SYNC CHECKING (machine rotating opposite to system) 1

[UPEE]

here's the scenario what would you expect to see?

a sync check relay monitors phase-phase voltage on both the incomming and running system. the incomming system is wired incorrectly with two of the generator wires crossed.

because of the crossed wires electrically the machine is rotating opposite to the system.

what would the single phase voltages look like at the sync relay? would you ever be able to match speeds and make the sync check relay think the system was actually in sync?

or would the single phase vectors still be going in opposite directions and not allow the sync relay to see a matching speed.

 

[Skogsgurra]

DiscoP,

You may be thinking of the two components, one rotating in positive direction and the other in negative direction, that I mentioned in the beginning of this thread: "Any single phase sine has two components; one positive and one negative. I think that a single phase sync relay will have a problem telling the difference between two voltages with same frequency and amplitudes if there is no phase information. The positive component in V1 will match the negative component in V2, and vice versa. I think"

A much simpler way of seing it is how jghrist says: " With only one phase, there is no rotation in this sense. The voltage increases and then decreases and then increases again, mathematically according to a sin function. I goes up and down because something is rotating (a generator rotor), but this is a different rotation"

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[Rodmcm]

GTStartup has the right idea, the secret is in the testing before you allow a synch relay to operate in anger. We always take the connections off the generator and close in the gen synch breaker. Everything can be checked, synch relays, metering, the works. Best test we ever do. Must admit though that we have had problems convincing some people that it should be done.

 

[sslobodan]

Well I run trough the subject so maybe someone suggested it already.
As I understand from original posting you try to sync two lines together with one rotating in different phase order So lets say You have L1,L2,L3 system at your side and L1',L2',L3' as incoming. Switch L1' and L3'( or L2' whatever is easier) and it should solve your problem with the rotation. I am writing this because I believe you have to connect primary and backup system to a pump (motor) and you want to have it turning in same direction all the time. I had similar problems, but I solved them in a few minutes doing that. You actually don't need to ask generator to switch its phase shift do it on your side. That will change the way of rotation and save you big thinking.....

P.S. If you have single phase rotation in wrong direction switch neutral and phase wire positions.

 

[UPEE]

this is the part that was confusing me:

"However the part I still don't understand is even if one side had an incorrect phase sequence, wouldn't the A phase vectors be rotating in the opposite direction to each other if the generator is rotating in the wrong direction, thus only in synch for a very small period of time - much lower than the time delay." -- DiscoP

In our situation we were actually reading Vab on the incoming bus and Vac on the running bus. (should have been Vab and Vab for both) So if you were to put both of these systems at exactly 60 HZ and look at the vector diagram Vab and Vac will never come together for any period of time. Because the incoming was rotating ABC and the running was rotating ACB.

It seems to me that if both speeds were actually matched these voltages would be 180 degrees out all of the time. Much like what was said earlier one would be positve and one would be negative.

If anyone is interested what actually happened here is we had a contractor swap some underground cables for us. They crossed the phases when reinstalling the new cable. We phased check across the open disconnects to the substation. This phase check was done wrong. We were doing some pre start checks on the turbine and noticed 3 motors rotating in the wrong direction.

So here's a second question to spur some discussion. If performing phase checks across open disconnects. What needs to be done to verify proper phasing on two systems that are not rotating at the same speed? I do know the answer to this one as we did this correctly the second time.

 

[jghrist]

So here's a second question to spur some discussion. If performing phase checks across open disconnects. What needs to be done to verify proper phasing on two systems that are not rotating at the same speed? I do know the answer to this one as we did this correctly the second time.

I'd use a phase rotation meter on each side. You could simultaneously measure the voltage across each open disconnect phase. If they are close to identical, then the rotation is the same. If the phase rotation were different, the voltages across each phase would be very different.

 

[GTstartup]

It always amazes me when this question of correct synchronizing comes up. There are a few ways to do it, but the accepted definitive way is as I described above and Stevenal pretty much repeated.

It's always difficult to measure if systems are in phase with disconnects open. What you need to do is actually close the disconnects, then feed either back from the utility or forward from the generator. (of course isolating the other system) Then take measurements at all devices involved for rotation and synch. The sych check should be happy, the scope at 12 o'clock voltages match etc etc.

 

[DiscoP]

Thanks for sticking with me - this is a little out of my comfort zone but I have learnt a lesson by getting involved.
I worked it out while in the shower last night - this is the place where most of my thoughts come together !
Generators produce sine waves - not vectors.
If you work with a mathematical tool for too long, sometimes you start to think they are real !

 

[Skogsgurra]

Very well put, DiscoP!

But still, the models describing all these things are very real and very valid. So, you shouldn't have a problem either way you look at it.

Isn't it funny how the shower brings out thoughts? If you don't sing...

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[prebellon]

I witnessed a situation years ago - two of the leads (A and C) in a 10 MW machine were inadvertently swapped while replacing the 4.16 kV terminal cables. Two attempts were made to close the synchronizing breaker, causing the generator differential relay to trip instantaneously in each attempt. After some troubleshooting we figured out the problem, swapped the two cables to their proper positions and fortunately were able to return the machine back to normal operation.

 

[GenSwgrEng]

A single-phase sync check relay will allow closure of two sources with opposing phase rotations. Most systems I've dealt with have two permanant sources. So once phase rotation has been verified, a single-phase sync check will suffice for future operations. If one of the sources is temporary, it is a bettery idea to use a three-phase sync check, or a combination of a sync check and a phase sequence relay.

For initial check-out, the easiest way to confirm matching phase rotation is to energize the same piece of equipment from each source, one at a time and check rotation there. Another way to check is to take two voltmeters, one across A-phase of both sources, and one across B-phase of both sources. If both meters are bottoming-out and topping out together as the sources come into and out of phase, your phase rotations match. This is easier to see using Simpsons, but works with digital meters as well.

Engineers are always honest in matters of technology and human relationships. That's why it's a good idea to keep engineers away from customers, romantic interests, and other people who can't handle the truth.

 

[ScottyUK]

GenSwgrEng,

Instead of voltmeters it was once common to use tungsten lamps. This is the real old fashioned way of doing things - many old plants still have synchronising lamps as a fallback to the auto synchroniser. There are variations of the theme which use three bright lamps to indicate synchronism or two lamps of equal brilliance and one dark lamp to indicate synchronism. With a little practice it is possible to have as good a synch as the auto system would give.


----------------------------------

Sometimes I only open my mouth to swap feet...

 

[slavag]

Hi All.
GenSwgrEng , you are right 100%.
I think we need add to GenSwgrEng post, post of Stevenal and
GTstartup ( what need for commissioning) and we have best and short manual for design and commissioning teams.
I would like only closed several small points.
1. Synchrocheck relay used for permanent source ( dU, dF and dFreq) or as voltage check ( U1>U2 and U1> Umax and U2<Umin or U1=U2=0 (dead bus-dead line)or other same options). For voltagecheck functionality prefer add VT MCB OK signals or/and dead source CB or Dis are open.
2. Synchrocheck used as additional check point after synchronazing relay for two rotate source. Series connection or DualChannel syncronazing relay.

 

[slavag]

Hi Skotty.
Your post we also must add to this manual.
it's old but work w/o any problem up today.

 

[UPEE]

GenSwgrEng - what you said in the last part of your post is what we did to check the system. It worked great and it was very obvious that everything was rotating the same direction.

"Another way to check is to take two voltmeters, one across A-phase of both sources, and one across B-phase of both sources. If both meters are bottoming-out and topping out together as the sources come into and out of phase, your phase rotations match. This is easier to see using Simpsons, but works with digital meters as well"

 

[GTstartup]

"Another way to check is to take two voltmeters, one across A-phase of both sources, and one across B-phase of both sources. If both meters are bottoming-out and topping out together as the sources come into and out of phase, your phase rotations match. This is easier to see using Simpsons, but works with digital meters as well"

This is only bullet proof if you can measure the primary voltages. If you are measuring PTs this is a disaster waiting to happen.

 

[UPEE]

"This is only bullet proof if you can measure the primary voltages. If you are measuring PTs this is a disaster waiting to happen."

- i agree and this is what we did using two sets of high voltage sticks and high voltage meters.

 

[GenSwgrEng]

Aboslutely, you'd want to put your voltage meters directly on the bus/cables. Of course, this is okay for low-voltage applications, but is more tricky for medium voltage applications. If lieu of meters, you could hot sticks and watch them grow dimmer and brighter as ScottyUK describes. I have resorted to using PT secondaries, but only after visually verifying the phasing of the buswork, and I put my meter directly on the PT, so crossed wiring is not an issue. BTW, we still put sync lights in addition to a sync scope in our paralleling switchgear.

Engineers are always honest in matters of technology and human relationships. That's why it's a good idea to keep engineers away from customers, romantic interests, and other people who can't handle the truth.

 

[Gatorsoft]

This subject is near and dear to me as I will be performing commisioning on a new generator installation in the next few days. I think the following procedure will ensure that all is correct. Please warn me if my thinking is flawed.

1) Break the connection at the generator and close the synch breaker.
2) Verify zero potential between generator and bus PT inputs for each phase (A and C)both at the auto-synchronizer and synch check relay.
3) Reconnect the generator (after opening the synch breaker of course)
4) With the generator rotating and excited, connect lamps between the bus and generator PT inputs for each phase at the auto-synchronizer.
5) If the lamps go bright and dim together, my rotation is correct.

Have I got it covered?

Thanks,
Gatorsoft

 

[ScottyUK]

Gatorsoft,

Your step 5 is ok, but my preferred method is to wire the lamps A-A', B-C' C-B'. Synchronism is indicated by a dark A lamp and equal brightness on the B and C lamps. It is easier to detect equal brilliance than it is to detect a true dark condition, unless you switch off the lights in the substation. If wired per your proposal the lamps appear dark for a fairly broad phase angle near the point of sychronism. You must use identical lamps and make sure they are rated for double the line-neutral voltage.



----------------------------------

Sometimes I only open my mouth to swap feet...