Optimum number of POLES on Bigger BLDC motor

[Tek]

I need some help understanding something about brushless DC motors.
For the example I would like us to imagine the motor size is constant, Inrunner, aluminum housing, and between 25-35kW desired output. Something that an electric dirtbike might use.
What are the factors that determine how many poles it should have? Wouldn't fewer poles and bigger magnets etc make for a higher mechanical RPM and why wouldn't that be desirable?

 

[Miles]

First you have to decide the maximum speed of the motor needed, then you can decide on the number of poles based on the acceptable fundamental frequency. For an inrunner, the biggest advantage from increasing the pole count is the reduction in the stator yoke section required. The extra space can be used for more copper or the airgap diameter can be increased.

 

[major]

I need some help understanding something about brushless DC motors.
For the example I would like us to imagine the motor size is constant, Inrunner, aluminum housing, and between 25-35kW desired output. Something that an electric dirtbike might use.
What are the factors that determine how many poles it should have? Wouldn't fewer poles and bigger magnets etc make for a higher mechanical RPM and why wouldn't that be desirable?

You can find volumes on the subject of optimum pole count. As Miles pointed out it does depend on the RPM. Lower RPM typically commands larger aspect ratio (diameter/length) and higher pole count. But higher RPM typically means a smaller motor with lower d/l and fewer poles. Higher pole count means higher frequency per RPM and therefore higher grade thinner (more expensive) laminations or coreless. Also the electronics must accommodate the higher frequency.

In the 25/35kW and up thru 200kW motors for EVs using radial air gaps and interior rotors, 4 to 10 pole machines appear to be norm nowadays with speeds up to 12kRPM range.

 

[Tek]

That's what I'm talking about 10,000-15,000rpm motors. Let's say 25+kW, 10,000+rpm...would axial flux inrunner design be inappropriate for this? How large a diameter could I make a 10 pole rotor?

 

[major]

..... axial flux inrunner design ....

Inrunner/outrunner refer to interior or exterior rotor on radial flux machines if I'm not mistaken. Axial flux machines place the rotor and stator adjacent to each other relative to the axis of rotation. Do you have an example of what you mean?

The design and making of a 25kW, 10kRPM motor suitable for a dirt bike is not an easy task. From your questions, I think you should reconsider attempting a DIY motor or at minimum, start with a more reasonable objective. Try a 100W, 2500 RPM motor first.

 

[Tek]

I hear what you're saying. My question about axial flux motor was merely in reference to the placement of the coils and the magnets. Really I'd like to build a motor for this bike. I have machining, winding of coils and all materials well sorted, I really just need a good design. Is there any open source motor design of 5kw-10kw that I can use?

 

[liveforphysics]

Ultimately, pole count matters very little. 4 poles or 40 poles can both make an excellent high performance efficient motor.

If you're looking to build your own motor, the geometric relationships between the magnet/tooth to not induce harmonics, and to be capable of easy control is the objective that matters so much more.


If you desire maximum efficiency (although not always the lowest mass solution), you want your motor's speed and torque to be as close as practical to the speed range and torque you wish to have at the wheel. There are no loss-less stages of power transfer, or power conversion. Very easily a few gearing/reduction stages can sum to more losses than the entire motor's inefficiency, and any mass that you add to an electric drivetrain that isn't aiding in the conversion of electrical power into mechanical power is parasitic.

Unless you think you have the skills to make a lightweight high-torque rear hubmotor (which is possible, but difficult to design, construct, and expensive), the sweet spot for a dirtbike tends to be a single stage reduction to the rear wheel to minimize unsprung mass by letting the motor be chassis mounted. You can work backwards from what practical sprocket tooth fitments are possible for your wheelsize and application, and work backwards from there to determine your motors rough RPM range, as well as torque requirements.

Those are the first two critical variables to know before you even start to explore anything related to the motors specific topology.

Ideally, you just run your battery power into your wheel and the wheel magically converts 100% of it into mechanical drive power for you. We don't live in an ideal world though, but the path to best results is to try to get as close as you can to that ideal model. This means minimizing/eliminating every source of loss you can, which means think simple and think of any method to remove parasitics.

 

[major]

.........I really just need a good design. Is there any open source motor design of 5kw-10kw that I can use?

Not that I am aware. Suggest you search for and review motor builds by member Farfle over the past year or two. Also member Miles has several design threads. Look at those.

 

[liveforphysics]

.........I really just need a good design. Is there any open source motor design of 5kw-10kw that I can use?

Not that I am aware. Suggest you search for and review motor builds by member Farfle over the past year or two. Also member Miles has several design threads. Look at those.

It's a pleasure to see you donating your time to the ES community Major. I've always enjoyed your input over the years in various EV communities. Part of the beauty of ES is the scale of motors/controllers is so much more DIY-budget-friendly when you're working in a range typically around or below 10's of kW's max rather than hundreds of kW's max, that it tends to enable more ambitious projects to be attempted from a home garage that would otherwise not be considered do to cost/practicalities for larger scale projects.

We also have the pleasure of a generous site admin who understands the criticality of remaining impartial and not-commercially influenced, so ideas and concepts are weighed largely on merit alone.

That's the recipe for awesome free EV technical discussion, development, design etc.

 

[major]

Thanks. You guys do things a bit differently than the big EVs, but the physics remain the same. And the ad-free atmosphere is certainly a pleasure.

 

[Tek]

I've been doing a lot of reading in the past week concerning my inane questions that I've posed to you more knowledgeable folks. I've started to favor an outrunner construction due to the high torque to space factor. Something similar to Farfle motor.
I'm curious about the use of an outrunner placed at the swingarm pivot and holding the pinion sprocket. This way, no unsprung weight but I am a bit worried about the gyro effect on the handling of the machine. Any thoughts from our experienced members?

 

[Miles]

I've been doing a lot of reading in the past week concerning my inane questions that I've posed to you more knowledgeable folks. I've started to favor an outrunner construction due to the high torque to space factor.

Yes, outrunners win on torque to gross volume. Torque to weight, it's more even - especially as the diameter increases.