Tie Breaker Closing Make Before Break 4

[rockman7892]

I have a situation at my facility where we have two plants and their respective 4.16kV power systems operating off of two different utility transformers. On the main switchgear lineup for each distribution system there is a tie breaker to conntect one power system to the other in the event that we loose one of our utilty transformers this allows us to run both plants at reduced load. The tie breaker system is designed only to have on transformer operating when the tie breakers are closed, or in other words the system is not intended to run with the utility transformer paralleled. The 2 main breakers and tie-breakers are vaccum breakers that have a kirk key interlock system in place to ensure that the ties and both mains are never all closed at once putting transformers in parallel.

I am aware of all the hazards and percautions involved with putting transforers in parallel (tap changes, circulating currents, increased fault capacity etc, phase seqencing etc...) Most of these percautions with the exception of phase synchronizing are for transformers paralled for an extended duration. I am curious what options I may have to set these tie-breakers up in a "make before break" transfer scheme in order to keep the power supply seemless to the plant that is already running. If I have a phase synchronizing relay on both voltage sources is it possible to momentarily parallel both sources before interrupting one of the mains only as quickly as it would take for the breakers to open and close from a relay operation?

One of the issues we have is that if we need to shut down one of the utility transformers in order to switch the plant load to the other transformer we have to shut the entire plant down in order to close the tie breakers and then bring the plant back up again. This interrupts production and is generally not acceptable. I am looking for what options may exist for making this transfer seemless by closing both tie breaker first before opening the main and briefly paralleling the two transformers under a relay controlled operation.

Is this even an option without getting into a larger order of other issues with the transformers themsleves?

 

[Zogzog]

Some schemes like this use "electronic kirk keys" which allows a closed transition for a few seconds.

 

[davidbeach]

Not a trivial task at all. First stop has to be the utility, and find out what their requirements are for the transformers to be paralleled. You may then give up on the whole idea, or it may be a reasonable project to attempt.

 

[waross]

If I have a phase synchronizing relay on both voltage sources

This is used when one of the sources may be a variable frequency such as a generator.
On a utility/utility the phase synchronizing relay will prove that incoming phases have not been switched during an outage.
Your two most important issues may be back feeding into a utility circuit and fault current levels.
Due to possible liability issues should an unauthorized back feed cause damage, injury or death, it is STRONGLY recommended that you follow David's advice and contact your supply utility.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[rbulsara]

Momentary closed transition transfer schemes using breakers is very common, although it takes appropriate controls and engineering, similar to those found in closed transition ATSs.

I have designed it many times, but on new or replacement systems. Retrofitting existing could be a challenge but can be done.

It is simple in theory but takes experience and care in practice to implement. Basically a phase check relay, transfer scheme with relays, timers and/or PLCs plus an independent watchdog timer circuit to separate the two sources, if the transition is not completed within a prescribed time, not exceeding 0.5 second. Other checks and balances have to be in place to avoid a mishap based on your system. Where it gets a little complex is allowing manual operation, when automation fails or in a bypass mode. Locking out after a fault, etc. Some hardwired interlocks still needs to be in place all the time.

You would need approval of the utility copay, if at least one of the sources is them.

Rafiq Bulsara

http://www.srengineersct.com

 

[davidbeach]

0.5 seconds is 0.4 seconds too long. Anything over 100ms is considered a sustained parallel connection and requires a full protection package.

 

[rbulsara]

davidbeach:

Yes, 80-100ms is the normal operating time. 0.5 second is the maximum permitted by utilities for the watchdog timer. Typically set at 0.3 sec. This is true for "standard" closed transition ATSs as well. Utility companies refers to them as Momentary Paralleling on their interconnection requirements.

Rafiq Bulsara

http://www.srengineersct.com

 

[davidbeach]

Our requirements are normal operation in less than 100ms, breaker failure tripping initiated at 10 cycles. IEEE 1547 divides the time spectrum into two at 100ms; less and it can't do much harm to the rest of the system (but can destroy the customer's equipment if done wrong), longer than 100ms and it is treated as paralleled indefinitely. We don't recognize 500ms as meaningful for anything in this regard.

 

[rbulsara]

david:
That is not surprising. In fact that would make your utility company more lenient, not requiring a watchdog timer, which is a backup and a last ditch effort to separate the two sources if the normal transfer mechanism get stuck. This is over and above normal operation required to be completed in 100ms.

Each utility co. has own rules. What I stated is the published policy of more than one local utility companies in Northeast, in fact available on their websites as part of the application for paralleling interconnections.

Rafiq Bulsara

http://www.srengineersct.com

 

[davidbeach]

I don't know about lenient. The breaker failure/stuck breaker tripping is mandatory and can't rely on trying to retrip the breaker that should have opened in the first place.

 

[rbulsara]

david:
It is not re-tripping the same device. It is an entirely independent circuit to trip a "different" device to break the circuit, IF the primary arrangement FAILS. This is a backup requirement on top of the a "normal" closed transition throw-over controls.

The different device typically is upstream breaker feeding an ATS or one of the other two breakers in a main-tie-main setup.

Plus I do not make the rules, nor do I work for a utility company. It is a requirement in our region, nonetheless. Most ATS mfr offer it as an option, the good ones offer it as a standard feature.

Rafiq Bulsara

http://www.srengineersct.com

 

[davidbeach]

OK, watchdog timer sounds to me like something in a PLC. As far as I'm concerned, PLCs don't do protection, relays do protection. But in any case, the secret to success is to get the buy-off of the serving utility.

 

[dagath]

Devices that do just this are readily available, making switchover times in ~30ms quite possible. A lot depends on the breaker switching times. High speed bus transfer it's called. It is quite common in power plants to switch the aux. power supply from an external transformer to the generator once it is online. Or in petrochemical industries to switch without interruption from one feeder to the other.

Especially in the last case, more interesting is switching over from supply A to B when A fails, provided you have separate feeders upstream. If you have enough spinning reserves in the form of motors to keep the voltage up for a very short time you can just synchronize on the fly onto the other feeder and resume operation. Depending on the inrush current you might have to load shed some, but it's better as going black.

The one I used once was a SUE3000 by ABB. And a REC316 once, but those are end-of-life by now. Not sure about other manufacturers. I've seen some hand build solutions in old power plants as well. From my experience setting the things up requires so much tweaking and timing that you're better off with something tried and tested.

 

[rockman7892]

Thanks for the great information guys. I agree that as a first step I need to discuss this with the utility to see what their requirments are. I am only in the brainstorming stages at this point.

One thing that I am concerned about with the momentary paralleling of the two transformers is the increased fault capacity. If there was a fault at a downstream location in one of the plants during the brief paralleling period this would subject our equipment to fault currents exceeding their ratings. I would think this is something to be genuinly concerned about? It will also increase Arc Flash levels but when looking at these two I would think equipment failure is the greater of the two evils.

After seeing some of the responses I did a little more research and looked into other "High Speed Bus Transfers". It looks like as an option to a closed transition transfer there are an "open transition fast transfer" and an "open transition in-phase transfer". Does anyone have any experience with these two methods in particular to an application where we are switching a whole plant as I desire? Would these open transfer schemes be more desirable over the closed transfer scheme? Is one scheme economically better as opposed to the other?

I appreciate everyone comments and input.

 

[dpc]

Fast bus transfer is a high-speed open transition typically used when there are large induction motors involved. I have experience with it at large fossil power plants where auxiliary loads are transferred between transformers after synchronizing.

This is a tricky scheme to implement and involves a fair amount of complex interlocking and special high-speed breaker status contacts. It is generally designed in from the ground up. The Beckwith Electric website used to have some good technical papers on this subject.



David Castor

http://www.cvoes.com

 

[rcw retired EE]

We've used the fast transfer schemes on several power plants. The first few were not successful because the motor loads were too small and drove high speed pumps with low inertia. The motors couldn't hold the voltage long enough for successful transfers.

The most recent one on a 7.2 kV Main-Tie-Main arrangement works well in one direction and OK in the other. Going A-B it is imperceptible - no lights blink and the UPS doesn't even alarm. B-A always gets a momentary alarm out of the downstream UPS and a noticeable THUMP from the four 2.5 MVA 600V dry transformers in the electrical room, but the motors don't seem to notice. We lost one 230 kV feeder which killed the A Main but no loads were lost.

Momentary paralleling is not allowed on this system, but we do use simultaneous trip and close signals to speed the transfer on power loss. (All upstream protection relays send a transfer initiate to the downstream gear.)

All transitions are simultaneous trip and close, supervised by several features in the two Beckwith relays. There is a lot of logic available for breaker failure and control schemes. For example, the transfer delay feature lets you initiate transfer but delay it for 15 seconds while you get away from the gear out of the blast zone. But it's not cheap.

I have installed many rudimentary Momentary Paralleling systems, with no known issues so far. (If all three breakers are closed for > 0.5 seconds, then trip the selected breaker.) But my esteemed colleagues on this site have convinced me that is not a good design.

 

[rockman7892]

rcwilson

Very intersting point about issues with the transfer on a system with small inertia loads. It sounds like part of the sucess for implementing an open transfer scheme depends on the motor loads holding the bus voltage to an appropriate level during the couple cycle transition? Is there a way of analytically determining before hand weather or not the motors on the system will be large enough to hold up bus voltage, or is this something that has to be determined through testing or trial and error. I have previous fault reports which show our bus voltage for several cycles after the feeder breaker has opened. Is this sort of information adequate for determining sucess?

It sounds like both closed transition and open transition may be equally as complex however an open transition may be the safer of the two options. Do others agree? Would the cost for implementing this on an existing system be about the same.

I have seen some of these Beckwith papers and understand that they make a pretty good relay for this fast transfer scheme. I'm not sure about others but I'm thinking ASCO may make something as well.

The one question that I have is how non-motor loads are effected during an open trasnition. Most of the papers I have seen talk about motor loads transfering seemlessly however I have not seen how this transfer can effect electronic controls,PLC's etc.. for such transfers. There was one comment above regarding a UPS but I'm curious if otehrs have seen such effects?

 

[86ed]

rockman7892

Try looking into this system:

http://www.energy.siemens.com/hq/en...busbar-transfer-system/standard-solutions.htm

I've seen this system in service and it is fairly elegant.

 

[rbulsara]

david:
No, it has to be independent of PLC, powered from engine battery power and a separate discrete digital timer. No controls that plays part in normal operation transfer are part of this watchdog timer circuit. As they could be the culprit of the malfunction to begin with. Visit website of UI/ Northeast Utilities. They include a circuit diagram in their application's appendix.

They call is "extended paralleling protection". It is primarily intended, I believe (you have to ask them), to prevent sitting in closed transition mode for extended period of time, when two sources are in sync by default, such as when two transformers fed by the same utility source (even if separate feeders), like a spot network.




Rafiq Bulsara

http://www.srengineersct.com

 

[rbulsara]

Rockman:

during the brief paralleling period this would subject our equipment to fault currents exceeding their ratings. I would think this is something to be genuinely concerned about?

.
Yes. This has been discussed several times here.

I would stay away from any type of fast open transition transfers, especially with transformers. As far as I am concerned there are only two type of transfers, open and closed.



Rafiq Bulsara

http://www.srengineersct.com