sudden Change of lube oil header temperature of Gas Turbine

Hello all !

In one of gas turbines (frame-V, 19.6 MW with mark-VIe control system) of our co-generation plant (with three gas turbines), today we have observed sudden change in lube oil header temperature from 64 deg C to 71 deg C, when the GT was running at a load of 19 MW. Other two gas turbine header temp remained same though the cooling water to all the gas turbine comes from the same cooling tower. We have checked it's cooling water pressure did not change and the operation of the main cooling Tower also remained same. The set temp of alarm is at 74 deg C and tripping at 79 deg C. Being electrical guy in operation, I was unable to understand this phenomena. May the esteemed forum please give some clue as to what should we check to troubleshoot about this issue .....thanks
 
Hello, jagriti!

First of all the L.O temperature of the gas turbine bearing header should be able to be regulated to no more than approximately 54 deg C (not greater than 130 deg F). So, a temperature of 64 deg C is too high to begin with.

Casting that aside, there is usually a T/C (thermocouple) in the bearing header AND a temperature gauge (thermometer) on the side of the unit also in the same run of bearing header piping. So, the way to verify the actual bearing header temperature is to WALK OUT TO THE LEFT SIDE OF THE ACCESSORY COMPARTMENT, BEND OVER AND LOOK AT THE TEMPERATURE GAUGE UNDER THE WALKWAY to check what the reading is on the temperature gauge (thermometer). If the two are relatively the same, well, then it's probably likely that the temperature is, indeed, much higher than it should be.

One side note though, it's kind of hard to understand why the L.O. bearing header temperature would change 'suddenly.' It would be very helpful if you could describe how fast it changed. Over a 5 minute period; or a 10 minute period; or a one hour period; or ??? Sudden can mean a lot of things. (To me it means, the operator didn't notice it increasing over some period of time (a couple of hours, maybe more) and SUDDENLY he notices it's higher than it was or should be. Operators are like that; to them, most everything is 'sudden.')

The typical L.O. Temperature regulating valve--which is NOT controlled by the Mark* is on the other side of the accessory base, and it has a handle on it. THE HANDLE IS NOT TO BE USED TO CHANGE THE AUTOMATIC TEMPERATURE REGULATING SETPOINT; IT IS FOR TEMPORARY MANUAL ADJUSTMENT OF THE VALVE'S ACTUATOR. There should be a metal disc on the valve stem and there should be an indicator of some type very close to the metal disc which tells if the valve is more open or more closed. (In this case, I believe, closing the valve increases the L.O. bearing header temperature, and opening the valve decreases the L.O. bearing header temperature.) If the valve is fully open and sending the maximum amount of water to the L.O. cooler, then there's likely some issue with the cooler, or possibly the valve is plugged/choked (poor quality cooling water can have this effect).

[As noted, the typically-provided L.O. temperature regulating valve is NOT connected to or controlled by the Mark* in any way whatsoever. This seems to induce anger in many plant managers, and so during a maintenance outage (or when the turbine control system is upgraded) they spend a LOT of money to get a pneumatically-operated valve that accepts a 4-20 mA signal from the Mark* to control L.L bearing header temperature. Also, some packagers of GE-design heavy duty gas turbines, including Frame 5s also seem to be providing a temperature regulating valve that is connected to and controlled by the Mark*. Unfortunately, we don't know which you have, and I'm only going to talk about the typical temperature regulating valve.]

You didn't say what fuel the unit was burning....

If it were me, the first thing I would do is make sure that only one L.O. cooler is in service, and that the manual cooling water isolation valves of the in-service cooler are both fully open (they are usually butterfly valves, and sometimes people throttle flow through the cooler for various and sundry odd reasons by closing one or the other valve (usually the wrong valve, too)). I'd then use the manual handle on the temperature regulating valve and crank it in the direction to try to increase the flow of cooling water to the L.O. cooler. Again, if the valve is already fully open it wouldn't hurt to try to open it a little more with the manual valve--just don't CRANK TOO HARD on the hand wheel.

If that doesn't work, and the unit has a second L.O. cooler which is not in operation (is in standy), I would first have the operators lower load to try to get the heat load down on the L.O. system for the next operations so that it doesn't trip. Then I would begin by using the transfer fill valve to fill the cooler with L.O. (NOT the TRANSFER valve--the transger FILL valve!!!) and at the same time crack open the manual isolation cooling water inlet valve to the standby cooler, and also crack open the manual isolation cooling water outlet valve. I would leave the outlet cooling water valve just slightly opened and then gradually fully open the cooling water inlet valve. After the cooler is filled with L.O. I would then close the transfer fill valve and be prepared with the proper wrench (ratchet and socket, probably) to operate the L.O. transfer valve. Then, I would begin using the wrench to, as quickly as possible, move the L.O. transfer valve to switch the L.O. flow to the standby cooler from the in-service cooler. (Don't dilly-dally while operating the L.O. transfer valve--there are O-rings which can be unseated from their positions and get cut as the valve is rotated--which means there's no way to isolate L.O. from either cooler...) Most units with dual L.O. coolers can switch coolers while the unit is running without any problems--AS LONG AS THE STANDBY COOLER IS COMPLETELY FILLED WITH L.O. BEFORE TRANSFERRING THE L.O. FLOW TO THE STANDBY COOLER. The above procedure is how to do that.

Once the standby cooler is in service, then I would close the manual cooling water isolation valve to the out-of-service cooler, and then close the inlet cooling water valve. If the temperature drops, then it would seem there is a problem with the L.O. cooler which was originally in service.

Another thing about GE-design L.O. systems (that is different from other manufacturers) is that GE cools the L.O. BEFORE it goes into the filter. SO, if the L.O. filter is dirty and has a high differential pressure, then the flow through the filter can be reduced to the point that the L.O. bearing header temperature can be affected.... So, it's a good idea to check the L.O. filter differential pressure before beginning any of the above. If it's high, really high, then that could be part of the problem.

Hope this helps! Please write back to let us know what you find and how you resolve the problem. Again, it's unusual for the L.O. bearing header temperature to increase "suddenly" so if the temperature gauge is reading lower than the HMI, then something's wrong with the T/C or the wiring between the Mark* and the T/C (or both).

Lots to consider! Best of luck!
 
CSA,
you said ::
The typical L.O. Temperature regulating valve--which is NOT controlled by the Mark* is on the other side of the accessory base, and it has a handle on it. THE HANDLE IS NOT TO BE USED TO CHANGE THE AUTOMATIC TEMPERATURE REGULATING SETPOINT; IT IS FOR TEMPORARY MANUAL ADJUSTMENT OF THE VALVE'S ACTUATOR. There should be a metal disc on the valve stem and there should be an indicator of some type very close to the metal disc which tells if the valve is more open or more closed.

I would never/ever let someone with little experience look at, touch, or think about Robertshaw valve adjustment. Even the manual adjustment can be left in a undesirable position. EVERYTHING else in post is spot on except " closing the valve decreases the L.O. bearing header temperature, and opening the valve decreases the L.O. bearing header temperature." .

Wish you well.
 
Curious_One,

Thank you for the correction, and I've edited the post.

The typical L.O. temperature regulating valve, manufactured by Robertshaw, is something which has to be experienced to be understood. Everyone has to start somewhere, right? jagriti has been around Control.com long enough; he should be capable. What I really wanted him, and others reading the post to understand is: There are many parameters which are not controlled by the Mark* (though that seems to be changing, and it really angers a lot of people when they are told that (Plant Managers, particularly, firmly believe the Mark* controls all parameters of turbine operation--or at least they think it does, and should). L.O. bearing header pressure and -temperature are two things (on non-F-class machines) which are not controlled by the Mark*. (And most people ass-u-me they both are, and are usually surprised to learn they are not.)

Anyway, thanks again for the correction. Be safe--and stay healthy!!!
 
jagriti,

Could you please tell us how quickly the L.O. temperature changed when it "suddenly" increased?

Could you tell us what the temperature gauge on the side of the L.O. tank (Accessory Base) reads and what the HMI L.O. Bearing Header Temperature value is at the same time (or within a minute or two of each other)?

Can you tell us what the position of the disc on the Robertshaw L.O. Temperature Regulating Valve is in?

Thank you!
 
You are right CSA, gotta learn sometime. Another note for others. The robertshaw valve works via a very large capiliary bulb immersed in lube oil . The robertshaw valve cannot do anything or change anything "suddenly". a perfect example of Slow process control !!
 
The L.O. bearing header temperature can change quickly by using the manual hand wheel. I hope jagriti, or whoever might try using the manual hand valve will return the assembly to AUTO when they're done. But, you're right, it's agonizingly slow under Automatic control. Which is one reason a lot of plants want so desperately to replace it with something that's controlled by the Mark*. That and they don't understand about the capillary tube, which usually gets damaged at some point in the turbine's lifetime because someone will step on it or drop something on it and even though the smell of the fluid leaking out of the capillary is pretty bad, it only lasts a short time and it's forgotten until the temperature gets out of "control."

That's why I asked about the time it took for the "sudden" change in L.O. temperature.... If the capillary tube broke and leaked, that would probably be the fastest change of temperature IF THE ROBERTSHAW WAS IN AUTO. (I find VERY few Robertshaws in AUTO.... VERY few.)

We all have to learn some time. That's one of the things I try to remember when replying to posts here on control.com: We all have to start somewhere and learn some time. I was extremely fortunate that I had a very patient Customer on my first start-up who taught me SO MUCH and helped my build my confidence. The difference for me was I was NOT afraid to study the P&IDs (Piping Schematics) and find every device and component on the drawings. (The L.O. tank had to be cleaned and re-painted on my first start-up, so I was able to get in and crawl around and locate things!) P&IDs are the key to understanding a lot about GE-design heavy duty gas turbines, but most people don't think they need to learn to read them or locate the devices and instruments on them.

Anyway, it's because of the man who gave so freely of his time and knowledge to me on my first start-up that I feel a strong need to give back and to "pay it forward" in the hopes others will do the same. So, I try very hard to remember what is was like when I was first learning.
 
I first learned my turbine experience on old oil burning 70s vintage Frame 7Bs with Mark I. I am starting to believe that the "new guys" simply do not care. Everything can be repaired from the keyboard according to them. Yet, they forget all the stuff that works without control system assistance. Until they encounter a robertshaw valve. LOL
 
jagriti,

Could you please tell us how quickly the L.O. temperature changed when it "suddenly" increased?

Could you tell us what the temperature gauge on the side of the L.O. tank (Accessory Base) reads and what the HMI L.O. Bearing Header Temperature value is at the same time (or within a minute or two of each other)?

Can you tell us what the position of the disc on the Robertshaw L.O. Temperature Regulating Valve is in?

Thank you!
It changes within 2-3 days. Yes the tank temperature gauge reads 82 DegC and the HMI value of L.O. header temp is 70 deg C.
Regarding the position of disc 'Robertshaw LO temp regulating Valve.....need to check. Will revert on this. Thank you, sorry for the delayed reply
 
jagriti,

Oh dear.

Oh my.

Some GE-design Frame 5 heavy duty gas turbines are different from others, especially if they were packaged by licensees of GE (John Brown; BHEL; etc.). So, sometimes the packages have different options or standard offerings.

Usually, on the left side of the Accessory Compartment, under the walkway, there are several temperature switches. Usually, at least three of the temperature switches are arranged in a horizontal line on the side of the I-beam web (which is the side of the L.O. tank). And, there is usually one temperature gauge in that same line, and often another junction box in that same line--with a dual-element T/C terminated inside of the fixture). These are all penetrating the web of the I-beam and into the L.O. Bearing Header (usually downstream of the Bearing Header Pressure Regulator. There are thermowells that protrude through the web of the I-beam and into the pipe that serves as the bearing header, and measures the temperature of the oil in that pipe.

Temperature sensors, when inserted in thermowells, do best when there is a heat-conductive grease or material around the T/C which fills any void (space) between the sensor and the wall of the thermowell. Even cooking oil works well as a conductive fluid--it doesn't have to be anything industrial or special.

Many times the temperature sensors inserted into a thermowell have a spring in them to make sure the tip of the temperature sensor bottoms out (touches the bottom of) in the thermowell. If the sensor is not properly inserted into the thermowell, it can't measure or sense the temperature it's trying to measure very well. The very same is true of T/Cs used to measure the oil temperature--they need to be properly inserted, and should be in contact with the bottom of the thermowell with some kind of heat-conducting medium surrounding them.

AND, very often, the T/C used to measure L.O. Bearing Header Temperature is a dual element T/C--with only one of the elements connected to the Mark*. (The other element (pair of leads) is a spare.)

Very often, there is also another temperature gauge on the side of web of the I-beam, and it measures the temperature of the L.O. in the tank. Sometimes it's on the same side as the L.O. Bearing Header Temperature Gauge; sometimes it's on the other (right) side of the Accessory Base.

So, it would seem like you have the reading from the temperature gauge for the L.O. tank--not the Bearing Header Temperature.

I would, personally, not call an increase of 10 degrees over 2-3 days "sudden." Gradual; steady; increasing; but not sudden.

AND if the temperature has stabilized over a couple of days, then it would seem that whatever caused the increase has stabilized.

It may be that the unit at your site is NOT equipped with a Bearing Header Temperature Gauge; only a L.O. Tank temperature gauge. That would be somewhat unusual, but not surprising or unexpected. Again, if there is one, and if it's working or not broken, it's going to be in the same horizontal plane as the L.O. Bearing Header Temperature switch sensors and the L.O. Bearing Header Temperature T/C(s).

Anyway, best of luck. If you find out what's going on, please write back to let us know.

[By the way, some machines only have one L.O. Cooler, but even ones with one L.O. Cooler will have a temperature regulating valve. You can take a (clear) picture of the temperature regulating valve's valve stem and disc and indicator and attach it to a response to this thread.)

Here's a snippet from the Robertshaw Self-Actuated Temperature Regulator Manual:

1604433808680.png

F is the scale for the indicator.

Here is a link to the manual:

https://www.bucksales.com/robertshaw/manuals/160_Regulators/MAN160_RT1010.pdf

Enjoy!
 
Top