11kV Motor failing to start 2

[allan789]

We have a 11kV, GEC, 6.5MW motor which is connected via a gearbox and pinion to a ball mill. The motor is soft started by a LRS.

About a year and a half ago, April and again now yesterday the mill has gone for a start and has sat there not turning. It has sat there pulling 1800A until the protection eventually triped.

In all cases the motor and LRS has been inspected with nothing found. The mill was then inched by a seperate inching motor with no apparent mechanical windup or resistance. The mill was able to be restart after the inspection each time.
The LRS conductivity was measured at 14mS as compared to a design value of 12.7mS.
We have also had the motor reciently tested by a motor test company with no obvious abnormailties.

We are at a loss to what could be the cause of this intermitant fail to starts and any suggestions would be greatly appriciated.

 

[electricpete]

"unless the CT's which are used to provide phase current are the ones on the neutral side (vs wye?)"
should have been
"unless the CT's which are used to provide phase current are the ones on the neutral side (vs line?)"
i.e. does the phase current indication that you provided come from the neutral side CT's or the line side CT's?

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

 

[electricpete]

Another thing that doesn't add up... the large differential current in phase A needs a return path either as differential current in another phase or as ground current... the other phase differential indications and the ground current indication show much lower magnitude than the A phase differential current.

The ground current does not resemble the sum of the three phase currents. How is the ground current signal developed? And from which end of the winding (neutral or line)?


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

 

[electricpete]

At the beginning of normal start, total (internal plus external) R2 <= |X1 + X2| (that is how we intend to set it up, right?)
|Znormal_start| = sqrt(2*(X1+X2))^2 = sqrt(2) * X1+X2

With resistors fully shorted out, R2 >=0.
Zshorted_minimum = X1 + X2

So the difference in starting stator current is at most sqrt(2)=1.4 if we have all or none of external resistinace in place, right? But we have somewhere around 2.5 difference. I think (although haven't confirmed) that the difference can be higher when there is an unbalance on the rotor side (?). Certainly the presence of 100hz in the Id as well as the A-phase differential current seems to suggest a side rotor unbalance.

If you haven't had a tester in since the differential trip, I'd think it is time to get him back in for thorough check of everything: stator, rotor, rotor resistors/controls, possibly stator CT's. One scenario is that you've had a lingering rotor-side problem which causes intermittent failure to start (either due to intermittent nature of rotor side fault, or due to fault which places starting torque just on the brink of where it needs to be to breakaway the machine, so it doesn't stop it every time)... which eventually took it's toll and led to a serious stator fault.


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

 

[waross]

I would suggest that the two most probable causes may be Lionel's suggestion that the LRS may not have been in starting position and dry bearings.
An hour for the pump seems excessive. I would expect that minutes may suffice.
How long will the motor be inched? We needed 20 or 30 turns to lubricate the bearings. If inching is very slow, it may require more turns.

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

 

[allan789]

The motor was tested after the first two failed started but before the last one we had a few days ago. The plant was installed around 2002, this motor was changed out around 2 years ago. We have had three failed starts, but other than that we havent had any other problems with this motor.

Im not a 100% sure how long it is inched, but around 30-40mins seems around the number, which is around 2 revolutions of the mill i believe. That would mean around 200 revolution of the motor. It seems like if there was some bearing bind or resistance we would see it in the inching motor which is i think 150kW (its the weekend so i cant check!). You can see the current on the motor change with the charge moving so its quite responsive to load. In saying that the inching current is not displayed so unless it tripped no one would notice high current draw.

Its funny you made that comment Lionel because this LRS is unique to our LRS fleet in that it is the only one where the LRS electrodes stay at the bottom and only return to the start when the motor stops. On its own that should not cause any problems but it does hint that there may be a gremlin hiding somewhere there.

I think a full inspection of the LRS may be in order. Check all of the proxies, connections and other gear inside and outside the tank. A change of LRS liquid would probably rid the myth onsite here that the sodium carbonate is falling out of solution as well.

 

[koizumi]

Dear all,
I've experienced a similar incident with a 6MW-6kV-6p wound rotor motor before.
We received that motor with some damages in the rotor bar connections due to some explosions.
Our customer also reported of some failed starts, happened randomly in a long time period before motor breakdown. After each failed starts, they'd checked the motor&LRS carefully but found nothing wrong. And the next starts were all OK, until the breakdown at the last start.
You all know that rotor windings mainly are wave type, consist of rotor bars formed and connected at two ends which we called rotor fins (see attached).
Examining that rotor, we found that there were some kind of arc appeared between some adjacent fins, which belonged to two different phases of rotor. It means there was a short circuit between, for example, phase A and phase B of rotor winding.
And we presumed that before, sometimes an electric arc also appeared at that point at starting time, but not to the level of causing any destruction to the winding. Just caused the motor a fail start.
To conjure an arc between two adjacent rotor fins (belong to different phases), there are some conditions:
-There are some damages inside the fin insulation.
-The moisture and/or dust buil-up in the space between the fins reach a certain level.
-Voltage and, especially current, high enough to create a strong magnetic field that could ionize the air and trigger an arc.
And when an arc appears, it connects two phases of rotor, and causing stator current unbalance, high starting current, rotor at a stand-still, etc. After its disappearance, everything might return to normal, and testing of the motor found nothing wrong. And it doesn't extinguished abruptly after CB cutout, so due to some inductions there's still some residual current inside stator winding.
And my advice for you is that you should take your motor to a qualified workshop as soon as possible for rotor winding check-up.

 

[electricpete]

Good comments.

I have been thinking about the way that IA and IC are equal/opposite decaying to zero.. and IB was zero at the time of termination.

I have described a model of a very simple model of a CT here:
thread238-217182

Attached I have applied the same model to 2 different scenario's of current dropping suddently to zero: one scenario is that we have a dc current that drops to zero, the 2nd scenario is that we have a sinusoid dropping to zero. In both cases, you can see that the the result is a CT output which jumps to the opposite side of zero and then decays away.

The whole calculation is done with Laplace Transforms, so if you are into that you should be able to follow it.

But if not, you can still scan the 4 graphs to get the big picture of the calculation.
Graph 1 - Primary DC input which drops to 0 at t=0.01
Graph 2 - Secondary output for the input described in graph 1
Graph 3 - Primary AC cos input which drops to 0 at t=20.004 seconds (not a natural current zero)
Graph 4 - Secondary output for the input described in graph 3

If anyone is interested, I tried adding a secondary inductance into the model, but it had no effect on this overshoot when L2<<Lm, so I took it back out to keep the model simple.

The bottom line, I am thinking this equal/opposite decaying Ia and Ic might be an expected result of chopping a sinusoid at somehwere other than it's natural current zero.

Then again, I am still unable to explain the differential current (unless there is an actual stator fault).

We don't have digital relays, and I have never had the fortune to monitor current during motor trip (plenty of starts, but don't know when the trip will happen), so I have never seen this. I am curious if anyone else has seen this as a normal expected CT response, or for that matter anyone has comments on the analysis attached.

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

 

[electricpete]

There was an error in what you saw displayed. I changed the paraemeters R2 and L2 but didn't re-execute the plot, so the last plot you saw didn't match the R2/L2 parameters.

Attached is a revised version, freshly executed the whole thing so all plots are updated correctly.

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

 

[electricpete]

And here is a powerpoint that summarizes the results of the pdf (perhaps more readible... doesn't include all the math details)

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

 

[electricpete]

Now a different subject.

I presume that some of the external rotor resistance is decreasing during the time period of the waveform you provided, correct?

I am not sure if it makes a huge difference, but my intuition is that what we are seeing is an open circuit in one phase of the rotor circuit (rather than shorted external resistor in one phase). The reason is that the Iq current shows increasing 2*LF = 100hz over time. That suggests an increase in imbalance over time (which would correspond to initial open circuit at inifnite impedance... other phase resistances get farther and farther away from that as the external resistance decreases) Rather than an initial short circuit on one phase external resistor (which corresponds to low initial resistance on defective phase, other resistors get closer and closer to that as the other external resistances decrease).


If I get energetic, I will try to do a numerical simulation of one open phase of a wye-connected rotor, to see if it acts like your waveforms.

The other much less likely alternative to explain apparent increasing rotor imbalance might be that you have shorted rotor turns and they are continuing to short more and more during the short... but doesn't seem likely that you would continue to operate long in that scenario.

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

 

[waross]

How about a position switch on the LRS and a latching relay to indicate if a start was attempted with the LRS electrodes in the wrong position?

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