E-bike mid-drive motors hardly outperforms hub motors

General Discussion about electric bicycles.
Grantmac   10 kW

10 kW
Posts: 873
Joined: Oct 22 2018 12:43pm
Location: Victoria, BC

Re: E-bike mid-drive motors hardly outperforms hub motors

Post by Grantmac » May 23 2021 7:00pm

neptronix wrote:
May 23 2021 12:26pm
Grantmac wrote:
May 23 2021 10:50am
It's a fallacy to think 10spd chains are weaker. If you actually look at their ratings the 10 and 11spd are stronger than the older designs. The only difference is the plate thickness and that isn't where chains wear, it's in the rollers.
Chain line and staying in the larger cogs is how you save drivetrains.
That makes sense that a narrower chain would actually be stronger if the rollers were the same size. Less width means less flexing means less fatigue, etc.
Are the sprockets of an 11 speed drivetrain the same width and strength?

..do you have any articles or tests to share on this?
Just read a few articles (not at my disposal) about ebike specific drivetrains. Microshift, SRAM and Box are taking the approach of using a higher speed chain on lower number of cog cassettes in order to keep the chain line straight.

What's failing is the big and small cogs but for different reasons. The big ones because of the leverage with the chain line crooked bending them, the small because not enough tooth engagement.

Grantmac   10 kW

10 kW
Posts: 873
Joined: Oct 22 2018 12:43pm
Location: Victoria, BC

Re: E-bike mid-drive motors hardly outperforms hub motors

Post by Grantmac » May 23 2021 7:01pm

Ianhill wrote:
May 23 2021 10:53am
I don't think ill be getting a denzel gb200, dont look like it will hold much power for long really theres 2 main design issues thats going to be a problem and thats the stator plastic inserts thats wound round and the copper wind is lose as a kiddys shoe laces theres a monster airgap between the rotor and stator could even fibreglass tape the rotor theres so much space and thats going to be a low efficiency motor as a result.
Every inrunner ive had is dead from demag and self destruct so to me ipm is what this should be.
Its sad they keep knocking this out rather than step up the game they will get left behind they dont sooner or later but I'm out on this one maybe mk2 version they will get it more robust and rev happy but till then ill be waiting.

Edit

Im talking shit just found the web page a and theres a gb150 ipm version 2 speed like they got a time traveling phone box.

unnamed (10).jpg
Link please!!!!

TrotterBob   1 W

1 W
Posts: 60
Joined: May 10 2016 11:21am

Re: E-bike mid-drive motors hardly outperforms hub motors

Post by TrotterBob » May 23 2021 7:07pm

Performance numbers, charts and graphs are all well and good. But I think that leaves out the bigger picture. A hub motor is a large thing occupying the middle of a bicycle wheel. Unless its geared internally its spinning at a 1:1 ratio which is not ideal for a brushless motor.

And just like the physics experiment that most of us have either seen or done, turning that spinning mass causes gyroscopic effects.

Hub motors seem cumbersome to me. Clunky even. I like the idea of a 2 stage drive. Or even a single stage drive with a large belt driven pulley attached to the rear wheel.

donn   10 kW

10 kW
Posts: 733
Joined: Aug 13 2018 10:30am
Location: Seattle

Re: E-bike mid-drive motors hardly outperforms hub motors

Post by donn » May 23 2021 8:56pm

Interesting philosophical question. I saw it more or less the opposite way - charts and graphs providing a single data point reality check to all the "big picture" theories that eventually get to be treated as fact - i.e., myth. Of course his hub motor was geared internally, as would I think typically be the case in the 250W world.

I'd like to see a 2 or 3 gear "real" mid drive (real, as opposed to crank drives that on my recumbent wouldn't be any kind of "mid.") I probably wouldn't buy one though, as happy as I have been with my direct drive hub.

sleepy_tired   100 W

100 W
Posts: 150
Joined: Jul 02 2019 11:46am

Re: E-bike mid-drive motors hardly outperforms hub motors

Post by sleepy_tired » May 23 2021 9:49pm

TrotterBob wrote:
May 23 2021 7:07pm
Unless its geared internally its spinning at a 1:1 ratio which is not ideal for a brushless motor.
It's not a issue of ideal or not. If your goal is the best power to weight ratio possible, then it's a lousy design. If you want high torque motor, then there just isn't very much room in bicycle wheel for fitting wide magnets, so that sucks. Also cooling is less then ideal because it's a outrunner motor. The copper windings are what gets hot and those separated from the outside air by the motor housing.

But you know what? Direct drives are awesome because they are solid state. The only mechanical parts are the bearings. They are used all over the place because of this, just not in EVs. The trade off for everything I described above is is increased weight. The mass is not a bad thing. It's just a thing.

But you know what (also)?

Weight is one of the least important things when it comes to e-bikes. Especially ones that go faster then 15mph. Once you get going very fast then upsrung weight starts to matter for hitting bumps and such things. For speeds from 5-30, however, it's irrelevant.

Unless you have to carry it up stairs to a studio apartment or something like that. Then it's a big problem.
And just like the physics experiment that most of us have either seen or done, turning that spinning mass causes gyroscopic effects.
If you are worried about gyroscopic effects then a big wheeled bike isn't for you. Maybe you should consider a child's kick scooter. Much less gyroscopic forces with those tiny 4 inch wheels. Personally I like gyroscopic forces. It helps keep the bike going straight at speed. That's pretty nice.

Go watch a dirt bike race sometime and then come back and tell me how gyroscopic forces ruin handling (or whatever). Those big wheels and tires provide a hell of a lot more forces then a hub motor. Even a big one.

Hub motors seem cumbersome to me. Clunky even. I like the idea of a 2 stage drive. Or even a single stage drive with a large belt driven pulley attached to the rear wheel.
Direct drive motors are a elegant design. There is no other vehicle on planet earth that has less moving parts.

The ideal motor for a bicycle would be mid-drive with left hand chain reduction. This is what motorcycles use. This is what scooters use. But I don't see anybody actually doing that because it's a major problem packaging that into bicycle sized vehicles.


Piggy backing the motor on the human-power drive is a miserable approach. It's clunky, very complex, inefficient, wears rapidly, and is expensive. So, of course, this is the preferred approach by companies and bicycle dealers that make most of their money on parts and servicing. For actual end users it's really quite a bad idea.
Unless you are doing some serious off-road stuff on fast groomed paths were unsprung mass is a priority.

TrotterBob   1 W

1 W
Posts: 60
Joined: May 10 2016 11:21am

Re: E-bike mid-drive motors hardly outperforms hub motors

Post by TrotterBob » May 23 2021 10:15pm

A few hundred grams per wheel is going to be a whole lot more agile than a 14 lb hub motor.

Watch a dirt bike race? Those guys arent pedaling and the mass of the bikes and wheels absolutely does affect handling. I just dont accept that a hub motor would be agile unless its rather small.

A fit cyclist can pedal and produce a few hundred watts. Thats a far cry from the hp rating of a dirt bike.

Direct drive motors are a simple design. Elegant? Hardly.

20mph on a bike with 26 inch wheels works out to about 330rpm or so. Again that is nowhere near the peak efficiency of those hub motors or the speed controllers that drive them. And at slower speeds and during speed changes is where your batteries will really take a hit. Speeding up and slowing down a 14 lb motor takes a lot more energy vs spinning a relatively small hub motor and its gear reductions.

1boris   100 W

100 W
Posts: 131
Joined: Dec 06 2013 3:46pm

Re: E-bike mid-drive motors hardly outperforms hub motors

Post by 1boris » May 24 2021 9:30am

dogman dan wrote:
May 05 2021 6:01am
I was only considering performance in terms of can it cruise 25 mph, and carry 300 pounds up 10% grade without motor damage. But yeah, gearing could definitely improve acceleration across an intersection.

Not that a 500w rated dd running on 1000w from the controller accelerates all that lame. It was enough to blow all the 50 cc gas bikes and many of the 90cc bikes away leaving the start line at the death race.
The Mxus XF 19c motor easily do this.it feels stronger than my bafang bbs02 at most hills at least up to 15 % and that is while running at 48volt 22a vs bafang bbs 48volt 25 a

silentflight   100 W

100 W
Posts: 143
Joined: May 07 2007 5:42pm
Location: cloud 9.5

Re: E-bike mid-drive motors hardly outperforms hub motors

Post by silentflight » May 25 2021 3:41am

TrotterBob wrote:
May 23 2021 7:07pm
Performance numbers, charts and graphs are all well and good. But I think that leaves out the bigger picture. A hub motor is a large thing occupying the middle of a bicycle wheel. Unless its geared internally its spinning at a 1:1 ratio which is not ideal for a brushless motor.

None of the charts or graphs in the report authored and linked to by the OP refer to direct drive hub motors.

TrotterBob   1 W

1 W
Posts: 60
Joined: May 10 2016 11:21am

Re: E-bike mid-drive motors hardly outperforms hub motors

Post by TrotterBob » May 25 2021 5:01am

My mistake.

sleepy_tired   100 W

100 W
Posts: 150
Joined: Jul 02 2019 11:46am

Re: E-bike mid-drive motors hardly outperforms hub motors

Post by sleepy_tired » May 25 2021 9:45am

Watch a dirt bike race? Those guys arent pedaling and the mass of the bikes and wheels absolutely does affect handling. I just dont accept that a hub motor would be agile unless its rather small.
What you need to learn to accept is that unless you are racing on groomed mountain bike trails with jumps and whoops and things like that... it doesn't actually matter. Just because something exists in theory doesn't mean that it's actually going to cause you any issues in practice. In street driven bicycles it's not a problem.

Direct drive motors are a simple design. Elegant? Hardly.
Absolutely.
20mph on a bike with 26 inch wheels works out to about 330rpm or so.
So?
Again that is nowhere near the peak efficiency of those hub motors or the speed controllers that drive them.
I don't think you understand how motors work. If you are willing to spend money on high quality motor then efficiencies of DD hub motors can easily be well over the 80% range in practice. Cheap motors are in the more 75-80% range. Not simulated, not theoretical, but something that you can actually achieve on the street with street tires. Maybe up to 90% in ideal conditions if you have a very top of the line motor.

With a single speed reduction mid-drive, like I described in my previous post as the ideal motor setup for EVs, you can use a chain drive. Chain drives, if properly aligned and sized and cleaned and lubricated are the most efficient form of gearing available. Looking at something like 95% in practice. So you lose as low as 5% over DD on flat ground. When you get the chain offset, like on derailleurs, then efficiency drops. Belt drives are worse, efficiency wise, but because of the low maintenance they make up for it. Chain drives require a lot of work, relatively, to maintain peak efficiency.

So in situations were you are always accelerating and stopping or climbing hills then the single speed reduction motor wins over a direct drive. Better torque. For a EV it's a obvious win. No question about it. It's really not debatable. The problem is how to fit it on a bicycle. On motorcycles it's not a problem. Which is why they use this approach for high quality motor bikes.

Luckily bicycles are not high performance machines. So the downsides of DD hub motor are far less pronounced.


You seem to think that a 330 rpm is ridiculously slow for a ebike motor, but what about 75 rpm?

Conventional Ebike mid-drives, like you see sold by companies like Bosch, have to go through at least one stage of reduction, but more likely 2, before they ever get to the chains. Because they need to match the arbitrary cadence of a bicycle rider, which is in the 50-100 rpm range. Gear reduction are much less efficient then chain drives.

Something like a 20:1 gear reduction from a buzzing motor to the crank is, at best and under ideal conditions, probably around 76% efficient. Then you need to go through a chain on top of that. Derailleurs probably range in the 85-95% efficiency zone, depending on how offset the chain is.

So you have a .85 efficient motor * 0.76 * .85 to .95....

That's 54% to 61% efficient. Maybe with special lubrication and highly optimized gears a very high quality motor might get upwards to 70% at perfect speeds and cadence, but that is pushing it.

That's elegant? No, more like "monstrously ugly".

On top of that you now have a bicycle that is miserable to pedal. If you have a geared hub motor you won't reach the same efficiencies as a DD hub motor (in practice it doesn't matter noticeably), but at least you can have a nice time pedaling. If you use a controller that is capable of "virtual electric freewheeling" then you can pedal a DD hub motor like it doesn't exist and make up for the lost energy in regen.

You really need a specific situation to make up for all these downsides. Hill climbing competitions and going fast on mountain bike trails are were these sorts of convention bicycle mid-drive wins over hub motors.

And at slower speeds and during speed changes is where your batteries will really take a hit.
The speed isn't as super important as you imagine. It's torque output that matters.

Torque = Amps = heat.

If you are hauling up a hill efficiency dives because torque is high. You need high amperage to get high torque out of a motor and that produces a lot of heat. This is the advantage of having a mechanical gear reduction is simple: it's a torque multiplier. However you can achieve the same mechanical advantage by simply having a larger motor (wider and longer). In practice on large EVs single speed reduction and large motor is the best trade off.

However If you are just riding at a steady state then you'll get good efficiencies on a hub motor. No question about it. There is no disadvantage in just going slower with a hub motor.

https://ebikes.ca/tools/simulator.html? ... b=B5220_GA

at 32mph this setup is 82.6% efficient.

Dropping to less then 15mph gets you a grand total of 2.6% drop in efficiency. That's a fart in the wind at these speeds.
Speeding up and slowing down a 14 lb motor takes a lot more energy vs spinning a relatively small hub motor and its gear reductions.
You can more then make it up with regen.

Regen isn't going to add tremendous range to any sort of electric bicycle, but it saves wear and tear on the brakes, which is it's best feature. And for the sort of stuff you are pointing out the benefits of regen dwarfs it's downsides.

User avatar
neptronix   100 GW

100 GW
Posts: 15679
Joined: Jun 15 2010 5:56pm
Location: Utah, USA
Contact:

Re: E-bike mid-drive motors hardly outperforms hub motors

Post by neptronix » May 25 2021 9:57am

TrotterBob wrote:
May 23 2021 10:15pm
Speeding up and slowing down a 14 lb motor takes a lot more energy vs spinning a relatively small hub motor and its gear reductions.
Not really. The rotating mass of a DD hub is a magnet ring and case.
The rotating mass of a small geared hub is gears, a clutch, the case, maybe an inrunner stator, etc.
Even a small geared hub has much more friction loss than a DD.

There's a reason why small geared hubs, big DDs, mid drives, etc still exist and one particular drivetrain has not won the battle.

The ideal motor depends on your target speed, terrain, power requirements, speed requirements, etc.

The closest thing to a drive unit that works for all situations would be a swingarm mounted RC motor drive.
Although those tend to be uncommon because they are noisy, expensive, and difficult to fit.
"Love and compassion are necessities, not luxuries. Without them, humanity cannot survive." - Dalai Lama

My first major build: 1.6kW 8T MAC motor on a Trek 4500 MTB.
Monster MTB: Leafmotor 1500w @ 4kW on a Turner O2 full suspension.
The monster scooter: Heavy duty Cannondale semi recumbent - under construction.
Blue Dream: Maxaraya FS semi recumbent and high efficiency mid-drive - under construction.

john61ct   100 GW

100 GW
Posts: 7806
Joined: Dec 18 2018 2:06pm

Re: E-bike mid-drive motors hardly outperforms hub motors

Post by john61ct » May 25 2021 10:22am

Fantastic stuff here!

"Elegance" in the engineering sense is far from an objective goal.

All depends on both the use case and what the owner values.

For me, reliability / longevity and ease of repair in remote third-world locations are critical.

For someone else low noise levels and aesthetic considerations.

Many here value the "fun factor", biking as sport rather than utility transportation.

So there is **no** set of general statements that are "true", apply to everyone.

Grantmac   10 kW

10 kW
Posts: 873
Joined: Oct 22 2018 12:43pm
Location: Victoria, BC

Re: E-bike mid-drive motors hardly outperforms hub motors

Post by Grantmac » May 25 2021 10:27am

The company which integrates a mid-drive with a central gearbox (like a pinion) is going to get close to the ultimate EMTB drivetrain.

User avatar
neptronix   100 GW

100 GW
Posts: 15679
Joined: Jun 15 2010 5:56pm
Location: Utah, USA
Contact:

Re: E-bike mid-drive motors hardly outperforms hub motors

Post by neptronix » May 25 2021 10:50am

Yeah.
I've experimented with every drive system available other than RC rear chain drives.
Nothing wins as a general purpose drive. Each one has upsides and downsides.

I thought dual reduction geared hubs might have settled the debate, but every one on the market has severe heat shedding limitations which make their amazing power density potential impossible to achieve.
This could be solved with a complete rethink of how the motor is designed, or liquid cooling.

Mid drives like the lightest.bike and CYC pro show us that extreme power density that is possible in ebike drive systems.
If only someone put that much focus into hub motor design!
"Love and compassion are necessities, not luxuries. Without them, humanity cannot survive." - Dalai Lama

My first major build: 1.6kW 8T MAC motor on a Trek 4500 MTB.
Monster MTB: Leafmotor 1500w @ 4kW on a Turner O2 full suspension.
The monster scooter: Heavy duty Cannondale semi recumbent - under construction.
Blue Dream: Maxaraya FS semi recumbent and high efficiency mid-drive - under construction.

qwerkus   10 kW

10 kW
Posts: 703
Joined: Jul 22 2017 4:00am

Re: E-bike mid-drive motors hardly outperforms hub motors

Post by qwerkus » Jun 09 2021 11:14am

neptronix wrote:
May 25 2021 10:50am
Yeah.
I've experimented with every drive system available other than RC rear chain drives.
Nothing wins as a general purpose drive. Each one has upsides and downsides.

I thought dual reduction geared hubs might have settled the debate, but every one on the market has severe heat shedding limitations which make their amazing power density potential impossible to achieve.
This could be solved with a complete rethink of how the motor is designed, or liquid cooling.

Mid drives like the lightest.bike and CYC pro show us that extreme power density that is possible in ebike drive systems.
If only someone put that much focus into hub motor design!
Well, heinzmann kind of did (https://www.heinzmann-electric-motors.c ... cargopower); unfortunately that motor never went commercial. I still hope that if we all pm that company asking for the motor, they might consider finishing it. That's the only high power hub motor with dedicated oil cooling I know of.
6.jpg
6.jpg (92.58 KiB) Viewed 197 times

john61ct   100 GW

100 GW
Posts: 7806
Joined: Dec 18 2018 2:06pm

Re: E-bike mid-drive motors hardly outperforms hub motors

Post by john61ct » Jun 09 2021 7:13pm

I thought the bigger QS have liquid cooling?

moto not bike I know

TDB   100 W

100 W
Posts: 129
Joined: Apr 17 2016 5:44am

Re: E-bike mid-drive motors hardly outperforms hub motors

Post by TDB » Jun 09 2021 9:35pm

Ianhill wrote:
May 21 2021 10:56am
Anyone seen theses floating about like they down a manhole with IT the clown.
Screenshot_20210521-165016_Alibabacom.jpg

3kw nominal 12kw peaks ive seen it wheelie on a youtube video guy flipped it looked like it had loads of bottom end and its 4 speed with clutch.

Not the best solution from engineering stand point but it looked a blast dip the clutch and drop a hurt bomb on the wheel.

They been used on big bikes with 19inch front tyres but id like to see this on a pitbike 14 inch front wheel 12 in the back the smaller wheel would make it mental and it would ride exactly like a normal bike rear foot brake etc.

Good job dragging the thread horribly off topic.

Ianhill   100 MW

100 MW
Posts: 2535
Joined: Sep 25 2015 5:55pm

Re: E-bike mid-drive motors hardly outperforms hub motors

Post by Ianhill » Jun 10 2021 3:03am

TDB wrote:
Jun 09 2021 9:35pm
Ianhill wrote:
May 21 2021 10:56am
Anyone seen theses floating about like they down a manhole with IT the clown.
Screenshot_20210521-165016_Alibabacom.jpg

3kw nominal 12kw peaks ive seen it wheelie on a youtube video guy flipped it looked like it had loads of bottom end and its 4 speed with clutch.

Not the best solution from engineering stand point but it looked a blast dip the clutch and drop a hurt bomb on the wheel.

They been used on big bikes with 19inch front tyres but id like to see this on a pitbike 14 inch front wheel 12 in the back the smaller wheel would make it mental and it would ride exactly like a normal bike rear foot brake etc.

Good job dragging the thread horribly off topic.
The thread was been engaged by others than yourself all the best child.

TDB   100 W

100 W
Posts: 129
Joined: Apr 17 2016 5:44am

Re: E-bike mid-drive motors hardly outperforms hub motors

Post by TDB » Jun 10 2021 2:18pm

Pffft... Merkins! lol. Excuse me...

It could have been an interesting thread if it stayed on topic.

Post Reply