In search of the illusive High-Torque, Low-Speed hub motor

Hey guys.

I'm starting a project that is going to require a high-torque, low-speed motor because I'm running a ~30 inch diameter tire.

To throw some numbers out, I'm looking for:
~100+ lb-ft of peak torque. (135 Nm) (90 lbs thrust)
~3000W of peak (motor) power.
~400 RPM (unloaded speed = ~35 MPH for my ~30" diameter tire application)

Justin suggested the GMAC 10T (clutchless grin-special), I may go with that in the end but I have a rightful fear of gears for my application. So I'd rather find a DD motor...

I thought I found the perfect motor:
QS 205 50H V3 3000W Hub Motor w/ a 6.25 Kv winding.



Here's the Grin Simulation showing it can eat 15% grade hills all day long (Cromo 5004 w/ KV adjusted accordingly):

https://www.ebikes.ca/tools/simulat...in&mass_b=110&hp_b=0&grade_b=15&k_b=.67&k=.67
That's a lot of awesome.

Only problem is:

According to a QS sales-rep:
QS doesn't actually make any of their 205 motors with anything less than 9.03 Kv.
(despite what it says on their website)

So, the quest for the illusive high-torque, low-speed hub motor continues...
Any suggestions?

Thanks.
-Josh!

With a motor that has a higher kV than you need, just run lower voltage and higher amps.
A MAC wouldn't cut the mustard if you really have 15% grades to climb in such a large wheel.

Here's an interesting idea for you. There's a 3000w cyclone motor out there that's geared for something like 500rpm output. On a 29er type bike, you could just run perhaps a 1:1 chain to a rear sprocket and be done. Of course, you'd need to come up with mounting brackets and other bits and pieces. But you'd save a hell of a lot of weight, versus an obese QS motor!

EBJ said:

According to a QS sales-rep:
QS doesn't actually make any of their 205 motors with anything less than 9.03 Kv.
(despite what it says on their website)

So, the quest for the illusive high-torque, low-speed hub motor continues...
Any suggestions?

Click to expand...

Buy one and rewind it for the kV you really want? It's a lot of work....

Simpler to do as neptronix suggests, of running it at lower voltage. I just don't know if it will do the same thing / peform the same way you want it to.

I don't know if this would do it either, but what I would do is run it at the full voltage you'd intended, so you can have as kickass a takeoff as you want, *but* make sure you throttle down so you don't exceed the speed you want. (or make it speed limited via a Cycle Analyst or something if you have to). I'd use a torque (current) throttle based controller, that directly senses phase current (rather than battery current), as it would make it more controllable.

neptronix said:

A MAC wouldn't cut the mustard if you really have 15% grades to climb in such a large wheel.

Click to expand...

Thanks for letting me know. (I noticed you ran a MAC in one of your builds)

neptronix said:

[...] higher kV [...] run lower voltage and higher amps.

Click to expand...

Thanks for the tip. I tried that in the simulator. But at 100A, even the low-voltage system isn't wanting much more than that. (not shown: but I gave the lower-voltage an extra digit of amps, and that didn't budge the performance much. Supplied amps doesn't seem to be a bottle-neck when we are talking about 100A)

Here's both KV values @ 72V:
(massive performance gain! but o.c. at the cost of heat)

https://www.ebikes.ca/tools/simulator.html?bopen=true&motor=MCRO5004_Sinks&cont=cust_100_300_0.03_V&batt=B7223_AC&axis=mph&black=load&blue=Lbs&hp=0&mass=110&frame=mountain&wheel=27.5i&grade=15&throt=100&autothrot=false&cont_b=cust_100_300_0.03_V&motor_b=MCRO5004_Sinks&batt_b=B7223_AC&wheel_b=27.5i&frame_b=mountain&mass_b=110&hp_b=0&grade_b=15&k_b=.67&k=.96

Here's with adjusted voltage to match the different Kv values:
(performance loss, even with ample amps!)

https://www.ebikes.ca/tools/simulator.html?bopen=true&motor=MCRO5004_Sinks&cont=cust_100_300_0.03_V&batt=B5216_GA&axis=mph&black=load&blue=Lbs&hp=0&mass=110&frame=mountain&wheel=27.5i&grade=15&throt=100&autothrot=false&cont_b=cust_100_300_0.03_V&motor_b=MCRO5004_Sinks&batt_b=B7223_AC&wheel_b=27.5i&frame_b=mountain&mass_b=110&hp_b=0&grade_b=15&k_b=.67&k=.96
*for above: I could probably do some middle-ground like 66V, but not really keen on that idea.

neptronix said:

[...] 3000w cyclone motor [...] a 1:1 chain to a rear sprocket and be done. [...] save a hell of a lot of weight, versus an obese QS motor!

Click to expand...

I love that idea, but my project is a self-balancing electric unicycle, so unfortunately I need to stick with the stupidly heavy hub-motor for space-saving reasons. It bums me out that there isn't a large diameter (~12-16"), light-weight (~20lb), high-torque, ~3,000W hub motor out there yet. But I can see how mid-drives and geared hubs kill the need for such a product.

amberwolf said:

Buy one and rewind it for the kV you really want? It's a lot of work....

Click to expand...

Thanks for the suggestion, but not really keen on diving in that deep.
I share that same thinking about it.
lol.

amberwolf said:

What I would do is run it at the full voltage you'd intended, so you can have as kickass a takeoff as you want, *but* make sure you throttle down so you don't exceed the speed you want.

Click to expand...

I like that idea. Especially because it means I have a lot of future un-tapped performance if I ever want to push it. Unfortunately (as shown in one of the simulator graphs above) this approach seems to take a hit to sustained hill-climbing ability.

Good news:

The QS motor sales-rep said they are willing to wind my motor to be ~6.7Kv.

That's close enough to my originally desired specs:

https://www.ebikes.ca/tools/simulator.html?bopen=true&motor=MCRO5004_Sinks&cont=cust_100_300_0.03_V&batt=B7223_AC&axis=mph&black=load&blue=Lbs&hp=0&mass=110&frame=mountain&wheel=27.5i&grade=15&throt=100&autothrot=false&cont_b=cust_100_300_0.03_V&motor_b=MCRO5004_Sinks&batt_b=B7223_AC&wheel_b=27.5i&frame_b=mountain&mass_b=110&hp_b=0&grade_b=15&k_b=.67&k=.70

So that's awesome.

Thanks for the helpful tips guys.
And thanks Justin (if you are reading this) for creating such an awesome motor simulator.

-Josh!

OOPS.

You were right Neptronix, (about just using the same system V and higher Kv with reduced performance) I just forgot to reduce throttle in the simulator. Doh!

Here it is again corrected:


https://www.ebikes.ca/tools/simulator.html?bopen=true&motor=MCRO5004_Sinks&cont=cust_100_300_0.03_V&batt=B7223_AC&axis=mph&black=load&blue=Lbs&hp=0&mass=110&frame=mountain&wheel=27.5i&grade=15&throt=72&autothrot=false&cont_b=cust_100_300_0.03_V&motor_b=MCRO5004_Sinks&batt_b=B7223_AC&wheel_b=27.5i&frame_b=mountain&mass_b=110&hp_b=0&grade_b=15&k_b=.67&k=.96

Now I just need to see if my motor controller can handle the increased AMPS. But I'm starting to like this "just reduce your throttle/performance" option more. Maybe I'll stick with the default 9.0 Kv after all.

Other than a higher Kv eating more Amps (and no significant effect on range (at reduced throttle)... which I still need to work out in my head as to why that is).
Why would someone want a lower Kv value motor?

To get a particular speed at a particular voltage.

To get a different inductance for controllers that need it for one reason or another. (some controllers don't do well with low inductance, for instance).

Sometimes a different kV also means a significantly different amount of copper fill in a particular motor because of the way it has to be wound, so it may have enough more copper mass in it to be worth other system compromises to get that.

Probably others, but those are off the top of my head.



Regarding running at higher voltage than the speed desired needs, to get better takeoff performance, that's part of what I do with the SB Cruiser trike. I don't need anything faster than 20MPH. But I do need to be able to reach taht speed very very quickly even with several hundred pounds of dog or cargo onboard (or on the trailer), for starting up from a stop at traffic controls to not be in the way of cars / etc., and also be able to accelerate out of the way of traffic that suddenly decides to be where I am in my lane, in the common case of braking being the wrong choice to get out of their way (due to traffic close behind me).

Are you sure you want a hub motor with that large wheel?
Normally the hub motors perform better with smaller wheels because the torque is proportional to the wheel diameter on the same grade hill and the copper loss is proportional to the square of the torque.
smaller wheel -> higher efficiency -> lower motor temperature. The simulator shows it, too, the difference is significant.