Phase Rotation Check - 3.3kv Submerged Generator

[BigMotorGuy]

I have a need to verify phase rotation on a submerged expander generator. We are in testing phase and need to verify the correct rotation prior to trying to synch with the power grid. This is a 2MW induction generator (2 pole) and we do not have access to the unit as it is submerged in LNG. We have markings on the leads (A, B, C) but I an hesitant to trust the markings as they have been wrong in the past on some units. The problem comes if the power grid rotation is backwards from the direction the expander is being driven hydraulically, we can crash the machine. I know of several motor phase meters, but they are all rated only up to 600v and this unit is a 3.3kv unit, so not sure what the best course of action is........any help would be greatly appreciated!

 

[oldfieldguy]

You can do it with two sets of analog meter sticks rated for the voltage.

Using the proper safety precautions, with the generator running, simultaneously measure between the A phase on the generator and the A phase on the bus, and between the B phase on the generator and the B phase on the bus. Because the frequencies will be slightly off, you will see the voltage indicated on the meters rise to phase -to-phase value when the two sources are out of synch and go bact to zero when in synch. If the two meters rise and fall together, then the rotation of the generator is correct. If one rises and the other falls, then the generator is backward. Roll two phases to correct this.

Yes, this can be done through potential transformers, but then you assume that the connections of your potential transformers is correct. This is not a good place for unfounded assumptions. Checking phase rotation at primary voltage takes this possibility away.

Most digital meters do not react fast enough for this test to be reliable.

old field guy

 

[ScottyUK]

OFG,

The rule on big stuff is to prove the instruments using a safe voltage, then trust them. At transmission voltages there aren't many alternatives; at MV where phasing sticks exist people tend to use them, but in my opinion it introduces avoidable risk where there is little to be gained.

 

[BigMotorGuy]

Oldfieldguy,

Thanks for the procedure......question, I assume when you say "with the generator running" you mean the generator being driven mechanically and with the power leads opened?

Also, do you only need to take one measurement between any 2 phases or must you measure 3 times between each pair combo?

As far as the mentioned transformers, we do have 3 identical individual PT's (supposedly 0 degree lag) that can be hooked up separately to each incoming phase, so as long as we tie each phase into the same primary terminal location on each PT while tying the other primary terminals together into a Y connection, should this not be trustworthy? We could probably verify the transformers by putting a low voltage power and checking the input to the output with an oscilloscope. I say this since I don't believe we have any analog meters rated up to 3.3kv.

I appreciate your help!

 

[davidbeach]

A permanently connected VT that supplies a meter or relay capable of capturing voltage magnitudes and angles may be all that's needed. Energize the VT from the system with an open between there and the generator. Then energize it from the generator with an open between there and the system. If both sets show the same phase rotation you're good to go. Getting a usable voltage from an isolated induction generator may be a bit of trick.

 

[electricpete]

If you do have access to rotate the shaft continuously by hand (I’m not sure whether you can), then you can use an oscilloscope to monitor the low-frequency low-magnitude voltage induced into stator by the rotor residual magnetism in order to determine direction of rotation (two channel measurement of phase to phase voltages is enough). I first heard it suggested by Gunnar. Here is the detailed procedure we use:

http://www.eng-tips.com/viewthread.cfm?qid=265331

=====================================
(2B)+(2B)' ?

 

[oldfieldguy]

Scotty-

I've done this at 69kV between bus and generator step-up transformer using the appropriate metering sticks.

DB-

As I said, **IF** i knew beyond a doubt that the PT's were correctly connected, testing at lower voltages would be suitable. I have seen too many (and ONE is too many) cases where PT phasing was wrong. Phasing at primary voltage eliminates this possibility.

BMG-

I missed the part about this being an induction generator. All bets are off. You have to provide voltage for excitation as well as mechanical energy to generate. That might be quite a trick.

old field guy

 

[Skogsgurra]

See electricpete above (18 Sep 14 14:16).

I do this routinely and there's always enough remanence in the rotor to give a reliable indication of phase sequence. No need for external excitation.

Any scope or recorder will do, but you can't use a logger that samples once every minute or so. You need at least around 1 kSa/s and that is easy for any scope or recorder.

I have a description somewhere. Shall try and find it for you.


Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[davidbeach]

Actually, as long as nothing changes between the two tests it doesn't matter how correctly the permanent VT is wired. If the first record shows the system one way, either "right" or "wrong" and the second record shows it going the same direction then you're good to go. Absolutely not possible with more than one VT or instrument; also not possible if anything has to be moved other than the opening/closing of breakers or switches.

The system is what it is; as monitored at a particular location it rotates the direction it is seen to rotate; as long as the generator doesn't rotate the other direction there won't be any conflicts. If a known ABC system shows as ACB on the recording, a generator also showing ACB will still have ABC at the point of interface to the system.

 

[ScottyUK]

OFG - really? Better you than me! I don't think they make them for 275kV and 400kV, and if they do make them then I'm damned sure I wouldn't want to use them.

Each to their own I guess, but I firmly believe that the less direct interaction with HV the better. It isn't hard to prove the VT phasing beyond any doubt if you have access to, and control of, both the bus and line sides of the breaker.

 

[Skogsgurra]

Found it! Or rather refreshed the old one:

http://gke.org/presentationer/files/Three Phase motor rotation direction september 2014.pdf

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[BigMotorGuy]

Thanks for all the info.....I cannot spin the rotor by hand, but can drive it at about 10Hz with our hydraulic drive pump (runs off of a VFD), so using a o-scope, I assume I will still be able to verify whether 120 degree or 240 degree separation between the phases on the motor leads.

 

[Skogsgurra]

Yes, BMG. That's a perfect speed to do the measurement at. I was running the motor at around 14% of rated speed (got 7.2 Hz from a 50 Hz machine). You can expect more voltage than I got out of "my" machine. I would guess anything between 5 and 50 V. And that is a perfect level for any scope or recorder.

You should not only verify 120 and 240 degrees (important also, yes) but it is equally important that the sequence A, B, C is correct. Please note that there are grids where the sequence is reversed. So, you have to keep that in mind, too.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.