DC VSD problem

[Besoeker]

Today I got a call from a paper mill that has a DC drive system with about 15 ndividual DC drives all on the same (paper making) machine. For those of you who have had experience on such systems, you will know that it's quite an exacting application. Typically, speed holding has to be within 0.1% otherwise the paper breaks and production stops. The motors are 600Vdc and are mostly from 100kW to 300kW. The machine, as a whole, usually runs at maximum speed or close to it so all the motors run at close to maximum speed. That's the background.

So to the reported problem.
One motor was running at an armature voltage 150Vdc instead of the usual 550V - 600V.
The machine was in production so this particular motor had to be running at the correct speed. I asked the electrical guy if he had checked the motor field voltage. He had. It was about 50% higher than normal.

Any thoughts on what the problem might be?

 

[beanland]

Help me...

Help me...

Isn't the torque of a DC motor controlled by the field and the speed by the armature voltage?

Is it possible that the motor was running at the required speed and torque for the load and application but that the motor was "misapplied" so that full typical armature voltage was not required?

It has been a few years for me and DC motors...

 

[Wire_nutz]

First check the field fuse on the drive of the motor that has the issue. If the fuse is ok, check the voltage at the field terminals of the other drives that are of the same horsepower as the motor that is having the problem, and compare voltages.

The motors most likely all work together on this machine with a controller. Each motor most likely has a encoder and a tachometer that provides position and feedback to the controller.

The controller monitors and regulates the speed and position to all the individual drives and motors on the machine.

If the field fuse on the drive, encoders and tachs. are ok, it may be a internal drive problem. I suggest swapping or trying another drive.

 

[Wire_nutz]

One other thing, also check the field board on the drive of the motor that has the issue.
The field board can be swapped or changed on some drives also. If this doesn't solve the problem, then swap drives.

 

[Besoeker]

Isn't the torque of a DC motor controlled by the field and the speed by the armature voltage?

Torque is a product of field flux and armature current.
Armature voltage is a product of speed and field flux.

Is it possible that the motor was running at the required speed and torque for the load and application but that the motor was "misapplied" so that full typical armature voltage was not required?

Unlikely.
It's a mature installation (c1991). Running conditions are monitored by a SCADA system and routinely checked at regular maintenance visits. The most recent was September 8th this year. Armature voltage was recorded as 590Vdc.

 

[Besoeker]

First check the field fuse on the drive of the motor that has the issue. If the fuse is ok, check the voltage at the field terminals of the other drives that are of the same horsepower as the motor that is having the problem, and compare voltages.

Field fuse failure would result in loss of field. Field loss is monitored, indicated and in the shutdown circuit. Fuse failure would have prevented the drive from running.

The motors most likely all work together on this machine with a controller. Each motor most likely has a encoder and a tachometer that provides position and feedback to the controller.

The controller monitors and regulates the speed and position to all the individual drives and motors on the machine.

Speed feedback is from an encoder.

If the field fuse on the drive, encoders and tachs. are ok, it may be a internal drive problem. I suggest swapping or trying another drive.

Or maybe a motor problem?

 

[gar]

081127-1543 EST

Besoeker:

I believe you know that motor speed is inversely proportional to field flux, and proportional to armature voltage.

Thus, a 50% increase in field flux would decrease the speed to 2/3 for the same armature voltage. Armature voltage has to decrease to 2/3 to maintain the same speed. But this does not account for an armature voltage that is 26% of its normal value, or a 1/4 ratio instead of an expected 2/3 ratio.


beanland:

Torque is determined by the speed torque characteristics of the load. At a fixed speed the motor will supply whatever torque the load requires at that speed within the capability of the motor system. If the load has a torque requirement that increases with speed and you exceed the torque capability of the motor, for example the motor has a current limiting feature, then the motor will slow down until the load torque equals the capability of the motor.

The output torque of the motor is the product of armature current and field flux. If you double the flux, then 1/2 the armature current will produce the same torque as before doubling the field flux.

Look at it from a power standpoint. If the output power is constant, constant speed and constant torque, then V*I input is input power and approximately equal to output power. Double the field flux and for the same speed the input voltage has to double. But doubling the flux only requires half the armature current for the same torque, we have (2V)*(I/2) equals input power and this is the same as before the flux was increased.

Hopefully i did not make a mistake in this description. These are approximate relationships for the operation of a real shunt wound DC motor and are very useful for a basic understanding. Because of differing losses the descriptions have to be considered approximations.

.

 

[gar]

081126-1726 EST

Besoeker:

I started my comments before your responses, then I went to dinner, came back, and finished without looking for other comments.

It would seem likely there is more than than a 50 % change in field flux, or a change in turns on the armature, unlikely (impossible).

Something does not compute.

Does the motor control have the ability to dynamically adjust field excitation?

.

 

[cadpoint]

I'm going to say that it is a stator, a brush, or the commutator, one of these are failing!

Nothing against the electrician, but what was isolated to read the windings? :roll:

 

[gar]

081127-1918 EST

cadpoint:

Fundamentally a DC motor operates on the basis of
e = K * dF/dt
where
e is the induced voltage in a coil,
K is a constant including the number of turns linking the flux
dF/dt is the rate of change of flux (proportional to speed)

A DC motor when rotating creates a counter EMF that opposes the applied voltage. The difference of the two voltages is the voltage drop in the internal resistance of the armature from the motor load current. In the normal operating region this is small and as a first order approximation can be assumed zero. Under these conditions counter EMF equals applied aramature voltage. Therefore, motor speed is directly proportional to armature voltage for a given field excitation.

The implication in the first post was that if any one motor was running at an incorrect speed that paper breakage would occur. I do not know if this means that if the motor was not supplying power and was just a small frictional load that the paper would break if the motor was being mechanically driven by the paper.

If speed is held constant and field flux is reduced, then armature counter EMF will be reduced, and thus applied voltage will be lower. This could occur if a high resistance developed in the field circuit. So maybe the field voltage went up because of high resistance in the field circuit thus lowering the flux and therefore the counter EMF for a given speed. Looks like current measurements on field excitation should be checked because we may not be able to assume flux increased simply because field voltage went up.

The following statement in my earlier post was wrong.
"Thus, a 50% increase in field flux would decrease the speed to 2/3 for the same armature voltage. Armature voltage has to decrease to 2/3 to maintain the same speed. But this does not account for an armature voltage that is 26% of its normal value, or a 1/4 ratio instead of an expected 2/3 ratio."

It should have said: Armature voltage has to increase to 3/2 to maintain the same speed.

Conclusion: Looks like field excitation was reduced even though the field voltage went up.

.

 

[Besoeker]

The implication in the first post was that if any one motor was running at an incorrect speed that paper breakage would occur. I do not know if this means that if the motor was not supplying power and was just a small frictional load that the paper would break if the motor was being mechanically driven by the paper.

The motor in question is at the "wet end" of the machine where the paper has no tensile strength, so being pulled round by the paper isn't really a consideration so it had to be running at the correct speed. And, even if it was being pulled round, as you earlier and correctly noted, back EMF would/should be proportional to its rotational speed. But it wasn't.

If speed is held constant and field flux is reduced, then armature counter EMF will be reduced, and thus applied voltage will be lower. This could occur if a high resistance developed in the field circuit. So maybe the field voltage went up because of high resistance in the field circuit thus lowering the flux and therefore the counter EMF for a given speed. Looks like current measurements on field excitation should be checked because we may not be able to assume flux increased simply because field voltage went up.

Conclusion: Looks like field excitation was reduced even though the field voltage went up.

I came to a similar conclusion. I'm pretty sure it's a fault with the motor field.
I asked the guy to compare field resistance on the operational motor with their spare. That sounds simple but, at $20,000 per hour lost output, it would take something like an act of Parliament (or Congress:smile to get production to agree to an outage.

I don't know whether the problem has been resolved but I guess the machine is still running. We have a 24/7 call-out contract with a dedicated support line and I don't think that has been used. It's now about 5 am here. I'll give them a call in a while.

 

[gar]

081128-0803 EST

Besoeker:

I would assume you could measure the field current and voltage to determine if the field circuit resistance is different than other identical motors, and thus no machine shutdown except to fix the problem.

Are you in the UK?

.

 

[Besoeker]

081128-0803 EST

Besoeker:

I would assume you could measure the field current and voltage to determine if the field circuit resistance is different than other identical motors, and thus no machine shutdown except to fix the problem.

That sounds obvious but...
...and there had to be a but, or several, of course.
The first is that, like most electrical panels here, there is a door interlocked main isolator. You can't gain access without first turning off the power.
It would be possible to take the cover off the motor terminal box but, given that it is in a wet area and the risk of contact with live conductors, it would contravene the electrical safety rules for most companies. Live working usually requires a live work permit and it probably wouldn't be granted.
In any case, I don't know if there are any other identical motors on the machine.
So yes, it would be an obvious thing to measure and would probably pinpoint the problem.

Are you in the UK?

Not only am I in the UK, but I'm a Brit too.
Here, health and safety seems to have been taken to levels that are .......well, getting to the unworkable silly levels.
As an example:
One of our customers has three of our 600kW 12-pulse inverters in a water pumping station. Each has six doors, three at the front, three at the rear and each door has a Castell mechanical locking system. To gain access, you first have to open the supply breaker which is in a distribution centre. That releases a key which is then used in a key exchange box to release the six unique keys for the six doors. In short, there is no way you can get into any panel with the power on.
Well, we now have a contract to put polycarbonate shrouding to cover all internal live parts that you can't actually get to when they are live.
And documentation to do even this simplest of work on site has become totally disproportionate to the job in hand.

Yes, measuring the field current should be a simple job.....

 

[Doug S.]

I got two beers that it's in the motor. Shorted/open winding some where.
I'd guess short in the armature, pulling the voltage down from the drive?

No math, no thinking about, just experience. BUT to verify my inclination, I'd see what the drive output voltage is with no load. (might be as easy as disconnecting the motor)

All that being said I'm curious what the outcome is.


Regards,

Doug S.

 

[Besoeker]

I got two beers that it's in the motor.

Well, thank you. I'll have Guinness extra cold, please.

Shorted/open winding some where.
I'd guess short in the armature, pulling the voltage down from the drive?

Possibly, but I'm inclined to think not. If it was pulling the voltage to that extent it would be sitting in current limit and running at the wrong speed. If that was happening the process, papermaking, wouldn't be working. But it is.

No math, no thinking about, just experience. BUT to verify my inclination, I'd see what the drive output voltage is with no load. (might be as easy as disconnecting the motor)

Not so easy for a number of reasons, the chief among them being that taking the motor out of service to do anything would result in about $20k per hour in lost production.

All that being said I'm curious what the outcome is.

Me too! A shutdown is scheduled for Wednesday 3rd December - if it lasts until then.
Watch this space........

 

[frenchelectrican]

R?v?lez alors le pop-corn.

( Then bring out the popcorn.)

Merci.

Marc

 

[Besoeker]

Well, as I said in post #11 I was pretty sure it was a fault in the motor field circuit.
It was.
A failed "temporary" repair at the motor terminal box.
The motor continued to run because it was a compound machine with both shunt and series fields.
I added this to my report:

The broken field wire connection accounts for the much lower than normal armature voltage.
It may be worth noting that this is one of the old Harland motors. It is a compound wound motor. It has both a shunt field and a series field to produce the excitation. It was the shunt field that had the faulty connection. The series field would have resulted in the 100Vdc observed and allowed continued running albeit at the expense of higher mechanical load on the Forward Drive.

 

[Doug S.]

Well, thank you. I'll have Guinness extra cold, please.

I don't believe you want Guinness from me, I've heard it's much better on the other side of the pond... that being said we have great local stuff that should accommodate you, should you happen upon the neighborhood.

I do understand that "hunches" are frowned upon when real $$$ is on the line, but I wasn't getting paid real $$$ so I went with a hunch. :grin:

Regards,

Doug S.