PSA: Why small hub motors overheat and fail

Many people on this forum will buy the cheapest parts they can find. This may be on ebay, or even known vendors who are on this forum.

Hub motors, being relatively new technology, is in high demand right now. They are simpler (less maintenance, which is almost none), can be just as efficient as belt drive setups (raptor 2 for example), and are stealthier (No belt and pulley system, and supports FOC better, since they run at a lower erpm, which means almost no noise).

There are cons to hubs also, such as the lack of gearing (which means for speed freaks like me, more wasted battery and heat, since we need higher kv hubs to reach the speeds we want) and less urethane on hub wheels.

For most people though, hubs have become a more appealing choice, especially for new builders. Theres one problem though: Almost all of the hub motors on the market today are simply too small.

The reasons are simple:

  • Small motors have small stators. The stator can become over saturated by the magnetic field, and they suffer from what I like to call the nuclear effect. More on this in a second…
  • Small motors have little copper in them. What this means is they heat up quicker.
  • Small motors use thinner copper strands, because of space issues. What this means is for the same fill, you have more insulation. The insulation on a 24, 22, and 20 awg wire are all the same, and despite a 20 awg only having 40% more copper than a 24 awg, the 20 awg is rated at double the amps. The 18 awg is double the copper of a 24 awg wire, but is rated to 4 times the amps of the 24 awg. Why? Lower resistance and resistance is directly proportional to the amount of heat in the motor.
  • Once over 130 F, your magnets strength drops quickly. Now you need more amps for the same power. More amps = more heat.
  • Because of the combination of these three facts, the nuclear effect will drive the motor off the charts. This is why we see this difference in testings hummies hubs:

-old hubs: small stator, 15 turns of 24 awg, max temp of 260 F measured, will go higher if they didn’t almost catch fire at this point -new hubs: double stator length, 15 turns of 20 awg, max temp of 131 F.

Only 40% more copper and double the stator size, and we see all heat issues go away.

So what does this mean? Small motors are simply too small.

Now what do I mean by small hubs? I’m talking about jacobs hubs, maytechs hubs, koowheel hubs, and all of the direct from china hubs.

Now your asking yourself, why do so many people now make these inferior hubs? The answer is the same reason we have hover boards blowing up. Many manufacturers don’t care about the quality of their product, and the ones that show they do, don’t do adequate testing. For example, one manufacturer posted a video on youtube of their testing. Their test: 8 hours on a rack at full speed. The issue here, is that the current is very low without a load, fractions of an amp. With that low current, theres no way for the motors to get hot without a short in the motor. Compare that to the 20 amps you’ll pull with them to go up a hill.

There’s more to it than they don’t care also. The terrain, riders weight, riding speed, and riding style play a huge role in how many amps you’ll draw. On hummies small motors, I used to pull 4 amps on flat, and 10-15 on small hills. When you live in an area full of hills, you can see you’ll generate 3-4 times the heat than when in a flat area. This means 20-25% of the range you’d expect to get also.

Some riders do ok with these small hubs, as they weigh 130 lb or less, and ride at speeds under 10 mph, on flats.

However, if you plan to ride on hills, are more than 130 lb, and like to go over 10 mph, you’ll find these small hubs heat up quickly.

Some, like NVG, claim to have mastered the small motor, but why? Because they can do high heat, not because they have some magic way of cooling them. This is no solution, because once they got hot (which is quickly because of the little copper), you need more amps to get the same power from the copper. The magnets strength weakens, and now you need more amps to get the same magnetic field. And finally, the stator gets over saturated very quickly, which now means heat goes off the charts. Running a motor hot is not a solution to anything, because heat, creates more heat, which creates more heat. For programmers like me, this is the classic case of recursion with no base case. you end up with an exponential function of heat generation.

The solution? Simple… Bigger hubs. With hub motors, bigger, is simply better.

Now, why aren’t there more big hubs on the market?

  • Those in china producing and seller straight from the manufacturer are trying to be first to the market. Even if that means they poorly engineer them, and do no real tests on them to see how they perform, only that they work on a bench.
  • Bigger hubs need a custom truck. Why? Because the wheel base will be wider, you need to push them towards the center on the hanger more or your going to have a ridiculously wide wheel base.
  • Designing a new truck, in addition to the longer or wider hub, takes time, and they don’t want to wait. They want to sell now. It also costs more, and looks less attractive to customers who don’t know any better about hubs.

Now what does this mean?

As builders who want to use hubs as this time, unless your a slow, light rider who rides short short distances on flat, you have 3 options for hub motors:

  • R-Spec hubs (not sold separately, must buy raptor 2 complete)
  • Hummies NEW double wide hub motors
  • Carvon v3 or v4

If you feel like wasting your time and money, go ahead and buy small motors from vendors, on ebay, or straight from china, but I’m tired of seeing threads on here and get messaged by people who buy these small motors from china and are surprised that they failed. Now you know why they fail, so either you head this advice, or you risk wasting your time and money. The choice is yours…

28 Likes

Bigger is always better!

One question though. What are the benefits of having more poles on the stator? I have noticed that the rspec hubs have a lot more than carvon’s.

Well put.

I have found the meepo hubs are pretty fkin good though. I am 260 and I have a 10s4p 8.8ah pack and I ran it from 100-0 mostly up and down 5-10 percent gradient till the battery died and the hubs where not that hot. These 90mm hubs are massive and the esc limit the speed to 22mph so they are not the best performer, but they are very good intro hubs.

I am buying 2 of the 3 hubs(Hummie/enertion) you mentioned cause I am heavy and do want to go fast, but I must say the 90mm meepo hub is pretty good for the money. It works for me at 260 where as the backfire 2 motor kinda doesn’t. For heavy guys looking for a cheaper option, these hubs work well.

2 Likes

The stator hummie is using now is big enough. It’s narrow though, and the space inside for copper grows exponentially when you increase diameter instead of length. But then your on thin thane… You can’t win.

Your also limited in the stators not as large, so it can’t produce the same level of magnetic field. At some point, more copper on the same sized stator causes the nuclear effect. There must be the right balance.

Not really sure yet… We know more poles means more iron losses due to the constant flipping. But we have had at least 5 serious engineers tell us that doing 24 instead of 12 does not mean more torque. Possibly, it will find it’s position better when starting (less cogging), but with sensors, that’s eliminated. We will work on that more and are open to any engineers who want to share there opinions…

4 Likes

Love all the data in your post dude!

Does anyone actually know what size motors NGV used? The wind your own scorp. kits when wound at the factory weigh 658/660/664g a piece, with heli. guy on the forums saying custom winds fit a bit more copper again ~700g…

How many poles and slots are you using now? 12n14p is the popular choice for aircraft and effeciency whereas 12n10p is used for a lot of the heli motors that focus on power/weight and then the 1/5 car guys are running 4 poles and 12-24 slots…

no real consensus

How would belt drives/gear reductuon help acheive high speeds? Seems hubs will always be fastesr at given KV unless people start building 1:1 or overdriven belt drives.

Why would a 50mph hub board be any less efficient than a 50mph belt drive? The belt drive would need double to triple the KV to achieve the same speed right? So wouldnt losses be similar?

1 Like

You are also forgetting something very important, There is very little heat dissipation in hub motors and little airflow. My aluminum motor mount is very hot always and no doubt my exposed SK3 motor can breath cool air and it carries all of my weight with a 3:1 reduction, You will want to over design any hub motors in order to overcome these flaws.

2 Likes

I agree with your statement about hubs trapping heat, what I was questioning was the idea that hubs lose more energy too heat than a belt drive geared for the same high speed.

You can do an efficient gearing to 40 mph as far as I’ve seen (look at chakras free ride for example, with those tiny motors.

Hubs have 2 issues, one being they need a higher kv to run more efficiently, and the other is they cool while spinning in te open air. Colin is hard in hubs, which creates a bigger issue.

As @NNGG said, hubs will always need to be over engineered. But over engineered doesn’t mean more exspensive. Take gummies hubs. We recently found mjnor changes to the design that resulted in huge reduction of machining costs. This means we can afford to give better prices than we should be charging for motors this size. Not everyone has a motor working well enough to optimize. It’s been years of refining, and t will take China years to do the same, as many are just starting. if they ever do catch on.

This smells like a preamble for an upcoming product launch

1 Like

What do you mean higher KV? Higher than what? What KV does the ollin board use to acheive 40mph? I hit 40mph on 110-130kv hubs, are you saying his motors are lower KV than that? Or am I missing the point?

You can call it what you want. I’m just trying to share my research and development with the community. Do don’t have anyone else here sharing their secrets… Enertion won’t even tell me the size of their stator, while carvon isn’t sharing anything about their research and development. We have spent 10’s of thousands of personal money developing this, and I’m giving china the answer to their problem. Because I’m honestly tired of all the junk that’s on the market today. The vast majority of it, especially outside of this community. Out profits on these motors is not high, it’s just enough to keep researching and developing. There’s no classes about how to build or assemble motors. It takes time, money, and a dedication. I belief we have one of the solutions. I’m not bashing everyone though, and saying we are the only option. I’m opening admitting that the carvon and r-spect raptor 2 motors are good also. You don’t and won’t see others doing the same.

7 Likes

well loopy has lead the way…

First up I’ll say I think Hummie works hard to make a better product, and he is doing a lot of things I like a lot! and as before thanks for all the solid info but jmasta does kind of have a point. From the tittle this does come across a bit like some sort of stunt Jason would pull. Maybe instead of bagging everyone else just let us all know why what you have discovered works… a bit more factual, less conjecture, think of it like a research article.

lastly and ill quote Hummie here:“kv is kv”

1 Like

With all the hype around the ngv rig I think this post was completely relevant. “Urethane on hub” is still a huge weak link though.

1 Like

agreed

but as per above the hubs might weigh 700g sans thane. is that small?

Certainly weighty.

1 Like

yes and no. If your hub is small and long, you can still fit 23mm of urethane on every side of the hub and have an 83mm wheel. 28mm and we have a 90mm wheel. Much better than a wide hub. This is why every solution is different. I believe we all have working options, but they are all very different at the same time. There’s no one size fits all solution, and everyone will need to find what works best for them.

3 Likes

have you thought of using some kind of bridged core like a tang, flywheel… it would help with reducing the transfer of heat… increase air flow… makes it bigger but if it solves a problem?

1 Like

I think we would lose too much space. Wheel size would need to go up to around 100mm and even then, thane would be super thin. I’ll do a better analysis though, thanks for the idea.

Really, the point of this was to share that bigger prevents heat in the first place. I think that it’s better to prevent heat, than to deal with dissipating it. Once you move up to a motor around this size, heat doesn’t build up the same way and active cooling is not needed. More thane :stuck_out_tongue_winking_eye:

2 Likes