No Bafang used by mainstream bike manufactures.

[Ken M]

Product longevity is a trade-off, especially powered equipment. You want longer service life? It will cost much more, try and be heavier too. Just look at the cost of pro grade power tools vs consumer grade, and their weight (unless you pay a huge premium for light weight...)
Give the industry time, though, and these things will sort out.
Remember too that it wasn't that long ago that cars needed much more regular maintenance than they do now. Oil every 3K. Plugs and points every 12K. Tires might last 20K. Engines were shot at 100K unless they got a ring job, valve job and maybe bearings. Remember those?
Finally, remember that we're dealing with a lot of prospective buyers who think they can get a great, long lasting bike for 2K $. Oh, it should have a 100 mile range, be under 40 lbs and capable of 30 or more MPH too.

I agree with your assessment but the weight paradigm that drove the bike industry for 50 years is far far less important with ebikes (especially urban mobility ebikes that do not need hyper handling like a great mountain ebike). Cost wise I think longevity isn't always more expensive - a cheaper Schimano XT last longer than an XTR because they use heavier more durable alloys - they go counter to the lighter is always better paradigm which can be done on many ebike components.

One simple example of a way to improve product lifecycle is adding a few more mms to the thickness of ebike tires. Sure it adds a few 100 grams to each tire but getting 10,000 miles from an ebike tire would be something 99% or riders would appreciate much more than the negative impact of that much additional weight. Oh but don't expect the tire company bean counters to allow that to happen.

I'm an engineer so I know the answer to this. How hard would it be to have a mid-drive motor with oil bath lubrication of the gears instead of grease? Not technically challenging at all but hey then the motors would last a lot longer and not really cost any more to produce (a bean counters nightmare). I think they put the controllers inside the motors just to have an excuse why they don't utilize oil lubrication. Brose uses an internal belt but I think they still have a grease lubricated planetary gear set to wear out fast enough for more sales / service.

I'm still amazed that some ebikers think they need 14 or more gears when they have a Bafang Ultra drive system. The past mindset is hard to break.

Note: Be sure to notice the belt drive ratio on most Bafang Ultra models with IGHs. The industry tends to protect the IGHs by reducing the drive torque to the rear wheel by about 50% (they claim it's only because of under drive ratios of the IGHs but some have unity 1:1 as the lowest gear ratio so that's simply a misleading claim).

 

[harryS]

I didn't mean that we topped out at 4K miles. Those are the two bikes with the highest miles. We've ridden close to 20K miles since 2015, but it's on a number of different ebike motors. We've not had one wear out yet.

 

[retiredNH]

Ken M, you're obviously on a mission (or crusade) of some sort, so I won't engage in an endless debate. I'll just point out that for most riders a 10,000 mile bike tire would rot out far before they rode 10,000 miles. And I could care less about mating a Bafang Ultra with an IGH as things stand now. It's about as marginal a situation as anyone could find.

 

[mschwett]

Ken M, you're obviously on a mission (or crusade) of some sort, so I won't engage in an endless debate. I'll just point out that for most riders a 10,000 mile bike tire would rot out far before they rode 10,000 miles. And I could care less about mating a Bafang Ultra with an IGH as things stand now. It's about as marginal a situation as anyone could find.

definitely on a mission.

the screed against the gearing of bicycles is only appropriate for someone who never intends to pedal. the gear ratios of a bicycle are to allow a non-olympian human to propel it from a stop, up shallow grades at speed, etc. it’s not a conspiracy theory. a vehicle whose drivetrain is designed exclusively around powered operation is called a moped, scooter, motorcycle, etc.

 

[DavidRvR]

I think its more Name recognition with Bosch or Yamaha...... Sell a junk bike and through a name brand motorcycle makers name on the side of a motor and people believe its a great product.. ( not saying it isnt). Would you have a Yamaha motorcycle or something like Haojin?

Name recognition sells... Harbor freight drill or Bosch .. You get the Oohs and AAhhhs with those names Not Bafang... unless you are on this site or talk to someone that knows..

Just my opinion.. But I have to agree it is a good question on why and I wonder if it will change in time? Power and Torque from the bafang is hard to beat but there are some up and comers ... but they wont be used on anyone bike either anytime soon..

 

[rich c]

I was in my local canadian tire yesterday and saw a Raleigh eBike with some type of Integrated Bafang motor (Had a Bafang Bottom Bracket looked like an ultra , obviously wasn't) i was pleasantly surprised to see it.

this well known bike company was bought by canadian tire a while back i think....

@Alex M - we build consumerism on the backs of a communist countries working class , that country now poses significant global threat , and growing military...... how could we ever have predicted that (lol)... also , the Made In America plan will never work ... people who make an average living can't afford to pay the difference and its gotten to the point they rely on the dollar tree and walmarts to allow them to scrape by. a reckoning is coming.

People have said a reckoning is coming for centuries.

 

[Ken M]

Ken M, you're obviously on a mission (or crusade) of some sort, so I won't engage in an endless debate. I'll just point out that for most riders a 10,000 mile bike tire would rot out far before they rode 10,000 miles. And I could care less about mating a Bafang Ultra with an IGH as things stand now. It's about as marginal a situation as anyone could find.

I understand. I rode my ebike over 6,000 miles one year and put another 6,000+ on my car. I thought it was strange I was putting more money/mile for bike tires than I was car tires. I do tend to believe ebike can be a viable transportation solution for many people so the old mindset that bikes are just for recreation and leisure should be pushed aside a bit.

Right now a Bafang Ultra will pretty much go thru cassettes and/or IGHs if ridden to it's potential. If an ebike transmission can tolerate about 350nM of torque it's pretty much "future proof" for ebikes because that is the torque at which most ebikes would begin to break loose the rear tire. That may explain why the Bafang is rated at 160nm as a 200 lb rider standing on the cranks generates another 160nm so much more torque from the Bafang and riders could do burnout on their ebike which makes no sense for a bike.

I'm working with a small team on a new IGH that will be rated at 350nm or higher in all gears and we are striving for 50,000 mile service life or longer so it will last the life of the bike. It should come in about 2000 grams (maybe a bit more) which is about what the Rohloff is now. Pretty much all current IGHs were designed for just human power bikes so they really don't hold up to the huge torque output of the Bafang Ultra.

 

[Ken M]

definitely on a mission.

the screed against the gearing of bicycles is only appropriate for someone who never intends to pedal. the gear ratios of a bicycle are to allow a non-olympian human to propel it from a stop, up shallow grades at speed, etc. it’s not a conspiracy theory. a vehicle whose drivetrain is designed exclusively around powered operation is called a moped, scooter, motorcycle, etc.

If you ride a Bafang Ultra ebike with a 14 speed Rohloff you would find that you rarely need to be below like 7th of 8th gear and if you start t in 1st gear you will be shifting thru the gears so fast given the acceleration it makes no sense. Obviously I'm only talking about riding the ebike under power (I'd still want gearing such that I can at least ride home if the battery dies). 3 speeds is probably enough gears but they need to be spaced to provide a minimum 400% range from unity so you could have reasonable cadence at 28mph.

 

[AHicks]

If you ride a Bafang Ultra ebike with a 14 speed Rohloff you would find that you rarely need to be below like 7th of 8th gear and if you start t in 1st gear you will be shifting thru the gears so fast given the acceleration it makes no sense. Obviously I'm only talking about riding the ebike under power (I'd still want gearing such that I can at least ride home if the battery dies). 3 speeds is probably enough gears but they need to be spaced to provide a minimum 400% range from unity so you could have reasonable cadence at 28mph.

To add to that just briefly, when it comes to e-bikes, I believe the bigger the motor, the less the need for multiple gears. If that's hard to wrap your head around, look into how many gears electric cars are (aren't!!) using.....

 

[Ken M]

To add to that just briefly, when it comes to e-bikes, I believe the bigger the motor, the less the need for multiple gears. If that's hard to wrap your head around, look into how many gears electric cars are (aren't!!) using.....

Yea...agreed. The gearing on an ebike with a powerful motor like the Bafang Ultra is not about enabling the human power to be effective but more about regulating cadence in a range that is comfortable for the rider. Motors are simply more efficient and effective at higher RPMs so their internal gear reduction makes the bike gearing pretty much redundant (ie rider only intent). I think many riders think back to the days of 24 and 27 speed bikes and just can't imagine an ebike with just 3 gears and a 400% range makes sense. With power comes less rider responsibility.

 

[retiredNH]

To add to that just briefly, when it comes to e-bikes, I believe the bigger the motor, the less the need for multiple gears. If that's hard to wrap your head around, look into how many gears electric cars are (aren't!!) using.....

Much depends on (1) use case and (2) design of the motor. Let's ignore hub vs mid drive for a moment.
An electric motor system can be designed with a broad power or torque curve (note the difference, though) but there will be limits, especially at low rpm, where you can have very high torque at modest power because the motor is turning slowly. Then add efficiency into the mix (or not...).
Now, note that hub motors ae indeed single speed. The gearing is for the human pedaling. If they could solve the cooling issue, you might see more of them used on hilly routes, but low speed power is their weakness.

 

[AHicks]

Much depends on (1) use case and (2) design of the motor. Let's ignore hub vs mid drive for a moment.
An electric motor system can be designed with a broad power or torque curve (note the difference, though) but there will be limits, especially at low rpm, where you can have very high torque at modest power because the motor is turning slowly. Then add efficiency into the mix (or not...).
Now, note that hub motors ae indeed single speed. The gearing is for the human pedaling. If they could solve the cooling issue, you might see more of them used on hilly routes, but low speed power is their weakness.

In the geared hub case, I agree the gearing is for humans, and I also recognize/agree with the cooling issue. When it comes to power and torque curves -

I have a favorite way of demonstrating motor design differences and what happens when you move the target rpm on otherwise identical motors. MAC motors offer a perfect scenario to demonstrate this point as they offer several different winding options on what are otherwise identical motors. If you're interested in what happens check this out. Here's 2 motors, a Mac 8t (speed) and a MAC 12t (torque) run/compared side by side (these are both very high quality geared hub motors). The 8t clearly makes it's power at higher rpm, but check out how much more power it requires when doing that! Check out the difference in available starting torque!

Motor Simulator - Tools

Our ebike motor simulator allows you to easily simulate the different performance characteristics of different ebike setups - with a wide selection of hub motors modeled, and the ability to add custom batteries and controllers and set a wide variety of vehicle parameters you'll be able to see...

ebikes.ca

 

[retiredNH]

In the geared hub case, I agree the gearing is for humans, and I also recognize/agree with the cooling issue. When it comes to power and torque curves -

I have a favorite way of demonstrating motor design differences and what happens when you move the target rpm on otherwise identical motors. MAC motors offer a perfect scenario to demonstrate this point as they offer several different winding options on what are otherwise identical motors. If you're interested in what happens check this out. Here's 2 motors, a Mac 8t (speed) and a MAC 12t (torque) run/compared side by side (these are both very high quality geared hub motors). The 8t clearly makes it's power at higher rpm, but check out how much more power it requires when doing that! Check out the difference in available starting torque!

Motor Simulator - Tools

Our ebike motor simulator allows you to easily simulate the different performance characteristics of different ebike setups - with a wide selection of hub motors modeled, and the ability to add custom batteries and controllers and set a wide variety of vehicle parameters you'll be able to see...

ebikes.ca

What's really notable in the link is how going 11 kph faster doubles power requirement and halves the range. IOW, going 1/3 faster exacts a huge penalty.
Yes, motor design matters a lot...

 

[AHicks]

What's really notable in the link is how going 11 kph faster doubles power requirement and halves the range. IOW, going 1/3 faster exacts a huge penalty.
Yes, motor design matters a lot...

Yes, exactly. The idea of building a "faster" motor needs to be carefully thought out/researched. There's a LOT to this!

 

[m@Robertson]

To add to that just briefly, when it comes to e-bikes, I believe the bigger the motor, the less the need for multiple gears. If that's hard to wrap your head around, look into how many gears electric cars are (aren't!!) using.....

Yeah but you aren't pedaling the electric car for exercise. I've found that more gears allow for more variance in my pedal cadence. So on a 9 or 11 speed bike, I am using all of them. This becomes even more important when I am hauling with cargo, where small changes in cadence are necessary to haul the grocery load up a hill. Pedaling under load is very much about cadence management as the bike at that point is a moving exercise machine.

If you just want to throttle the bike, then the fewer-gears proposition is absolutely correct. If you want to exercise by tailoring your cadence so the gearing is just enough to let you push hard on the pedals while maintaining forward progress, then more is better. Motor power is irrelevant as its way more than you need no matter what if pedaling. the assumption for a powerful motor seems to be the user is always going for full power. that is seldom the case in real life.

 

[Ken M]

In the geared hub case, I agree the gearing is for humans, and I also recognize/agree with the cooling issue. When it comes to power and torque curves -

I have a favorite way of demonstrating motor design differences and what happens when you move the target rpm on otherwise identical motors. MAC motors offer a perfect scenario to demonstrate this point as they offer several different winding options on what are otherwise identical motors. If you're interested in what happens check this out. Here's 2 motors, a Mac 8t (speed) and a MAC 12t (torque) run/compared side by side (these are both very high quality geared hub motors). The 8t clearly makes it's power at higher rpm, but check out how much more power it requires when doing that! Check out the difference in available starting torque!

Motor Simulator - Tools

Our ebike motor simulator allows you to easily simulate the different performance characteristics of different ebike setups - with a wide selection of hub motors modeled, and the ability to add custom batteries and controllers and set a wide variety of vehicle parameters you'll be able to see...

ebikes.ca

The motors can be optimized for how an ebike is primarily used - mainly about slow speed climbing or faster speed urban mobility. With parallel winding some of that trade-off can be eliminated as the "trick" is to make sure the Kv and battery voltage provides efficient performance at the typical higher riding speeds.

There is a lot said about mid-drives having an advantage because they can leverage the bike gearing (we all agree this is for rider optimization) but that advantage is really only limited to very low speed climbing when the gearing magnifies the motor torque (the reason why mtn bikers truly benefit from mid-drives besides the CG and reduced unsprung wheel weight). The advantage of a hub drive is that both the motor and rider effectiveness can be optimized independently. That said, if you have an ebike with a powerful mid drive like the Bafang Ultra it's best to have a drive system focused on optimium motor performance as the rider input is much less important especially at higher speeds (say over 20mph).

 

[Ken M]

Yeah but you aren't pedaling the electric car for exercise. I've found that more gears allow for more variance in my pedal cadence. So on a 9 or 11 speed bike, I am using all of them. This becomes even more important when I am hauling with cargo, where small changes in cadence are necessary to haul the grocery load up a hill. Pedaling under load is very much about cadence management as the bike at that point is a moving exercise machine.

If you just want to throttle the bike, then the fewer-gears proposition is absolutely correct. If you want to exercise by tailoring your cadence so the gearing is just enough to let you push hard on the pedals while maintaining forward progress, then more is better. Motor power is irrelevant as its way more than you need no matter what if pedaling. the assumption for a powerful motor seems to be the user is always going for full power. that is seldom the case in real life.

You don't need a lot of gears to get that exercise. If you are comfortable say with a cadence range of 40-100rpm an ebike is probably OK with just 3 gears. If a human/rider was capable of say a 1000rpm then we'd have bikes with small front chain rings and large rear cogs to always magnify the torque to the rear wheel. There really is a lot to this subject but few ebike companies really talk about what are weaknesses to a mid-drive for urban mobility when using gearing optimized for the rider. One virtually every mopen or motorcycle you will see smaller front chain rings because that increases the torque to the rear wheel and the motor has no issue with powerband at the higher drive shaft RPMs.

 

[m@Robertson]

I can exceed the BBSHD's (and by extension the Ultra's) ability to power the bike easily. Put a 52T cog on the front and try running it in 11T in the back. On paper this can power the bike into the mid to upper 30 mph range. In practice, not so much. I've done 60T front chainrings to maintain cadence on a powerful hub. If it made sense I could do that on a mid.

Comments stating a given mid drive has a limitation of X insofar as the rider is concerned don't take into account you can change the gears in one direction or the other.

 

[m@Robertson]

You don't need a lot of gears to get that exercise. (etc etc)

Thats just flat out wrong. It fails to understand how cyclists use cadence and fails to take into account what is needed to avoid ghost pedaling. A 40-100 rpm range is colossal and totally wrong thinking. Reality is a rider's preferred cadence is much closer to a single rpm number, and the preferred variation is very small. To maintain your comfortable cadence you need gears so that, as terrain varies, you can make small changes to maintain your preferred range.

This is the opposite of what throttlers prefer, but they are not cyclists riding for exercise. So they don't necessarily understand the techniques used in riding.

With all of this said, I hate torque sensing and won't use it. I much prefer a carefully dialed in cadence based system with the kind of variations Bafang allows (or at least used to) with regard to pulling power as cadence levels increase, slow-start settings etc etc. Probably the best system out there is the Cyc system that combines the two. Torque for appling a jolt off the line (think a start from a stop when stopped on a steep hill), then it switches to cadence for regular cruise.

If I could boil down where you are going wrong: You are treating this as an engineering topic. Its not and its not about what the motor likes. Its about the rider and how riders like to ride. Take speed completely out of your vocabulary and concentrate solely on cadence.

 

[Ken M]

Thats just flat out wrong. It fails to understand how cyclists use cadence and fails to take into account what is needed to avoid ghost pedaling. A 40-100 rpm range is colossal and totally wrong thinking. Reality is a rider's preferred cadence is much closer to a single rpm number, and the preferred variation is very small. To maintain your comfortable cadence you need gears so that, as terrain varies, you can make small changes to maintain your preferred range.

This is the opposite of what throttlers prefer, but they are not cyclists riding for exercise. So they don't necessarily understand the techniques used in riding.

With all of this said, I hate torque sensing and won't use it. I much prefer a carefully dialed in cadence based system with the kind of variations Bafang allows (or at least used to) with regard to pulling power as cadence levels increase, slow-start settings etc etc. Probably the best system out there is the Cyc system that combines the two. Torque for appling a jolt off the line (think a start from a stop when stopped on a steep hill), then it switches to cadence for regular cruise.

If I could boil down where you are going wrong: You are treating this as an engineering topic. Its not and its not about what the motor likes. Its about the rider and how riders like to ride. Take speed completely out of your vocabulary and concentrate solely on cadence.f

I agree and disagree at the same time. I do agree that preferred cadence typically falls within a smaller range but there are great riders that truly love their single speed bikes where the cadence range is huge and they obviously get effective exercise.

That said "ghost pedaling" is entirely elective on an ebike. Sure you can ride a throttle ebike without pedaling at all or you can choose to ride it with a high exertion level and throttle only to achieve the speed you want to maintain. So many riders tend to think the use of throttle automatically means the rider isn't pedaling or putting out any effort but that is elective.

I do think the combination of torque sensing at low speeds with cadence bias at the higher speeds makes sense for control systems but I do have a preference for the simplicity of a throttle. I just see the merits of keeping a bike simple and allowing programmers to decide the assist level based on sensor parameters (torque, speed, cadence, gear, hearth rate, slope, etc.) is at best a guessing game because they do not have a way to know exactly what assist the rider wants continuously. Their egos think they can achieve this but they can't. It's a fools errand but I'm not saying their are not merits to PAS systems (they can achieve more traditional bike-like feeling).

There is no doubt that gearing on a traditional bike was important because we humans have limited capability. Bikes are incredibly good at amplifying out capably. Theres a great short video of Steve Jobs talking about this. But ebikes are the most efficient form of mobility ever ... they are more efficient than fully loaded passenger trains or even someone walking from point A-to-B so I do tend to consider it an engineering topic because that is how some ebike (urban mobility focused ebikes) should be designed. For example, for sustaining a speed over 20mph the rider may be only be capable of providing a fraction of the continuous power needed so doesn't it make sense to focus on motor drive system optimization for urban mobility. We are truly very capable of getting a bike moving from a dead stop and at sustaining speeds in the 10-15mph range which is where motors are less effective.