3.5 MW motor starting problems continue... 7

[ters]

A while ago, I started a thread

http://eng-tips.com/viewthread.cfm?qid=235885&page=1

which many of you contributed to, but that story has not ended yet...

So the issue is that a 3.5 MW motor driving a large fan has problems starting. The starting method is the soft starter. Rotational inertia is high.

We discussed several causes, such as inadequate motor, inadequate starter, week and inadequate power system, etc. But calculations show that although things are marginal this still should work, but it does not.

Today we managed for the first time to put two identical 13/20MVA transformer in parallel. Their impedance is 10.5% at 20MVA but each transformer also has a current limiting reactor 6%. However, 20+20MVA system still would no start a 4MVA motor...

System: two 13/20 MVA transformers in parallel, + about 5 MW of additional load shared between two transformers.

Motor: asynchronous, TECO-Westinghouse 3.5MW, In = 385Amp, 6kV, 50Hz, 1485 rmp, locked rotor current 2300Amp.

Fan: Large rotational inertia. The fan has no load at the present (no air, dampers closed). The Impeller weight is 15,000 lbs, and the Impeller moment of inertia (WR^2) is 66,000 lb-ft^2.)

Starter: constant current, voltage ramp ~10 sec, current limit 470%, starting time 50 sec, after which bypass contactor closes (when then speed is about 80%).

We had some luck starting and running the motor using one 13/20 MVA dedicated transformer but now with 2 transformers in parallel it would not work. We tried three times and none of the starts was successful. Each time the soft starter was manually shut down after about 20 sec since motor was creating too much vibrations and was not accelerating any more.

Problems still could be electrical, but for a change we started to suspect that there may be some mechanical problems, perhaps related to the motor rotor axial movement. If the rotor indeed does try to move, either it may not have enough space to move to the electrical center or the shaft is moving back and forth around its electrical center for some unknown reason and hitting the trust.

The attached charts provide more info. During the voltage rump time, which was about 10 sec, motor runs smooth and with two transformers in parallel the initial voltage drop is rather modest - falls from 6.4kV to 6kV. This is good news, with one transfomer we had much larger initial voltage drops. However, the bad news is that after the voltage is ramped up high enough that the current reached the preset limit (470% In), some vibrations start.

As you can see, the current and voltage outline forms on the charts start at some point having "teeth" like a saw. Once stabilized there is about 3 - 4 “teeth”/second. Just judging by the ear, the frequency of mechanical vibrations is in the same range, meaning current spikes likely correspond with frequency of mechanical vibrations.

Any idea what might be happening?

 

[ters]

Alehman, what you see on our diagrams is just one phase voltage and current forms. The chart recorder paper speed is 50 mm/sec, which means each 50Hz cycle is 1mm of paper. We cannot provide a better resolution, but even with this you can see cycle half period forms as narrow positive and negative spikes.

As you said, the strange thing is that current measured from pick to pick (positive to negative) is always the same, but the amplitudes of positive an negative half periods are not – higher amplitude of the positive half period is followed by the lower amplitude if the negative one and vice versa.

Or one could put this another way around and say that amplitudes of both cycle half period are the same, but the reference point is modulated somehow and floats up and down making so its own sinusoidal curve where it frequency is proportional to the slip...

Based on what Electricpete underlined several times here, and some other engineers did it during earlier discussions, and also some other people outside this form stated as well, the likelihood of damaged rotor is high, but then as you wondered why it would create such a phenomena that negative and positive half periods would seem to differ in amplitude from 1 ~ 1 (normal symmetrical values) to something like 0.5 ~ 1.5 the worst case which happens after every several cycles?

Edison 123, as getting manufactures involved... well.. it sends me back to square one... I started the initial thread about this problem on the forum about two months ago since we were unable to get ABB help for the soft starter at that time. After waiting forever, ABB USA sent a specialist from Egypt who after a lot of struggle made the soft starter working somehow but left blaming all problems to the weak power system… OK, maybe it was sort of weak then, but now we can provide 40MVA with reduced impedance to deal with the motor which starts at 2000A and about 5000V (average reduced voltage) so 40MVA should be enough for this…

As for getting TECO involved, as I said, their real experts may only reside in Taiwan and so far we have never had a direct contact with them, but we will try to do it this week.

That is why there is so much trial and error, it still may cheaper to just replace the soft starter then wait another forever...

 

[electricpete]

If you have a motor shop close by, they should be very familiar with off-line tests for rotor bar problems. If they have a pdma box, there is a pretty quick R.I.C. test that can be done.

Even more effective, but more work is a single phase test. It is described on page 8/26 here.

Which figure are you guys looking at to see that the positive and negative peaks don't line up? I only see a good view of the oscillation in the very first attachment, and the scale is too scrunched to make any judgement on lining up of the peaks for me.

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

 

[electricpete]

I forgot the link. Single phase testing to check rotor bars is described on page 8/26 here"

http://www.pumpingmachinery.com/pump_magazine/pump_articles/article_29/rotortesting.pdf

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

 

[LionelHutz]

I simulated your data;
350% current = 85 second start
450% current = 43 second start
The 350% current simulation is attached.

So, it appears you definately have a problem with the starter or the motor/load. It should start just fine at 450% current. There should not be an oscillation caused by limiting the current. The motor should stay smooth and just quit accelerating if you limit the current too much and it stalls. It is either a mechanical or electrical resonance. Any yes, it could be the circuit impedance causing the current limit loop in the starter to oscillate like you earlier asked.

We have been building soft-starters for applications >5kV for years and yet we still dump a ton of money into R&D each year to further understand these units when connected in a real world application. Building a soft-starter to reliably and properly start that motor is much more complex then it looks from the outside.

 

[LionelHutz]

I forgot to point out. At 350% current you go well past the stall-damage curve and at 450% current you will just pass the stall-damage curve. Even at full-voltage it appears you will be very close to the limit.

You'd be surprized how often we do run into places that push these large motors like this and they don't seem to have a real concern about it. It's a lucky thing for this planet that 3-phase asynchronous motors are very rugged devices.

 

[ters]

Here are some more data of todays testing, a better chart taken by a digital recorder.

Please refer to the attached chart called Test No6 – non-symmetrical SCR firing. This was an attempt to start the motor with the fan coupled, which failed - the motor stalled at several hundred RPM and was manually shut down after abut 30 sec as it was apparently not accelerating any more - the speed stalls after about 10 sec and just stays there. The soft starter settings were: starting voltage 40%, ramp time 8 sec, current limit 470%

What you see on the diagram is just one phase voltage (red) and current forms (green). The chart on the first page shows about 18 50Hz cycles (recorded using a multichannel scope), while the second chart captures the same values but zoomed out to show more cycles.

Each positive or negative spike on the chart is supposed to represent a half of 50Hz period, whereas all half periods should be normally chopped symmetrically, and be symmetrical by the width and amplitude.

However, you can see that the amplitude of voltage and current half periods are very uneven – some spikes are 2-3 times higher than the other. Also please look at the wide negative half periods which look like a letter W. This cannot be a normal voltage form since a positive half period in the middle of "W"is completely in the negative area – below the mid reference line.

Similar for the current - higher amplitude of the positive half period is followed by the lower amplitude if the negative one and vice versa.

 

[ters]

And thank you again Electricpete and LionelHutz for very valuable comments.

 

[edison123]

Sorry that you got a raw deal from the OEM's.

You mentioned about 400 meters of cable. Did you consider the possibility of a major voltage drop across the cable, especially when starting ?

 

[ters]

LionelHutz what we described may as well be what happened here. Something like in the old days when the farmer was trying to make his horse to pull a heavy load uphill. So he takes a big stick and beats the horse with no mercy until it pulls the cart for a few meters. As soon as the horse stops exhausted, farmer’s son who walks behind inserts a stone under the rear wheel to prevent the cart going back. Then the farmer beats the horse again, etc, until they finally make on the top. Then the horse dies... Looks like that is what we did ..

Edison 123, sorry I mistyped, should say 300m of cable. No, there is no big voltage drop on the cable. The cable is oversized, we have much more of the voltage drop before the cable. The system is: 13/20MA transformer + (or two of them, if in parallel) at 10.5% impedance at 20MVA (or less at 13MVA) plus a 6% reactor (or two of them in parallel if transformers are paralleled) + 300 m of cable. Most impedance is on the transformer(s) and reactor(s), but we tried everything including shorting the reactor, made no difference. Line voltage measured just before the soft starter, which is only 10m from the motor, when the motor starts, and if two transformers are paralleled, is 5.8kV. Not a big drop from the initial 6.3kV.

 

[edison123]

5.8 to 6.3 KV, that's a 8% voltage drop with nearly 15% drop in torque. I wouldn't dismiss cable-as-culprit just yet.

 

[edison123]

Do your trafos have primary tapping so that you can boost the secondary voltage by 5%, which is quite normal ?

 

[ters]

OK, maybe the cable does contribute, but the measured voltage drop from end to end of the cable is only about 100-200V when the motor starts (there is a switchgear and voltmeters at both cable ends). So looks like most of the voltage drop from 6.3 vs 5.8 kV still happens on reactors and transformers. We also did calculations which show that the cable is not creating a major voltage drop.

Also, the voltage measured on the motor terminals during the start is usually not higher than 5000V (with the fan coupled), so the soft starter itself, as presently set(to let no more than 470% of the current) is actually a major “system impedance” where the voltage drops from 5.8kV (before the soft starter) to about 5000 kV (after the soft starter) - if one can trust to the RMS measurement of that funny voltage form on the motor terminals.

 

[ters]

Yes,they do - there are primary tap changers. We ususally boost the voltage to about 6.5kV (nominal 6.3kV)

 

[edison123]

Gotcha. OK. I am done but will keep an eye on this thread on how this turns out.

Best wishes. You seem to need them.

 

[LionelHutz]

When I look at your data, I see a chart recording where the number of samples per second is not high enough to get a proper cycle by cycle picture of what is happening. You're missing what happens between each of those "peaks" on your waveforms and each of those peaks might not be where the peak really is.

If you are hitting the 470% current limit then I don't see the voltage drop as being an issue. Your data says the motor will start just fine at 450% current so 470% should work just as well. Heck, we've gone to 20% plus before and been fine because we were still able to give the motor enough current to start.

 

[ters]

I agree that the chart doesn't provide details such as how the chopped wave form looks like, but it still seems to me that there are no picks/half periods missing - the last chart shows a period of about 0.5 sec and all half periods seem to be there. But as you can see, most of the time, higher amplitude positive half period is followed by lower amplitude negative one.

 

[panter]

Hello, everyone!!
I am worker in local repair electric shop and we are trying to help Mr Ters with this big challenge.
Pete, I am familiar with motor current signature analysis because I have worked more then 5 years with SKF instrument CMVA 60 and I found more then 30 motors with rotor problems but, unfortunately, it was out of the service few days while motor worked loaded. We were also concerned about rotor because motor has a lot of unsuccessful starts last two months and I have watched two ampermeters in 6 kV station and I didnt see swinging needles of ampermetrers which is usually sign of broken rotor bars.
We have worked in the past single phase rotor test on smaller motors but this rotor is problem to rotate because it has 8 tons weight!!
I noticed that strange sound, vibration, ect. motor have had from start of commissioning and speed of the motor where problem starting depends of starting current limits. I have attached oscilloscope recording one of start, when the motor runs rough and it looks like that motor receive double frequency from soft starter.

 

[electricpete]

Hi Panters - It looks like a voltage waveform which has been regulated by electronically chopping a portion of each cycle - I think that is what we expect to find on the output of a device that electronically controls its output voltage. Do you have the current waveform?

We have a 7000hp 2-pole horizontal motor with sleeve bearings that I assume is similar size/construction to this one. We have manually rotated with a strap wrench - the toughest part was breaking free but then easy to keep rotating once started. As a conservative measure to avoid sleeve bearing damage when rotating motors by hand, some people recommend to prelubricate the bearing by spinning the oil ring a few times by hand before you rotate the shaft.

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

 

[ters]

Let me mention that Mr. Panter is actually the Client and I'm just a pure consultant...

We did not record the current form with such details -zooming into several cycles - the only current form we have is on the charts I posted, which is more crude but still indicative. Also, the chart Panter posted was recorded about two months ago and at that time we occasionally managed to make a successful start with the fan, but now that is not the case any more (last 4 starts failed), perhaps suggesting that some problems with the soft started developed meanwhile or worsened if always existed.

About 2 month ago we sent Panter's chart to the manufacturer (Motortronics) and they commented it looked OK. Also some people here on the forum said the same.

However, I'm still puzzled with the created voltage form. When, say, the positive half period is chopped, then the voltage goes to the negative before it recovers. The same applied for the chooping of the negative half period. So if you count all positive and negative picks in period (regardless of their amplitude and duration) you will actually count 6 picks instead 2, which means we have voltage of some sort of 150Hz frequency... Why that would be still considered normal? I know that the current fills the gap and still remains 50Hz, but would not it be more natural to chop the wave form at the beginning and or end, rather then in the middle?

 

[LionelHutz]

To comment again on the current waveform. It appears there is something like 2 or 3 samples per period. The samples are not always taken at the peaks of the signal and so it appears it is oscillating. Honestly, I would need a current recording like the voltage waveforms if I wanted to determine anything useful.

To comment on the voltage waveform. It looks fine. When the SCR is off the voltage on the off phase will go to the voltage as dictated by the motor and the voltage applied to the other 2 phases. The off period is in the middle of the waveform becuase the current is lagging the voltage. This means, for example, that the negative SCR will turn off after the voltage has crossed 0 and become positive.