Gas turbine protection - motoring

[ters]

This LM6000 project, with a gearbox (50 Hz). Suppose the following scenario:

The turbine shuts down for some reason other than electrical fault on the generator side. The GT control system will then trip the generator protection and the generator breaker. However, generator breaker fails to open. The generator remains on the grid since breaker failure zone trip scheme is not implemented. Generator happily continues to run as a an induction motor while the turbine is shut down. This condition persist until other breakers contributing to the same bus are open manually.

What are mechanical consequences in this case - may any major turbine or gearbox damage occur? Thank you.

 

[stratford]

Hi ters. The turbine mechanical parts will start overheating. Also the generator will draw big amount of reactive power. This is the reason that if the generator circut breaker fails then thne Circuit breaker function shoulf trip the overhead breaker, on the grid.
Also there should be thrust couplings that should brekak their bolts when such abnormalities occur and let the turbine shut down.

 

[ters]

Thank you very much statford. Sounds that there would be no catastrofic mechanical failures, but looks like I still need to insist on breaker fail protection anyway.

 

[GTstartup]

Hi Ters,

It's not correct that as stratford says "the turbine mechanical parts will start overheating" or that there should be "thrust couplings" that should "break their bolts"
(Never heard of such a thing)

I have seen this happen a few times and the turbine and compressor will not be damaged.

However your statement that:

"Generator happily continues to run as a an induction motor while the turbine is shut down. This condition persist until other breakers contributing to the same bus are open manually."

is wrong.

The generator will not "happily" continue to run as an induction motor. If the breaker fails to open and the excitation is off, the generator rotor will overheat and will be soon damaged. If the excitation stays on, then it's not so critical and is mainly just a matter of manually opening the breaker and shutting the GT down.

 

[slavag]

Hi.
I'm agree with GTstartup.
W/O excitation your rotor will damage, generator will recieve reactive power from grid.
What do you think about "some idea", in this case send to field breaker command close or add logic "in case of turbine trip send trip to field CB only after main CB open position ".
But Ters,in all cases you must add BFP in your system,
additional logics are only game.

 

[ters]

Thank you very much gents. This forum is really great. Re excitation, I suspect that control system logic is such that if the turbine shuts down, it will shut down the excitation too.

Seems that I have to convince "higher levels" to spend some money on the breaker failure - it is actualy available as a part of generator protection but the problem is that both generator breaker or further upstream breaker(s) are something like 500 m away. Long & thick cable are needed or other means of transferring the trip...

 

[davidbeach]

Long and thick cable??? Two strands of fiber optic is all that is needed for the transfer trip.

 

[ters]

Probably so. But by the time we provide end terminal devices for the FO links (relays cannot accept FO cable directly) it same cases it will still be cheaper to pull 2c cable - not all breakers are that far as I stated above, same are as "close" as 200 m.

 

[GTstartup]

Breaker failure is not normally provided for this situation. It's not the same as clearing a fault and the fault clearing breaker does not open.

The generator breaker should have 2 trip coils, preferably one AC and one DC. The field breaker should be hardwire interlocked so that it is not possible for the field breaker to open if the generator breaker is closed.

Normal breaker failure philosophy is not suitable because in actual fact an electrical fault has not occurred and as I stated, there is no damage likely to a GT in this scenario.

Steam turbine is a different story altogether.

 

[rmw]

While a large proportion of GT driven generators that operate as synchronous condensers have clutches that declutch the turbine when the generator is being motored in the condensing mode; some do not and the power turbine is motored by the generator.

When the turbine tripped, I assume the lube oil system stay energized because you aren't writing about trashing the generator bearings (unless you have jacking oil and it came on and saved you).

Check your generator O&M manual to see if it provides for your generator being operated as a Sync Cond as many are purchased for that purpose and some combustion turbine generators get more hours as Sync Condensers than power generators.

Ask the grid operator how much they are going to pay you for eating all those vars when this occurred.

rmw

 

[slavag]

Hi.
1. For my opinion you must use BFP in all cases, possible use several criterions for BFP trip:
1.1 For trip with high level of current.
1.2 For trip with low level of current ( turbine trip, reverse power, bucholz of trafo, etc).
If turbine control or AVR system sended trip , protection system must provide this trip and open main CB or upstream CB. If you don't want BFP trip in some cases, please separate signals from turbine control system and send to protection several signals or send them to shut down turbine only.
2. Protection sended trip to field breaker and main CB simultanies, that means actually we don't have any interlocks between field breaker and main CB.
3. Rmw are right, possible work in absorbtion condition (if generator rotor build for this prupose), but it decide only operator of power plant with grid operator and not some logic in protection system.
4.GTstartup, of course CB have two trip coils ( Ters we hope) and chance for BFP is around zero today. But cost of BFP logic (or relay ) and cable (or FO)is not issue.
Regards.
Slava

 

[ters]

Thank you all again for very constructive discussion. The way how the turbine protection seems to operate is:

Most turbine trip conditions or abnormalities will trigger a lockout relay 86. Then, 86 will trip the breaker and at the same time another contract from 86 is used an input to the generator protection to initiate the breaker fail function. Upon a time delay, another contact assigned to “breaker fail" will close.

So it all seems to be there already, it is just the matter if and how we use it externally.

 

[GTstartup]

Please let me reiterate;

There is no need to declutch the turbine, because the turbine will not be damaged in reverse power mode. Breaker failure is not required because no electrical fault has ocurred. Also 60/50Hz LM600 has (normally) no clutch and therefore cannot be declutched, the gearbox is only to match the 60Hz GT to 50Hz generator

In case you insist on a breaker failure scheme, I offer the following suggestion.

When GT trips with generator breaker closed, send operator to upstream breaker to manually open breaker.

Pray that all DC systems work because no AC is available.

Cost:

One pat on back for operator for opening correct breaker
Mvar import = $0.00 for because excitation is still on.

A couple of points to back up my opinion:

Operating a synchronous generator in reverse power mode with excitation on will not harm the generator

Operating a gas turbine in reverse power will not harm the turbine (it couldn't care less)

Cost in MVARS is not relevant because normally excitation is on.

GTs are normally run in reverse power (if driven by LCI)for startup and during shutdown to about -5MW.

Bottom line is, why add protection when there is nothing to protect?

 

[GTstartup]

Small correction to my earlier post

I stated:

"Operating a gas turbine in reverse power will not harm the turbine (it couldn't care less)"

There is a concern that motoring a GT can cause forced cooling and maybe blade rubbing could occur. This is not, however, reason for immediate action but certainly action as soon as practicable with operator intervention

 

[ters]

Thanks again GTstartup for very illustrative description of BF protection in the form of a lazy operator walking down to the switchgear .

Well, this is in the ME, where the operator will be sitting in a nice air-conditioned control room some 500 m or so from the next upstream breaker. Therefore, I suspect that he may not be very well motivated to stretch the legs for that much in July, when the outside temperature may be about 50 C or even more... So I would rather rely on something else than him .

Also, the control system appears to be so called "GE standard packge", which seems to use the following logic: mechanical trips will trigger a lockout relay (86) which will (in addition to what I listed above) also shut down the excitation (AVR).

This is not to say that the AVR logic cannot be easily changed, but as it is designed now, it seems that the GE intent was to clear any reverse power condition almost instantly by the generator protection which is equipped and will be set to provide a BF contact. It is just the matter if we use it or not. Given the circumstance, looks like we should.

Related to this, why the gen. breaker may fail to open? That usually does not happen, but on this project there is an additional factor which may icrease chances for that to occur – the medium voltage switchgear is not fed by DC from a battery but rather just from an AC/DC power supply, basically a rectifiers!

OK, the AC comes from an UPS, so there is some kind of battery in an indirect form, but still I have never seen this and if that fancy converter fails, the breaker will fail to open as well.

 

[ters]

Just to clarify, in some, but not all cases, the operator will be able to open the corresponding upstream breaker from his room. In that case my am argument about his motivation does not apply. In other cases, however, he may either have to walk or phone another operator somewhere to do it. The medium/high voltage scheme is rather complex and unusual (and is still being developed).

 

[GTstartup]

Hi Ters,

Glad you liked my toungue-in-cheek operator based BF scheme.

One small point about your latest post.

It may be GE's standard package to trip the excitaton breaker with an 86 but the AVR trip circuit should be wired through an open contact ("B" contact) of the generator breaker. Under no circumstances should the field breaker be open and the generator breaker closed.

Similarly, there should be a interlock so that the generator breaker should not close if the field breaker is open

Ine event you DO install breaker failure for this event. please be aware that probably tripping an upstream breaker automatilcally will result in the GT losing it's auxiliary power and relying on DC/Emergency power for shutdown. If you leave it as an operator function you have at least the opportunity to mechanically trip the generator breaker and/or to do some tie bus switching to keep auxiliaries on.

Nevertheless, good luck with whatever scheme you decide.

 

[ters]

OK, GTstartup, thank you, will take your point about AVR logic seriously. However, I have to deal with controls guys whose logic is: don’t even think about touching my panel... it is standard, it works everywhere, so why it shouldn't work here

The possibility of loosing GT auxiliary power is not that big. The MCC is fed from a switchgear with double feed and an automatic transfer, whereas the second feed may have nothing to do with the upstream breaker (again depending on the case - there are 6 units) so that should be satisfactory.

 

[slavag]

Hi Ters and GTstartup.
Very intresting discussion.
GTstartup thanks for good explanation. I'm many learn from it.
Several points.
1. Send operator open HV breaker is not so easy.
Please take in account safety requrements.
You must two person for work in HV area (of course is depend on rules of country). In many case is not possible open HV breaker locally ( also safety) and you must write some procedure. And usally operator of power plant not electrical person.
2. I many think ( becouse I have now more or less same project) about other solution.
2.1 Disconnect turbine trip from 86 LOR.
2.2 Connect this trip to protective relay.
2.2.1. Connect status of main CB to prtective relay
2.3 Programme separate trip output to field breaker.
2.3.1 In case of electrical trip send trip to main CB and AVR simultanies.
2.3.2. In case of turbine trip only after staus " Main CB open" send trip to AVR.
2.4 In case of electrical fault and BF situation send BFP trip to upsteram CB after some delay (about 200ms).
2.4.1. In case of turbine trip and BF situation send alarm signal only to operator and after about 30min send BFP trip to upstream CB.
Need think about reset of this delay ( maybe status of upstream CB ?)and about other criterions.
O.K. it's only idea.
Ters if have "good" relay for generator protection it's not problem build this logic.
Wait for your opinion.
Regards.
Slava

 

[ters]

Thank you very much Slava for your detailed elaboration of protection scheme. That all seems to be possible, the generator protection is Beckwith M-3425A which appear to be “good” in terms of flexibility, although I have never worked with that particular Beckwith model before, still have to learn what it can do.

As for AVR, I still don’t know what make and model it is.

My position is awkward, I’m electrical and trying to put the whole MV system to work together, while in case of so called “standard GT package”, the AVR and generator protection are not part of electrical panels, but rather part of the GT control panel with PLC and other stuff, what makes my life more difficult - it is under the jurisdiction of control engineers which don’t appear to be easy to work with... Or, maybe I’m just an old complainer . I’m not sure they will allow digging that much into their panel to rewire turbine trip from 86. We might get away changing only AVR logic if needed, and implementing BF. I do not have AVR operational diagrams yet, controls guys say it always trips when 86 trips, but they are known to be wrong form before (as everyone else is . The CB status already goes to Beckwith, which is needed for BF.

The means of transferring the trip from generator protection to generator breaker is also unusual:

Generators are away from breakers so 200-500m. There is MV cable between. The cable has its own pilot differential protection with one relay at each end (lets call them A and B) and the fiber optic cable between. It was decided to use the following scheme: the trip output contact of Beckwith relay will not trip gen. breaker directly but will rather be wired as an input to A diff relay. Then the A relay will transfer the trip signal to diff relay B via FO, and finally B will trip the breaker. Elegant, but the trip now relies on three magic boxes instead just one.

Since our upstream breakers are not far away from the relay B, then A-FO-B (happens-to-be-handy) communication link could be used for BF as well, although I’m of an opinion that using hard wires for BF would be more feasible.