250W E-bike mid-drive motors hardly outperforms 250W hub motors

[avandalen]

I wrote an article where you can read that a mid-drive motor hardly outperforms a hub motor.

Compared: mid-drive motor, hub motor, Xiongda 2-speed hub motor

Hub motor simulation

www.avdweb.nl

In practice there is no significant difference in performance. You can clearly experience this yourself during test drives with both kind of motors, I did this myself too. Because I wanted to know the cause, I researched this subject extensively. The conclusion is that the advantage of mid-drive motors is only noticeable in extreme situations, on steep slopes while the cyclist pedals only a little or not.

I would like to know how others think about this.

 

[K PierreR]

On pavement, I think a hub motor is probably a better choice. I ride about 80% of the time on pavement. At times I am riding at 8-10 mph and my power consumption is on the order of 50-60 watt hours per mile. I will take my mid drive in those situations.
The odd pavement situation that I have found a mid drive to be better at is into a very stiff headwind. The mid drive does not outshine the hub motor until you quit pedaling, drop to a low gear and use the throttle. The mid drive is very efficient into a headwind in the cadence levels that I can't sustain. We are talking wind levels that top out your sustainable power limits before you get to an efficient speed for that power on a hub motor. I get better mileage than if I pedaled.

I have not tried the high cadence thing without a lot of headwind. You might try equivalent bikes at about 16 -18 mph. One bike a hub motor with no pedaling and the other a mid drive cadence sensed bike in a low enough gear that the speed is sustained with the throttle on full. The power drops significantly as the max rpm is reached. I think I will play with that on the simulator.

 

[fooferdoggie]

it depends on what your after. if you like to keep your cadence up and work then a mid drive will be the only way to go. the gear range the natural feel and in the case of the name brands smaller batteries and more effect setups. If you want to ride a bike like a bike then a mid drive is the most practical way to go. to be able to use the greartrain to your advance is key fore me to be able to put out 200 watts while riding in changing road surfaces and hills.

 

[theemartymac]

It's hard to say "Hardly" outperforms, but I'm not disagreeing with you.

I have two very similar bikes. Both Rize with nearly identical frames and accessories outside the drive system. One is a 750w hub, the other a 1000w mid-drive. In my use case of west coast hilly terrain and high-speed long-distance commuting, there is no caparison. The mid-drive handily trounces the hub in all aspects of performance, and by far wider margin that the 250w rating difference should suggest.

But you are also spot-on that the downsides of the mid are not to be undervalued. The resistance level is stupid unpowered (future versions will almost certainly add a clutch at some point), and they efficiency of the hub in it's peak range is generally more than adequate for many, if not most, riders. The hub is MUCH easier on the drivetrain (personally witnessed), and easier/lazier to ride in general. Of course there are other mid-drive benefits like simple rear wheel maintenance, overall balance, etc. So that is partly why it's such a hot debate.

I think I would be very happy with a geared hub that could give me perhaps 1250-1500W performance (to widen the power curve), but no such monster exists in the wild just yet - and it seems like the manufacturers are happy focusing on the mid-drive market. I think there are obvious ways to further improve both designs, but based on my non-scientific but "real world" experiences, I would give up my hub long before my mid. It would make me sad to do it though...

 

[Captain Slow]

My first ebike had a Bionx 250 watt direct drive hub motor. It absolutely sucked. It was terrible. It was really heavy and it didn't help that much going up long steep hills and I also found that at times I could smell it as it started to get hot. The DD hub was really heavy and it threw off the weight distribution of the bike. I could not take corners very fast because of it. At the time I thought to myself that hub bikes absolutely suck.

Then I bought a Juiced CCS and it was so much better than the Bionx. I can see why Bionx went out of business. They were expensive, underperformed and had the stupid proprietary battery to motor communication that made battery replacement restrictive and expensive.

The hub motor on my CCS was so much better than the Bionx and it changed my opinion of hub motors. I'm not sure if it was because it was a geared hub motor or that it was more powerful. I'm guessing it was both. Hills that the Bionx struggled on were no problem at all for the CCS. Handling was infinitely better. I could ride the bike normally through corners.

But I still bought a mid-drive road bike (Creo) and I wanted a mid-drive for that. I didn't want to be packing big wrenches with me in case I got a flat tire and on the road bike sometimes I'm 50 to 60 km's from home so a flat tire with a hub motor that has a wheel that can't be removed by hand was a non-starter. That and wanting a more natural feel were the reasons I went with mid-drive for that bike. I wouldn't rule out a hub motor in the future, but because of the possible flat tire scenario I'll be biased towards mid-drive. If someone comes out with a light hub motor that has an integrated torque arm and uses a QR skewer or thru axle then I'd probably be indifferent.

 

[indianajo]

it depends on what your after. if you like to keep your cadence up and work then a mid drive will be the only way to go.

False. My throttle allows me to add as little power as necessary to overcome the headwind, even 25 watts. Too fast PAS is useless, I deleted the crank pickup.
Mac stated their motors will burn a winding if run at low speeds full throttle on a steep grade for as much as an hour. I suppose all the other geared hub motors are equally sensitive. Not living in the sierras or the rockies, not a problem. I cross >77 hill on my 3.7 hour commute. I do get 5000 miles/chain, reaping one advantage of the hub drive. The other advantage, when motor wears out, trash it and install another in 2 to 4 hours. No waiting on parts or a dealership. My $221 hub drive wore out in ~4500 miles. I had one waiting in the garage, lost only one day. & bike was rideable pedaling unpowered with the worn out motor. Pedaled it out & home again 37 miles broken, in fact, no drag.
I deleted the insta-steal axle off my bike the first week after it was delivered. I use wrenches to change tires & tubes. Saw 4 bikes chained up missing front wheels the Sunday after Derby 4 years ago. People selling those wheels? Nah, just creating chaos & destruction. Missing seats, too. Deleted that quick adjustment, too.

 

[Captain Slow]

I just clicked on your link and see you compared 250 watt motors. That's interesting because my Bionx was 250 watts and my Creo is 240 watts. So basically the same power levels as in your comparison. Yours is a simulation and not real world experience so I'm not sure how you know your simulation would be accurate in real world testing or that it's actually representative.

In my experience the Creo outperformed the Bionx hub motor by an incredible margin. If you used my two motors as a comparison the conclusion would be very different. I live near Simon Fraser University and will ride to the top at times. The ride on the Creo was far easier and I'd go up much faster. It would not even be close.

Then if you look at range. I'm not sure how big the battery on the Bionx was but it was sure much bigger and heavier than the battery on my Creo. Even with regen the range of my Bionx bike was much, much less than on the Creo. Last summer I did a 122 km ride on my Creo with a small amount of battery left. On the Bionx bike range would not even be half of that.

But I understand that the differences are not solely related to hub vs. mid and that the Bionx is technology that was developed probably about 15+ years ago and the Creo is quite recent, so it's not a fair comparison.

But I also think that the Creo is an example of how the big bike companies are pouring R&D into mid-drives and to me at least it doesn't appear that R&D is going into hub motors. If that continues then I think you'll see mid drives noticeably pull ahead of hub motors. I mean the motor on my Creo has no resistance when pedaled with the motor off. I have even had times when I forgot to turn on the motor and it took me a couple of km's to realize that the motor wasn't on. So your blanket statement about hub motors having huge resistance with the motor off is simply wrong when applied to the Creo.

Having said that, I'm not sure how much better they can make a mid-drive motor. It's already pretty small and efficient. So maybe I'll have to retract my above statement since it's hard for me to see how much better they can make mid-drives. I think the big gains are to be had in the batteries. I have a small 320 wh battery in my downtube and if they could increase the capacity to 500 wh but not make the battery any heavier that would allow some really long rides or let me go a lot faster!

The statement you make about mid-drives being more expensive, I completely agree with that. I think pretty much everyone would agree.

 

[Captain Slow]

Wow 5,000 miles on a chain!!!! That is amazing.

On my Cervelo that doesn't have a motor I change my chains every 3,000 to 4,000 km's so that my cassette and chain rings will last longer. Having the hub move the bike along instead of the drivetrain sure helps the drivetrain last longer. Wow, that's a nice benefit.

 

[m@Robertson]

@avandalen: Having extensive experience riding both types, as well as having built those bikes, I think you haven't done enough real world riding on actual bicycles as, if you did, you could not come to the same conclusion. I'll admit straight away I have absolutely zero experience with motor simulations or how to make them. I can use the Grin motor tool but only as an end user. On the other hand, I ride daily, build my bikes, and do so having learned from previous experience - and mistakes. So the next bike tends to eliminate the weaknesses of its predecessor. In my case I specifically learned the hard way that hub drives are inferior - in certain common use cases. So with that out of the way...

My daily driver commuter for some time has been until recently a 2wd geared hub bike that I ride daily on a 15.5 mile route, so 31 miles (47.9 km) total daily. this bike has a 31ah 52v battery with a 90a continuous BMS, connected to twin 35a peak controllers. So peak power on this bike is over 4kw. Max speed on flat ground with a bit of a tailwind is about 32-34 mph, with typical cruise being 30 mph. Both motors are hooked up to a single pedal assist sensor so PAS is active on the two motors and in fact I use it as a speed pedelec despite having throttles on the bike. I geared it up so I could pedal - with effort - up past the bike's max speed. Big chainrings and small cogs on a fat bike are unusual, but this bike was purpose-built for paved, flat, long streets. I have configured it in a number of ways over a period of years and at present the bike is running on smooth summer 26x4.0" tires. Pictured are fat 4.8" and smooth 2.5" (never again).

These are two spots along my commute. Note both speed limit signs are in mph not km/h, and the first pic has a separate Class 1/2 path off the road, while both have an on-road path which is where I ride. Laws in my area allow assist limits of 28 mph on that street path, and local law enforcement has no qualms whatsoever about my pedaling hard along a street at higher speeds.

So with all of the above conditions, THIS is the ideal environment for a hub bike. Flat, smooth, straight. However despite my work being in this flat terrain, my actual home is in hilly coastal terrain. I took this bike to that home expecting to be able to ride it. I found that, with its exceptional power levels, it could climb anything (and anything in this case is what could be called a San-Francisco-style hills, which are the rule in that area. Nothing is flat until you get to the beach). However I also found that - given the inherent single-speed nature of the hub - it was laboring severely. It could climb, but it was obviously not happy doing it, and while I had thousands of flatland miles on the bike, if I tried to use the bike for an extended period in this terrain - all paved, mind you - it would not survive essentially forever as it was destined to in the California Central Valley. Those gears would give out eventually.

What kind of terrain is this? In discussions like this people point to pictures of horrifically steep hills, but its these sorts of long, still-fairly-steep grinds that are more typical here, and these do in the single-speed hub as the bike slowly loses momentum and lugs its sorry self moreand more slowly up that hill, turning the power it once used to rotate the motor casing, instead, to heat. I'll link this typical example so you can see it on Google Maps and look around.

Google Maps

Find local businesses, view maps and get driving directions in Google Maps.

www.google.com

This one is on another side of the same hill. I positioned the viewer in the middle so spin around. Not visible at the top of this hill is the sort of severe grade mid drive riders like to point to and say "you can't climb that". Well, I can but my poor singlespeed hubs with no mechanical advantages will die trying sooner, not later.

Google Maps

Find local businesses, view maps and get driving directions in Google Maps.

www.google.com

So, faced with this experience, I built a follow-on bike - for hills. Believe it or not this bike has the same frame as the orange one, except its in titanium this time. This bike has none of the weaknesses of the hub on the rear wheel.

I'll make this short: Problem solved. Bike goes up hills easily. For reasons that should be clear. There is still the anomaly of a front hub but I lower its output to a sweet spot of traction vs. battery drain to about 250w delivered.

Life sucks for a singlespeed hub going up a hill for the same reason it sucks for a human trying to do the same thing. Don't sell a transmission short. They serve a powerful purpose.

 

[mclewis1]

I ride about 80% of the time on pavement. At times I am riding at 8-10 mph and my power consumption is on the order of 50-60 watt hours per mile.

Do you mean 5-6 watt hours per mile? Or are you referring to some really serious watt sucking riding like serious hill climbing? 50-60 watts/mile would be hugely inefficient (on the order of getting 10 miles out of a typical battery).

 

[m@Robertson]

Wow 5,000 miles on a chain!!!! That is amazing.

I get about 3000 miles (cheating... a front motor reduces the shock the mid delivers to the drivetrain, which is why I did it). I think a lot of the chain wear / drivetrain wear issues are on the builder/designer. Make a bike with good chainline and it doesn't chainsaw the chainrings and cogs. Use steel rear clusters that are welded together (Shimano HG400-9, Sunrace CSMS7 11s) so they distribute the load across the entire cassette body. Spend the admittedly exorbitant sum on a proper mid-drive stressed chain. Forged front chainring. Ratchet engagement rear hub. But the rider has to do their part as well. Mid drives are not without their down sides.

How To Ride a Mid Drive Ebike Without Breaking It

If you can ride a bicycle you already know how to ride a hub drive ebike. Not so a mid drive. Particularly a powerful one that can tear your chain apart. Fear not. The rules are simple.

talesontwowheels.com

Also to make another comment on the hub vs. mid performance subject: If I want a bike that accelerates quickly and makes me giggle... its the hub bike. A mid cannot compete on raw acceleration on flat ground. I have mids as strong as 4kw (a Cyc X1) and to keep the motor from destroying the chain you have to gear it down so it accelerates from a standing start without putting colossal stress on just a few teeth in the back. That means you reach your max speed pretty quick because gears. That in turn means you have to upshift. As you row thru the gears, a fast hub bike will leave you in the dust since all the hub rider needs to do is stab and steer. The acceleration curve is unbroken. So... no contest there. Hub wins a drag race.

 

[Mr. Coffee]

Request to the moderators: can we make a hub-drive versus mid-drive section so all of these conversations can be in one place?

 

[AHicks]

With the huge differences in available power within the "mid drives", as well as the huge differences available within the term "hub drives" I find any comparison between "hub drives" and "mid drives" pretty much useless. You need to break these down MUCH more to gain ANY insight into their differences.

 

[fooferdoggie]

Mac stated their motors will burn a winding if run at low speeds full throttle on a steep grade for as much as an hour. I suppose all the other geared hub motors are equally sensitive. Not living in the sierras or the rockies, not a problem. I cross >77 hill on my 3.7 hour commute. I do get 5000 miles/chain, reaping one advantage of the hub drive. The other advantage, when motor wears out, trash it and install another in 2 to 4 hours. No waiting on parts or a dealership. My $221 hub drive wore out in ~4500 miles. I had one waiting in the garage, lost only one day. & bike was rideable pedaling unpowered with the worn out motor. Pedaled it out & home again 37 miles broken, in fact, no drag.

well I get 3000 miles on a chain on my bosch powered bike on a 10 speed chain. having a motor wear out in 4500 miles means i would lose days of riding and I would wear out two motors a year (yes I ride 9000 miles a year). so to me thats a big negative. I don't want my bike to be down because I wore it out so fast.

 

[K PierreR]

Do you mean 5-6 watt hours per mile? Or are you referring to some really serious watt sucking riding like serious hill climbing? 50-60 watts/mile would be hugely inefficient (on the order of getting 10 miles out of a typical battery).

50-60 is right with a 49Ah battery

 

[Mr. Coffee]

Do you mean 5-6 watt hours per mile? Or are you referring to some really serious watt sucking riding like serious hill climbing? 50-60 watts/mile would be hugely inefficient (on the order of getting 10 miles out of a typical battery).

I have rides I routinely do that burn 100wh/mile on the uphill. Of course on the ride back it is quite a bit less.

 

[theemartymac]

Request to the moderators: can we make a hub-drive versus mid-drive section so all of these conversations can be in one place?

And then apply a giant automatic e-stick to beat it with for us.... ;-)

 

[Alaskan]

As the great Yogi Berra once said, "In theory, there is no difference between theory and practice...in practice there is."

Those of us who have actually ridden variety of ebikes under many different conditions (practice) will instantly realize that many, if not all, of the theoretical conclusions drawn from this simulation study are fatally flawed...frankly too numerous to debunk one by one.

I suggest you get out of the lab and ride a few ebikes under a variety of conditions over a variety of terrains.

The next report you will be writing after that will be detailing, what factors were omitted and why your simulation led to so many flawed conclusions.

 

[harryS]

"The conclusion is that the advantage of mid-drive motors is only noticeable in extreme situations, on steep slopes while the cyclist pedals only a little or not."

Last line in your study. I would add though that the Grin Simulator says an EZ250 melts after 16-20 minutes of such abuse while it would never do so in a mid drive configuration.

In addition, Grin does the thermal modeling and power modling of the off-the-shelf motors in their simulator catalog. I would have to think that the thermal parameters for a true hub motor, without the need to accomodate planetary gears inside the motor housing would be far better.

I've converted maybe a dozen bikes to ebikes using geared hubmotors and mid drives. I've never gone as small as a QX-85, only the more powerful Q100's, The latter is a nice, mild mannered little hub motor. I've climbed short 10% gradients with mine, They are not going up any steep slopes w/o rider assistance.

 

[Mr. Coffee]

I suggest you get out of the lab and ride a few ebikes under a variety of conditions over a variety of terrains.

I think the kindest word in the English language is "look".

When I look at the photo threads on this site one thing that is very striking is that all of the people who ride a lot and are living the dream with their e-bike are doing so on mid-drives. Just saying...