hill climbing? 750W hub versus 250W mid drive

[tomjasz]

Why Hub Motors - Learn

Learn why we chose to keep our focus primarily on hub motor drive systems over mid-drive conversions.

ebikes.ca

Why don't you carry the Bafang BBSXX mid-drive kits?​

We get this question several times a week and feel that it's worth devoting a page to the subject.

The Bafang mid-drives have become quite popular since they were first introduced as an aftermarket kit option around 2013. Prior to that, most of the mid-drive systms were designed for OEM bike frames and so not accessible to DIY conversions, or those that were meant for aftermarket installations were usually interesting but mechanically complex. The BBS01 design (which Bafang cloned from Sunstar) was neatly integrated as a self contained bottom bracket replacement without all the external mounting hardware, chains, pulleys etc normally associated with mid-motors.

In any case we've had samples of all the BBSxx motors and many variants from other manufacturers but have never felt that they hit the right note for what the majority of our customers are actually after. Integrated controllers limit your upgrade and repair options, wide 'Q' factor cranks to clear the motor can make for awkward pedaling, there have been limited chainring options, not to mention some messy politics in the distribution channel if you've been following that. We have been actively involved in mid-drive solutions for cargo ebikes with the Stokemonkey and varients, just not at the moment for regular bicycles.

Where do BBSxx and similar mid-drives shine?​

There are a certain areas where bottom bracket middrive motors are hands down better suited than hub motors. Offroad mountain biking is one great example. The motor weight is low and on the unsprung frame so it has minimal effect on handling even in full suspension bikes. In these situations you are often going slowly uphill through trails with short steep sections and a small motor working through the drivechain in the granny gears of the bike can do this with excellent efficiency. Another example would be say fat bikes riding in sandy beaches or through fresh snow. These are areas where the bike is moving at lowish speeds through high resistance terrain, and the mid-motor in an easy gearing will do this with better efficiency than most hubs drives, while still performing just as well on the faster roads.

For us though, a majority of our customers are using their ebikes for daily use commuting, riding on roads where maintaining a steady speed regardless of the hill grade is important. For that requirement a mid-drive offers almost no advantage and we continue to believe that hub motor kits shine the strongest for reliable and low maintenance commuter ebike conversions, and that's the bulk of who we serve.

Hub Motor Advantages​

Here are a few benefits to a hub motor system to keep in mind and help inform your decision. Please don't see this as us taking sides and saying that hub motors are universally better than mid-drive motors, all we want to do is balance out the discussion a little by highlighting the many benefits of a hub motor drive which are frequently missing from popular discourse at the moment.

1) Independent propulsion​

This point is often overlooked, but a hub motor has significant robustness in being totally independent of the human drivechain. If you've ever had a situation where a chain breaks, derailleur gets jammed, wet snow packs into the sprockets, freewheel busts apart etc. you become quite grateful that the hub motor can continue to propel the bike to get you home. With a mid-drive motor, anything that takes the drivechain out of commission stops the bike completely as both the motor and pedals become useless.

2) Less stress and wear on drivechain​

With a hub motor bike, the use of the motor lessens the mechanical wear and tear on the chain, cogs, and derailleur system, since it decreases pedal stress that the rider would otherwise put on the transmission. A mid-drive motor does the opposite, putting much more force on the chain and cogs with the motor propulsion now added on top of the rider's input. A normal bike chain drive is fairly robust and can usually handle this extra load without failure, but will wear out at a much faster rate, requiring more vigilant attention to chain stretch, cog wear, shifter alignment etc.

3) Ability to use internal gear hubs (IGH)​

This is somewhat related to the above point, but one of the great innovations in bicycle hardware in the past decade has been the production of compact internal gear hubs that are 7-14 speeds with wide gear ranges. With a few exceptions (Rohlhoff and the original Nuvinci N171) these hubs are not rated for the extra stress of a motor drive and will usually have a short service life with more than an extra couple hundred watts from a motor. If you want to use a mid-drive with an internal gear hub, be sure to check whether the hub is rated for tandem and/or cargo use, most explicitly say no.

4) Higher peak power capability​

The transmission through your bicycle drivechain also limits the maximum motor power that can be coupled to the wheels. Power levels in the 250-750 watt range are usually fine just with increased wear and tear, but when you are looking for multiple kW of power then you start seeing catastrophic failures with snapped chains, sheared freewheels, broken spokes etc. A hub motor with a properly coupled torque arm has no such limitation.

5) Potential for regenerative braking​

With a direct drive hub motor or a geared hub motor that has a locked clutch, the motor can act as a perfectly controllable brake that isn't affected by rain, pad wear, cable adjustment or anything like that. In a similar manner to how a hub motor reduces wear and maintenance on the pedal drive chain, it can reduce and even eliminate brake pad maintenance as well. You can easily do 90% of your braking electronically with only the odd emergency stop needing the mechanical engagement. On a mid-drive motor, there is really no potential for regen (unless of course you electrify a fixie, but we don't see much of that).

6) Ease of installation​

Admittedly the BBSXX systems and clones are pretty straightforward to install if you have bottom bracket removal tools and crank tools. But nothing gets quite as easy to fit as a front hub motor, where you are just taking off the original wheel and replacing it with a motorized one. In general we wouldn't emphasize this too much, as the time spent installing any conversion kit is way less than the time you save from riding it and you should install the system that is right for you. If you are concerned about installation and the facility of moving the assist to different bicycles and such, it's hard to top a front hub motor for minimal hassle.

7) Torque sensors options​

Because of the independent propulsion, the hub motor drive leaves any number of sensor technologies available for measuring rider pedal input, including chain tension (BeamTS), rear dropout (TMM4), Rear axle flex (eg BionX), and the many bottom bracket torque sensors (NCTE, Thun, TDCM, Sempu etc.). With a mid-drive, the motor couples through the drivechain so these sensors can'd distinguish pedal power from motor power, and so you are stuck with what the mid-drive system comes with which has so far been pretty limited. Some (like lightningrods and Tongsheng) have a torque sensing, while on the vast majority they are have just a basic pedal cadence sensor.

Now there is no reason why more aftermarket bbs style mid-drive can't have an integrated torque sensor, it's just that so far this mostly exists only on OEM bottom bracket drives like the Bosch. .

8) Simpler operation with shifting​

For a mid-motor to work well the rider needs to shift through the gears as they speed up and slow down, and similarly back off on the motor power while shifting to avoid seriously harsh shift transitions. This constant backing off and reapplying motor power between each change of gears can get taxing and results in some speed loss too, while with a hub motor you can continue to have uninterrupted power through all your gear transitions. There are shift sensor products to make these motor cutouts automatic, but there is nothing you can do if you come to a stop in a mid-drive system and forget to downshift first. The motor will start off in a high gear with low acceleration and efficiency. For regular cyclists this is probably no issue, but for those without that background, the simplicity of not having to shift gears all the time is one big appeal of an ebike. The rider can generally leave the bike in a high gear and use the hub motor to quickly get up to speed so there is no need to go through the motions of downshifting and up-shifting at each stop and go.

Remarks​

So, we hope that this sheds some more light on why we've largely stuck to hub motor systems since first becoming enamoured with the Crystalyte 400 motors in 2003. If you've read on websites or forums that hub motors are "yesterday's news" with poor efficiency and handling, and mid-drive systems are the only way to go for an ebike, perhaps you'll have a more informed outlook after reading this.

Mid-drive systems are not some new technology (most ebikes from the 1990's and early 2000's like the Merida Power Cycle and Giant Lafree were mid-drives), and hub motors continue to evolve to higher levels of power density, efficiency, and bike compatibility since they first rose to prominence in the mid to late 2000's.

 

[Stefan Mikes]

Why do they call the Specialized 2.2 a 250 W motor?

It is for legal reasons. It's called "nominal continuous motor power". 565 W is the peak power. Indeed, a Specialized 2.2 motor in Turbo mode feels like it were lifting you up a steep hill! The motor in my Vado 6.0 is 520 W max mechanical peak power. As I was climbing an overpass im Turbo mode, there was a traditional cyclist who was climbing faster than I could. After we stopped, I asked 'An impressive feat! What is your peak leg power?' to which he smiled and said '800 W?'

 

[teskow]

My mid drive may climb steeper hills than my hub drive but at a lower gear which results in a slower bike speed. Keep bike speed and gear up and the 52v battery 750 w hub motor and it will take me up a hill faster and easier. If the wheel speed gets to slow though it looses torque.

 

[tgianco]

Well, I know what is "supposed" to be, but my old 750w hub drive Himiway Cruiser flew up steep hills while my new mid-drive Himiway A7 Pro struggles... let me amend that, I am struggling with it. Maybe I AM a lazier rider than I thought I was.

 

[spokewrench]

Well, I know what is "supposed" to be, but my old 750w hub drive Himiway Cruiser flew up steep hills while my new mid-drive Himiway A7 Pro struggles... let me amend that, I am struggling with it. Maybe I AM a lazier rider than I thought I was.

I wonder if it's the torque sensor. The one on my Aventon Abound was made by M&S E-Driving Systems. After 2 months the right bearing was making a rubbing sound. The cage hides the balls, but a gram of oil quieted it for 10 days or so. When the noise returned, I packed in what bearing grease I could. Now it pedaled so much more cleanly that I didn't need PAS on a 2-mile circuit where I'd needed it in 5 places. I took the BB out to see if there was some way to get the old lube out of the bearing. Now that I could handle the bearing directly, it felt gummed up. I got it turning freely with brake cleaner and WD 40, then put in fresh grease, hoping some would wick past the cage. It was great for a couple of days. Once again, when I pedal it feels as if there's a layer of mud on the pavement. Since I can't clean up the bearing, I ordered a plain BB. It arrives tomorrow, and maybe the Abound will pedal as freely as other bikes. If I need assistance, I'll add the watts I need with the throttle.

 

[m@Robertson]

This is one of those arguments where people jump through hoops to cheer for their favorite team. But real money is being spent by the person asking the question, so its important they get an answer that doesn't knowingly encourage a waste of money by trusting a bad source.

Here is an objective test of 14 hub motor'd ebikes. I have it tee'd up past the initial intro speech to launch straight into the test description and the tests themselves. Just watch. You will see how hub motors behave in hill country. You should get an idea real fast as to whether this is what you want out of an ebike. I would stick with it for all 14 tests because it shows you runs with big wheels, small ones that do give a torque advantage, big watts etc. Its also worth noting that the measure of grunt (torque) is "continuous" amp output on the controller, and you can't guarantee what a given factory bike has inside. So a 1000w motor with 20" wheels and a 52v battery could be saddled with a 15a controller and you are ... well, screwed.

Note these are geared hub motors, and direct drive hubs have less torque, until you supersize them with something like a QS203 v3 pump like 3000w+ thru it.

 

[Chargeride]

I climb my geared 750w hub to pretty well its limit off-road up mountains and run it at 2kw, this requires short fast bursts of around 30 seconds before it overheats, but it does cover a lot of ground during that.
The same track, my stock bbshd just tootles up in the same time, is far more controllable as I'm not desperately trying to keep the speed going, not the slightest bit of overheating and uses less battery doing it.
I can also run full size wheels with a larger contact area.
The bit starting at 5.40 in this video would be impossible on a hub, you need crawler gear control.
TBh, I couldn't possibly imagine taking my hub up there, the effort reqd to keep it spinning is beyond me.

 

[Stefan Mikes]

I climb my geared 750w hub to pretty well its limit off-road up mountains and run it at 2kw, this requires short fast bursts of around 30 seconds before it overheats, but it does cover a lot of ground during that.
The same track, my stock bbshd just tootles up in the same time, is far more controllable as I'm not desperately trying to keep the speed going, not the slightest bit of overheating and uses less battery doing it.
I can also run full size wheels with a larger contact area.
The bit starting at 5.40 in this video would be impossible on a hub, you need crawler gear control.
TBh, I couldn't possibly imagine taking my hub up there, the effort reqd to keep it spinning is beyond me.

Lovely!

Charge, how about climbing Mt. Snowdon from Llanberis in the cold season? As I can understand, cycling is banned in the warm season there. Would the trip be doable in some late Autumn or early Spring?

 

[Chargeride]

Yes, it's quite popular, but most people carry their bikes up, even electric versions, I have a plan to take a mad creation up there.
You can ride all year round, but there are strict time constraints in the popular seasons, but this being Britain, you just constantly apologise if riding out of hours.
Embn took a 300 quid hub drive supermarket bike up there.

 

[kevinmccune]

nterstingly enough had a ful EBR reviewier said that about the 2020 Radrunner. He also said that putting the battery under the seat balanced the bike, and fat tires made it stable. I was skeptical, but I bought one based on the performance he demonstrated.

Mine climbed poorly. I went back to the part of the video where he said he was climbing a hill on throttle alone because hisl suspension knees couldn't handle pedaling. I determined the grade when he turned his camera to a house whose front yard was on the same slope as the street. I determined his speed by the cadence of two Radrunners he passed. He had to be using a controller of at least 35 amps, more than twice what the OE controller put out. A 35 amp controller vastly improved climbing.

Here's why I was skeptical when he claimedrstingly the 20-inch wheels meant more torque. The diameter designation really means the tire fits a 16 inch rim. I found that the center of the axle was 11.5 inches from the ground, so the diameter was 23 inches on account of the fat tires. That's only 11.5% smaller than a conventional 26 inch tire. and it might be 4.2% smaller than your 24-inch tires.

had a "fatbear" 20 incher claimed 500 watts the controller was so restrictive it would barely help me up the grade going to my house have had better results with 250 watt cheap bikes that would actually help you to"lug" had a 750 watt with a 25 amp controller that felt like it had wings a hikers legs(26x4) interestingly enough had a full suspension 500 watt "truckrun" powered bike that would peak at 625 that pulled really well it actualled helped me up a driveway we had built (without cutting out) the driveway was so steep it was difficult to walk up, barely made it. and had a middrive torque sensor bike that probably wouldnt have went half way up that hill( small battery hurt that bike i believe, for whatever reason i prefer cadence sensors.i didnt get ebikes for a "natural feel" i got ebikes so i could make the highs again.

 

[kevinmccune]

Yes, it's quite popular, but most people carry their bikes up, even electric versions, I have a plan to take a mad creation up there.
You can ride all year round, but there are strict time constraints in the popular seasons, but this being Britain, you just constantly apologise if riding out of hours.
Embn took a 300 quid hub drive supermarket bike up there.
lllls? too much

you have to really want it, to try it on a bike like that,wonder how hot the hub became on the pull? too much heat and the magnets can fall out of place,surprised he had any battery left at the top.

 

[spokewrench]

Yes, it's quite popular, but most people carry their bikes up, even electric versions, I have a plan to take a mad creation up there.
You can ride all year round, but there are strict time constraints in the popular seasons, but this being Britain, you just constantly apologise if riding out of hours.
Embn took a 300 quid hub drive supermarket bike up there.

Norwegian Fred Hansen won the Mount Snowdon hill climb in 1907, I believe, on a Norton driven by a leather belt around a pulley by the back rim. He told me that 67 years ago. He sometimes came to my house for breakfast.

 

[spokewrench]

This is one of those arguments where people jump through hoops to cheer for their favorite team. But real money is being spent by the person asking the question, so its important they get an answer that doesn't knowingly encourage a waste of money by trusting a bad source.

Here is an objective test of 14 hub motor'd ebikes. I have it tee'd up past the initial intro speech to launch straight into the test description and the tests themselves. Just watch. You will see how hub motors behave in hill country. You should get an idea real fast as to whether this is what you want out of an ebike. I would stick with it for all 14 tests because it shows you runs with big wheels, small ones that do give a torque advantage, big watts etc. Its also worth noting that the measure of grunt (torque) is "continuous" amp output on the controller, and you can't guarantee what a given factory bike has inside. So a 1000w motor with 20" wheels and a 52v battery could be saddled with a 15a controller and you are ... well, screwed.

Note these are geared hub motors, and direct drive hubs have less torque, until you supersize them with something like a QS203 v3 pump like 3000w+ thru it.

He calls these the most powerful 14 for under $2000. Is it coincidence that each of them agreed to pay him a kickback on sales? Some don't even have adjustable saddle heights.

In a little box on a bike, I can carry the stuff I need to measure a grade in a couple of minutes, but he didn't measure the grade. None or all those bikes could be adequate for Rexlion. Anyway, the test used throttle only, but Rexlion asked about PAS.

By "steep," Rexlion means he'd rather use PAS than tire himself out. He says the problem is that his torque-sensor ebike is hard to pedal without power on any grade, compared to his push bike. My Abound is like that. Even 2% can tire me so that I turn on PAS, but if I manage to free up the right BB bearing temporarily, 5% will be fine without PAS.

Gears can be used to get torque out of a light motor. A couple of years ago I bought a Sixthreezero with a 500 watt direct-drive motor. The back end of the bike was unexpectedly heavy. In torque, the motor compared favorably with the geared 750-watt motor on my Radrunner.

I bought a 500 watt Radmission with a similar direct-drive motor. The back wheel was very heavy, and I knew the rim and tire were light. Initially, the 10 amp controller limited its climbing ability. With a 25-amp controller, it outperforms a stock 750 watt geared Radunner.

Torque is a selling point of the Abound. In more than one video, a 200-pound reviewer would throttle up a grassy hill with 100 pounds of bottled water on the back. I had high expectations and wasn't disappointed. The motor is heavy and measures only 6 cm between flanges, so it seems to be direct drive.

 

[m@Robertson]

He calls these the most powerful 14 for under $2000. Is it coincidence that each of them agreed to pay him a kickback on sales?

None of that matters, though. Objectively you can just sit and watch how a hub motor'd bike operates on a steady, fairly steep hill. And if you stick with the video long enough you can see the same story repeat itself 14 times. If thats what someone wants in a hill-climbing ebike then by all means go buy one.

Some don't even have adjustable saddle heights.

Irrelevant. Saddle height affects pedaling which is not a part of the test of the bike's ability, on its own. Thats why using the same hill every time, and flat out throttle-only, is so important. Pedaling and muscle introduces variables that ruin the test of the bike.

In a little box on a bike, I can carry the stuff I need to measure a grade in a couple of minutes, but he didn't measure the grade. None or all those bikes could be adequate for Rexlion.

This too seems irrelevant. The point is you are seeing how a singlespeed motor handles a steady grade (over and over again).

Anyway, the test used throttle only, but Rexlion asked about PAS.

He sure did, and he said

I enjoy bicycling, but if I wanted to work my tail off during the climbs I'd just ride an acoustic!

Which seems very clear.

By "steep," Rexlion means he'd rather use PAS than tire himself out.

No. PAS does not give more assist than full throttle, and if someone wants more than what you see in the video, they will have to work their tail off. Simple.

@Chargeride posted an intelligent, experience-based post above on one versus the other. With more video.

But for the people who just want to learn a straight answer, and could care less about the cheerleading and typical internet arguments over alternative facts... tune out the argument and just watch the video. You can believe your eyes.

 

[spokewrench]

None of that matters, though. Objectively you can just sit and watch how a hub motor'd bike operates on a steady, fairly steep hill. And if you stick with the video long enough you can see the same story repeat itself 14 times. If thats what someone wants in a hill-climbing ebike then by all means go buy one.

Is "fairly steep" objective? Ten percent is unusually steep for a public street. Some in San Francisco are 37%, requiring 3.5 times more torque on the wheel. Without knowing the grade, how does one know if any of those bikes would suit one's needs or how they compare to a bike not on the list of advertisers?

Irrelevant. Saddle height affects pedaling which is not a part of the test of the bike's ability, on its own. Thats why using the same hill every time, and flat out throttle-only, is so important. Pedaling and muscle introduces variables that ruin the test of the bike.

Rexlion said his bike climbs very well on throttle but not so well on PAS. PAS requires pedaling. Pedaling would be difficult with a seat at scooter height.

No. PAS does not give more assist than full throttle, and if someone wants more than what you see in the video, they will have to work their tail off. Simple.

Rexlion says his bike climbs well on throttle, but he's looking for something better on PAS. Therefore, I infer that PAS gives him much less assistance than full throttle. He says unpowered pedaling is terrible. That's why I think the bottom bracket may be the culprit, as it was for me.

@Chargeride posted an intelligent, experience-based post above on one versus the other. With more video.

But for the people who just want to learn a straight answer, and could care less about the cheerleading and typical internet arguments over alternative facts... tune out the argument and just watch the video. You can believe your eyes.

Rexlion asked if mid drive gave better PAS than hub. You dismiss mid drive and PAS as alternative facts.

 

[m@Robertson]

14 bikes go up hill. See and hear obvious results. Use brain. Decide. The end.

 

[Rexlion]

I would be pedaling with PAS, not using the throttle. I don't mind working along with the motor, I just don't want to work hard enough to make my legs feel like jelly.

I've been thinking about a trip to Acadia NP and in my imagination I could picture myself pedaling to the top of Cadillac Mountain. But who am I kidding. Probably my legs would feel like jelly no matter what ebike I rode up that slope. Maybe if I were 35, and not 65.

 

[spokewrench]

I would be pedaling with PAS, not using the throttle. I don't mind working along with the motor, I just don't want to work hard enough to make my legs feel like jelly.

I've been thinking about a trip to Acadia NP and in my imagination I could picture myself pedaling to the top of Cadillac Mountain. But who am I kidding. Probably my legs would feel like jelly no matter what ebike I rode up that slope. Maybe if I were 35, and not 65.

I seem to have solved my problem by replacing the torque sensor with a conventional bottom bracket. Inadequate lubrication on a ball bearing may not be obvious if you crank by hand. As the load increases, the balls are squeezed harder between the races, and that means the cage has to push harder to keep them evenly spaced.

I became aware that without PAS, my Abound didn't pedal like a bicycle should. My grassy driveway puts up more rolling resistance than pavement, but I can get underway without assistance on my Radmission, with 86 gear inches. To do it on my Abound, I had to shift down to 4, or 51 gear inches. Was it getting 70% more rolling resistance? With a conventional BB, I can now get underway in top gear: 74 gear inches. I think that's evidence that putting the conventional BB under a load doesn't cause as much friction.

My torque sensor BB was made for Aventon by M&S E-Driving Systems. Maybe the grease they chose turned out to be inadequate. Maybe all torque sensors of this type are vulnerable. The rotating part has a bigger diameter, so the bearing has to be smaller, and maybe it can't handle the load as well. The shafts on a conventional BB stick out 28mm on each end. A torque sensor requires each end to stick out 39 mm, meaning a bigger load on the bearing.

I'd thought using a throttle lever for assistance would be clumsy. Then I figured it out. I noted where my thumb could lie along the bar without disturbing my wrist or fingers. I moved the throttle to where the end of my thumb would be. It's easy to feather in whatever assistance I want. With the side of my thumb against the throttle's clamp, it's easy to hold it steady.

 

[Rexlion]

That's an interesting situation with the ball bearing. Hmm.

 

[m@Robertson]

Test both hub and mid-drive systems to see which feels better for climbing.

exactly this!

I would be pedaling with PAS, not using the throttle. I don't mind working along with the motor, I just don't want to work hard enough to make my legs feel like jelly.

It *should* be possible to put the mid on the highest PAS which will in turn make the climb easy. That will also eat more battery so the decision on-bike if its the right one for the job be centered around dialing back the assist to the point where you have the option to get the effort output you choose to exert, and not drain the bejesus out of the battery doing it.

I try and set the gear for pedaling to be one more gear lower (bigger cog) than I would normally use without assist. Then sit there without shifting up the hill and adjust my PAS setting a click up or down as the slope lessens or increases from one city block up the hill to the next. Going one gear lower means I never bog the bike down and can always keep pedaling. I use the PAS output instead of the derailleur to deal with the changing hill slope. I go a bit faster or slower as a result but my cadence and effort stay pretty much the same which is what I want. This is a city street so on an mtb on singletrack a gear sensor and shifting would probably be smarter.