER RE-INSERTION, simply trend the wheelspace temperatures until the next time they are removed and re-inserted. The two will track each other, even if they are radically different, and if there is truly a problem they both will start trending up or down together.

Sixth of all, the temporary solution is to change the setpoints of that wheelspace to be just slightly higher than the current (erroneous) reading so if this wheelspace or any other wheelspace goes high the operator will get an alarm (since GE doesn't alarm individual wheelspace temperature problems as they should).

Does this answer your question, GT?

 

Dear Mr. CSA!

First, I am sorry. After CI, the temp. higher is the same side before CI.

Second, the temp. sensor have ferrules "set". my team work have be inserted very carefully, also exactly.

So what any reason except "the problem is IMPROPER RE-INSERTION"?

Thanks for your support!

 

GT,

The next time your machine is open and the rotor has been removed, have a look at how the wheelspace T/Cs are inserted through the outside casing of the machine, through guide tubes, and into "nipples" (thermowells)in the nozzle segment. Then ask yourself, and anyone observing the arrangement with you, "Is it <b>REALLY</b> possible to re-insert the thermocouples <b>EXACTLY</b> to the same position as when they were removed?"

Given the length of the thermocouple, and the way in which they have to be "snaked" through the guide tubes and then make what is essentially a 90-degree turn into the nozzle segment thermowell ("nipple), it is physically impossible to get the tip of the thermocouple to reside in exactly the same position as it was in prior to removal.

The ferrule, when set, may not have been in the proper position, either. The fact the temperature, as <b>now</b> reported, is almost the same as before the T/Cs were removed and re-inserted reinforces that fact. If the temperature being reported is relatively the same as before the CI and the temperature differential with the other T/C is relatively the same as before the CI, then it's likely the T/Cs were re-inserted nearly identically as before the CI.

It's EXTREMELY unfortunate that GE and their packagers don't put better illustrations of wheelspaces in their manuals. The distance between the stationary nozzle segment/diaphragm and the rotating turbine wheel is just a few mm, and the wheel (on your machine) is rotating at approximately 5100 RPM. So, how can there be a high differential between the temperatures on the two sides of the unit in the same wheelspace? Even if there is a leak of hot combustion gases into the wheelspace on one side of the machine, the turbine wheel rotating at 5100 RPM is going to mix the hot gases and cause the high temperature to be experienced by ALL the metal in the same wheelspace. Full stop. Period.

It's physically impossible for there to be a high differential in the same wheelspace (between the two T/Cs in the same wheelspace) unless there is something radically wrong, and if something was radically wrong there would be high vibration, loss of performance (low power output) and the effects would be very noticeable.

You say there was a high differential before the CI and again after the CI. Now, you didn't have the machine apart with the rotor and nozzles segments removed during the CI, but if the performance (power output) of the unit after the CI is roughly the same (hopefully better) than before the CI, and the vibrations are about the same as before the CI, then the problem is IMPROPER RE-INSERTION. And, if there was a high differential before the CI I will even go so far as to say the problem is IMPROPER INSERTION.

Look, the technicians <b>ALWAYS</b> do their best when re-inserting wheelspace T/Cs. But, the nature of the way they are inserted and how they have to fit into the thermowells in the nozzle segments just does not allow for positioning the T/C tip in the exactly the same position. The fact that you have determined the differential stayed on the same side of the unit and is about the same as before the CI means they did a pretty good job of re-inserting the T/C. But, if it wasn't inserted properly when the ferrule was originally set then even if the T/C was re-inserted nearly identically the same as before it was removed there will still be the same problems as before the T/C was removed.

I've had this same discussion at least 200 times over my career. I've had Customers refuse to pay for the maintenance outage because the wheelspace T/Cs have a higher (or lower!) differential than before the outage. And, this when the Customer's own personnel removed and re-inserted the T/Cs! Of course, they continue to run the machine and make power and steam (if there is an HRSG), even though the wheelspace T/C temperatures are not perfectly matched on both sides of the machine.

Rest assured: The actual temperature in the wheelspace is uniform on both sides of the machine (again, unless there is something radically wrong which would have other symptoms as well). The turbine shaft and turbine wheels rotating at 5100 RPM in close proximity to the nozzle segments/diaphragms ensure the actual temperatures are uniform. But, if the instruments being used to measure the actual temperature are not properly inserted then the reported/indicated temperatures will not be a proper reflection of the actual temperatures.

Next time your machine is open and the rotor is removed, have a close look at how wheelspace T/Cs are routed from the compressor casing into the thermowells in the nozzle segments. Have a look at the thermowells in the nozzle segments. If you have spare nozzle segments in your parts storage warehouse, have a look at the thermowells now and try to imagine how the T/C tip has to be inserted from 1.5 or 2 meters away, through tubes, into the thermowell.

When there are problems like this (high differential in the same wheelspace) and there is no loss of performance and no high vibration, then one has to <b>TREND</b> the indicated wheelspace temperatures to observe any change and the rate of change. That's how you will determine when and if there is a future problem in that wheelspace (or any wheelspace, for that matter)--by observing the change and rate of change of indicated wheelspace temperatures.

So, <b>based on the information provided,</b> it would indeed seem the T/Cs were re-inserted nearly identically but that they weren't inserted properly to begin with. Even when T/Cs come from the supplier with ferrules already set in place they can be set wrong. I've seen factory-set ferrules be incorrectly set by as much as 15 mm--both too long and too short!

If you have the ability to trend parameters on your machine, trend each of the pairs of wheelspace T/C s upstream and downstream of the wheelspace that has the "problem" during a start from a cold condition. Trend the problem wheelspace T/C temp's also. You will see they all increase and decrease at approximately the same rate, even if they aren't exactly the same value. That's because the rotating shaft/turbine wheels mix the gases in the wheelspaces and that makes the actual temperatures in the wheelspace uniform and causes the indicated temperatures to change at approximately the same rate even if they aren't exactly the same values (as the actual wheelspace temperatures are).

But, given what wheelspaces are, and how wheelspace T/Cs are inserted, the only cause (unless there is low power output of the turbine and high vibration) of a high differential <b>in the same wheelspace</b> is improper re-insertion.