VFD's will not run a PM motor, they are designed specifically for Induction style motors, trust me, I have tried it!
That pretty much includes most.

 

You must use a variable frequency drive with a PM mode.

https://www.yaskawa.com/products/drives/industrial-ac-drives/microdrives/v1000-drive

• Induction motor (IM) or permanent magnet motor (PM) operation

 

Please let me say again that I appreciate the responses which have been made since I last wrote to this forum. On the one hand I feel that this quest is justified, because I know from experience that this particular F&P motor design is capable of good performance in a washing machine. I wrote to Fisher and Paykel asking for advice on how the F&P controller might be adapted for my intended purpose. I also included the following request:

“Another solution might be if I could obtain a general-purpose, variable speed ESC BLDC controller which uses 240 volt AC as its power source, which is designed to operate at the power level of the Aquasmart washing machine motor. I sincerely hope that you will assist me in this quest.”

On the other hand, I feel at least a little disappointed at the reply from F&P, which was brief and direct, “Unfortunately, we are not able to assist you in this query.”

Yaakov commented, “See this post here for one scheme powering an F&P motor”, but that particular website acknowledged that the Coil configuration needed to be changed, for battery operation, and that the results were inefficient. Yaakov went on to say, “Unfortunately, it appears you are not in for an easy time whatever you do”, and in a subsequent post he said, “Yes, it seems that driving F&P motors is widely desired but rarely done well”. I think these words are very true.

The suggestion by Shortbus, to use a breakout board which uses a MC33035 integrated circuit may be a possible solution. Since this is a driver board which requires additional components to be added to make a functioning controller, I wonder whether anyone can identify a suitable circuit diagram, and component list, etc, which could use this approach?

MisterBill2 suggested “to use one of the variable-speed drives intended to run a 3-phase motor from single phase mains”. I contacted Automation Direct, the company he suggested, but they advised that the controllers which they sell would definitely not be suitable for use with the F&P permanent magnet motor.

MaxHeadRoom said, “VFD's will not run a PM motor, they are designed specifically for Induction style motors, trust me, I have tried it!”

Nsaspook included two links for internet sites, the first for a variable frequency drive, and the second for the YASKAWA AC Drive-V1000. The first link is for a Non-Enclosure Variable Frequency Drive (VFD), but that website says, “this non-enclosure VFD is 1 phase input, but the output is 3 phase, so it only suitable for 3 phase induction motors”, so that does not seem to be the way to go.

The other link, for YASKAWA AC Drive-V1000, may be a possibility, but I have not been able yet to contact a vendor for this equipment. Hopefully I will be able to do that soon. The technical manual for this controller, which I have found on the internet, is 512 pages!!! I hope that I do not have to read all of them!

I have been persisting with my enquiries because I really hope that a practical (and affordable) option will be discovered. There certainly have been many enquiries, over a number of years now, by people who have hoped for a result. I wonder whether any of them have had a good success story. It would seem that the F&P BLDC motors have been fairly easily adapted for use as alternators which are wind-powered, but that is not what I wish to achieve. I do truly appreciate the efforts made by the various contributors to this discussion, and hope that we can work out a solution. I wonder where each of you live, who have responded so far. I am in Melbourne, Australia.

 

Perhaps I was unclear, and even goofed. VSDs, (Variable Speed drives.Able to output different waveforms. VFD is not the best description. You got me on that one, Max.

Really, it should be possible to use a portion of the original drive system to run the motor, since it already has the power supply and driver circuits intended for that motor. The challenge will be upstream from the driver circuits, in the control area. If the circuit board is singe-sided then tracing the circuit will not be so very bad, but if it is double sides or multi-layer, it would be a pain.
Possibly the circuit is available from one of the many websites claiming to provide circuits. Or maybe an alternative driver circuit can be located.

 

Perhaps I was unclear, and even goofed. VSDs, (Variable Speed drives.Able to output different waveforms. VFD is not the best description. You got me on that one, Max.

Really, it should be possible to use a portion of the original drive system to run the motor, since it already has the power supply and driver circuits intended for that motor.

VFD, Variable Frequency Drive sums it up exactly?
Also universally used throughout the industry.
I contacted one Manuf. of VFD's, namely Hitachi and they stated a VFD will not run a PM 3ph motor, out of the box, It would require a specialized adaption on the standard models in order to do it.

 

The suggestion by Shortbus, to use a breakout board which uses a MC33035 integrated circuit may be a possible solution. Since this is a driver board which requires additional components to be added to make a functioning controller, I wonder whether anyone can identify a suitable circuit diagram, and component list, etc, which could use this approach?

Go to the link and read it, down deeper in the link it shows the schematic . You would need to have a suitable DC power supply to power the break out board, then choose the mosfets to match the voltage and amperage of the motor. These breakout boards make things way easier to use the MC33035, which was the way BLDC motors were originally run.

 

The original Motorola app note is still available for the MC33035 circuit, AN1046 (now Onsemi),
The PCT board shown used to be available from Motorola , I still have mine.

 

https://spares.bigwarehouse.com.au/product_info.php?cPath=64658_64721_64733&&products_id=1730231
https://spares.bigwarehouse.com.au/images/files/33736_b2d8dd05fa3d22faec9876e77b901f74.pdf
If the motor is like this type there's nothing really special about this motor, simple 6 trapezoid step 3-phase system. Hall position sensors and phase coils that most general brushless controllers could handle if they could output the voltage/current needed for full speed for a High Voltage Brushless Motor.


A cheap 50VDC controller will likely get it running (in sensor-less mode) slowly.
https://myosuploads3.banggood.com/products/20190508/20190508220426SKU600181WS55-1801.pdf

Examples of HV BLDC controllers:
https://www.aliexpress.com/item/33026024537.html
https://www.ebay.com/itm/233997459843

 

This is why my most recent suggestion is to see how much of the existing board, which includes drivers, power supply, sensor interfaces, as well as extra stuff. is worth investigating some more.

 

There still seems to be confusion out there and/or the mixing of terms, the BLDC motor requires rotor position detectors, typically previously done with three hall effect devices, or if there is an encoder on the motor, via commutation tracks included, and only two stator windings are under power at any one time.
The PMSM, or Permanent Magnet Synchronous Motor, although identical in construction to the BLDC, is driven with a simulated 3phase power source, virtually identical to the VFD used for induction motors.
Often the initial rotor position at power up is determined by exercising the rotor and detecting its position,.

 

This is why my most recent suggestion is to see how much of the existing board, which includes drivers, power supply, sensor interfaces, as well as extra stuff. is worth investigating some more.

Like most things, easier said than done.

It's interesting to see what people have done with the PM motor to alternator conversion. Most rewire the stator coils to something more conventional.
https://www.thebackshed.com/windmill/FPRewire.asp

 

There still seems to be confusion out there and/or the mixing of terms, the BLDC motor requires rotor position detectors, typically previously done with three hall effect devices, or if there is an encoder on the motor, via commutation tracks included, and only two stator windings are under power at any one time.
The PMSM, or Permanent Magnet Synchronous Motor, although identical in construction to the BLDC, is driven with a simulated 3phase power source, virtually identical to the VFD used for induction motors.
Often the initial rotor position at power up is determined by exercising the rotor and detecting its position,.

Electromagnetically there is little actual difference between BLDC and PMSM motors but there are differences in stator and magnet design to optimize for the type of drive waveform to the Back EMF generated. 6 step Square, trapezoidal or sine with sensor or sensorless rotor position detection.

https://forum.allaboutcircuits.com/threads/pic32mk-mc-qei-example.150351/post-1550621

 

Those motors CAN be driven open loop, but that is a bit less efficient. Coser to a stepper at that point, but stil driving.

 

Electromagnetically there is little actual difference between BLDC and PMSM motors but there are differences in stator and magnet design to optimize for the type of drive waveform.

Aware of it , see #30
Attempt to explain the difference in the commutation.

@MisterBill2 IF BLDC, cannot be driven open loop

 

Aware of it , see #30
Attempt to explain the difference in the commutation.

@MisterBill2 IF BLDC, cannot be driven open loop

Can't is too strong a term here. You statement on #30 needed context.

A BLDC can be driven open loop (position feedback to drive waveform) at slow(er) speed, it's a matter of stability under dynamic load conditions. Most sensorless systems start with no EMF or positional information. The trick to to provide adequate electromagnetic torque (current) to maintain rotor lock within a few degrees of the rotating waveform as V/f ramps while keeping the commutation delta angle slightly retarded (stator current lags the back EMF) from zero. Mathematically it's not too complex to build a static motor load model as I've done it several times with different systems I've designed and built.

Not optimal for most applications but it works just fine for very slow speed 3-ph sinewave drive systems that need very smooth movements at high torque. This might mean a specialized controller type but I can design and built one from scratch.

 

Interesting video, shows the limitations, but I think it would be even more illuminating if a significant load was applied.

 

Interesting video, shows the limitations, but I think it would be even more illuminating if a significant load was applied.

It's more a factor of dynamic loads causing the drive waveform delta to lead making the system unstable. A static load can be calculated into the motor control program as a mass/speed/inertia calculation in the motor model for start/run/stop drive delta angles and the required drive currents. Closed-loop positional feedback at the needed resolution for the drive waveform is usually better but it's not a absolute requirement of rotation. Open-loop is like controlling a dog on a long leash. It works but you need to know the nature of the beast for precise control as things go faster. For the software system we need an adaptive observer of the position and speed of the rotor. The observer's task is to calculate the position of the back-EMF vector so other functions can make stabilizing corrections to the current drive waveform until the system is calculated to be stable at X set-point.

 

All of my application experiences have been with CNC positioning control, so accuracy is imperative.

 

All of my application experiences have been with CNC positioning control, so accuracy is imperative.

Your positional control loop to zero encoder read-backs counts can be closed-loop ensuring accuracy while the drive control loop (for motion) to correct for positional errors from the desired position can be open loop. That's not usually the case because if you have high resolution encoder data it would be foolish IMO not to let the adaptive observer use that in calculations of speed and position.

 

All of my application experiences have been with CNC positioning control, so accuracy is imperative.

CNC positioning is about the most demanding of all position control applications in the production machining realm.
IC fab position control is a whole different world, that is certain.