Trouble diagnosing controller problem.

[e-beach]

Been modifying and tinkering for years myself. I like your spirit!

In case you haven't seen these pages yet, they could help you out some.

https://ebikes.ca/learn/power-ratings.html

https://ebikes.ca/tools/simulator.html

Good luck with it, and let us know how it goes.

 

[markz]

Thats why Endless Sphere Exists!

Sure we could have purchased a complete ebike, but wheres the fun in that!

Good luck indeed rewinding that motor, jamiejackherer

Been modifying and tinkering for years myself. I like your spirit!


Good luck with it, and let us know how it goes.

 

[mxlemming]

http://www.bavaria-direct.co.za/scheme/calculator/

How many magnets you got? Alternating polarity?

I'm interested. As you've said, 1500W motors are far cheaper and lighter...

You might be able to get better torque with a different windings pattern but that's quite dependent on the exact arrangement whether it's possible atall.

 

[jamiejackherer]

Been modifying and tinkering for years myself. I like your spirit!

In case you haven't seen these pages yet, they could help you out some.

https://ebikes.ca/learn/power-ratings.html

https://ebikes.ca/tools/simulator.html

Good luck with it, and let us know how it goes.

Yeah I've come across them in the past but actually forgot about them.. I am computer programmer and software developer so I actually built my own custom tools for stuff but it was only basic stuff when I was starting out. 18650 configurations, range calculators etc.. but don't have the code any more and dont have time for coding ATM anyway [emoji23].

Cheers man

Sent from my SM-G970F using Tapatalk

 

[jamiejackherer]

Thought I'd do a quick update before work [emoji2].

I've decided to rewind the stator, it's still in excellent, like new condition. I have just about unwound the old copper, being careful not to bend the corners of the stator laminations.

Here's what I've noticed so far;

The turns are not consistent I have counted random turns of either 5, 7 or 6.5 [emoji848]. I think this might be a bad winding? Maybe an inexperienced builder in the factory..

The materials used in the motors construction are very poor quality and not designed for high power motoring applications.

Here's how I'm marking the slots and direction of turns;

[

Sent from somewhere using Tapatalk

 

[Thecoco974]

It look like the old winding was already quite compact ? If you want to increase copper section for more AMP you will be limited in space for extra turns. For a test you can rewind two neighbours slots with the old copper (same number of strand) and see if you can increase the number of turns or not.

Another thing to check before doing all the work is if the high temps have hurt the magnets, usually those magnet have a Currie temp at 80°c past wich they loose some magnetism.
You can try to feel if some are weaker than others with a Steel rod or by tuning it with the stator in and feel for cogging consistency.

Envoyé de mon Redmi Note 9S en utilisant Tapatalk

 

[mxlemming]

Another thought, it's failed below the hall sensor board. My hypothesis is that under this board there's substantially less cooling... Coil to air to outer plate... Air gets stirred for fairly good transfer, FR4 less so...

Might be worth re thinking the hall sensors a bit. Run the cables all the way over the coils with PTFE/silicone insulated wires and put the board somewhere in the middle.

@Thecoco974... Any idea what the magnet temp is compared to the coils? Obviously the magnets are strapped to the shell which has the most cooling by far. But do they get heated a lot by eddie currents?

 

[SlowCo]

So I burnt yet another motor out today [emoji91][emoji91]. It's a Voilamart 48v 1500w 31a with internal controller. Luckily I only paid £50 for it.

You need to get an external programmable controller and some cooling mods (Statorade and maybe Hubsinks) for the motor after rewinding!
Thread about cooling:https://endless-sphere.com/forums/viewtopic.php?f=2&t=48753

 

[Thecoco974]

To my understanding magnets Eddy current losses are considered negligeable in standard brushless motor. I'm only worried of the magnets temp through heat conduction, since the coils are known to have reach 200+°c we can assume that by conduction the core has reach 150+°c and through a 1mm air gap it's not impossible to have heated the magnets to much. It's only guessed numbers but i've already burned a small motor the same way and magnets were trashed.

Envoyé de mon Redmi Note 9S en utilisant Tapatalk

 

[jamiejackherer]

So I burnt yet another motor out today [emoji91][emoji91]. It's a Voilamart 48v 1500w 31a with internal controller. Luckily I only paid £50 for it.

You need to get an external programmable controller and some cooling mods (Statorade and maybe Hubsinks) for the motor after rewinding!
Thread about cooling:https://endless-sphere.com/forums/viewtopic.php?f=2&t=48753

Im now realising that cooling is very important [emoji23][emoji23].. looking at ferrofluid and fins.. can't do holes unless I fully waterproof the inside of the hub.. which is viable tbh.

And yeah I'm on a very tight budget but Im looking at sourcing a second hand sabvaton or something similar.

Sent from my SM-G970F using Tapatalk

 

[jamiejackherer]

So guys.. its rewind day!!

I have wound the first part of phase A.. I have started with 3 strands of 0.75mm. because I have 3 rolls it was going to be easier to wind 3 strands at a time, but I don't think I'll get the full 6 strands I wanted to fit in.. but on the up side I have managed to get 5.5 turns in. I haven't yet "squashed" the wire into each slot, so I may be able to fit in more than i think I will..

Sent from my SM-G970F using Tapatalk

 

[Lewp91]

Thinner why is better As the ennamel is also thinner trust me been there.. I wanted to use 0.71. then I found 3 calculators that proved me wrong.. the 0.31 (so long as its the thin 150c coat) will give the highest copper fill every time. (For amature hand winding anyway)

 

[larsb]

Most of you are wrong here, in several ways. Torque capacity of a motor is the same for all windings if you have battery and controller to drive it with current as needed.

There’s no such thing as a torque winding or slow winding either (what are you comparing with when you’re calling it slow? A 10-turn is slower than a 5-turn but is fast compared to a 20-turn at same voltage)

It all depends on the balance of drive voltage, drive current and having a winding that matches these two.

Then what thickness of wire to wind is a balance of
- increased eddy losses with thickness of wires due to skin effect
- increased winding time due to stiff wires
but also
- increased copper fill and because of this higher current capability.
- loss of turns for thicker wires since that last wire can’t fit in the last available space (normally the multiwire crossings can take more space than this)

Single wire winds with thick wire can be done without getting the wires laying crossed on the stator teeth (impossible when winding many strands of tiny parallell wires) so there’s no lost fill percent. One crossing equals one lost strand turn at a minimum. This is why single thicker wire winding can often be a superior if done correctly.

Thin wires are not better since they have thinner enamel. Yes, they are damaged a lot easier..
In relation to their copper area they have relatively more enamel than thicker wires and that’s what counts.

 

[Chalo]

Most of you are wrong here, in several ways. Torque capacity of a motor is the same for all windings if you have battery and controller to drive it with current as needed.

This is true, but high turn counts give more torque per phase amp. Increasing phase current is usually more expensive and difficult than increasing voltage, so if you want to maximize torque, a high turn count can be a more feasible way to get there than a low turn count with high phase amps.

 

[larsb]

Agreed :thumb:

 

[John in CR]

If you overheated a motor already, then the only way to get better performance is to increase cooling, change how your ride to make less heat, and/or include temperature sensing so your controller will automatically cut back current at a given motor temp and cut power completely if temps continue to climb.

Personally I find increasing cooling the easiest and most effective route, but invest in a better motor first.

edit- Improved motor control can also reduce heat through greater efficiency, but you generally won't feel a big improvement in the raw torque of a cool motor. It's more of a trimming of the waveform to make it match better to exactly what the motor needs for less waste electricity sent through it.

 

[jamiejackherer]

Most of you are wrong here, in several ways. Torque capacity of a motor is the same for all windings if you have battery and controller to drive it with current as needed.

There’s no such thing as a torque winding or slow winding either (what are you comparing with when you’re calling it slow? A 10-turn is slower than a 5-turn but is fast compared to a 20-turn at same voltage)

It all depends on the balance of drive voltage, drive current and having a winding that matches these two.

Then what thickness of wire to wind is a balance of
- increased eddy losses with thickness of wires due to skin effect
- increased winding time due to stiff wires
but also
- increased copper fill and because of this higher current capability.
- loss of turns for thicker wires since that last wire can’t fit in the last available space (normally the multiwire crossings can take more space than this)

Single wire winds with thick wire can be done without getting the wires laying crossed on the stator teeth (impossible when winding many strands of tiny parallell wires) so there’s no lost fill percent. One crossing equals one lost strand turn at a minimum. This is why single thicker wire winding can often be a superior if done correctly.

Thin wires are not better since they have thinner enamel. Yes, they are damaged a lot easier..
In relation to their copper area they have relatively more enamel than thicker wires and that’s what counts.

I have ridden bikes with different number of windings. The 3.5T motors are slower to accelerate but have a higher top speed. 5T motors have a much lower top speed but have really brutally high torque when setting off from a standing start or when climbing hills.

A perfect example is that the 3.5T motor was not able to climb my testing hill (40% incline) but the 5T motor actually accelerated up the same hill. This is using the same battery, controller and bike. The only change was the number of winding turns in the motor.

How can this be if what you say is correct?

Sent from my SM-G970F using Tapatalk

 

[jamiejackherer]

If you overheated a motor already, then the only way to get better performance is to increase cooling, change how your ride to make less heat, and/or include temperature sensing so your controller will automatically cut back current at a given motor temp and cut power completely if temps continue to climb.

Personally I find increasing cooling the easiest and most effective route, but invest in a better motor first.

edit- Improved motor control can also reduce heat through greater efficiency, but you generally won't feel a big improvement in the raw torque of a cool motor. It's more of a trimming of the waveform to make it match better to exactly what the motor needs for less waste electricity sent through it.

It frustrates me greatly when I hear this.

Do you honestly believe that a motor cannot be rewound to handle more voltage and/or current?

The materials used in these motors is very basic and cheap materials. Like the insulating paper that I took out actually sets on fire when applying a flame. And will begin to melt at 130°C. The enamel on the copper wire is very low quality/thin and is easily scratched off.

So by using high quality materials, it is obvs possible to run more power and not experience the materials melting or burning.

Sent from my SM-G970F using Tapatalk

 

[j bjork]

Most of you are wrong here, in several ways. Torque capacity of a motor is the same for all windings if you have battery and controller to drive it with current as needed.

There’s no such thing as a torque winding or slow winding either (what are you comparing with when you’re calling it slow? A 10-turn is slower than a 5-turn but is fast compared to a 20-turn at same voltage)

It all depends on the balance of drive voltage, drive current and having a winding that matches these two.

I have ridden bikes with different number of windings. The 3.5T motors are slower to accelerate but have a higher top speed. 5T motors have a much lower top speed but have really brutally high torque when setting off from a standing start or when climbing hills.

A perfect example is that the 3.5T motor was not able to climb my testing hill (40% incline) but the 5T motor actually accelerated up the same hill. This is using the same battery, controller and bike. The only change was the number of winding turns in the motor.

How can this be if what you say is correct?

Sent from my SM-G970F using Tapatalk

How does that make anything in larsb:s post incorrect? Just as he said it is a balance of phase amp and volt. If you use two different windings on the same controller you will get different result. Your controller simply cant feed the 3,5turn motor enough phase amp.

Btw, I think it looks like a few strands in your motor was shorted to the stator lamination's or something. If you really overheated the motor I think it should not be just a few strands. But maybe I am wrong, as no one else said anything

 

[jamiejackherer]

Most of you are wrong here, in several ways. Torque capacity of a motor is the same for all windings if you have battery and controller to drive it with current as needed.

There’s no such thing as a torque winding or slow winding either (what are you comparing with when you’re calling it slow? A 10-turn is slower than a 5-turn but is fast compared to a 20-turn at same voltage)

It all depends on the balance of drive voltage, drive current and having a winding that matches these two.

I have ridden bikes with different number of windings. The 3.5T motors are slower to accelerate but have a higher top speed. 5T motors have a much lower top speed but have really brutally high torque when setting off from a standing start or when climbing hills.

A perfect example is that the 3.5T motor was not able to climb my testing hill (40% incline) but the 5T motor actually accelerated up the same hill. This is using the same battery, controller and bike. The only change was the number of winding turns in the motor.

How can this be if what you say is correct?

Sent from my SM-G970F using Tapatalk

How does that make anything in larsb:s post incorrect? Just as he said it is a balance of phase amp and volt. If you use two different windings on the same controller you will get different result. Your controller simply cant feed the 3,5turn motor enough phase amp.

Btw, I think it looks like a few strands in your motor was shorted to the stator lamination's or something. If you really overheated the motor I think it should not be just a few strands. But maybe I am wrong, as no one else said anything

Firstly, I never said "incorrect" I asked how it "is correct".

I'm still in the early stages of learning so hopefully you explain this to me. Let's put very technical stuff to one side for a moment.

Phase amp/current = the current pulled by each phase.

So if my ammeter is reading 100A, that means I have 100A being pulled from the battery. And in a 3 phase motor only 2 phases are ever active at one time. That means the phase amperage is 50A. So as long as the parallel strands are able to carry those 50A per phase, whether we have 3.5T or 5T, we still have 50A in each phase?

It all very well and good telling me it this way or that way, or that doesn't work like that etc. But people should really be explaining why....

Now additionally, I wonder, if my controller has a max amp rating of 105A but I'm only pulling 50A in total, surely that means I still have 55A available to be pulled? Surely the number of turns only affects how that power is used?

And additionally again... You said that a different number of turns will give a different result when using the same controller.... Ok cool.. thats my aim here. I want a different result! I want to have higher torque.

So can anybody explain how I can achieve this? One thing I find on this forum is that many people are able to give their opinion, but rarely back this opinion up with real knowledge. And they seem to know everything but are unable to answer the simple questions.


I want to increase the torque of the motor.

1. Why do motors come in 5T and 3.5T variants that will have more torque and less RPM, and less torque and more rpm respectively, if number of turns is irrelevant?

2. How can I increase the torque of a given motor when rewinding it? Thicker wire? More turns? Both?

P.S. nothing shorted to the stator. The wires enamel got so hot it burned away which caused the shorting to another phase that was not shown in the pictures. Besides the wires you see are not next to the stator but in the outside so it's not possible to short to the stator I'm afraid.

Sent from my SM-G970F using Tapatalk

 

[larsb]

So can anybody explain how I can achieve this? One thing I find on this forum is that many people are able to give their opinion, but rarely back this opinion up with real knowledge. And they seem to know everything but are unable to answer the simple questions.

We have enough knowledge to explain it but what we don’t have is the job to explain it to you. Just read some motor theory, it’s in the educational links in the motor section, or you can search among my threads for ”the myth”. There are plenty of explanations here on ES already, no need for new ones. It isn’t deep knowledge either, if you can’t be bothered to read up on the basics yourself then, in my view, you can hardly be helped.

One simple pointer:
Torque is proportional to coil field.
coil fields can be measured in ampereturns. That’s ampere*turns
Higher amperage, higher field
Higher turn count, higher field

This goes proportional up to where the steel saturates and flux starts to go outside of the correct magnetic circuit. This limit is fixed for any stator design and dependent on the steel and quality of laminations. If max field is 100 ampereturns then you can achieve the same max through 10A*10turns or with 100A*1turn, remember same max field, same max torque.

As for the coil copper losses: they are only dependent on the copper fill percent, not on the winding dimension.
Why? Ohms law.
a 4 turn winding has 4 times the length of copper and 1/4 of the cross section but needs only a quarter of the current to create the same field as a 1 turn winding.

Thats:
4x higher losses due to higher length
4x higher losses due to lower cross section
but 1/4 current needed to get same ampereturns

As resistive losses are squared proportional to current following ohms law the comparable losses are —> 4*4/4*4=1 —> so losses are the same regardless of winding if winding is done to same fill percent.

The motor on your pic has really poor copper fill so yes, it can be improved if you can wind.

 

[John in CR]

If you overheated a motor already, then the only way to get better performance is to increase cooling, change how your ride to make less heat, and/or include temperature sensing so your controller will automatically cut back current at a given motor temp and cut power completely if temps continue to climb.

Personally I find increasing cooling the easiest and most effective route, but invest in a better motor first.

edit- Improved motor control can also reduce heat through greater efficiency, but you generally won't feel a big improvement in the raw torque of a cool motor. It's more of a trimming of the waveform to make it match better to exactly what the motor needs for less waste electricity sent through it.

It frustrates me greatly when I hear this.

Do you honestly believe that a motor cannot be rewound to handle more voltage and/or current?

The materials used in these motors is very basic and cheap materials. Like the insulating paper that I took out actually sets on fire when applying a flame. And will begin to melt at 130°C. The enamel on the copper wire is very low quality/thin and is easily scratched off.

So by using high quality materials, it is obvs possible to run more power and not experience the materials melting or burning.

Sent from my SM-G970F using Tapatalk

You're not nearly as frustrated as I am for having to correct people on this topic for more than 10 years on this forum. The only way you can improve the performance of a motor other than to improve heat dissipation is to fit more total copper on the windings. With a change in the turn count on the stator teeth windings all you accomplish is a change in the combination of voltage and current required to achieve the same result. The rpm, torque and power limits remain the same. If you increase the turn count to get more torque per amp, you also decrease current handling by the same % since the copper windings are thinner in total and longer. Power = Torque X RPM , and if you try to increase power by the other route, increasing RPM/volt , you end up decreasing torque/amp with less turns of shorter thicker copper.

Sure you may be able to get some improvement using better materials, as long as you can wind at least as well as the robotic winding machines or experience professional winders in China, but again why go to all that trouble putting that effort into a motor design to be as cheap as possible and as a result are only moderately efficient at best with stators made with the cheapest steel laminations available.

Spending hard earned treasure and valuable time to make a hubmotor able to run hotter is a bad idea sure to result in another failure. A far more effective route is make your motor run cooler.

 

[j bjork]

The myth:
https://endless-sphere.com/forums/viewtopic.php?f=2&t=64907

And the battery amp and phase amp is not the same. You can have a lot higher phase amp at a lower voltage than the battery amp and voltage.

 

[larsb]

but again why go to all that trouble putting that effort into a motor design to be as cheap as possible and as a result are only moderately efficient at best with stators made with the cheapest steel laminations available.

Agreed, by the look of that winding one might assume that the rest of the motor is indeed made of the cheapest possible materials - it might be turd polishing to rewind it. If OP has the cash it could be better to just buy a better motor.

 

[jamiejackherer]

if you can’t be bothered to read up on the basics yourself then, in my view, you can hardly be helped.

I really don't appreciate being told I can't be bothered to do something!!!

I have 2 jobs, 6 kids and I am studying a degree aswell... It's not a case of "can't be bothered". Its a case of literally not having the time.. I don't have transport until I can get this done, and I am waiting on some insulating materials currently.

But please don't assume that I can't be bothered! It's just plain rude.

While I appreciate what information you have given me, if you cant help then dont. I'm asking for help so I don't have learn everything about motors just to do a simple rewinding.

Sent from my SM-G970F using Tapatalk