motor current spike

[drbond24]

All:

What would cause a spike in the running current of a 3 phase motor? The motor in question is 5000 hp, 4 kV and powers a fan. It will be running along just fine pulling around 570 amps, then suddenly spike to about 630 amps for a second or two, then return to normal for hours before doing it again. Once last week, one of the motor's differential relays engaged during one of these spikes and tripped the fan. For that to happen there would have to be disparity between phase currents. The ammeter that is recording the spikes is on phase 2 and I do not have an indication on either of the other two phases. So, perhaps my real question is: what would cause a spike in amps on only one or two phases of a running motor?

I checked our amp trends for all other running equipment fed from the same buss and did not see similar current spikes, but I am not entirely sure which phase(s) are being monitored on all of the other equipment. I can find out, but I haven't invested the time to do that as of yet.

Can anyone suggest something else I can check?

This motor/fan is one of three identical units and the other two do not have this problem, but each of them is fed from a separate buss.

 

[Jraef]

Since it isn't continuous, the first thing I would suspect is a connection issue, i.e. a loose connection or an intermittent grounding. I've seen liquid in a conduit do this on a large refrigeration compressor. Condensation was forming and dripping into the underground (in concrete) feeder conduit to the motor. After a while it would build up to some point where there was a flaw in the cable insulation and there would be a high resistance ground fault too low to cause the fuses to clear, but would show up as an imbalance. Then apparently the heat would evaporate the water, the problem would go away for a while until conditions were right for a lot of condensation again. We really didn't find it until we programmed the protection relay to trip on imbalance instead of just warn, that way the water didn't evaporate and we could see the cable leakage with the megger.

 

[Sahib]

I think insulation breakdown of the motor is imminent Meggaring the motor may reveal any defects.

 

[Jraef]

I think insulation breakdown of the motor is imminent Meggaring the motor may reveal any defects.

I would not think that would be intermittent though. Might be, but it's hard to imagine how that intermittent occurrence would take place. As I see it, the insulation is either breaking down or it isn't. Heat maybe?

 

[Phil Corso]

Drbond24...

In addition to observing ammmeters connected to feeders on the same 4kV bus, are you able to view upstream ammeters for additional info? For example the Xfmr's secondary connection or the its primary
line?

Regards, Phil Corso

 

[gar]

130507-1553 EDT

drbond24:

Is the load a wye or delta? Same question for the source?

What is the duration of the spike? You mentioned 1 to 2 seconds. What is the time resolution of your current recorder.

Is the spike always on the same phase?

I am assuming both the source and load are wyes. Thus, you might not see any current change on the other two phases. It would be important to sense the voltage on the problem phase at the time of the spike and correlate with before and after the spike. A 1 to 2 second spike is a very long disturbance when there is no change on the other phases.

For the current spike mentioned the voltage also has to increase, or the load has to become a lower impedance.

.

 

[beanland]

Intermittant Fault

Intermittant Fault

I would be very suspicious that there is an intermittent fault. It could be that voltage rises just a little to initiate the arcing fault. Then the trace is burned up or the voltage drops and the arc extinguishes. Similarly a turn-turn fault could cause a slight increase in load and then burn itself clear. Given the cost/value of the motor, performing tests to identify the cause is highly recommended. Bad things can happen if the fault is ignored.

 

[GoldDigger]

Similarly a turn-turn fault could cause a slight increase in load and then burn itself clear. Given the cost/value of the motor, performing tests to identify the cause is highly recommended.

A continuous turn-to-turn fault would be pretty easy to detect by doing an impedance (not resistance) measurement of each coil.
A turn to frame or phase fault which is intermittent could probably be found with a megger (but it would have to produce more than the 4Kv line voltage. Maybe a hi-pot tester would be more suitable?)
But how do you get an intermittent turn-to-turn fault to show itself? A test stand which can apply higher than normal voltage to the unloaded motor?

Or would you just call in someone who specializes in MV motors and go from there?

 

[drbond24]

The motor was meggared after it tripped and was good.

In addition to observing ammmeters connected to feeders on the same 4kV bus, are you able to view upstream ammeters for additional info? For example the Xfmr's secondary connection or the its primary
line?

There are analog ammeters and voltmeters for everything, but they are not connected to a datalogger. You'd have to be staring at just the right one at just the right time to catch anything happening.

Is the load a wye or delta? Same question for the source?

What is the duration of the spike? You mentioned 1 to 2 seconds. What is the time resolution of your current recorder.

Is the spike always on the same phase?

I am assuming both the source and load are wyes. Thus, you might not see any current change on the other two phases. It would be important to sense the voltage on the problem phase at the time of the spike and correlate with before and after the spike. A 1 to 2 second spike is a very long disturbance when there is no change on the other phases.

For the current spike mentioned the voltage also has to increase, or the load has to become a lower impedance.

Source is definitely wye. I am not sure about the load, but I'd be surprised if it wasn't wye as well.

Time resolution on my recorder is terrible. It scans once per second I think, and it only catches one scan with the higher current. One scan is normal, next scan is high, next scan is normal again. The spike is lasting long enough to get caught, so it isn't terribly fast, but it is less than 2 seconds or I'd catch it on more than one scan.

I only have an ammeter on phase 2. I have no idea what the other phases are doing.

 

[drbond24]

More information: The transformer that feeds the buss for the fan in question feeds another buss as well. I have noticed on our datalogger trends that the amp spikes on this fan correlate to starting other equipment on these two busses. There are two busses, A and B. The fan is on buss A. Transformer AB feeds both busses. Anytime another motor is started on either buss A or B, there is an amp spike on this fan (the fan runs continously but other equipment stops and starts variably). The magnitude of the amp spike does not appear to depend on a specfic piece of the other equipment or even a motor size. Other motors ranging from 400 hp to 1250 hp starting will cause the spike, but the larger motor starting does not necessarily create a larger spike in the fan.

 

[gar]

130508-1233 EDT

drbond24:

This is a whole new story.

If when the other equipment starts there is a source voltage drop, then with constant load on the fan motor, which will be the case for a short time, then you would expect the fan motor current to increase.

You need more complete instrumentation. Need to log all three phases, voltages to neutral and each phase current, with a resolution of possibly 0.1 second. Would be useful to use a power monitor that included the voltage and currents.

The couple second duration would possibly correlate with a motor start up if the inertia load is not too large.

.

 

[drbond24]

We just started noticing this current spike in the one fan motor the last week or so. It has been wired the same, along with all of the other equipment on these two busses, since 1973. Something has changed. What would amplify the effect of another piece of equipment starting?

 

[ATSman]

Motor Protection

Motor Protection

Like others have mentioned, we need more data.
A motor of this size and voltage should have all kinds of protection: Phase differential, phase imbalance, phase overcurrent, undervoltage, overvoltage, groundfault, etc.
You mentioned a differential relay tripped the motor off line. If my memory serves me, this relay looks at current into a winding and compares it to the current out. If there is an intermittent internal winding fault then it should also show up as a ground fault which is usually set more sensitive than a phase current device. Are there any other protective devices showing a trip target?

You may end up having a line disturbance analyzer installed (Dranetz, Fluke, etc) to look at voltage and current events to track this one down.
I recently worked on a 2000A, 12KV synchronous motor that was tripping off line. It turned out to be a faulty (misfiring) SCR in the DC rotor field circuit.

 

[gar]

130508-2349 EDT

drbond24:

You need better instrumentation or you are likely to spend a lot of time guessing.

.

 

[drbond24]

I understand. Most of my instrumentation is 40 years old just like the equipment. I will see if we have, or can acquire, a three phase datalogger to use on this fan. Thank you all for the advice. If I can manage to learn more I will post it.

 

[Sahib]

I would not think that would be intermittent though. Might be, but it's hard to imagine how that intermittent occurrence would take place. As I see it, the insulation is either breaking down or it isn't. Heat maybe?

Could any partial discharges in the solid insulation of the HV motor be associated with observed current spikes?

 

[rcwilson]

Could any partial discharges in the solid insulation of the HV motor be associated with observed current spikes?

That is doubtful. Partial dsicharge is microamps or nanoamps. That is why it is "partial". When it leads to full discharge, the insulation is completely broken through and full fault current flows, provided there is a good ground return path, the system is not high resistance grounded, etc.

I would suspect:
1. Something mechanical
2. Intermittant loose connection,
3. Supply voltage issue due to another load,
4. Excitation diode issue, if it is a synchronous motor.

 

[Jraef]

We just started noticing this current spike in the one fan motor the last week or so. It has been wired the same, along with all of the other equipment on these two busses, since 1973. Something has changed. What would amplify the effect of another piece of equipment starting?

Given that it WAS working fine for 40 years and has BEGUN to have these intermittent current surges that are now deemed to be associated with starting of other equipment fed from the same transformer, there is a possibility that you are seeing the early stages of a failure in that transformer. It could possibly be indicating a breakdown of the winding insulation that results in a loss of the ability for the transformer to supply surge current. You not only need better instrumentation, I think it's time to call in a NETA certified testing agency to look at your entire system before you have a catastrophic failure.

When the bean counters whine about the cost, remind them of the cost of an unscheduled and protracted loss of power for whatever is being done by this equipment. With large MV motors, this is often the heart of an operation and a catastrophic breakdown can put a company out of business!

 

[weressl]

All:

What would cause a spike in the running current of a 3 phase motor? The motor in question is 5000 hp, 4 kV and powers a fan. It will be running along just fine pulling around 570 amps, then suddenly spike to about 630 amps for a second or two, then return to normal for hours before doing it again. Once last week, one of the motor's differential relays engaged during one of these spikes and tripped the fan. For that to happen there would have to be disparity between phase currents. The ammeter that is recording the spikes is on phase 2 and I do not have an indication on either of the other two phases. So, perhaps my real question is: what would cause a spike in amps on only one or two phases of a running motor?

I checked our amp trends for all other running equipment fed from the same buss and did not see similar current spikes, but I am not entirely sure which phase(s) are being monitored on all of the other equipment. I can find out, but I haven't invested the time to do that as of yet.

Can anyone suggest something else I can check?

This motor/fan is one of three identical units and the other two do not have this problem, but each of them is fed from a separate buss.

Look for mechanical issues, overload, a buffer leak, etc. if you can trend the pressure and corelate it to the overcurrent then you found your culprit. (A loose flap opens up, drops the pressure, then it falls back again when the pressure demand can not be sustained by the fan.)

 

[weressl]

We just started noticing this current spike in the one fan motor the last week or so. It has been wired the same, along with all of the other equipment on these two busses, since 1973. Something has changed. What would amplify the effect of another piece of equipment starting?

Other loads were added, or operational mode had increased the base load significantly?