Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

Oversized VSD speed accuracy 4

Status
Not open for further replies.

123MB

Electrical
Apr 25, 2008
265
0
0
AU
Hi All

Was sure I would find this in the search, but no luck.

Have an application where we have a 30kW motor with a 45kW vector drive. The motor is a fan which is controlling flow rate and requires good accuracy at very low shaft power (Its duty points are from 20kW to 3kW shaft power).

I'm almost embarrassed to ask this as it seems so simple but I can't work out how to determine the impact of the larger drive on the accuracy of the flow rate control.

I know that for vector drives, it is not good to oversize the drive too much since the accuracy of the current measurement is reduced, but it just doesn't make sense to me considering these drives should be able to operate in the full power range.

For example, to put it simply, if a 100A drive was used on a 100A motor, it would still be required to have sufficient accuracy and performance at power conditions resulting in 5A current. So why would there be any issues in connecting, say, a 50A motor to the 100A drive, the accuracy at 5A would still be the same.

Sorry if this is very simple I am just not getting it and I think I've had too many coffees.

Cheers

 
Replies continue below

Recommended for you

Accuracy of application control (output/measured variable - in this case air flow or presure) it's one problem and vector (or similar name) drive/inverter it's other even if the first may be influenced by second one.
Vector drive use induction motor model, some parameters are measured in autotune procedure (Rs, Xs) and other are estimated (rotor values); if no sensor speed, this is estimated also with negative impact to motor model and lower dynamic performance especially at low speed.
Main problem is not current sensor accuracy in a wide range but motor model difference in high range power/current.
Inverter firmware is optimized/designed for a narrow power range motor model.
So more than 1:3 power range may induce issues in motor control in vector mode.
The first one issue tie to control loop optimization/tunning... either use inverter controller (PID or similar) or separate one, both need same attention in tunning parameters.
 
Great response. I’ll just throw in that most drives are fine with the motor being 50% of the drive rating. It’s when it gets lower than that where this becomes potentially problematic (in terms of maintaining accuracy). But I have used drives as much as 10x the motor size (100HP VFD on a 10HP motor) and nobody has complained. Yes, the motor model is so off kilter that the speed accuracy was likely no better than 1-2% at low speeds, but this was on a pump (in an emergency) and we couldn’t detect any difference.

I doubt that the difference in accuracy you would experience on a fan would be noticeable either by the way.


" We are all here on earth to help others; what on earth the others are here for I don't know." -- W. H. Auden
 
Thanks guys for the quick responses

We have an option here of going for a 45kW drive or a 22kW drive (both vector drives) for the 30kW motor - with our operating point so low in the power range I wonder if the 22kW would provide better control - I think this is really the question I am struggling with.

However, overall, I don't think there should be any issues operating at low power with a 45kW vector drive with the 30kW motor.
 
I've used drives that simply told me to piss-up-a-rope when I I tried to move them from V/Hz to vector and the motor was 50% of the drive. It would not allow that parameter number to be set.


I'm also not following your logic. If you use a motor 50% the size of the drive the controller is going to be faced with half the signal resolution to work with. Same as using a 200A shunt on a 100A circuit, you're throwing away half the information that would otherwise be available. The signal-to-noise-ratio will be halved.

Keith Cress
kcress -
 
Yeah, I’m not running a smaller motor on a larger drive, I’m running the motor at duty point which is at very low power and expecting very good speed regulation. It’s a 30kW motor operating at a 2kW duty point with a 45kW drive
 
123MB said:
It’s a 30kW motor operating at a 2kW duty point with a 45kW drive
That's fine. Don't get hung up on the duty point, the issue is that you have the motor model of a 30kW motor being run by a VFD designed for 45kW, that's well within the acceptable range of any drive I am aware of, so it should be capable of SVC level accuracy down to around 1-2Hz.

The problems you will have with trying tp use the 22kW drive will be that the drive will not be capable of being programmed for a 30kW motor, so you will have to "lie" to it by telling it you have a 22kW motor, and that will likely affect the motor model more than the other direction. It will still work, I've done that too, but if you are concerned about accuracy at the low speed, this would be MORE of a problem.


" We are all here on earth to help others; what on earth the others are here for I don't know." -- W. H. Auden
 
If you really need speed accuracy, add encoder feedback. Given a motor selected for the service, it is possible, with encoder feedback to get full torque at zero speed, with no drift.
 
Yes, but for a fan?


" We are all here on earth to help others; what on earth the others are here for I don't know." -- W. H. Auden
 
At some point slowing this down you will start to get motor cogging.
You won't notice it in the fan output.
But you notice it when you start getting torsional fatigue shaft failures.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed
 
Use encoder feedback if you really need to operate at
OP said:
so it should be capable of SVC level accuracy down to around 1-2Hz.
At that low speed small changes in friction external to the fan could cause fairly large flow rate changes, so this might end up not being the whole story.

Ed's concerns are valid. they can be designed around, but that gets expensive. I would not use encoders for a fan application as it is better to design out the need for very low speed fan operation.
 
Status
Not open for further replies.
Back
Top