Good day to the control.com community,

I am asking for some education on MKVI generator synchronization. We operate 2 GE 7EA MKVI TMR EX2000 gas turbines, circa 2005, baseload operations. We came offline this past weekend for an offline wash and replacement of a failed VSVO card. We encountered trouble trying to sync after our short outage. We have had problems similar to this twice in the past on both units, but never fully figured out the cause.

Typically when we reach FSNL we engage autosync and the unit will sync within 1 revolution of the sync scope, voltage and speed are steady. We have never used manual sync. We have no aux relays inline with the generator close coil, we let the MKVI control of K25P, K25A and K25 govern sync.

This past weekend the unit reached FSNL and was ready to sync, there were no diagnostic alarms on the MKVI. The operator selected auto sync, but the unit did not close the breaker. We started a trend and were able to verify that the MKVI was trying to close the breaker by energizing the K25 relay. Investigation showed that we had failed to properly rack in the breaker, the unit was kept at FSNL and the breaker condition was resolved. The VTUR card set a diagnostic for “breaker slower than adjustment limit allows”, which I expected.

Following this, several attempts to sync were unsuccessful, several different diagnostics were set and sync lockout alarms were generated. Alarm from the VTUR, GEN_SYNC_Lo would set, causing the “sync lockout”. As well, 2 to 3 VPRO cores would set diagnostic fault 90, indicating trouble with the K25A relay. Reading the Speedtronic system guide GEH-6421 chapter 7, Applications, it indicates that GEN_SYNC_LO will set if the K25 contacts are closed when they should not be, OR the K25A is not picked up before K25. Although the alarm in the MKVI says this alarm is generated by “K25 relay contacts stuck closed or welded”.

This same unit had this same problem almost 2 years back. While a GE TA was en route to diagnose the problem he suggested replacing the TTUR card. The unit was shut down, card replaced and the unit restarted and synced without any trouble, as I am told. Whether the card fixed it or it was just coincidence I can’t say.

This weekend, due to the faults stored, we replaced all 3 relays on the TTUR card. The next sync attempt was successful, although a single VPRO core stored the diagnostic fault 90. I still believe that the only reason the unit synced was that 2-3 VPROs did not set a diagnostic as they did on the previous attempts, and that there is still something going on with the timing of the K25 relay group. Our second unit has set this same diagnostic at times when syncing but again usually only 1 core. I did find that quite a few signals from the TTUR that indicate the command and feedback status of the K25 K25A and K25P relays are “not used” so I can’t trend them. I will fix this at our next outage.

I am curious if CSA or another group member might have been down this road before, the archive did not provide any of the in depth answers I want. I did compare card settings to some .m6b files I have from some other 7EAs and the voltage and slip settings are the same, 10 deg slip, etc.

I thank any and all in advance for their time and comments. I would like to make sure I fully understand this system, so I can be better prepared for my next battle.

 

MIKEVI,

Through all the troubleshooting you did this past weekend, you didn't mention any Master Resets (though I feel confident that several were performed; most operators click on Master Reset even before they click on Alarm Silence!). AutoSynch Lockouts and many synchronizing-related diagnostics are latched in with a Master Reset logic. Enough about that.

I have, in the last few days, been told that the Mark VI synchronizing circuit is polarity sensitive. Yes, that's right; a polarity sensitive synchronizing circuit though that's not made clear in the documentation *EXCEPT* with the nomenclature on the diagrams of the TTUR card and the synch circuitry/terminations of the card. Others may have known this, but it was news to me and to the people I learned of it from. If you look at the figure 'TTUR Control and I/O and VTUR board, TMR' in the Mark VI System Guide, Vol. II, it shows that P125Gen is connected to TTUR terminal 03 and N125Gen is connected to TTUR terminal 08. There have been reports of lots of similar problems: immediate closure or lots of diagnostics and lockouts and failures to synch. My SWAG here is that the circuitry that's monitoring contact closures is polarity sensitive, and if it sees incorrect polarity on a 'sweep' of the synch 'scope' when one of the contacts doesn't permit closure (speed matching or voltage matching not within limits, for example; or, for sites with additional synch check relays in series which haven't had their synch permissives satisfied; etc.) that the monitor logic prevents breaker closure until a Master Reset is issued "in time" for a successful synch attempt.

I'm also aware that some installations don't seem to have any wire terminated on TTUR terminal 08, and I'm wondering if that's also part of the problem but I don't have access to a unit to do any testing at this time.

Some of the breakers used on recent Frame 7EA installations have experienced problems with the racking mechanism, which can result in the secondary disconnects not properly making up, which I surmise can lead to some issues with the polarity sensitive synch monitor circuitry of the TTUR card. I've been at a couple of sites and heard of many more that just lowered the breaker and then racked it back up (in) and the units successfully synchronized on the next and several successive attempts. Again, this is just a presumption, but if this polarity sensitive synch circuit is indeed true, it could explain a lot about many synch problems. To my knowledge, none of the previous Speedtronic turbine control systems with auto synch functions in the turbine control panel were not polarity sensitive, so this is something of a departure in my opinion.

I have been aware of lots of complaints about slow breaker closure alarms and excessively long synchronization times with Mark VIs for some time. The breakers always seem to close sooner or later, and for most sites that's enough to stop all troubleshooting. So, I don't know if this is "the" issue or just one of several (my feeling is there is more than one issue, particularly with the 'breaker slower than expected' diag alarms). But it's worth investigating; we look forward to hearing what you find if you check the polarities of the voltages on TTUR terminals 03 and 07/08. I'd also be very curious to know if there is(are) any wire(s) terminated on TTUR terminal 08, or even just a jumper from 08 to 07.

I don't know if this will help; I don't really know if this is what you're looking for because I'm not exactly clear what you're unclear about. And, again, most of the above was just recently learned and I haven't had any opportunity to make any personal observations or tests. It might be difficult to measure the polarity of the voltages at the TTUR without putting the generator breaker in the TEST/DISConnect position and doing some testing. Also, be careful because the generator breaker closure circuit fuses are generally 25A or larger in some cases.

I don't think the problem is with the TTUR or even the relays themselves. I think it's something much more "subtle" than that, like a polarity sensitive synch circuit. But, I've been wrong many times in the past, and odds are very high I'll be wrong many times in the future!

 

Thank you CSA for your time and thoughts. To answer some of your questions and comments:

Yes, Master resets were necessary, thankfully our operators are good about asking before this is done. In our case it was necessary to perform a master reset to clear the GEN_SYNC_LO alarm for auto sync lockout, so that we could attempt another auto sync breaker closure.

I have been reading the “applications” section of the MKVI system guide Volume #1 rev H over and over in the past few days and each time I read it I discover something I did not see the time before. I will be examining the wiring of our units and will report back on my findings.

In closing to answer, “I don't know if this will help; I don't really know if this is what you're looking for because I'm not exactly clear what you're unclear about.” I had hopes that someone might respond with “oh yeah we have seen this many times and here is what you need to do to fix it.” I didn’t think this would happen, but it was worth a try.

The second possibility was that someone such as yourself would really look at the system operation/manuals, give me their full attention to the subject, and possibly learn something in the process. I think I definitely got this.

My next course of action will be to: review wiring and terminations, set up a high speed trend with the necessary signals, and be ready for the next attempt.

If anyone in the group has any comments, or may have had some of the same intermittent problems without a full resolution, I would love to hear about them. Otherwise I will return with more answers to add to the archive that can hopefully help someone else someday.

 

Thanks for the feedback. And it's really good to hear that your operators ask before performing a Master Reset! Most don't and lots of valuable troubleshooting information can be lost.

There's even been speculation that there are "unannunciated" diagnostics that can be set (latched) after a successful synch attempt which took a long time to be successful that might be solved with a Master Reset just before enabling auto synch. But, that's just speculation. GE is pretty good about annunciating diagnostic alarms; wouldn't you agree? (Even if they don't make a lot of sense!)

You didn't mention anything about the breaker racking mechanism, but you did say that you initially found your breaker not correctly racked-in. I think that when the Mark VI is trying to close the breaker and it doesn't get a signal the breaker has closed and this happens on more than one attempt that the Mark VI seems to not like this very much.

Again, I think this is not a unique problem, but it's one that people have learned to live with or become accustomed to.

We'll be really interested to hear what you find after you enable some of the diagnostic signals and get some data from a high speed trend.

There really is some good information in the GE manuals, but many times it doesn't jump out at you in the first reading.

Let us know what you find; I'm especially curious about the synch circuit voltage polarity and if the same (correct or incorrect) on both the units at your site.

 

As our systems get smarter, and our techs learn to trend data. Many forget to check the simple stuff.

Amazingly, you pay a huge amount of dollars for a generator breaker and receive 2 dollar fuse holders.

Usually, a generator breaker has 2 trip coils and sometimes two close coils. Since this is not a breaker open problem, I will ask "Has anyone ensured that the close signal is actually reaching the breaker?"

In my experience, intermidate problems are usually a loose wire, terminal block problem etc while everyone blames the control system.

 

Dear CSA,
It has taken me a while, but I wanted to report back on the status of this issue for your thoughts at least. We have had the chance to synchronize this unit 3 times since the last problem, and have not encountered any issues. I have set up a 40ms trend for each sync event, but all have been successful, so at least I have saved trends of what a good sync looks like.

Unfortunately I have not discovered any smoking gun at this point. I did verify that our closing coil polarity appears correct from the MKVI perspective, the wires at terminals #3 and #8 are correct. I have not yet verified all wires to the secondary disconnects at the breaker(but I will!!)

Wiring is also the same on both units, wires are landed at terminals #3, #7 and #8 on the TTUR board. I have spoken to several GE field engineers but unfortunately none of them had any educated thoughts for this problem. So to quote you at this point (I think that when the Mark VI is trying to close the breaker and it doesn't get a signal the breaker has closed and this happens on more than one attempt that the Mark VI seems to not like this very much). And for now I will have to live with (The breakers always seem to close sooner or later, and for most sites that's enough to stop all troubleshooting).

I will not stop all troubleshooting, but will move on to other issues until I can work on this one again, hopefully with more information/ammunition. Thank you CSA for your time and thoughts, I cannot fathom the time and effort you spend in this group, but myself and others benefit from it for sure.

 

MIKEVI,

Thanks for the status report. I would venture that there is an intermittent connection somewhere in the secondary disconnect circuit that may or may not be immediately apparent. I believe you said that after racking the breaker out then back in the unit successfully synchronized; that's the basis for my position.

As for the time spent here, I learn as much as I contribute. It's been a great experience; I only wish there were more contributors. I've been holding back from immediately responding to some threads hoping others would provide the benefit of their experience and knowledge. But, on the whole, the community seems to be growing nicely, if not as "evenly" as hoped. You've been one of the more consistent contributors and thank you for that!

Thanks for the feedback, again. And, should you find anything obvious in your investigations please let us know!

(OOPS; I need to control the use of exclamation points lest I drift to the curmudgeonly side.)

 

I see this thread is quite old but I will take my chances.

I have a similar problem right now and no clue where it comes from.

As a background, we made hot gas path inspection on 5 Hitachi H25 units (similar to Frame5, Siemens generators and excitation system, MarkVI control panel) after 4 years of operation. No such problem was encountered in the past. After the inspections, I had the auto-synch problem in 3 units out of 5.

One funny behavior is that the first time after inspection it takes a very long time to match the generator and the bus voltages. Once we got to FSNL, excitation kicks in, generator voltage goes to 90% of nominal value in some seconds and then, when we try to synchronize, it takes minutes to cover the remaining 10%. In the beginning I thought this might be the problem. But I have 2 units that synchronized under the same conditions.

I did a manual synchronization to clear the lockout and tried again. This time the voltage difference was very small, I had permissive but... surprise... again failure. Manually synched again, left the unit operating for few weeks and tried again without any problem. Since then, I did the automatic synch 3 more times without any problem.

Under these conditions I do suspect the problem is the K25A relay on the TTUR board. I think the long synch time in the beginning is not the root cause but only brings to surface the real one, an aged K25A relay.

Anyway, right now I am not sure of anything. So, if you have some new insights on the subject, please update.

Thanks.