FluxEst Output Torque Error

cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 

FluxEst Output Torque Error

933 Views
derekcook
Senior Contributor I

Hello, 

I am seeing some error in my motorTorque given as the output of the FluxEst. I have a dynamometer that is controlling to torque on my motor. When I set the torque to 50Nm, the FluxEst output reads 70Nm. When I set the torque to 140Nm, the torque from the FluxEst is 222Nm. The torque seems off by around ~1.5. Is this due to the 1.5 scale factor in the power? Any other reason why the torque would be off from what the measured torque is?

pastedImage_1.png

Labels (1)
0 Kudos
5 Replies

768 Views
linestream-adam
Senior Contributor I

Derek,

FULL_SCALE_POWER is used to normalize the maximum possible power in the system. It won't be the source of that discrepency.

The difference between the KMS measured torque and the dynamometer measured torque is that KMS can only estimate the torque based on the measured motor parameters, device model information, and measured motor currents. It doesn't take any loss of efficiency into account. It also assumes that the motor parameters are perfectly accurate. It is a good rough measure, but obviously isn't a perfect measure of how much torque the motor is producing.

768 Views
derekcook
Senior Contributor I

Hey Adam, 

It looks like my motor torque is high by a sqrt(2). I divided my reported motor torque by sqrt(2) before reporting it to the user. Maybe this is converting RMS to Peak Current?

0 Kudos

768 Views
linestream-adam
Senior Contributor I

I ran back through the code and it is using peak current for the torque calculations. It is possible that the flux value is estimated high in sensored mode. You would need to compare the estimated flux in sensored and sensorless to see how the values compare. 

It won't be a 100% match due to the inability for KMS to measure the efficiency loss in your motor/inverter. KMS estimates the electrical torque while the dyno measures mechanical torque. 

768 Views
derekcook
Senior Contributor I

Thanks for the help Adam. The torque measurement is now within 2 Nm after converting for peak current. I think that's good enough for an estimation. 

0 Kudos

768 Views
derekcook
Senior Contributor I

Thanks Adam for the explanation here. I am using the datasheet values, not the values determined by the GUI. I could try putting the GUI values in and seeing what the torque output I get is.

0 Kudos