DD Hub motors: Why only 23 pole pairs?

Get all your technical information about electric bikes here.
Post Reply
User avatar
mrbill   1 kW

1 kW
Posts: 499
Joined: Jun 10 2008 6:47pm
Location: Silicon Valley, California
Contact:

DD Hub motors: Why only 23 pole pairs?

Post by mrbill » Feb 22 2018 7:00pm

Over the last couple of years I have been experimenting with direct-drive (DD) hub motors on my bikes, building them into ISO 559 wheels (26" MTB wheels). Since I live in an area where there are many hills, I have found the advent of Statorade, Hubsinks, and motor braking to be beneficial.

My recent experiments have been with the Edge1500 hub motor as sold by LA E-bike. I laced one into a 559 wheel, then tested it on a specific loop ride in its stock configuration, a second trial with Statorade added, then a third trial with Hubsinks added. My fourth trial has taken another instance of this motor and laced it into a smaller ISO 406 (20", BMX) wheel, running the same test course and conditions. The details of my experiments using this motor can be read here:

https://mrbill.homeip.net/hybridBike.ph ... 500Testing

In summary, the net efficiency of this motor laced into a 406 wheel appears to be about 20% greater, and motor overheating, even with the insufficient amount of Statorade in my 406 wheel, has not yet occurred. Some of the of the efficiency gain is due to the motor operating at a higher electrical RPM, but most of the efficiency gain appears to be from the greater amount of regenerated energy that can be captured by the system.

In a larger wheel the motor spins more slowly, leading to lower energy recapture with a given controller. To achieve the same braking force at slower speeds, the controller transitions from regenerating to plugging as speed slows from 30 to 12 kph. Using a 559 wheel, motor braking at speeds over 30 kph is regenerative, while motor braking at speeds under 12 kph is mostly plugging. Plugging consumes energy from the battery, converting it into heat in the motor instead of charging the battery.

That leads me to believe that a motor with, say, 29 pole pairs laced into a 559 wheel would appear to the controller indistinguishable from the 23 pole pair motor laced into a 406 wheel with both configurations enjoying the same improved efficiency and lower operating temperatures.

My experiments suggest a couple of questions that I submit to the E-S community:

1) Why do we not see a wide selection of DD hub motors with more than 23 pole pairs, e.g. 28-30 pole pairs, that allow such a DD hub motor in a 559 wheel to enjoy the same efficiency and lower operating temperatures one sees with an otherwise identical 23 pole pair motor laced into a 406 wheel?

2) Why does there exist a lower practical RPM limit (~20-30kph in a 559 wheel) at which a controller can recapture regenerative energy yet still maintain high braking force without plugging and generating heat? Is this a controller design limit, a device design limit, or a Law of Physics?

Thanks.
Bill Bushnell
http://mrbill.homeip.net/
Bike build thread can be seen here.

John in CR   100 GW

100 GW
Posts: 13510
Joined: May 20 2008 12:58am
Location: Paradise

Re: DD Hub motors: Why only 23 pole pairs?

Post by John in CR » Feb 22 2018 11:19pm

While you can get some extra regen with the smaller wheel, your big efficiency gain is coming from less energy used during acceleration. For the same acceleration you use less current due to the lower gearing of the motor, and/or you use greater acceleration using the same current levels but for a shorter duration.

If you want more pole pairs, then get yourself a MagicPie, but you'll quickly find they aren't more efficient. High pole count hubbies dominate the market, because they're cheaper to make due to less stator lamination steel, less copper, and less magnet for the same torque as the magnetic circuit is working at a greater radius.

For 10 years I've had the highest performing hubmotor driven street ebikes. That's because I learned the lesson very early about smaller wheels on hubbies. For the past 6 years I really stepped up the performance using superior high efficiency hubmotors, and they're only 10 pole pair count motors. Those motors are no longer in production, because they were too costly to make. Premium materials and a better design gives them unheard of efficiency, which enables run a bone stock hubbie at 16kw peak power limits pushing an all up load of 400lbs at up to 70mph with no heat issues. Despite the mountainous terrain where I live.

User avatar
fechter   100 GW

100 GW
Posts: 12907
Joined: Dec 31 2006 3:23pm
Location: California Bay Area, USA

Re: DD Hub motors: Why only 23 pole pairs?

Post by fechter » Feb 23 2018 3:05pm

I don't think the number of pole pairs is going to have a dramatic effect on the performance, though obviously there must be some optimum number. Using a smaller wheel is effectively gearing the motor lower and allowing it to spend less time in the inefficient range.

If you played with the number of poles and number of turns in the windings, you could optimize things for a given wheel diameter, but I don't think you could ever get a 26" wheel to perform as well as a 20" wheel with that motor size. If the motor had a larger diameter stator, then you might be able to match the performance. Saturation of the iron is the limiting factor.
"One test is worth a thousand opinions"

User avatar
mrbill   1 kW

1 kW
Posts: 499
Joined: Jun 10 2008 6:47pm
Location: Silicon Valley, California
Contact:

Re: DD Hub motors: Why only 23 pole pairs?

Post by mrbill » Feb 24 2018 11:59am

Hi fetcher and John in CR:

Thanks for your replies.

I understand that a given DD hub motor in a smaller wheel is effectively "geared down", and runs more efficiently at lower speed. In my data this is reflected in the total watt-hour figure: 867 wh declining to 815 wh to move the bike around the loop course. This figure read off the CA3 does not include the effect of regeneration and suggests that the effect of "gearing down" the motor saves 6% of the energy.

The wh/km figure does include the effect of regeneration. In the larger (559) wheel the loop could be completed at 11.0 wh/km, but using the smaller (406) wheel the loop can be completed at 9.3 wh/km, a 15% energy saving.

This suggests that more than half the savings is from the additional regeneration opportunities available with the smaller wheel. I know that I observe much cooler temperatures when using the motor to hold the bike speed on a steep (>12%) downgrade, and I observe higher peak regeneration currents. This also suggests to me that the way to achieve higher regeneration currents is to maintain higher electrical RPM, at least with my present controller.

It seems to me that a motor such as the Edge1500 could be designed with more poles and correspondingly more stator teeth while holding other design attributes constant (lamination thickness, wire thickness, etc.) to accomplish the same effect in a large wheel that I have observed in a small wheel, to electrically gear-down the motor.

My question was whether anyone had considered designing such a DD hub motor, one that is optimized for hilly terrain while keeping stator width 35mm or less (so that it can fit a standard 135-140mm dropout) and without demanding several kw of power to achieve the needed torque, one that could be laced into a standard-sized wheel (559-622) yet enjoy the efficiency gains one sees in the smaller wheel.
Bill Bushnell
http://mrbill.homeip.net/
Bike build thread can be seen here.

Post Reply