We have a 28MW gas turbine generator and using MarkVIe control system. (We have a load shedding system)
And there are four breakers. (GTG - 52G - 52L - 52B - 52M - Grid)
1) 52G : Generator circuit breaker (from GTG)
2) 52L : Line circuit breaker (from 52G)
3) 52M : Main circuit breaker (from grid)
4) 52B : Bus tie circuit breaker (from grid and 52L)

The generator has been operated in droop mode, and suddenly the mode changed to isochronous mode.
And load downed to 0MW and then 52G generator breaker tripped. TNKRIPR (Isochronous setpoint preset) was 100% and grid frequency was 100.3%.
I guess generator followed the grid frequency, because generator was operating synchronized with grid.
So, the generator speed did not changed and control system continously load downed. And finally ouput went to 0MW.

Finally I found some resonable reasons. (Maybe 52M or 52B breaker's status relay malfunction. Actually breakers did not opened)
Logically, if 52M or 52B breaker open, generator goes to isochronous mode.

If the generator trip, it cause a lot of electric charges loss.
So we need to figure out making up for logical weak point or hardware sequence.

Is there any good idea ?

Thank you.
(I attatcted logic and sequence drawing)

 

Can you share a kind of chronology /historical trip log event...

Looks like something happened to get that Breakers open ...did you investigate on Line /bus tie/generator protection relays trip /sequence of event

What Load shedding system is supposed to do in such case..Is it like a PMS power management system..

You shoudl be alerted before that breakers tripped/opened before to get that unit go to isochronous MODE..

As you mentionned there can be a malfunction of Auxiliary contacts of these breakers..hum ..maybe...

What kind of GCP (AVR+PROTECTION RELAY(s))are installed..

We can try to provide an assistance for getingt the best solution for a smooth operation of the plant if you can provide us such informations...

Also what you descibed is like Line /bus was dead during a time and unit controls system/PMS load shedding should be working altogether to get the situation better than described..

Again whithout knowing trip log event /sequence of event it would be difficult to state more on that case

Also a SLD would be welcome to be attached on that thread...

I will review the document taht you share and see if controls philosophy can be modified for a smooth operations of that plant
Any time!

James

 

The plant load frequency controlled by load shedding system. When frequency decreases, load shedding system open the load breakers to maintain the frquency.

AVR : EX2100e (BHGE), Protection Relay : G60 (GE) are installed.

The plant load frequency was holding at 100.3% so laod shedding system didn't operate.
(52M and 52B breakers actually didn't operate. So, I guessed malfuction of auxiliary contacts. But, I coudn't verify the log. Both of contact inputs are not alarmed setting.)

 

Attatched alarms

 

Looks like there was kind of FSR limited due to sensor failure..
And MWA transducer signal trouble.. fault
The thing is that 52B 52M status /positions are not shown

 

jackson,

You are most likely correct with your presumption that somehow the turbine control system believed one or more of the tie breakers was opened which is what caused the turbine control to switch to Isoch from Droop. Running in Droop under the conditions you cited (TNH at 100.3%--which is an "unusual" number for a couple of reasons) the turbine control responded precisely as it should have--reducing the load on the Isoch unit to try to reduce the frequency to 100.0% (or, based on typical default numbers to something between 99.83% and 100.17% TNH/frequency).

If you can't find out why the turbine control believed one or more of the tie breakers opened then you are going to have to go through each of the circuits for each of the tie breaker status contacts and make certain all of the terminations are tight (NOT gorilla tight! just tight) and that all wire crimp-on terminals are tight, and perhaps you should have someone look over the breaker status contacts of each of the breakers.

As I was writing this, I thought, "What if the tie breaker status contacts ARE NOT connected to the turbine control system via discrete contact inputs?" In other words, what if some genius is sending the signals to the turbine control system via MODBUS or GSM or via a DCS or the load shedding system--instead of connecting the breaker status contacts from each of the breakers DIRECTLY to the turbine control system?

For me, without understanding how the breaker status information is transmitted/sent to the turbine control system means we can't provide any more assistance. Full stop. Period. Not understanding the load shedding system is also reason for not being able to provide any further assistance. These two things are just very "variable" and without understanding them completely (seeing drawings and configuration and programming) we just can't make any further suggestions (guesses, really).

Make sure those breaker status signals are "solid" and not subject to any possible interruption. A lot of times, plant designers (and load shedding system programmers....!) make a lot of not-so-good choices, which introduces one--or more--single points of failure.

And, a loose termination or a loose wire crimp or a loose breaker status contact assembly or bad secondary contacts on a breaker can be the difference, too. Secondary disconnects require proper maintenance and greasing. Dry contacts, or too much grease, can be a problem over time. Easier to check ALL of the "simple" stuff first, and eliminate that as a possible cause--than to go through a lot of other trouble (time and MONEY!) just to find the problem is still there, and it was a really simple problem to solve. (I have seen some very creative report writing just to hide such a find--and more than several times in more than 30-year career and experience. In fact, one of my first assignments as a young field engineer was to investigate a field ground error which tripped a 300 MW steam turbine-generator off line, and the plant had NO alarm lists to indicate that was the cause of the trip (they didn't have ANY alarm lists at--that they were willing to show me, anyway.... read on). After a day-and-a-half of digging and searching and scratching my head, a nice, older board operator came over to me and said, "I see you're new. We always blame "nuisance" trips on generator field ground. Just write your report to say you checked all of the circuits and wires and couldn't find anything, and that's all you need to do." So, the likely reason they wouldn't share any alarm lists with me (and they had a VERY good SOE (Sequence-of-Event) recorder, too!) was because someone touched something they shouldn't have or didn't respond to another alarm they should have, and the unit tripped. In fact, they restarted the unit the next day--before I arrived and could perform any testing on the generator field.... So, my point is: They spent a LOT of money to get an OEM Field Engineer to come out to "investigate" and write a report, to cover up some other niggling little thing which caused a very large natural gas-fired boiler and steam turbine-generator to trip off line--and which takes about 24 hours to get back on line. Report written, accepted, filed, and money received for the Company I was working for. Everyone was happy!)

Check the little stuff. Even if it seems to be a nuisance.

Hope this helps!