Motor Wiring Design Questions

[cjb80]

Hello,

I recently found this forum and I have been reading through books on motor and ESC designs. I am not familiar with the forum rules and I am a little concerned by the lack of new topics (i.e., most topics are quite old but keep getting new posts in them); so feel free to nuke this post if this is inappropriate. I had a few questions that I haven't found answers to:

(Note: all of these questions are in the context of an axial flux motor design. I don't know if that has any impact on the questions or not)

1) I am trying to understand the design differences and benefits to 6 phase motors over 3 phase. The best article that I have found explaining 6-phase designs is located here: https://ietresearch.onlinelibrary.wiley.com/doi/full/10.1049/elp2.12133 (there is a wiring diagram in there as well; I assume the A+ and A- actually use the same wire and the difference is the timing of the gates). Is this a typical 6-phase design or is this different/novel? If this is different, than could someone explain or point me toward a more typical wiring design?

1a) If the linked-to design is typical, it would seem them that one could have 9 phases or 12 phases with a sufficiently large (diameter) motor design. The relative benefit would be reduced amperage per phase at the expense of an unusual or complex ESC.

1b) I have been independently trying to imagine other 6-phase designs and what seems like an obvious approach would be to have a multi-stator design where the second stator is offset (rotated) by half a slot width. This would obviously require a second set of writing with independent timing that would ultimately become phase shifted similar to the design shown at the link above. However, this would require a second stator.

2) For many of the motor wiring diagrams that I have seen, it seems like the windings are in series (for a 3 phase design, for example). Wiring the stator windings in series would require larger wire size relative to a parallel design (e.g., with a thicker gauge "bus" wire around the motor perimeter). Is there a good reason not to do this?

3) In one of the books that I am reading there is a diagram showing a relative phase shift between the PM rotors in a two rotor, interior stator design. However, the book does not discuss why (or why not) one would decide to pursue this design. (An example of this is shown in Axial Flux Permananet Brushless Machines, pg 11, variable "Xo").

4) I have seen numerous posts using VESC Tool. Is VESC Tool a common tool used by many different controllers? Is it up to the controller to support the software? I am trying to understand the ecosystem there.

Thank you and sorry if any of these questions are redundant.

 

[amberwolf]

I recently found this forum and I have been reading through books on motor and ESC designs. I am not familiar with the forum rules and I am a little concerned by the lack of new topics (i.e., most topics are quite old but keep getting new posts in them); so feel free to nuke this post if this is inappropriate.

Most of the "motor" posts end up in other technical forum areas, often buried within someone's build thread or troubleshooting thread. But there are not many people that go into motor design or theory, so there's not a lot of threads about it in any of the subforums.

I don't have answers to all the questions, so anything I didn't quote below is something someone else will have to chime in on.

1a) If the linked-to design is typical, it would seem them that one could have 9 phases or 12 phases with a sufficiently large (diameter) motor design. The relative benefit would be reduced amperage per phase at the expense of an unusual or complex ESC.

I think that the more phases, the more complicated the timing, and the more perfect the software would need to be to prevent incorrect current flows. You also then have more hardware sets that have to match each other (if you want them all to perform equally).

I'm not sure how much advantage there is to increase the number of phases this way; is there a specific goal you have that doing this would assist with?

There used to be a 5-phase motor that did things differently than the multiple-of-three types, and that apparently had some advantages that made it worth doing. There are some threads about the Falco motor system from back when it was an available system. There's one member here that is a big fan of them (I've never had the opportunity to test them out).

1b) I have been independently trying to imagine other 6-phase designs and what seems like an obvious approach would be to have a multi-stator design where the second stator is offset (rotated) by half a slot width. This would obviously require a second set of writing with independent timing that would ultimately become phase shifted similar to the design shown at the link above. However, this would require a second stator.

If your goal is torque, you'll get more usable torque per weight of motor, more efficiently (less end turn losses) by just using a single wider stator with non-offset slots.

The design you need will be determined by your goal and limitations.

2) For many of the motor wiring diagrams that I have seen, it seems like the windings are in series (for a 3 phase design, for example). Wiring the stator windings in series would require larger wire size relative to a parallel design (e.g., with a thicker gauge "bus" wire around the motor perimeter). Is there a good reason not to do this?

If you parallel all the windings you have a very low resistance and very low inductance motor, and it is hard to drive those; they tend to blow controllers up, especially if there is any imperfection in signal timing or position detection, currents, etc.

4) I have seen numerous posts using VESC Tool. Is VESC Tool a common tool used by many different controllers? Is it up to the controller to support the software? I am trying to understand the ecosystem there.

VESC is specifically it's own software and hardware set; it is an FOC controller design for three phase motors, originally designed for RC motors but expanded over the years to include many kinds. Since it is open source, there are many many variations, including a number of them designed here on ES (most of those threas are probably in this subforum).

So the VESC tool works on compatible versions of the VESC firmware running on compatible versions of the VESC hardware.

VESC Project is the main site for the project, but there are numerous repositories around the web of various versions of the code for subprojects.