GE Frame 7EA Combustion Turbine Starting Motor


Thread Starter

Bill Futch

We have a 4160V starting motor with torque converter for our GE frame 7EA combustion turbine. We recently upgraded the Mark IV controls to the Emerson Ovation system but it appears that the problem would have existed in the Mark IV as well. When a start is cancelled with a normal shutdown before the turbine reaches starting motor cutout speed of 60%, the turbine does not fire down because the starting motor is still engaged at full torque and holding the speed at about 1100 rpm's. This appears to overheat the starting motor windings with the long duration at full torque. It seems that the starting motor should cutout if it hasn't already when a normal stop is given. My first concern is if we modifiy the logic to do this, will the fuel have enough time to ramp down to flame out before the speed drops too low? Does anyone know why the MarkIV would not already do this?
Bill Futch,

I don't have any Frame 7EA sequencing to look at as of this writing, but I can say I'm not sure if a normal shutdown (STOP) is initiated during a START the unit would go on a fired shutdown (that would be good idea, but I don't recall ever seeing logic for that). It should basically just close the fuel stop valve because the START permissive is lost before the unit reaches complete sequence/FSNL.

I'd have to look at the L94T shutdown logic and the L4CR logic and the L20TU1X logic to be sure, but a normal STOP during an "incomplete" START should not initiate a fired shutdown where the fuel would be stepped down as it is during a fired shutdown.

STOPs before FSNL are pretty rare--or, I should say, I've only done it or seen it done a couple of times when the HRSG had problems and the operator had to stop the turbine and chose not to press the E-stop pushbutton and initiated a STOP.

But, if memory serves me correctly when L1S drops out because L94X picks up it should drop out L20TU-1 and also shut down the starting motor. It would be better if the starting motor kept running (cooldown), but to reduce the current draw on the starting motor 20TU-1 should be de-energized on a STOP. The last Frame 9E I worked on (which is similar in many respects to a 7EA) did just that--20TU-1 dropped out when a STOP was given during a start (before 14HA).

So, it would seem something is amiss. Again, the preferable situation would be for the cranking motor to go through it's cooldown after the torque converter is "de-pressurized" by de-energizing 20TU-1 to unload the starting motor.

And, the turbine should not necessarily go on a fired shutdown, but if it did that be even better. My recollection is that a STOP before complete sequence/FSNL should just drop out L4 (through L94T).

But, again, I don't have any Frame 7EA sequencing to review at this time.
I have not checked the logic either, but I experienced this problem on a MK1 controller.

The maintenance personnel wished to lubricate the starting motor.

They gave the turbine a Start and quickly gave the turbine a Stop before 14HR was picked up.

The 4CR relays did not drop out.

14HR was the key to the problem. Although your problem is different, I thought I would share this.

The loss of 52CR or the pickup of 94X should of dropped out the 4 master protective and shutdown the starting motor after the cooldown cycle timer expires.
Bill, like CuriousOne I have seen this before on older units, it is the unfortunate part of Emerson copying rung for rung logic from old units. It sounds to me your issue lies with logic that controls the torque converter fill solenoid 20TU, possibly how the cranking motor is issued a shutdown, and lastly I would wonder if you have any relay protection setup for cranking motor protection. Logic I am looking at was developed for newer units with the Voith converter, with ratchet driving an air cooled generator.

Logic is such that if a stop is issued with the unit above 14HM, that should latch an aux. shutdown signal(L94XX) which should drain the converter and allow the cranking motor to run for the duration of its cooldown timer. In the case the unit is firing it should go to a minimum FSR value and be allowed to decelerate normally. Once the unit reaches blowout speed of 20% the fuel is tripped.

Lastly you really should have relay protection for the cranking motor to protect it from a thermal event. Since these motors are typically operating at 150% of capacity during the acceleration portion of the start sequence, there is typically protection to trip the motor during an event like this.

Again this is logic developed that most likely does not match your existing Emerson logic. But is sounds like you should be looking at several items to protect the unit. It would be interesting to hear how you proceed, I look forward to your reply.
Thank you all for your replies. I apologize for this long post but I have a lot to talk about.

First of all let me explain what we have and how our units are operated. They are GE Frame 7EA units originally installed in 1989 with Mark IV controls. In 2011 we replace the control system with Emerson Ovation. Yes, they basically copied the logic rung for rung but there are still may differences. The starting motor is a 4160 motor with Voith torque converter. The 4 units that we operate are simple cycle, with no hrsg. They are in crank mode and cranking 24/7 at about 450 rpms (to satisfy purge). The torque converter guide vanes are positioned to hold this speed. This is done so that when they are called to run they can be at full load with a fast start within 10 minutes. They are started remotely by our dispatchers but they can't shut them down so they call our operators to shut them down when the peak is over.

There are many times when they start a unit because the mw prices spike but before they hit line the mw price drops back down and they call to cancel the start. This usually happens above 60% speed.

Recently, they cancelled a start and the operator initiated a normal stop before the unit reached 60% speed. The unit without a doubt does go into a fired shutdown mode. The starting motor does not shut down and 20TU-1 remains energized. What happens is that the fuel decreases at a predetermined rate and the speed decreases until it gets to a point where the starting motor alone is holding speed at about 30% (Torque converter at max torque). At this point, the fuel stops decreasing because there is a minimum fuel requirement based on speed. No more decrease in speed, no more decrease in fuel. It then sits there at 30% speed until a max time during normal shutdown timer expires (8 minutes) and trips the firing system through 94T. During the last event, the firing did trip but we also got a high motor winding temperature alarm just before that. There is no relay protection for a thermal event that I know of. There is relay protection, but it is for the 4160 bus and would probably open the breaker after the starting motor was damaged.

I have tried to contact the Emerson design engineers about this sequencing problem but after 3 months they have not responded (that's another story). Before I make logic changes, I was trying to figure out why it operates this way and maybe there was a good reason. My concern was if I dropped the starting motor out during a normal shutdown below 60%, would the fuel have enough time to decrease to a flameout condition before the speed dropped too low and didn't provide enough air flow. My planned logic changes are to do the following:

When a normal shutdown is initiated before 60% speed, allow the normal firing decrease rate to drop turbine speed (starting motor and 20TU-1 still engaged at this point). The moment that firing reaches its minimum value based on speed, a logic bit is set and will drop out 20TU-1 and the starting motor. Now as speed drops, the fuel will be allowed to decrease further and should flame out soon. This lines the fuel and speed up at a point that matches a normal shutdown initiated above 60% speed with the starting motor out during the entire shutdown. I had not considered any cooldown period for the starting motor, because there has never been any in the GE logic or the Emerson logic. At 60% speed, the starting motor and 20TU-1 drop out at the same time. What is a typical cooldown time for the starting motor? Is that something that I should add? Like I stated earlier, I do not believe there is any relay protection for a starting motor thermal event. However, we have recently added motor winding temperature and motor amp indication to our control system. A long time ago we burned up a starting motor because the torque converter motor failed while at max torque and on crank. The operator didn't notice the high speed and the motor eventually burned up. We have created logic to alert the operator to high temperatures or high/prolonged amps but there is nothing that automatically shuts it down at this time. I will be looking into that though.

I would be very interested in how your starting motor sequencing works. Maybe ours is different because we are simple cycle?

Thank you all very much for reading my long post and discussing with me. If you need more information for further discussions, please let me know.
One correction, when the turbine speed was holding 30% and the starting motor didn't drop out, the torque converter was at Acceleration torque (about 150 amps) not max torque.

I am working on a reply to your post. Sounds like an interesting site, can't imagine what is costs to run your 4160 motors 24/7. Definitely a different market out where you are.
So Bill here are my suggestions and thoughts on your issue.

Some further clarification would be helpful in our discussion of your units. The sequencing I am most familiar with assumes your 7EA is driving an air cooled generator and uses a Voith converter that is equipped with a ratchet assembly for cooldown and shaft break-away. Sequencing I am also familiar with but is different is a 7EA driving a hydrogen cooled generator, using a Voith converter without ratchet assembly, configured for the converter to "slow roll" the unit for cooldown. These units do not have a ratchet assembly for shaft breakaway and cooldown. Can you confirm which of these matches your units configuration? My suggestions assume your units are fitted with the ratchet assembly.

My suggestion would be that anytime the unit is in a fired shutdown with L94X being true, that L20TU1 should be false. If the unit is in a shutdown there is no reason for the torque converter to be transmitting torque to the unit.

For the cranking motor I would make the same suggestion that if the unit is in a fired shutdown that the motor should not be running to provide torque. If you have never had logic for a motor cooldown timer that is your call to look at adding it. I do not know when GE implemented this change. I think it is a good idea but not required. Either of these changes should take care of the current issue of the unit not decelerating properly during a shutdown initiated prior to a complete sequence.

Some abstract thoughts of mine. If you have not already done it I think it would be very worthwhile to review your MKIV elementaries to determine if any errors were made when logic was converted specific to sequencing for the torque converter fill solenoid and for the start/stop of the cranking motor.

The units I am familiar with have 3 settings for the torque converter. Setting #1 is the minimum vane setting. This is usually a hard setting from Voith that usually results in a speed ~10-12% with the converter filled and the cranking motor running. Setting #2 is usually the purge torque setting, with a speed around ~20% and a motor load of ~115%. Last setting is acceleration of max torque setting resulting in a motor load of 150%. With most units I am familiar with they are using some sort of motor protection relay, usually GE Multilin or SEL to protect the motor. This might be something for you to look at as an upgrade to your units.
Our units have air cooled generators but we do not have a ratchet assembly for slow roll or break away. Our torque settings for vane positioning are; max torque-guide vanes fully open for break away (170 amps, short duration), minimum torque-guide vanes set at near minimum to achieve firing speed after purge complete, also used for continuous crank (50-60 amps), acceleration torque-guide vanes set for purge and to assist with acceleration up to 60% speed (140-150 amps). While on cooldown (50-80 rpm’s), the starting motor is shutdown and the torque converter alone maintains speed with guide vanes at max, 20TU-2 porting oil to the vanes, and 20TU-1 draining. We don’t see cooldown very much since the units are immediately placed back into crank after a run.

My first thought was to simply de-energize the starting motor and 20TU-1 once the unit enters a fired shutdown mode. I was just concerned that since I don’t know where the shutdown may take place, 20% speed or 59% speed, or during a warm up period, that I give the fuel enough time to integrate down to a level that allows normal flame out at about 20-25% speed . I don’t want too much fuel with not enough air flow. Maybe that should not be a concern and that is why I wanted to talk to the experts.

I would be interested in what a typical cooldown period should be for the starting motor but ours does not have one. I have looked closely at the Mark IV elementaries for starting means logic and it appears that Emerson is performing the same sequences.

As for motor protection, I will be in contact with our meter and relay protection folks about what protections the starting motor has.
Thanks for discussing with me, I look forward to hearing back from you.
Bill, s

orry for the long delay in answering, I do not have a good excuse for you.

So your machine is one of the few air cooled without ratchet, good to know.

For your issue I would suggest that de-energizing 20TU1 and or shutting down the starting motor will be fine at any speed less than 60% if the unit goes into a fired shutdown. Assuming your min FSR curve is tuned well the unit should have no problem maintaining flame with the loss of torque input from the cranking motor. I have done this with several frame 5 units and 7ea units with no issue.

As CSA said this used to be somewhat uncommon, but not so much anymore. At one site I assisted with replacement of some older jaw clutch system with newer SSS clutches so the customer did not have to wait for the unit to coast to zero speed. This was just so they could declare the unit available or start it back up before it reached zero speed.

As far as "typical" cooldown times for the cranking motor I have seen from 5-15 minutes. I prefer the lower number so you are not running the cranking motor with the converter "spoiled" for long periods of time. You may want to check with Voith just to make sure your model will not experience any negative effects if you decide to add a cooldown cycle for the motor.
Bill Futch,

I second everything MIKEVI said; it's not necessary to ramp fuel down before de-pressurizing ("spoiling") the torque converter. Just drop out 20TU-1, and when L94X picks up it will select MINFSR for FSRSD and the unit should coast down to "blowout" (L60RB), presuming the MINFSR Constants are good.

Adding a cranking motor cooldown (L62CD) is a good idea; most of the units I have worked on only had a two minute cooldown from "the factory." I also concur with Mike, it's probably not a great idea to run the cranking motor for longer than, say, five minutes while the torque converter is "spoiled" (de-pressurized--both 20TU-1 and 20TU-2 de-energized).

Once the unit reaches L14HM drop-out you should be able to start cranking the unit again (since that seems to be how you keep your machines at the ready for a short acceleration to FSNL after firing). Cranking usually occurs at about 60-70% of rated cranking motor nameplate current, so that's not hurting the cranking motor at all (just the "bill" for the electricity for running the cranking motor(s)). But, in some case--such as yours, it seem--that makes sense for the capacity payments for a fast run-up to load.

Let us know how you fare in resolving this problem.

I do have a few questions, please: Since you have air-cooled generators and use slowroll for cooldown, what size is the Aux. L.O. Pump Motor (HP)?

Oh, and, how do you keep your L.O. Header temperature above about 80-90 deg F? Do you cycle the cooling water pumps and/or fans, or have you replaced the actuator on VTR1-1? If you've replaced the VTR1-1 actuator, what kind of actuator are you using?

Thanks; sorry for all the questions.
Sorry I have been away for so long. I honestly don't think I have had more than 5 minutes in my office for the last two weeks. Thanks for the information.

To answer some of your questions;

Our lube oil pump motor is a 60 hp motor.

We have a closed cooling water system. While on crank, the cooling water pump is running all the time and the fans cycle. VTR1-1 is a self regulating valve controlling the lube oil header temperature. With the turbine turning at 450 rpm's, cooling will be required so there is no problem keeping the lube oil temperature above 90 degf.

I do have a desire to replace the self regulating actuator with a small motor actuator but I haven't found a good replacement solution. If you know of a replacement or upgrade option I would love to hear about it.