Motor VFD open loop no-load current is much higher than closed loop no-load current. Why? 2

[edison123]

We ran a 490 V, 2095 KW, 3000 A, 100 RPM Permanent Magnet rotor motor on no-load via a VFD.
In open loop, the motor no-load current is 210 amps while in closed loop (with shaft mounted encoder), it is only 16 Amps.
Why this vast difference between open loop and closed loop currents?

Edit: It's a PM synch motor. Frequency 53.3 Hz. Would this be the reason for such a vast difference between open loop and closed loop currents?

 

[waross]

Will you take a WAG, Muthu?
In closed loop, only enough current is provided to supply the losses.
In open loop, enough current is supplied to develop rated torque whether a load is present or not.
With no load, much of that current will be reactive, to increase the flux density beyond what is actually needed.
That will lead also to an increase in the rel current to supply the increased I sqared R losses.

 

[LionelHutz]

Was it a PM capable VFD setup in PM mode? A standard VFD can do strange things on a PM motor.

 

[edison123]

Lionel - Yes, the custom designed drive is set for PM mode.

Bill - The very idea of VFD is to have constant flux density V/Hz through the entire speed range.

 

[LionelHutz]

You're probably not supposed to run a PM motor in open loop on the VFD.

 

[scalablewindpower]

The back EMF voltage of a PM motor is directly determined by the rotational speed so RPM is determined by the voltage. This requires the VFD to run with the correct Volts/RPM or Volts/HZ ratio. If that is set too high, the motor will consume additional current until the resistive drops equalize the input voltage with the EMF. You might need to look in the VFD manual for advanced features such as automatic tuning with back EMF measurements and even compensation for resistive voltage drops under load.

 

[waross]

Bill - The very idea of VFD is to have constant flux density V/Hz through the entire speed range

As I understand it, the VFD looks at the characteristics of the current drawn by the motor to infer the slip angle of the rotor and set the output accordingly.
I suspect that in open loop mode, the VFD is drawing the wrong inference and is over compensating.
I hope that Marke sees this thread. He may be able to shed some light on what is happening.

 

[jpts]

Did you check what was the used terminal voltage in each mode? I would guess that open loop had much lower or higher terminal voltage than closed-loop mode. Closed-loop control of PM machines often use so-called id=0 control which adjusts the terminal voltage in each load point so that the current will be minimized. Open loop typically uses constant u/f ratio which does not necessarily minimize the current. Usually minimum current is achieved when the terminal voltage is close to back-emf. If the u/f ratio is not correct, it will result in higher current. Depending on the VFD manufacturer, u/f ratio is usually set by defining the field-weakening point (speed and voltage). You could try playing with this parameter to see if it helps

 

[edison123]

V/Hz is set correctly in open loop. Drive is set for PM synch motor mode.

In open loop, the motor had odd electromagnetic noises on top of high current. In closed loop, low current and smooth running.

Guess, PM synch motors are odd beasts.

 

[jpts]

V/Hz is set correctly in open loop. Drive is set for PM synch motor mode.

In open loop, the motor had odd electromagnetic noises on top of high current. In closed loop, low current and smooth running.

Guess, PM synch motors are odd beasts.

Ok, but did you actually measure what was the terminal voltage in both open- and closed-loop operation? Even a rather small difference can mean relatively big difference on current especially in no-load. No-load current of a PM machine as a function of terminal voltage is a V-shape curve, where the minimum current is reached when the terminal voltage equals back-emf.

Even a small deviation in terminal voltage into smaller or larger rapidly increases the current. Closed loop control should automatically adjust the terminal voltage into this minimum current point (which open loop does not)

 

[iop95]

What kind of PM motor is? Interior or surface magnet?
What kind of rotor alignment method was set for open-loop?
Such high current difference indicate an incorrect VFD setup or may be some error/missing in firmware control routine in open-loop.

 

[lmarcantonio]

You're probably not supposed to run a PM motor in open loop on the VFD.

AFAIK unless you are running with the simplest V/f control either and a really slow acceleration ramp you are not supposed to run a PM motor on a VFD without PM capability... and good luck when it goes out of sync for whatever reason

 

[edison123]

jpts - I don't have the voltages in open loop and closed loop, only the currents.

iop - It's embedded round PM rotor.

 

[jraef]

Bill had it correct in the first response. The main difference in closed loop is that by knowing the exact rotor position and any given moment, the current used for establishing stator flux can be minimized to exactly what it requires at that moment, leaving the rest of the available current free to produce torque. When the motor is unloaded, it doesn’t need a lot of torque, so current remains low. In open loop, the drive just puts out the V/Hz required at that speed and even if full flux isn’t needed, it does it anyway resulting in high reactive current. But when measuring overall current, there is no difference so it looks high. The same happens in a standard induction motor, but in the PM motor none of the current needs to go into establishing rotor flux in either mode so the difference is more dramatic.

 

[edison123]

Thanks, Jeff.

Forget PM synch motor, last week I asked to demo a 125 KW drive (reputed Japanese brand) for no-load testing of four normal cage motors with no-load currents varying from 90 to 250 Amps in my shop and the drive dealer team couldn't do it for any of them without the drive tripping on current limit in open loop. They had to lower V/Hz to 1/4th get the motors running. They were talking about V curves for a cage motor !!! Now, they are trying to get the OEM drive guru to do the demo. (Closed loop is not an option for me since we test a wide range of motors and encoders are not possible).

I am this close going back to my MG set VFD. Except for the frigging noise and 50/60 Hz frequency limit, it works like a charm every time with smooth acceleration from zero speed to full speed and zero in-rush current.