Which Controller (edit: hub motor) for my daughter's wheelchair?

Glyn

10 W
Joined
Oct 7, 2017
Messages
70
Following on from a previous post I'm building my daughter a six-wheel-drive mountain going wheelchair.

The motors I plan to use are typical Chinese motorised hubs that are used on e-bikes.

There will be six brushless motors rated at 1500w each.
rated current is 25a
max current is 45a

At the moment, I'm thinking three motors will be wired in series on each side and I'm now searching for a suitable dual channel controller (I'm assuming a six channel controller doesn't exist).

The controller needs to be programmable.
I need to be able to hook up an RC receiver, as well as be able to operate it with joysticks.
Although not absolutely essential, it would be good if I could hook up a Raspberry Pi or Arduino too.
Networking is not important.
Max draw per channel, if I'm not mistaken, needs to be at least 150a.

The only suitable controller I've come accross is the Roboteq GBL260 but I wondered if anybody on here could suggest any alternatives?

I suppose it would be possible to just get a bunch of cheap Chinese controllers (one for each wheel) and hook them all up to two joysticks (one for left and one for right) but a more elegant solution would be preferred.
 
Actually - now that i've posted that - I've just had the idea of being able to switch between parallel and series wiring on the fly. It could be wired in parallel when you're plodding along (thus preserving battery capacity) and then at the flick of a switch, wired in series for those moments when extra torque is needed. Does this seem viable to you?
 
I'm not sure if I understand your desired setup here, but if you're talking about putting brushless motors in series, you can't do that. You can put them in parallel, but only if they are mechanically locked together so they both motors are always synced with each other.

In this case it sounds like you would need 6 controllers. Are you sure you need 6 wheels? Do all 6 wheels need to be powered individually?
 
Addy is right. Each (brushless) motor requires its own controller. You can have them all on one battery (as long as it will withstand the maximum current demand).
 
I did not know that. Good job I asked.
(All wheels need individual power)

Does anybody know if such a thing as a 6 channel (50a+ per channel) brushless controller exists? Or even a 3 channel for that matter? If not, what are these generic controllers like?

I've no experience of brushless motors. On a brushed motor you can just switch the polarity for forwards and reverse. Is the same true for brushless motors? Will those cheap Chinese controllers allow forwards and reverse?
 
Glyn said:
On a brushed motor you can just switch the polarity for forwards and reverse. Is the same true for brushless motors?

It's not that simple. You can rearrange the phase wires and sensor leads to get correct reverse operation, but it's not usually a simple switch-two-wires procedure. Some controllers have a self-teach mode that lets you reverse direction easily and makes the controller configure its own Hall sensors.
 
I can appreciate your enthusiasm for this project, and as much as I am a fan of high-powered builds, the 1500W Direct Drive (DD) motors are very much overkill for this. They are heavier and more expensive than necessary. If all of them were powered to their full capacity (the only reason to buy hubmotors that beefy), then it would require a battery capable of 9,000W...

For street use, smaller diameter wheels are quite useful, but it sounds like you are hoping to drive off road? I recommend using the hubmotors that are the smallest which you can attach the fat tires to them. 20-inch bicycle wheels come to mind, but if you can find fairly fat tires that will mount to a 16-inch, I think those would be better. Also, moped tires are much tougher than bicycle tires. They are heavier, but roughly the same price.

When ebikes have been adapted to be a 2WD, there is a significant boost in performance, with only a small extra draw in watts. The total power needed to accelerate at the demand that the throttle is asking for is divided between two motors, so heat-shedding is dramatically improved, plus you have the added traction of two tires gripping the pavement instead of one. Burn-outs are fun and impressive, but they do not help acceleration.

My suggestion is to buy one of the Magic Pie motors from Golden Motor. They have the controller built into the center of the DD hubmotor, which saves a lot of hassle during the build, plus is simplifies the wiring. They have a 16-inch version, and also a 20-inch version.

Of course those models are designed for normal-width bicycle tires, but you can easily bond a second bicycle rim to the outer edge of the Magic Pie in order to widen it so that it will accept 3-to-4 inch wide tires.

If you are set on having six wheels, might I suggest that the four wheels in the back half be powered? The two front wheels can be idlers added only for downhill stability, but...should test runs indicate that more traction is needed on the uphills, they can be upgraded to a powered wheel if desired.

The "smart pie" version has a smaller motor, and if you are driving four of them (or even six), I think you will be surprised at how much total hill-climbing power they provide. In fact if you are certain that you will power all six wheels, I would go straight to the smart pie version. At least buy one for testing purposes. if it proves to be inadequate, add it to a bicycle and sell it locally to recoup some of your investment.

MagicPie%20Series.jpg
 
If you are going with generic controller then I'd suggest ebay seller hksunwin as lots of people have opened them up, took photos and analyzed.

https://endless-sphere.com/forums/viewtopic.php?f=2&t=71128&hilit=sunwin

https://www.ebay.ca/sch/m.html?_ssn=hksunwin&_armrs=1&_from=R40&_sacat=0&_nkw=ebike&_sop=15

I would look at the 60V units, as sunwins 48V would have capacitors for 60V, while the 60V units would have much higher voltage. Plus you can desolder or solder on your appropriate low voltage cutoff. I am unsure if sunwins wattage ratings really mean anything, so a simple 60V 500W $30cdn controller would probably be the same as 60V 1000W $52cdn, $1cdn = $0.78usd = 0.65euro = 0.58 british pounds. So yeah they are cheap, and you can beef up the traces for even more power.
 
Thanks for all the comments guys. As I've said, I have near zero experience of brushless motors so your advice is much appreciated.

Chalo said:
It's not that simple.
Thank you. I've since learned that the controllers are cable of driving the motors in reverse. I didn't know if this would be the case (e-bikes don't usually need to reverse).

spinningmagnets said:
I can appreciate your enthusiasm for this project, and as much as I am a fan of high-powered builds, the 1500W Direct Drive (DD) motors are very much overkill for this.
I've already ordered these motors.

The reason I went with 1500w motors is because there will be times when only one wheel has traction and therefore a single motor will need to be able to move the weight of the vehicle and passenger.
To give you some context, here are a couple of pictures of the path I intend to take it on (click for full size):


I would have liked to have gone with fat-bike wheels but I have a limited width constraint and the forks to accommodate fat-bike wheels would have taken up too much of the available space. In the end I opted for 20" BMX wheels and the widest tyre I could find for them (2.25"). Originally my plan was to go with 12" wheels as they offered more torque and added less weight but the wheelchair needs to be able to climb over boulders up to 2' in height. I would have liked larger diameter wheels but I was worried about losing torque and it would have increased the length of the wheelchair which also has a constraint. 20" wheels seemed like a good middle ground.

One of the reasons I bought these particular motors is because they offer 1500w with a 100mm axle and help to keep the overall width down - Most 1500w motors I've looked at require 135mm.

All 6 wheels need to be powered otherwise it won't go where I plan to take it. If you take a look at the above photos and this video of a very early concept model of the wheelchair's drive mechanism hopefully it will begin to explain why.

You're the second person to recommend the GoldenMotor Magic Pie though. I've already purchased the 6 motors now but I got a good price for them so I could easily recoup most of the costs. My only concern, as I've suggested above, is that there will be times when only one wheel will have to move the entrie combined weight and the magic Pie only goes up to 1000w. I've heard you can increase this to 1500w with a 60v battery but this would significantly increase both the weight and cost of the wheelchair too - I'm half-expecting to need a 100ah battery pack, but will start out with 30ah and add more cells as required. To do this at 60v would be too much.

markz said:
If you are going with generic controller then I'd suggest ebay seller hksunwin as lots of people have opened them up, took photos and analyzed.
Wow, they're certainly affordable. I guess my main concern with those is driving the motors in both forwards and reverse directions.

Thanks again. All comments are much appreciated.
 
Too bad you didn't listen to the advise you were given in your first thread.
Because these DD motors in a 20" bicycle rim are about the worst choice you could have made for this application. At the speeds (<2mph?) your daughter will be climbing that path for hours on end those DD won't take more than a few hundred watts before you'll burn them up. And most of the energy will be lost on heat production instead of propulsion. They will be only turning at around 33rpm which is not (absolutely NOT!) what they were designed for. You'll end up needing a large battery just to power 6 large very slowly rotating heaters. It's a shame as I really like your efforts to let your daughter climb that hill.
 
SlowCo said:
Too bad you didn't listen to the advise you were given in your first thread.
I'm still open to advice. As I say, nothing is set in stone and I don't think it would be too difficult to recoup my expenses if I chose to sell the motors, in fact, if I acted quickly I imagine I could even cancel the order, but I'd have to act quickly and make a rushed a decision :? . When you say "Too bad you didn't listen to the advise you were given in your first thread", is there something in particular that you're referring to?

SlowCo said:
Because these DD motors in a 20" bicycle rim are about the worst choice you could have made for this application. At the speeds (<2mph?) your daughter will be climbing that path for hours on end those DD won't take more than a few hundred watts before you'll burn them up.
Yes, it will be hours on end - I've estimated 6 hours.
Perhaps I'm showing my ignorance here, as you can tell I have limited experience, but would you mind explaining, or pointing me towards an article that explains why they'd burn up?

SlowCo said:
They will be only turning at around 33rpm which is not (absolutely NOT!) what they were designed for.
What would you recommend instead?

SlowCo said:
You'll end up needing a large battery.
Agreed.

SlowCo said:
I really like your efforts to let your daughter climb that hill.
Thank you :)
 
Glyn said:
SlowCo said:
Too bad you didn't listen to the advise you were given in your first thread.
I'm still open to advice. As I say, nothing is set in stone and I don't think it would be too difficult to recoup my expenses if I chose to sell the motors, in fact, if I acted quickly I imagine I could even cancel the order, but I'd have to act quickly and make a rushed a decision :? . When you say "Too bad you didn't listen to the advise you were given in your first thread", is there something in particular that you're referring to?

SlowCo said:
Because these DD motors in a 20" bicycle rim are about the worst choice you could have made for this application. At the speeds (<2mph?) your daughter will be climbing that path for hours on end those DD won't take more than a few hundred watts before you'll burn them up.
What does DD mean? Direct Drive?
Yes, it will be hours on end - I've estimated 6 hours.
Perhaps I'm showing my ignorance here, as you can tell I have limited experience, but would you mind explaining, or pointing me towards an article that explains why they'd burn up?

SlowCo said:
They will be only turning at around 33rpm which is not (absolutely NOT!) what they were designed for.
What would you recommend instead?

Based upon my 'vast experience' :) of 3 rides totalling 31k, on my new Q128H, I recon that if you fitted your daughter's chair with 4 of them -- high torque, 800w geared hub motors -- she could probably climb the North face of the Eiger.

I just checked and they still have them on 19% discount, meaning 4 shipped (to the UK) would be £330:
4xQ128H.jpg

I wonder if they could be run using one of those quad-copter ESC setups?
 
I'm not an expert but I know that climbing a hill is all about gearing. If you ride up a hill on a normal MTB you will go down in gearing to be able to keep pedaling. The steeper the incline the lower you have to be in the gearing range to keep moving. A gearless hub motor in a 20" rim will be the same as not shifting down and needing more and more power to keep going. At 33rpm or even less with a direct drive brushless motor is like trying to stall that motor.

If you Google "electric wheelchair motor" and look at the pictures you'll see that the most common used motors are brushed motors with a high ratio worm drive gearbox attached. A brushed motor has more torque from a start and the worm drive will allow very low gearing and have the tendency to act as a brake because of the high ratio and construction. So these electric wheelchair motors are universally chosen for walking speeds and would be suitable for the climbing at even lower speeds and more load if you use them at all 6 wheels. And being simple brushed motors you can put them in parallel each side and control each side with one controller. Also reversing would be easier. You could even use "tank" style steering by reversing only one side. Hey that sounds like an electric wheelchair... :wink:

Even better would be to find industrial 24V brushed motors and worm drive gear boxes with an even higher gear ratio so you can mount larger wheels to them. Motorcross wheels and tires in 18" or 21" size would allow much better rolling over the big rocks and would be much stronger than bicycle wheels. Especially as they will see some side loading.
 
Buk___ said:
Based upon my 'vast experience' :) of 3 rides totalling 31k, on my new Q128H, I recon that if you fitted your daughter's chair with 4 of them -- high torque, 800w geared hub motors -- she could probably climb the North face of the Eiger.

I just checked and they still have them on 19% discount, meaning 4 shipped (to the UK) would be £330:
4xQ128H.jpg

Even the 201rpm Q128 motors would burn up at the <33rpm he needs. The guy that had the electric wheelchair with 4 specially wound (16T!) 1500W geared MAC motors in 20" fat bike rims had heat problems at the low speeds his wife was using it. And that was not even on a serious steep and long climb like the path this is intended for.

By the way, I don't think I would let my daughter climb up a path like that in a vehicle on wheels and especially knowing she would need to roll down again :shock:
That's another big concern: you need the vehicle to descend safely (regenerative) braking for hours. So you'll need serious braking capability and heat shedding.
 
SlowCo said:
Even the 201rpm Q128 motors would burn up at the <33rpm he needs.

Hm.
5kph = 5000/60/(16"*25.4*pi/1000)= 65rpm at the wheels at walking pace. 15:1 internal gear ratio gives motor speed at walking pace = 979 motor rpm.

As indicated above, I'm far from an expert; just thrilled with the climbing ability of my new Q128. One of my tests was to stop on the steepest part of the second steepest hill around me and then engage the 6kph 'walk assist' feature from stationary, whilst sat on the bike no pedaling. It pulled me (~120kg all up) to the top at just over 5kph, despite the battery being less than 50%; and the indicated power draw never more than 100W. At 36V that's under 3 amps.


SlowCo said:
By the way, I don't think I would let my daughter climb up a path like that in a vehicle on wheels and especially knowing she would need to roll down again :shock:
That's another big concern: you need the vehicle to descend safely (regenerative) braking for hours. So you'll need serious braking capability and heat shedding.

Given the battery would have plenty spare of capacity from the climb up, you'd have plenty spare to absorb the regen energy. Of course, you'd have to lock out the clutch for regen from Q128s :( (I've no idea if that is possible.)

But IF you have variable regen braking, a couple of 203mm disks would be more than man enough to ensure stopping safety. It's not like you're trying to pull up from 50kph.
 
Buk___ said:
I wonder if they could be run using one of those quad-copter ESC setups?
I did think about that and you can even get them with 6 channels, but unsurprisingly I wasn't able to find anything with the required amperage.
 
SlowCo said:
I'm not an expert but I know that climbing a hill is all about gearing.
I would have liked to have used a geared setup and have considered chain, toothed belt, and hydraulic setups but all were dismissed for one reason or another. The worm gear setup I think is the best solution of all though but for the moment, due to my limited experience, I'm trying to keep it as simple as possible.

SlowCo said:
a 20" bicycle rim are about the worst choice you could have made
...
Motorcross wheels and tires in 18" or 21" size would allow much better rolling over the big rocks and would be much stronger than bicycle wheels. Especially as they will see some side loading.
There will be some side loading and I agree that stronger wheels would have been better. In an ideal world it would be petrol driven with quad-bike tyres or similar, but as I've mentioned above, I have a strict width constraint so have no room for motorbike wheels. The wheels I'm using are Trials bike wheels so they should cope better than standard BMX wheels.

SlowCo said:
Even better would be to find industrial 24V brushed motors and worm drive gear boxes with an even higher gear ratio
I don't suppose you could provide an example of the sort of thing you're thinking of?

Can anybody point me towards an article that explains why, with my current plan, the motors would burn up?
 
Buk___ said:
5kph = 5000/60/(16"*25.4*pi/1000)= 65rpm at the wheels at walking pace. 15:1 internal gear ratio gives motor speed at walking pace = 979 motor rpm.
Thanks for this Buk. To make sure I understand correctly, are you saying that your Q128 has an internal gear ratio of 15:1? Sounds awesome if it does.

Buk___ said:
As indicated above, I'm far from an expert; just thrilled with the climbing ability of my new Q128. One of my tests was to stop on the steepest part of the second steepest hill around me and then engage the 6kph 'walk assist' feature from stationary, whilst sat on the bike no pedaling. It pulled me (~120kg all up) to the top at just over 5kph, despite the battery being less than 50%; and the indicated power draw never more than 100W. At 36V that's under 3 amps.
Sweet :)

SlowCo said:
Given the battery would have plenty spare of capacity from the climb up, you'd have plenty spare to absorb the regen energy. Of course, you'd have to lock out the clutch for regen from Q128s :( (I've no idea if that is possible.)
But IF you have variable regen braking, a couple of 203mm disks would be more than man enough to ensure stopping safety. It's not like you're trying to pull up from 50kph.
I plan on having 6 disc brakes - one for each wheel. I haven't looked into it yet but in my mind I think this would be quite easy with the cable type and it doesn't seem too far fetched to think it could be achieved with hydraulic disc brakes.
Buk, would you mind explaining what "variable" regenerative braking is please?
 
It looks like other people have been impressed with the hill climbing ability of the Q128H too: http://www.pedelecs.co.uk/forum/threads/q128h.20980/

The thread suggests they have an internal 16:1 gear ratio :shock:
 
Glyn said:
Buk___ said:
5kph = 5000/60/(16"*25.4*pi/1000)= 65rpm at the wheels at walking pace. 15:1 internal gear ratio gives motor speed at walking pace = 979 motor rpm.
Thanks for this Buk. To make sure I understand correctly, are you saying that your Q128 has an internal gear ratio of 15:1? Sounds awesome if it does.
According to this post (by motomech) here at E-S, the 201rpm wind is actually 16:1 (I remembered wrongly apparently).

There's another variant which is 13.2:1

Buk___ said:
As indicated above, I'm far from an expert; just thrilled with the climbing ability of my new Q128. One of my tests was to stop on the steepest part of the second steepest hill around me and then engage the 6kph 'walk assist' feature from stationary, whilst sat on the bike no pedaling. It pulled me (~120kg all up) to the top at just over 5kph, despite the battery being less than 50%; and the indicated power draw never more than 100W. At 36V that's under 3 amps.
Sweet :)
[/quote]

I'd need to verify that more carefully; it's a fairly busy road so I only had half an eye on the display.

I'll try it again tomorrow on the steepest hill (>10:1) which is further away, but much less busy.

Glyn said:
SlowCo said:
Given the battery would have plenty spare of capacity from the climb up, you'd have plenty spare to absorb the regen energy. Of course, you'd have to lock out the clutch for regen from Q128s :( (I've no idea if that is possible.)
But IF you have variable regen braking, a couple of 203mm disks would be more than man enough to ensure stopping safety. It's not like you're trying to pull up from 50kph.


I plan on having 6 disc brakes - one for each wheel. I haven't looked into it yet but in my mind I think this would be quite easy with the cable type and it doesn't seem too far fetched to think it could be achieved with hydraulic disc brakes.

Buk, would you mind explaining what "variable" regenerative braking is please?

You'd need to talk to/get a response from one of the many proper experts here for a full explanation; but in simple terms it means that rather than drawing the full potential of the motor as a generator -- akin to slamming the brakes full on -- it allows a controller configured for it to modulate the amount of power drawn from the motor (acting as a generator) and so allow the rider to control the amount of braking.

Like I say, that's my simplistic view of it; hopefully others will step in with a full explanation.
 
Thanks Buk, that explanation was perfect for my brain ;)
Am I right in thinking that this is the same product but for the front forks? https://bmsbattery.com/ebike-kit/397-q128-48v400w-500w-front-driving-v-brake-hub-motor-ebike-kit.html

There's a distinct lack of specs given on the website.
For the Bafang front motor the website specifies that the reduction ratio is 5:1 but no reduction ratio is given for the Q128.
 
Back
Top