Modifying a Hub Motor

EpicShelter

10 µW
Joined
Mar 22, 2021
Messages
5
Hello, I'm a bit of a noob in regards to brushless motors so bear with me. I do know my way around a workshop though, so I feel confident embarking on this mission.

A couple years ago I put together an ebike in a very spartan way. I connected 3x massive 6S LiPo packs in series, connected the directly to a Controller and hooked it all up to a 1500W hub motor. Funnily enough (not) the electronics stopped working and soon after I moved away and completely forgot about the project.

Now I found all the parts again and I want to rebuild. Here is my parts list

18S LiPo (=72V) ±1kWh battery pack which would be able to provide upwards of 25kW
1500W no-name Hub Motor and Rim
Fat Bike

The rest of the electronics don't work.

Now, with the bike having 4.5" wide tires and me riding on snow, I remember the bike being very fast but not very torque-y at all.

If I'm not mistaken, I should be able to rewind the hub motor to draw more current at 72V effectively making it more powerful (increasing its wattage).
Because I need to replace all electronics anyways, I thought I may as well go all out and do the whole job. And yes, this time I would like to use a BMS.

Does anyone have any resources on how to rewind a motor for increased current draw?

I would also appreciate any alternative input on how I can make this e-bike as torque-y and powerful as possible.

Thanks in advance
EpicShelter
 
Your query lacks specifics, but by "no name" motor, I assume you are talking about a direct drive 9C or 9c clone motor. No "rewind" (a difficult and tedious process), is going to change the fundamental nature of a direct drive motor. A DD motor w/ lots of Volts is a common route to speed, but not acceleration.
For what you want, you need a geared hub motor (size dependent on the size of your wallet), reduce the Volts to a more reasonable 48 to 52 Volt range and use a large controller matched to the motor.
As to the rest of your post, I'm not sure I can decipher what you are asking;
By "electronics", are you referencing the controller?
"18S LiPo (=72V) ±1kWh battery pack which would be able to provide upwards of 25kW"
If I'm not mistaken, 25 kW is about what a house might use in 24 hours and batteries in that range cost around $25,000.
Perhaps someone w/ more knowledge of BIG batteries than myself can comment.
At this point, you might want to do some research here, and look at some "high-powered hub motor, fat bike builds to see what some of the experienced builders are doing. There are plenty here.
 
Hi, thanks for the response.

By "electronics" I mean controller, screen, PAS, speed limiters and such.

Yes, the motor is a direct drive 9C motor so I realise I'm fairly limited in how much I can modify the specs.

As for the batteries I am using high-discharge LiPos like you'd see in a large RC project which can do upwards of 30C, meaning that they can discharge in 1/30th. With the pack being roughly 1kWh my peak discharge rate would be around the 25-30kW mark.
The point I was trying to make was that my batteries can easily handle the increased load.

Thanks,
ES
 
motomech said:
If I'm not mistaken, 25 kW is about what a house might use in 24 hours
Watts, kW and amps are all flow rates, like gallons per hour

therefore cannot be use as units "per 24 hours" of any time period, just no meaning.

Wh, kWh and Ah are quantities of energy in the static sense, analogous to gallons.

Wh per minute, kWh per hour, Ah per 24 hours

are all correct examples.

 
EpicShelter said:
Hi, thanks for the response.

By "electronics" I mean controller, screen, PAS, speed limiters and such.

Yes, the motor is a direct drive 9C motor so I realise I'm fairly limited in how much I can modify the specs.

As for the batteries I am using high-discharge LiPos like you'd see in a large RC project which can do upwards of 30C, meaning that they can discharge in 1/30th. With the pack being roughly 1kWh my peak discharge rate would be around the 25-30kW mark.
The point I was trying to make was that my batteries can easily handle the increased load.

Thanks,
ES

Seems like before you get to 25kW, a 9C motor will saturate, current will spike, followed by melt down.
 
Seems like before you get to 25kW, a 9C motor will saturate, current will spike, followed by melt down.

I wasn't claiming it wouldn't, I was just pointing out that my batteries can handle the increased load.

To be clear, I'm not trying to increase my motor power to anything near 25kW...
 
EpicShelter said:
Does anyone have any resources on how to rewind a motor for increased current draw?
The controller is the thing that dictates current flow, not the motor.

You can rewind the motor to produce more torque per amp (Kt=Nm/Amp), at the cost of a lower speed per volt (Kv=rpm/Volt).

https://endless-sphere.com/forums/viewtopic.php?t=39872
https://endless-sphere.com/forums/viewtopic.php?t=69358

Most importantly though, you haven't told us anything about what controller you will use. This is the key to getting your motor to output more torque.
 
Most importantly though, you haven't told us anything about what controller you will use. This is the key to getting your motor to output more torque.

Im open to suggestions.
 
If you want 25kw you will need a bigger motor then the 9C.
You will also need a very expensive controller to match your battery that can handle 25kw which is what? 96V 250A?


Ah my bad, so 18S is 72V and 25kw means 350A, so thats gotta be an awful lot of parallel groups in that battery to provide that kind of power. Say 25R's @ 14P, which would mean 35Ah.
18S LiPo (=72V) ±1kWh battery pack which would be able to provide upwards of 25kW
1500W no-name Hub Motor and Rim
Fat Bike
 
This is all a bit of a misunderstanding. I never claimed I wanted anything near 25kW. I was just pointing out that my batteries could handle the increased load.

I'm looking for a way to rewire the 9C motor for less RPM and possibly more power draw in the 2-3kW range.
 
EpicShelter said:
18S LiPo (=72V) ±1kWh battery pack which would be able to provide upwards of 25kW
The rest of the electronics don't work.
Because I need to replace all electronics anyways...
If you have to get a new controller anyway, then just get a higher current controller. Plenty of options at 72V.

EpicShelter said:
If I'm not mistaken, I should be able to rewind the hub motor to draw more current at 72V effectively making it more powerful (increasing its wattage).
Does anyone have any resources on how to rewind a motor for increased current draw?
No, as was correctly posted above, you can only trade speed for torque.

EpicShelter said:
I would also appreciate any alternative input on how I can make this e-bike as torque-y and powerful as possible.
Without changing the battery or motor? A high current controller, and increase the current until it's too hot to touch or until you burn it out. Then get a higher power, lower kV, higher kT motor and repeat.
 
You can't rewind a motor to make it able to make more torque at any voltage. A rewind accomplishes nothing that you can't accomplish by changing the voltage and current, unless you are able to fit more total copper around the stator teeth than the factory.

If you want better performance from the same motor, the best way to accomplish that is increasing voltage and decreasing wheel diameter.
 
John is absolutely right but you can rewind a motor to make more torque if you have a suitable controller and a crap ton of spare current from the battery but are limited by voltage, equally you could change the voltage so that isn't limiting things

The right answer for you depends on the reason for the question.

Rambling detail:
It may depend on what you prefer working on, which bits you prefer to keep, whether this is just down to performance vs budget and what the limiting factor was, what performance increase you want and what your budget is

If you don't want to change the motor and it was the limiting factor then yes, you can get more power through that motor by rewinding it or more easily by raising the voltage

fatty said:
EpicShelter said:
If I'm not mistaken, I should be able to rewind the hub motor to draw more current at 72V effectively making it more powerful (increasing its wattage).
Does anyone have any resources on how to rewind a motor for increased current draw?
No, as was correctly posted above, you can only trade speed for torque.
Nope,you were right, fatty is wrong, you could rewind the motor to get more power with that battery and a good enough controller. it just may not be worth doing.
With sufficient current and controller, then with the same copper fill, power is the same for a 3T 4S or a 4T 3S.
Fatty will come back and say that's what he meant but in this example of a fixed battery voltage & suffiecient current then the voltage is the limiting factor. so thicker windings are the answer and a controller to match, or more voltage to match the thinner windings for the same effect.
Depends on whether you can reconfigure the battery for higher voltage.

It's pretty simple concptually and you were absolutely right, more current, thicker wires works, thinner wire, more voltage works. This is what John was saying, it is easier to change the battery than rewind the motor.
EpicShelter said:
I would also appreciate any alternative input on how I can make this e-bike as torque-y and powerful as possible.
fatty said:
Without changing the battery or motor? A high current controller, and increase the current until it's too hot to touch or until you burn it out. Then get a higher power, lower kV, higher kT motor and repeat.
Nope, I mean, if the controller was the problem then sure, turn the power up. If that overheats it then you can get some cooling on the motor and don't need anything else but maybe 72V doesn't overhat the motor, even with a sufficient controller.
fatty doesn't like high voltage so he won't suggest that. if the controller was the problem then get a new controller but if not then you need either higher voltage or a different motor (or re-winding)
This would mean you need more voltage to drive more curent through those thin wires.
A higher current controller isn't going to help unless the controller current was the limiting factor(see before) The batteries have way more current than needed.

If the motor wasn't getting hot, then the battery or the controller were limiting the motor.
If it was just an underpowered controller then a better controller may be all you need
If it's the battery, and as you say it has plenty of spare amps, then higher voltage is the answer and the easiest solution is probably another pack in series.

At more than 96V, controllers get rare and more expensive so you'd probably be better off with a better motor cost-wise.

If you've got a 1.5 KW motor but want more, then it will overheat, so cooling fins & Statoraid will give you some small gains.

If your question is "The batteries are the expensive bit and have plenty of power so why not re-wind the motor?"
Probably no because it's a pain in the wotsits sort of job but maybe yes if you want to, conceivably for cost vs time reasons though your hourly rate will not be good
It might be that you want to re-wind a motor and so great, go for it.
I'd say doing it well can be tricky and the chance that you damage something during the process would make most people bin it part way but it might be your thing.

If for example it's a $200 motor and a $1000 battery pack, you can't reconfigure the battery pack, a bigger one wouldn't fit etc then a new motor may make more sense than more voltage, unlikely but possible.

Bear in mind that you won't get maximum mid range power, maximum torque bottom end, and maximum top speed along with maximum efficiency.
Have a look at the Grin simulator, have a look at battery configuration, controllers and what has what effect: https://ebikes.ca/tools/simulator.html

There are plenty of ways to skin the cat and I wish you all the best :)
 
BobBob said:
John is absolutely right but you can rewind a motor to make more torque if you have a suitable controller and a crap ton of spare current from the battery but are limited by voltage, equally you could change the voltage so that isn't limiting things...[etc]

Once again, this is incorrect, and has been explained/refuted in previous threads.

Rewinding a motor for more torque (higher kT) in the direct, textbook meaning of "rewind" is simply increasing turn count to trade kV for kT. It has nothing to do with controller or "spare current".

Rewiring with thicker wire windings can reduce electrical loss in the stator windings, which can then permit higher current at the same electrical loss, but
1) as above, this decreases kT for kV, so net torque may be equivalent to before rewiring or even lower if not overcompensating with higher current. And unlike a geared mid-drive, the higher kV in a hub motor may not even be useful,
2) this only applies at the ragged edge of maximum motor torque, where the motor will burn up with the thinner windings but not the thicker windings, instead of being controller current-limited, and
3) this only applies to current/temperature-limited motor designs, rather than flux-limited motor designs

While it's good to understand this mechanism (though it would be better to first understand motor constants), none of these apply to the OP (who doesn't even have a working controller), or indeed any poster asking such introductory questions.
 
https://youtu.be/c96n0Ma2rLY?t=8174

https://ebikes.ca/tools/simulator.html?motor=MX4506&batt=B7208_DT&cont=cust_21_42_0.035_V&hp=50&cont_b=cust_40.5_81_0.035_V&motor_b=MX4503&batt_b=cust_45_0.1_8&hp_b=50&bopen=true&throt_b=92&throt=100&grade_b=0&autothrot=true&autothrot_b=true
**********Range is different because the batteries are different total Wh.
 
fatty said:
BobBob said:
John is absolutely right but you can rewind a motor to make more torque if you have a suitable controller and a crap ton of spare current from the battery but are limited by voltage, equally you could change the voltage so that isn't limiting things...[etc]
Once again, this is incorrect, and has been explained/refuted in previous threads.
I'll do some simulations to illustrate later
To be clear, you are not rewinding the motor to be more "powerful" unless you achieve more copper density but you are changing the amount of voltage and current to suit your power supply, which in this example where there is more current than you can use but insufficient voltage will result in more available power from the motor.
fatty said:
Rewinding a motor for more torque (higher kT) in the direct, textbook meaning of "rewind" is simply increasing turn count to trade kV for kT. It has nothing to do with controller or "spare current".
If you want more current you need to decrease the turn count using thicker wire, not thinner wire, you need to go to a 3T from a 4T for example, whilst of course maintianing maximum copper fill.
Assuming the cross section of wire (copper fill) is the same and you have spare current capacity then this will have the same effect as going from three battery packs to four.

You are making use of the spare current available from the battery packs that is not being used, IE current was not a limiting factor so you can use more of it to get more power as the voltage will not change

This has nothing to do with the controller so long as the controller is able to handle the additional current in the former or voltage in the latter example.

If this is still unclear to you, I will use the following analogy and please don't then come back and tell me i was suggesting this is proposed design, it is a means of picturing what is happening:
Think of it being multistrand wire in the motor, the total current produces flux, this is how strongly the magnet pulls. It doesn't matter whether it goes through one strand or 12.

To further illustrate let's say that for each of the 12 strands we have one cell. You could make three strand "wires" and go round the motor 4 times or 4 strand wires and do 3 turns, (or 1 of 12, 2 of 6 etc)

Now you configure your battery so that for every strand there is a cell, so your 4S3P configuration powers a 3T (4strand per turn) motor or a 3S4P powers a 3T (or 12P1S for 1T 12S1P for 12T etc)

It doesn't matter whether you have two cells in series driving two strands in series or two in parallel driving two in parallel
Each strand has exactly the same current and exactly the same voltage in it in all configurations.

Grin Simulations in later to further illustrate if you're still unclear.
Gotta go, will re-edit later but you've got your calcs the wrong way round.
fatty said:
Rewiring with thicker wire windings can reduce electrical loss in the stator windings, which can then permit higher current at the same electrical loss, but
1) as above, this decreases kT for kV, so net torque may be equivalent to before rewiring or even lower if not overcompensating with higher current. And unlike a geared mid-drive, the higher kV in a hub motor may not even be useful,
2) this only applies at the ragged edge of maximum motor torque, where the motor will burn up with the thinner windings but not the thicker windings, instead of being controller current-limited, and
3) this only applies to current/temperature-limited motor designs, rather than flux-limited motor designs

While it's good to understand this mechanism (though it would be better to first understand motor constants), none of these apply to the OP (who doesn't even have a working controller), or indeed any poster asking such introductory questions.
 
BobBob said:
To be clear, you are not rewinding the motor to be more "powerful" unless you achieve more copper density but you are changing the amount of voltage and current to suit your power supply
This is nonsensical. Motor winding is independent of supply voltage and current. As explained above, rewinding only exchanges kV for kT.

BobBob said:
If you want more current you need thicker wire, not thinner wire, you need to go to a 3T from a 4T or 5T.
Gotta go, will re-edit later but you've got your calcs the wrong way round.
Read my post again -- I address both these points. You continue confusing two different procedures/mechanisms:
Rewinding for higher kT = lower kV, requires more turns, and possibly thinner wire with side effect of higher electrical losses -- relevant only at maximum motor torque (not maximum controller current)
Rewinding for higher kV = lower kT, requires less turns, can possibly use thicker wire with side effect of lower electrical losses -- relevant only at maximum motor torque (not maximum controller current)

Distinct from
Rewiring with thicker wires because you burned out your old windings with too much current in a current/temperature-limited (not flux-limited) motor design. Thicker wires fit less turns, with side effect of reducing kT, thus requiring more current for the same (previous) torque, thus requiring much thicker wires to increase current capacity to result in net torque increase. Much thicker wires decrease electrical fill factor.

Please take a second to re-read through this carefully, so we don't continue to talk past each other. You are overgeneralizing a rare corner case (rewiring) and thus missing the big (motor constant) picture.
 
Ok, what you have is a motor that in general, tolerates up to 2000w pretty well. What likely happened is you ran 18s lipo, which is 60v through a 48v controller. full charged, about 72v went through the capacitors in your controller, which are typically 63v capacitors.

So what you want now, is a 60v controller. If you get one with 30 amps limiter inside it, you will have just about 2000w max, and a top speed of around 35 mph. It may not have the torque you desire for crawling up a steep hill, or riding in deep sugar sand, but it will do without you having to buy different motors.

Modifying the motor itself is not possible, at least in terms of more power. The amount of copper that fits in the hole is already maximum. You cant make the motor bigger, without making the shell bigger. Rewinding the motor, or switching to a different wound motor of same size, only gets you a motor that is slower, for going slower. That's it. The slow motor is often sold as a higher torque motor, but its not. Its just slower. Go that route ONLY if slow is what you want. Slower can be good, especially for range, but its never more power or more torque.

A bigger motor is definitely possible, along with going to bigger battery, such as 24s instead of 18s. But your best bet for the least money, is to just get a 60v 30 amps controller. BTW, for short duration rides, those motors can take 3000w, so a 60v or 72v 40 amps controller is possible. Just keep the battery capacity 10 amp hours or less, so it runs out just before the motor melts.
 
markz said:
https://youtu.be/c96n0Ma2rLY?t=8174

https://ebikes.ca/tools/simulator.html?motor=MX4506&batt=B7208_DT&cont=cust_21_42_0.035_V&hp=50&cont_b=cust_40.5_81_0.035_V&motor_b=MX4503&batt_b=cust_45_0.1_8&hp_b=50&bopen=true&throt_b=92&throt=100&grade_b=0&autothrot=true&autothrot_b=true
**********Range is different because the batteries are different total Wh.

Exactly :thumb:
Here is a customized chart with the OP's fatbike and battery pack.

This is why I always recommend selecting the lowest kV suitable to reach desired maximum speed:
lower kV = higher kT
= lower current for same torque = lower electrical losses = higher efficiency (as seen in simulator above)
AND
= higher torque from same current
 
fatty said:
BobBob said:
To be clear, you are not rewinding the motor to be more "powerful" unless you achieve more copper density but you are changing the amount of voltage and current to suit your power supply
This is nonsensical. Motor winding is independent of supply voltage and current. As explained above, rewinding only exchanges kV for kT.
Ahem...
Different motor winding, same everything else. 7kW vs 8.5kW peak power available, winding dependent.
https://ebikes.ca/tools/simulator.h...X4503&batt_b=cust_84_.02_15&hp_b=0&bopen=true
Everything is constrained by real world conditions. It's all well and good to stick to the theoretical "same copper = same motor, irrespective of winding" spiel. BUT, in the real world, we are limited by the available controller specifications (and to a certain extent, the battery). Everything is conditional on these constraints. Controllers are not infinitely variable to adapt to different windings. Different windings will have different performance for a fixed controller specification. This is what BobBob is saying.
compare.jpg
 
dogman dan said:
Modifying the motor itself is not possible, at least in terms of more power. The amount of copper that fits in the hole is already maximum. You cant make the motor bigger, without making the shell bigger. Rewinding the motor, or switching to a different wound motor of same size, only gets you a motor that is slower, for going slower.
Correct, but slower or faster -- can rewind with more or fewer turns, respectively.

dogman dan said:
The slow motor is often sold as a higher torque motor, but its not. Its just slower. Go that route ONLY if slow is what you want. Slower can be good, especially for range, but its never more power or more torque.
This is incorrect. kV and kT are inverses by kT=60/(2*PI())/kV
Like Bob, you are overgeneralizing a current/temperature-limited (not flux-limited) motor at maximum motor torque. This does not apply to 30-40A controllers.
 
serious_sam said:
Ahem...
Different motor winding, same everything else. 7kW vs 8.5kW peak power available, winding dependent.
https://ebikes.ca/tools/simulator.h...X4503&batt_b=cust_84_.02_15&hp_b=0&bopen=true
It would be helpful to use the OP's constraints (as given in my simulator link) to illustrate your point.

serious_sam said:
Everything is constrained by real world conditions.
Agreed. This includes maximum desired speed as limited by speed limits, as well as controller voltage and current limits. Your simulator pushing 97V hot charged (23s), 100+A and 6kW into the OP's 1.5kW motor on a fatbike for 90k/h is not a real world condition.

serious_sam said:
It's all well and good to stick to the theoretical "same copper = same motor, irrespective of winding" spiel.
Actually, I think this is the opposite of what I'm saying, except for diminished returns in the corner case of rewiring for maximum motor current.

serious_sam said:
BUT, in the real world, we are limited by the available controller specifications (and to a certain extent, the battery). Everything is conditional on these constraints. Controllers are not infinitely variable to adapt to different windings.
Different windings will have different performance for a fixed controller specification.
Agreed. This is why I recommend cohesive system design: maximizing controller power density with highest voltage, then selecting the lowest kV to meet the desired maximum speed, then matching pack current to controller battery current.

serious_sam said:
This is what BobBob is saying.
That's a charitable re-interpretation. As we agree, you can change supply voltage and current to fit the motor, or you can change the motor kV/kT to fit the supply voltage and current. But as written, rewinding a motor to change voltage and current, is nonsensical in the literal meaning.


Again, I'm not disputing that higher kV = more power at fixed maximum motor torque (obviously, power = torque * rpm). But nobody asking such beginner questions is holding their motor at the ragged edge of burning up.
 
fatty said:
serious_sam said:
Ahem...
Different motor winding, same everything else. 7kW vs 8.5kW peak power available, winding dependent.
https://ebikes.ca/tools/simulator.h...X4503&batt_b=cust_84_.02_15&hp_b=0&bopen=true
It would be helpful to use the OP's constraints (as given in my simulator link) to illustrate your point.
The OP didn't specify a controler. So you're cherry picking a case to try to prove your point. I just used a different controller spec to disprove your point. Here it is with your battery and motor combo, but a different controller.
https://ebikes.ca/tools/simulator.h...=false&autothrot_b=true&frame=fat&frame_b=fat
 
serious_sam said:
The OP didn't specify a controler. So you're cherry picking a case to try to prove your point. I just used a different controller spec to disprove your point. Here it is with your battery and motor combo, but a different controller.
https://ebikes.ca/tools/simulator.h...=false&autothrot_b=true&frame=fat&frame_b=fat
I genuinely don't follow you. What point?
Your chart shows the same as mine: at a fixed speed, lower kV = higher efficiency. This is expected. The advantage of higher kV is higher maximum speed.
 
fatty said:
serious_sam said:
The OP didn't specify a controler. So you're cherry picking a case to try to prove your point. I just used a different controller spec to disprove your point. Here it is with your battery and motor combo, but a different controller.
https://ebikes.ca/tools/simulator.h...=false&autothrot_b=true&frame=fat&frame_b=fat
I genuinely don't follow you. What point?
Your chart shows the same as mine: at a fixed speed, lower kV = higher efficiency. This is expected. The advantage of higher kV is higher maximum speed.
Power.
 
Back
Top