Unbalanced Rotor Currents ? 1

[edison123]

We recently overhauled & tested a 600 HP slipring motor with following nameplate:

Output - 600 HP Speed - 990 RPM (50 Hz)
Stator Volts - 3300 V Rotor Volts - 720 V
Stator Current - 96 A Rotor Current - 390 A

During open shaft run, with the rotor energized at 400 V, we found unbalanced and hunting rotor currents of 70-75, 85-95 and 44-46 Amps. Suspecting rotor winding problem, I did a surge comparison test but it did not reveal any rotor winding problem (no phase unbalance or reversed connections etc.)

On open shaft run with stator energized at 400 V, the stator currents were equal and steady at 5.5 Amps.

The open shaft start up with stator energized was quicker and smoother as compared to start up with rotor energized (when it should have been the other way about) thus confirming that the unbalanced rotor currents were producing negative sequence torque resulting in net lower starting torque when rotor is energized.

Apparently both the stator & rotor were rewound some 12 years back and it was running without any problem. The motor came to me for conversion of sleeve bearings to anti-friction type.

What could cause these unbalanced and swinging rotor currents ? If the rotor winding is wrongly connected (reversed connections or unbalanced winding etc.) should not the surge test reveal it ?

I welcome any opinions on this strange phenomenon.

 

[Shortstub]

ep,
Thanks for the advisory. The problem now worsens. If you model the motor with the stator and rotor parameters interchanged, the rotor as the armature connected to a supply voltage of 400V, and the stator leads shorted, the magnetizing current is even less stable.

My point is... if, prior to the mechanical work, there was no noticable problem with the electrics, then the test stated showing rotor current variation is... a witch hunt. Especially if there is no prior comparable data.

 

[electricpete]

I agree I personally wouldn't do that test because I don't know anything about it. I wouldn't call it a witchhunt... all edison asked is why it acts that way.

Can you explain more why we would expect unstable magnetizing current?

I agree on the basis of different equivalent circuit parameters we expect different torque speed. Edison has already responded to that comment by pointing out that the pu voltage on the rotor is much higher, which should certainly make the average torque speed higher when energized by the rotor (since torque ~ Vpu^2), even though we don't know the exact shape.

There is one other aspect: harmonic torques. As we know long ago motors had problem with locking, cogging, syncronous torques etc. Careful selection of rotor and stator slots takes care of that. I am positive that there is very substantial change in the harmonic fluxes when we energize from rotor. This may undo the careful design of the OEM and reintroduce torque cogging, locking, etc



=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[edison123]

ss,

Wound rotor motors can be run with rotor energized and stator shorted and it behaves exactly the same way if it is energized other way about. No load currents (with stator or rotor energized) are always equal for balanced voltages. I have tested many WRIM's with rotor energized without any problem. This is the first time I am seeing a rotor, which passed a surge test, showing unbalanced no-load rotor currents.

pete,

cogging, crawling, harmonic induction and synchronous torques etc. are all connected with cage rotors and not with WRIM's.

 

[electricpete]

If you say torque cusps etc are not a problem on wound rotor motors, I will take your word for it (I don't know much about that).

What about if there were a rotor turn fault far from the slip ring connection down in the middle of the winding. Is it possible surge test would miss it? Maybe a rotor winding resistance test would be worthwhile? I don't think it's a likely (for a rotor turn fault to persist without escalating rapidly to phase or ground fault when energized at rotor side) but you never know.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[jbartos]

Question: Have slip rings, brushes, connections, etc. that may be producing unbalanced and hunting motor rotor currents, been checked?

 

[edison123]

pete & jb,

The rotor winding resistances are equal both at sliprings and at the winding take off leads (which were newly provided due to the bearing alteration). Brushes are all seated well on the sliprings. Connections from the brushes to the external terminals were also found ok.

 

[aolalde]

Edison123

Do you have winding data and lamination details for stator and rotor? 600 HP regularly had a wave winding rotor can you provide the connection details?

Why do you affirm that wound rotors do not have harmonic, cogging or noise problems? Do those motors develop a perfect sinusoidal flux distribution and the air gap reluctance is totally constant in spite of slots?

 

[Shortstub]

Back to the original problem!
Regardless of which winding is used as the armature, and which the "rotor", and although I agree either condition will result in rotation, you are really comparing apples to oranges.
If either the stator or rotor is connected to the source, its current and voltage measurements are at line frequency. The other measurements, ignoring startup, are at slip frequency. So my questions...

1) What was the test setup for each condition!

2) Was the instrumentation adequate for the frequency of the circuit in which they were connected?

 

[electricpete]

SS
#2 is answered by the original post.
Rotor currents measured when rotor was externally powered.
Stator currents measured when stator was externally powered.
=> All measured currents are at 50hz. There is no reason to suspect current measurement error associated with frequency.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[aolalde]

I thought that may be a short circuit develops in the rotor winding only when it spins, due to centrifugal forces. Static tests will not detect such a condition.

 

[jbartos]

Suggestion: Magnetic circuits should not be overlooked. There may be some problems with the magnetic circuits, e.g. poles, gaps, etc.

 

[Shortstub]

EP,

You misunderstood my caveat. Line frequency parameters are only present in the "rotor" at standstill, otherwise they are at a much lower frequency. The following will explain my "inductive" reasoning (excuse the pun):

My first engineering job was in an industry that had numerous WRIM's. Most were GE, Westinghouse, and Western Electric (yes, I said Western Electric) machines, manufactured in the early 20th century (as I was). (An aside... they were monstrous in size compared to today's designs!) I remember several salient points:

1) The reverse open-circuit, blocked-rotor, test, i.e., the rotor as the primary winding, and the stator as the secondary winding, could uncover iron damage, if significant. However, a more useful aspect of this test was that turns-ratio could be determined with greater accuracy than possible with the conventional test, i.e., stator as primary, rotor as secondary.

2) Either test was more useful if power measurements were included.

3) If rotation was a requirement for the reverse test, then short-circuiting the stator winding was done, but only if the stator was wye-connected. If the winding was delta-connected, the terminals were not shorted!

4) If rotation was a requirement for either test, the instrumentation on the secondary side had to be capable of working at slip frequency, i.e., sf, where f represents line frequency. Then, for no-load tests, secondary current and voltage parameters have an operational frequency less than 1 Hz. I know of no portable instrumentation with such a capability, unless it is of the moving iron-vane type. A standing joke at the time was that it took two men to carry the tong-type ammeter. Today's instrumentation with a "true" RMS capability may be able to carryout such a low frequency measurement, but only if so specified by the manufacturer.

ED123,
What say you regarding your finding/conclusion! Is the machine in service?

 

[edison123]

ss,

The motor had been in service for nearly 12 years before it came to me for bearing conversion. Now the motor is in service again and the client has not reported any problem.

In re point 3 of your last post, why would not a delta wound stator (shorted across terminals) work in the reverse mode ? Am I missing something ?

 

[jbartos]

Comment on the Delta wound stator shorted: If the stator is Delta-Wound and rotor is Wye-wound, there will be voltage shifts affecting the reverse test results.

 

[aolalde]

For WRIM, Coupling between rotor and stator windings is purely magnetic. Then the windings connection (wye or delta) do not change any electrical performance nor for the stator neither for the rotor.

 

[Shortstub]

ed,
I didn't say it wouldn't work for the delta connection. It just isn't necessary. A delta connection is already a closed circuit. However, the jumpers used between the stator terminals could influence operation depending on their size.

Transform the delta to an equivalent wye. Then the jumpers are analogous to the external resistors of a WRIM. For no-load operation when slip is near zero, then the jumper resistance magnitude compared to the stator winding resistance magnitude, alters torque output.

Perhaps it was done as routine in your facility because the stator winding connection was unknown. Do you know the wire size of the jumper? BTW, measurement of stator current and voltage, would probably have been more revealing.

jb,
Anything is possible. But a mix is not done as a rule for viable commercial design, unless the WRIM stator is wired for different speeds. (doug, any input?)

aol,
Hopefully my response to ed's query, should suffice.

 

[edison123]

ss,

Just trying to understand your post.

Do you mean that if the stator winding is delta connected, then the motor will run (under open shaft) without shorting the line terminals ?

If yes, then no. The motor will not run because it will still be in open circuit state and will see only induced voltage across stator terminals.

If no, then I misunderstood your post (again!).

As Aolalde says, winding connection does not change the basic physics of the induction motors.

 

[edison123]

aolalde,

You had mentioned an interesting point about "dynamic" rotor winding short.

Suppose, if I had mechanically driven this motor with another motor and then applied surge testing to the rotor sliprings, do you think

(a) I can see the rotor short or

(b) I would need to buy another (expensive!) surge tester.

 

[aolalde]

Edison:

I think your proposed test a) must work, since the rotor connection will be seen steady by the surge tester through the brushes and if your pony motor is close to full speed that kind of short should be shown. I have seen wound rotor windings with faults or wrong connection working if not at optimum conditions.

As far as I know, other ‘expensive’ test equipment should be a current signature analyzer, but those work better with the motor loaded so the slip shows the rotor winding performance.

 

[electricpete]

With the rotor mechanically driven and stator deenergized/open-circuited, you could apply line frequency to the rotor and measure current to determine impedance. Repeat between each pair of slip rings. Shorted turn should cause some unbalance (sort of similar to pole drop test). As I'm sure you know inductance varies with turns-squared and you don't have as much interference from various contact resistances as you would with dc test or low-frequency rotor currents during operation.

Then again, this is now a helluva lot different than your original test powered from the rotor (only difference is stator open-circuited).

=====================================
Eng-tips forums: The best place on the web for engineering discussions.