Hub motor industry dying?

[Grantmac]

The Aden is interesting simply because of how easily it can be added to an existing bike. In a way it's similar to a friction drive with a large helical gear instead of the tire.
Throwing 4# onto the swingarm is still a bit questionable from a handling perspective but not an appalling idea.

1300w isn't exactly huge power but would easily "self-shuttle" back up most hills. Looks like they built it around a backpack battery as well.

 

[PaulD]

The Aden is interesting simply because of how easily it can be added to an existing bike. In a way it's similar to a friction drive with a large helical gear instead of the tire.
Throwing 4# onto the swingarm is still a bit questionable from a handling perspective but not an appalling idea.

1300w isn't exactly huge power but would easily "self-shuttle" back up most hills. Looks like they built it around a backpack battery as well.

I think 1300w with a large gear reduction is a lot of power. Maybe 1300w would be not enough for dirt bike speeds if you had a heavy DD hub motor bike where most of it would go into heat climbing up a steep hill.

I'm interested to see how well they clamp onto the seat stay. There are lots of different shapes of tubing, and I imagine you need a good solid connection to prevent the motor from rotating.

 

[flat tire]

To make a productive contribution to this thread, I see hub motors as the best case when you want a bike that will mostly ride on smooth roads, has great acceleration, and most of all is really quiet.

They're not good for offroad or racing because the bike handles funny at the limits. But hubs definitely have their place. You need more than one bike anyway to adequately exploit your surroundings.

Also, China sells tens of millions of hub motors every year, which is insane.

 

[Grantmac]

Given the size of the driven gear I'd say they have a pretty high reduction going on. I wonder if they are using an inrunner or outrunner? I wouldn't want to introduce a bunch of mud to that gear.
The concept is definitely very cool though. An easy bolt on 1300w is nothing to sneeze at.

I bet I rarely use more than 1kw on MTB paths. It's when I'm riding where you'd take a motorcycle that 2.5kw feels underpowered.

 

[neptronix]

Hubs are still a fantastic option, but it seems like we're close to reaching the limits of how much better they can get without the cost approaching that of a mid drive. A big selling point is still less moving parts = less things to break, and low cost. hub motors aren't going anywhere.

The first dual reduction geared motors have been a huge disappointment but i believe there is room to improve them.

DD hubs' biggest limit is the slow RPM they turn at. You get poor power density in anything larger than a 20" wheel. Yet the fact that people buy the heaviest DDs ever made - the QS motors - shows that power density doesn't matter so much to some.

Direct to rear wheel chain drive is still the ultimate solution that gives the best of all worlds.

 

[PaulD]

The first dual reduction geared motors have been a huge disappointment but i believe there is room to improve them.

Are you referring to the "first gen" geared hub motors like Cute or Mac? or the newer Bafang GXXX hubs? I haven't heard much long term test info on the Bafangs. Is it the durability or noise or performance that's disappointing?

 

[John in CR]

What if we not only kept the gears, we added more, but pulled them out of the motor body as a separate transmission component? Arguably the reason for DD weight and inefficiency and geared hub fragility is that they have to operate at suboptimal speeds, and hence the attraction of crank drives. A lightweight variable speed transmission on that Aden system, and you'd really have something, though of course it wouldn't be a hub motor.

Properly sized and geared electric motors don't need multiple gears. The best example is trains, which have electric motors but no multi-speed gearboxes.

 

[neptronix]

Are you referring to the "first gen" geared hub motors like Cute or Mac? or the newer Bafang GXXX hubs? I haven't heard much long term test info on the Bafangs. Is it the durability or noise or performance that's disappointing?

Key term here is 'dual reduction'. So i mean the Cute 'H' motors ( blah - they're using thickish lams ), the Bafang Gxx ( 0.35mm lams and lots of design mistakes ), Xiongda YTW-06 ( just total garbage ), etc.

I believe that these hubs could use a combination of using thinner lams + change the casing for better heat dispersion, and in that case, they'd make killer mid drive and also rear wheel motors..

The Bafang G310 barely nudges ahead of the MAC in terms of continuous power per watt. It should be able to do better than that.

 

[efMX Trials Electric Freeride]

What if we not only kept the gears, we added more, but pulled them out of the motor body as a separate transmission component? Arguably the reason for DD weight and inefficiency and geared hub fragility is that they have to operate at suboptimal speeds, and hence the attraction of crank drives. A lightweight variable speed transmission on that Aden system, and you'd really have something, though of course it wouldn't be a hub motor.

Properly sized and geared electric motors don't need multiple gears. The best example is trains, which have electric motors but no multi-speed gearboxes.

trains run on well graded tracks, not usually very steep terrain where gears can have more benefits..

 

[sleepy_tired]

Hubs are still a fantastic option, but it seems like we're close to reaching the limits of how much better they can get without the cost approaching that of a mid drive. A big selling point is still less moving parts = less things to break, and low cost. hub motors aren't going anywhere.

From a engineering point of view direct drive hub motors are pure sex. What other vehicle can you get away with having what amounts to a solid state drive train? That has to be somebody's holy grail somewhere.

The fact that they are cheap and mundane and heavy are all good things. Yeah sure they can be improved and I really want to see that, but I am happy right now for what they are. But when you try to tell some guy that spent a extra 900 dollars on his bike to save 2 pounds of weight that he should slap a 14 pound motor to his bicycle.. yeah I can see why that would be unattractive to some people.

Are you referring to the "first gen" geared hub motors like Cute or Mac? or the newer Bafang GXXX hubs? I haven't heard much long term test info on the Bafangs. Is it the durability or noise or performance that's disappointing?

The Achilles heal of geared hub motors is cooling. You have a small motor spinning inside of a sealed steel case. There is nowhere for that heat to go. This limits the amount of torque they can produce... torque is amps and amps is heat. For example: The Bafang 31x motors are awesome, but can't really handle continuous speed up past 25mph by themselves due to the thermal limitations. Hill climbing they can provide a lot of boost, but you are going to need to pedal and keep the speed up on big hills or you are going to hit their thermal limits after a few minutes.

When you take a 9c-style direct drive hub motor and throw statorade oil cooling at it then you boost the torque potential by a easy 50%. This means that right now a 300 dollar high efficiency 1500w DD hub motor with statorade and a 20 inch wheel will out climb any other type of ebike out there. People talk about needing mid-drives to climb hills, but I doubt most common mid-drives people use on bicycles will be able to match that.

Now entering fantasy land...

If you can solve the cooling issues with geared hub motors then there wouldn't really be any reason to use anything else for 90% of people out there, I am guessing. If you can get a similar 50% boost then a little 250w class motor like the Bafang suddenly turns into a 400w class motor, and peak torque goes from 45nm to almost 70nm.

A bigger high-efficiency motor like the Em3ev-sold 'Improved' Mac motor or Grin's GMAC motor would go from something like 80nm peak torque to 120nm and be rated at 1500 watts continuous.

So you end up with a bike that would be able to tackle most hills pretty easily and provide a moped-like experience for people who want that... yet be still almost 100% pedal-friendly.

 

[sleepy_tired]

trains run on well graded tracks, not usually very steep terrain where gears can have more benefits..

They also pull trains measured that can be measured in miles and can travel in excess of 75mph. The main limitation on the size and speed of trains is braking, not motor power.

What about those monster dump trucks that are used in quarries? Those are are not on nicely graded roads and (I am guessing) most are diesel-electric and run single speed reduction (multiple levels of reduction, though) gearboxes and carry gigantic loads.

I find it useful to tthink of electric motors as transmissions. They are not like ICE motors and shouldn't be thought of in the same manner. Their job is to take amperage and turn it into rotational torque. Based on their windings and physical dimensions they can be 'geared' to different loads and amperages. Their ratings are much more akin to the type of ratings you see on 4 speed automatic transmissions then in a V6 motor.

The purpose of the geared reduction is to reduce the size and weight of the electric motor to fit vehicle packaging constraints. Otherwise there wouldn't be any point to it.

 

[efMX Trials Electric Freeride]

mines and quarries roads are also graded to the degree possible.. not usually comparable to steep mtn bike and motorcycle off-road trail and trials terrain imo..

 

[neptronix]

From a engineering point of view direct drive hub motors are pure sex. What other vehicle can you get away with having what amounts to a solid state drive train? That has to be somebody's holy grail somewhere.

Yup. Zero drivetrain losses. They're a thing of beauty. Especially that CSIRO 97% efficient hub
You could never get that kind of fantastic efficiency out of the best possible geared motor design.

DDs of today barely pass the 90% efficiency mark but IMHO we haven't seen a mid drive that can even peak at 90% yet - maybe some of the astro mid drives do? ( $$$$ )

When you take a 9c-style direct drive hub motor and throw statorade oil cooling at it then you boost the torque potential by a easy 50%. This means that right now a 300 dollar high efficiency 1500w DD hub motor with statorade and a 20 inch wheel will out climb any other type of ebike out there. People talk about needing mid-drives to climb hills, but I doubt most common mid-drives people use on bicycles will be able to match that.

No kidding, i calculated that i can climb up the rockies on a 20" without statorade.. :lol:

Now entering fantasy land...

If you can solve the cooling issues with geared hub motors then there wouldn't really be any reason to use anything else for 90% of people out there, I am guessing. If you can get a similar 50% boost then a little 250w class motor like the Bafang suddenly turns into a 400w class motor, and peak torque goes from 45nm to almost 70nm.

I think geared hubs could probably do even better than that. They'd have to be designed completely different, using something like bearings that are 160mm in diameter, and having the center of the motor casing being non-rotating, but used as a heatsink for the stator. You would build an inrunnner type system and the dual gear reductions would take up one side of the motor, the stator would be a pancake shape and sit on the other side. The stator and magnet assembly could be sitting in a bath of oil separate from the gears. Or you bathe both. Minus the friction of spinning the oil around, the continuous power of such a motor would be completely nuts.

I still have in mind a kit for driving the rear wheel directly, that'd adjustable to fit most hardtail MTBs. I'm actually surprised that rear chain drive isn't more common. A lot of mid drives can have the final gear reduction moved, and then you just route a chain to the rear wheel and tada..

 

[sleepy_tired]

mines and quarries roads are also graded to the degree possible.. not usually comparable to steep mtn bike and motorcycle off-road trail and trials terrain imo..

You have it backwards. Nothing ebikes do compare to anything the quarry trucks must handle.


I suspect the main reason we see companies like Bosch pumping out mid-drives has more to do with the same sort of reason they lock controller settings down, restrict diagnostic tools and software to officially licensed dealers and technicians, and prevent third party batteries from working. Dealerships are extremely important to that class of bicycle market and most dealers are not going to push bikes that are not going to make them any money. So the bicycles are designed to keep people returning to dealerships for service and support.

That's the way it is for automobiles, anyways. I've read that spare parts and maintenance can amount over 30% of the dealership profits. The other major profit center is financing. They make more money from upgrade options and services then they do from the actual selling cars in the first place. The cars are designed to protect dealer profitability, otherwise dealers won't direct customers to buy them. I don't see why bicycle dealerships are any different.

Note that I don't think that this is a evil system or anything. A bit predatory, but what isn't.

I tell people looking to buy ebikes that if they are not going to do DIY and they want something that is reliable they are a lot better of getting some Shimano-based ebike from the LBS then the cheapest thing from Amazon or Aliexpress. This way when the bike has problems they can get support and parts. Virtually guaranteed fix in a few days. They don't even need to visit the same store... anybody who sells that style of bike would be able to help them. I think that is well worth the price of admission if you need to have something that 'just works'.

I suspect that most bottom-barrel bikes that people buy over the internet end up being ridden a few times for a few months until some wire gets pulled out or the battery craps out and then it just ends up as more junk in the back yard or hidden in the back of the garage.

 

[PaulD]

I believe that these hubs could use a combination of using thinner lams + change the casing for better heat dispersion, and in that case, they'd make killer mid drive and also rear wheel motors..

Couldn't have said it better Thin lams, custom winding and direct thermal path from stator to shell.

 

[efMX Trials Electric Freeride]

mines and quarries roads are also graded to the degree possible.. not usually comparable to steep mtn bike and motorcycle off-road trail and trials terrain imo..

You have it backwards. Nothing ebikes do compare to anything the quarry trucks must handle. <snip>

I disagree, quarries use roads which are by definition graded.. steep off-road trails not necessarily.. and don't forget the throttle ebikes, so not limited to just low power pedal assist bikes.. and quarry trucks use reduction gearing as you mention..

 

[donn]

trains run on well graded tracks, not usually very steep terrain where gears can have more benefits..

Yeah. I'm no expert at this, but for my own amusement I would call up a modest size geared hub on the Motor Simulator, configured appropriately (like if it's supposed to be a 350W motor, 36V battery and 10A controller.) Put a good load on it and a non-trivial grade -> meltdown, in X minutes.

Change the wheel size from 26" to 16" -> goes faster, never overheats. That strongly suggests to me that an extra gear or two would go a long ways towards satisfactory performance under a reasonable range of bicycle loads. If we had a light, robust solution for that, I suspect we'd quickly start thinking maybe the hub is a crazy place to put a motor.

 

[neptronix]

Change the wheel size from 26" to 16" -> goes faster, never overheats. That strongly suggests to me that an extra gear or two would go a long ways towards satisfactory performance under a reasonable range of bicycle loads. If we had a light, robust solution for that, I suspect we'd quickly start thinking maybe the hub is a crazy place to put a motor.

That's why mid drives exist, but then you get systemic drivetrain problems beyond a certain wattage, so unless you do something like build a BMX chain based derailleur, or very tough internal geared hub, you're gonna have a bad time.

I would rather have a DD in a tiny wheel with razor thin 0.2mm laminations to reduce the eddy currents to as close to zero as they're going to get, and have a huge surplus of power without the rolling resistance penalty. But i can't convince anyone to make me such a beast. Rear chain driven systems can get close to that ideal of having a single ratio, but operators are standing by for a huge burst of torque. unfortunately you take a hit with transmission losses from running such a low ratio of load to friction in that case. Which is why we go back to DD.

I'd be singing a different tune for sure if mid drive transmissions were better. I'd e on the mid drive train, hard!

 

[MadRhino]

...But i can't convince anyone to make me such a beast.

Draw it. Label it private and confidential. Let it leak to China. They’re gonna make it for sure if they believe they stole it. :wink:

 

[rbrak29]

IMO the hub motor vs mid-drive debate as far as the main stream US bicycle industry is concerned is whether to keep the old diamond frame or develop a new frame around the mid-drive for the e-bike. Many companies will wait until a clear winner emerges or not even enter the e-bike market at all given the fragmented US e-bike laws. Apparently Specialized did not think hub motors were the answer. Mid-drives have expensive upfront costs and costs down the road with increased drive train wear and tear. They look nice, better frame integration (drive and battery), weight distribution, etc. In my case give me a quality, robust yet efficient, lighter hub motor, lighter battery with long range from a company that will support what it sells with a network of supporting bike shops. I am comfortable with DIY solutions, but a vast majority of people are not.

 

[MadRhino]

I like riding single tracks with a modern, fancy BB drive bike, but I kill it faster than tires. And, overall I am slower because of the speed my hub can achieve in nice sections, and climbing especially.

The frame is not a problem. The light weight BB drive is.

 

[flat tire]

DDs of today barely pass the 90% efficiency mark but IMHO we haven't seen a mid drive that can even peak at 90% yet - maybe some of the astro mid drives do? ( $$$$ )

Astro can do that driving a propeller but on a ebike the reduction eats that up to the point where you're back at a good hub motor.

 

[John in CR]

DDs of today barely pass the 90% efficiency mark but IMHO we haven't seen a mid drive that can even peak at 90% yet - maybe some of the astro mid drives do? ( $$$$ )

Astro can do that driving a propeller but on a ebike the reduction eats that up to the point where you're back at a good hub motor.

Just to put what is the efficiency king to rest, the DD hubmotors I use have a peak efficiency above 95% at the voltages I run, and they have a combination of thin lams and low pole count to accomplish it. Then there's the $10k+ hand made DD hubbies used for solar competitions that hit over 98% efficiency.

As far as the multi-speed gearbox thing, in addition to locomotives and giant dump trucks previously mentioned, don't forget Tesla's. Tesla even tried 2 speed trannies in the earliest test Roadsters, but transmissions just don't stand up to the torque of a properly sized motor. Electric motors have a wide torque band and don't need multi-speeds unlike ICE's which have quite narrow torque bands.

 

[donn]

As far as the multi-speed gearbox thing, in addition to locomotives and giant dump trucks previously mentioned, don't forget Tesla's. Tesla even tried 2 speed trannies in the earliest test Roadsters, but transmissions just don't stand up to the torque of a properly sized motor. Electric motors have a wide torque band and don't need multi-speeds unlike ICE's which have quite narrow torque bands.

Can't make multi-speed work, is different from don't need multi-speed. If Tesla tried it, I think we can assume they did so for a reason, maybe for the same reason that small hub motors burn up on heavy loads.

 

[efMX Trials Electric Freeride]

when tesla makes a jeep it will have a transmission..