Any IGHs with over 350% range, >20,000km service life & 300nM torque rating???

[mschwett]

You are not correct on this. Motors like to spin fast so the best way to provide the motor power to the rear wheel would be a small front cog and a larger rear cog. Check out the configuration of Luna's Sur-Ron electric cycle (very small front and large rear and single speed). This is pretty much how every electric motorcycle is configured. Sure mid-drives when in the lower gears (ie getting a mechanical boost from the gearing can climb well but there is a compromise taking place when you combine human and motor thru one drive system). Think about a rider putting their power thru that gearing you mention and they get a good boost at a desirable cadence but if the motor is separated a small front to a very large rear will significantly boost the motor torque more than sharing what is best for us weak pedaling humans. If you are riding a say 20mph with a 44T front chainring and an 11T rear chain ring on a mid-drive the combined human and motor torque is dropped 75% just to provide reasonable cadence for the rider. If the motor was on a separate drive the torque would still be magnified (there are other factors such as motor Kv and Kt that play into this but just given that motors are more efficient spinning fast I don't think there is much chance that integrating a motor and rider via a mid-drive is optimized. I don't think there is any technical merit to saying a combine drive system is best. Simple not true based on everything I know but I'm wanting to know if there is any chance this is wrong.

the solution is very simple though: you optimize the electric motor for the weight, size, and power targets of the bike/vehicle and that have internal gears to match it to a speed which is compatible with the human rider. it’s a bike. not a motorcycle. the efficiency of a derailleur drivetrain is incredibly high, and the latest IGH are getting there as well.

as you know, most all hub motors are also geared (again, size and weight, which is now rotating mass!) which further illustrates why the quest for an arbitrary 1:1 relationship between motor and axle speed is quixotic.

the bafang motor you’re talking about is also geared internally, to produce optimum power in the RPM range expected of the crank of a bicycle. if they had expected it to drive the rear axle at 1:1 at the “middle” of the range, they’d undoubtedly have geared it differently. for most bike drivetrains 1:1 is the extreme, with 1:2 in the middle and 1:4 at the other end.

 

[Ken M]

the solution is very simple though: you optimize the electric motor for the weight, size, and power targets of the bike/vehicle and that have internal gears to match it to a speed which is compatible with the human rider. it’s a bike. not a motorcycle. the efficiency of a derailleur drivetrain is incredibly high, and the latest IGH are getting there as well.

as you know, most all hub motors are also geared (again, size and weight, which is now rotating mass!) which further illustrates why the quest for an arbitrary 1:1 relationship between motor and axle speed is quixotic.

the bafang motor you’re talking about is also geared internally, to produce optimum power in the RPM range expected of the crank of a bicycle. if they had expected it to drive the rear axle at 1:1 at the “middle” of the range, they’d undoubtedly have geared it differently. for most bike drivetrains 1:1 is the extreme, with 1:2 in the middle and 1:4 at the other end.

Huh....Most of the modern mtn bike chain 1x drivetrains have rear 50T spockets and fronts in the 32-36T range so that is under 1:1. It's not accurate to say "most bike drivetrains 1:1 is the extreme" because with IGHs you actually have two drivetrains (one just happens to be a fixed ratio). I am struggling trying to grasp how the gearing in the motor, drive ratio from crank to rear axle, and finally the IGH range and ratios all play out but it sure seems like the IGHs are being protected from too much torque when the Bafang M620 is used.

You mention that the 1:1 relationship is arbitrary but it is not consistent with the mid drive internal gearing to allow the motor to spin faster and then reduce that spin speed and torque by 50% via the final ratio from the crank to rear axle. A 1:1 would allows the motor to be be spinning faster. I understand that the IGH gearing is a factor in all this but if this drive ratio is done to protect the IGH from too much torque then I ??? the practice.

The reason why there is so much gear reduction in mid-drives is because they need it to increase the torque produced and the motors are more efficient when spinning faster. But the maximum torque a motor produces is at stall so I'm sure the design intent is to "optimize" the motor performance over the range of ebike speeds but you really can not achieve that if both the rider and motor utilize the same drivetrain ... that just seems self evident to me as a motor doesn't need a slower cadence gearing provides for the rider at say 20mph.

That said I just question that it's ideal to have a 1:2 - 1:4 ratio on IGHs. Seems to me the best would be to transmit as much of the human and motor torque to the rear assuming the range and ratios of the IGH provide the speed and cadence range desired and it can take the torque loads if not reduced via gearing.

The efficiencies of a derailleur and IGH is actually a mutually exclusive topic.

 

[jabberwocky]

Strange here's the delite with a Nuvinci with a 26T front and 28T rear. All I did was do a quick search on EBR for all the R&M reviews and then looked at the cogs when they had an IGH installed

Thats weird. They don't show those ratios on the models on their website with either hub, but their bikes are (I think) pretty much custom built so you could probably get them that way if you asked. The Rohloff has a ratio of ~0.3-1.5, and the Enviolo is 0.5-1.9. I'd recommend converting your current gearing to gear inches (which accounts for drive ratio as well as wheel diameter and tire size) and then figure out what front/rear combo puts the IGH where you want it in comparison.

A 1:1 ratio with the Rohloff would be hilarious to ride. You could probably ride up a wall with the torque it would give but you'd be spinning like crazy just to go fast enough to stay upright!

 

[Ken M]

Thats weird. They don't show those ratios on the models on their website with either hub, but their bikes are (I think) pretty much custom built so you could probably get them that way if you asked. The Rohloff has a ratio of ~0.3-1.5, and the Enviolo is 0.5-1.9. I'd recommend converting your current gearing to gear inches (which accounts for drive ratio as well as wheel diameter and tire size) and then figure out what front/rear combo puts the IGH where you want it in comparison.
.
A 1:1 ratio with the Rohloff would be hilarious to ride. You could probably ride up a wall with the torque it would give but you'd be spinning like crazy just to go fast enough to stay upright!

I just realized that maybe some / most of the EBR reviews were with the earlier Bosch mid-drives that had the 2x chain ring to crank gearing. I always thought this was a superior system for S-pedelecs and it seemed to come out in some reviews but Bosch still ended up changing that design on the newest models. This just added another layer of complexity so I want to look at the Bafang M620 ebikes with IGHs again because I know it does not have any chain ring multiplying of the crank cadence.

 

[mclewis1]

... Can you imagine a car with transmission or rear end differential not strong enough to reliably take the full power of the motor it's integrated to (obviously muscle cars can do burnouts without damaging anything in the drive train)....

That is exactly the situation with just about any large displacement engine and automatic transmission combination. There are indeed a number of known weak points in the drive trains, but what prevents them from failing is built in "slippage" when too much torque is applied on start up. It also allows an engine to stay in a more efficient rpm range as the vehicle starts to move. That slippage is provided by either the viscous coupling in a torque converter or a dry multi stage clutch. Want to find the weak points? Simply remove or bind that slippage point, remove the geared slippage in the differential (a locked differential) and use tires with a higher friction coefficient (wider with less tread). Then you'll quickly find which parts of the automatic transmission or differential that can't take the torque of an unmodified engine on startup.

----------------------

Cargo ebikes are entering into much the same environment. We're now capable of building higher torque motors and the various components in the rest of the drivetrain are coming into question (there's no engineered "slippage" to limit the application of torque on less than capable components). As Mr Robertson suggested it's actual riding style that's usually preventing substantial failures. A nice slow application of power will indeed protect the drivetrain. Unfortunately as with any 2 wheeled bike you need to quickly get up to a reasonable speed to allow for stable operation. With a rotational torque (pedal pressure) based control of power you're initially on the fine line of balance and the extra weight of the cargo makes this even worse.

One option is to move to a 3 wheeled configuration where low starting speed/balance isn't an issue and then gear everything down to make the initial movement easier. Another is to just engineer in more motor torque potential on startup but this puts more drivetrain components at risk and changes the riding feel.

Another method of reducing the risk of component failure is to use a throttle with controller programming that applies power/torque smoothly on startup. This alleviates the uneven application of power from the combination of rider applied torque (initially standing on the pedals) and the geared motor, and simply gets the whole setup moving efficiently. Once moving at a reasonable rate the throttle's use is over and a rider's pedaling comfortably adds to or even replaces the power from the motor. Unfortunately a throttle usually also introduces higher top speeds with the resulting regulatory issues. But IMHO a properly configured throttle would make any cargo bike safer, easier to ride, and more reliable.

 

[jabberwocky]

I just realized that maybe some / most of the EBR reviews were with the earlier Bosch mid-drives that had the 2x chain ring to crank gearing. I always thought this was a superior system for S-pedelecs and it seemed to come out in some reviews but Bosch still ended up changing that design on the newest models. This just added another layer of complexity so I want to look at the Bafang M620 ebikes with IGHs again because I know it does not have any chain ring multiplying of the crank cadence.

Ahhhh, that explains it. Planetary geared cranksets have been around for a while (I knew someone who had a Schlumpf on a singlespeed mtb like 15 years ago).

 

[jabberwocky]

Cargo ebikes are entering into much the same environment. We're now capable of building higher torque motors and the various components in the rest of the drivetrain are coming into question (there's no engineered "slippage" to limit the application of torque on less than capable components). As Mr Robertson suggested it's actual riding style that's usually preventing substantial failures. A nice slow application of power will indeed protect the drivetrain. Unfortunately as with any 2 wheeled bike you need to quickly get up to a reasonable speed to allow for stable operation. With a rotational torque (pedal pressure) based control of power you're initially on the fine line of balance and the extra weight of the cargo makes this even worse.

Yeah, we are in this weird area where ebikes are exploding in popularity, but component mfgs haven't really caught up. Cargo bikes were already bit of a niche before ebikes came along, especially ones that can carry a substantial load. The bottom line is there isn't really anything out there designed around a several hundred pound bike and not much designed around the higher power mid-drive motors. Put the two together and you're in waaaay outside design limits land. I mean, your average Harley Davidson weighs 550-900 pounds; take a look at the drivetrain on one and you can see whats required to couple high power to that weight regularly.

 

[mschwett]

… I mean, your average Harley Davidson weighs 550-900 pounds; take a look at the drivetrain on one and you can see whats required to couple high power to that weight regularly.

…and why a harley shouldn’t be in the bike lane

 

[Ken M]

Some really good points have been made in the recent posts. Thanks for the feedback on this. I thought there seemed to be a potential risk when integrating the Bafang Ultra with an IGH on some ebikes, especially cargo ebikes. I think the tendency for someone to stand on the pedals when starting which typically maxes out the torque on pedelecs can be problematic for most IGHs so riders just have to be aware of this is my current understanding.

 

[m@Robertson]

You are not correct on this. Motors like to spin fast so the best way to provide the motor power to the rear wheel would be a small front cog and a larger rear cog.

You are stating the obvious, but its only a part of the story. Here's reality: When starting from a stop, the motor either lugs in a cog too small for it, or spins fast in a gear properly sized for it. (Fix the lugging in the small cog so it doesn't need to shift? Use more torque, which tears apart the drivetrain). If you do it right with a granny gear, you need to shift to continue accelerating or the motor reaches peak rpm and you are riding a bike at 15 mph at full rpms. Which is stupid.

At a guess you're unaware of how low the peak rpms are on most of these motors. There's the big limiting factor.

I did say there is a level where gears are not necessary. Bigass 3-6kw hubs like a QSv3 are how that is usually done. But we are talking mids here. I have an enduro bike with a Cyc X1 Pro. That motor as-configured peaks at 3000w, its controller (a BAC800 done for me by Electric Race Technologies) has shown 60a output and climbing before I chickened out and lifted. Oh and the motor can sustain 10000 rpms. Ten Thousand. It outputs 180 Nm. My front ring is 32T (there's your small ring), and my EX1 tool steel chainring in the back goes as high as I think 46, although I don't go past the one just under that to keep chain alignment safe on a motor whizzing that fast.

THAT motor can skip the gear changes. But its not practical. Think about what I am listening to as a motor runs up at rpms like that. Consider for a moment what you are up against with respect to the chain, chainrings, cogs. Can a belt do better? Maybe. I only know of a Lightning Rods Big Block that has used them though. And every one of those are effectively custom builds.

Another issue with current single drivetrain mid drive, is that because of the pedaling gear ratio needed to provide proper pedaling cadence range, the gearing for the motor is not ideal, hence the need to use the gears for the motor too.
If you were to give these eBike a separate drivetrain with a ratio suited to the motor, I suspect the motor could handle the power without gears. It would just spins at higher rpms at the top of the range.

Yeah but then its no longer a bicycle. Doing that with a motor is what is commonly called a jackshaft, if I am reading you right. They are some really high power builds, typically, but mostly eMTBs. I have seen a Big Fat Dummy that used a LR big block ... but it was a train wreck that never worked right. Look to E-S for the build saga and aftermath on it.

Nobody does this with cargo bikes, I assume, because of the lack of reliability. My Cyclone and Cyc builds are a kick in the ass, but they have too much spinning steel, extra freewheels and bearings to make them practical daily drivers.

The torque that would break loose a rear tire isn't that complex of a calculation but it's only an estimate. I was just suggesting that an IGH that could reliably handle that much torque is at the limit that would ever be needed on any ebike regardless of the motor power. I asked Kindernay if they would guarantee their IGH with a Bafang G620 (assuming 1:1 belt/chain ratio) and they said that was torque that they would not consider a normal bike. I assume that meant no.

Keep in mind I am not considering if the IGH from any of the brands are proven reliable with the 250W EU mid drives. I want to know if they will have the same success when they are integrated on an ebike with a 750W (or higher peak) Bafang M620 including the possibility of say a 200lb rider standing on the cranks. Oh, and not geared down to reduce the torque from the mid drive axle to the rear IGH. I think there is clear concern in the industry that the Bafang M620 is a bit much for the current IGHs (and not just with typical cargo bike loads).

Big surprise... the manufacturer would not warranty it. If you build smart, and you ride smart, it will work fine. This is a thing I saw when I was building hot rod cars. Some guys wanted guarantees and warranties. But if you are going to play in the big leagues, you have to acknowledge that your wallet is the warranty and your brain is the guarantee.

Its been figured out and done before with success. You are far from the first person to think on this and won't be the first build of one, by any stretch. Should it be done on a cargo application? Only field experience will tell you this. So put the keyboard down, wipe off the donut crumbs and get out into the garage.

 

[m@Robertson]

Another method of reducing the risk of component failure is to use a throttle with controller programming that applies power/torque smoothly on startup. This alleviates the uneven application of power from the combination of rider applied torque (initially standing on the pedals) and the geared motor, and simply gets the whole setup moving efficiently. Once moving at a reasonable rate the throttle's use is over and a rider's pedaling comfortably adds to or even replaces the power from the motor. Unfortunately a throttle usually also introduces higher top speeds with the resulting regulatory issues. But IMHO a properly configured throttle would make any cargo bike safer, easier to ride, and more reliable.

I have done this on all three of mine, and its critical to their survival. Its worth noting the older UART M620's had a very similar settings interface that allowed the same thing: Slow application of initial power that ramps up gently. Also heavy reduction of power as cadence increases past different points at different assist levels. If you can spin the crankarms you don't need heavy electrical assist.

I did take an entirely different approach on my Bullitt, that was a follow-on to earlier efforts that had great success: A front hub motor can be used to get the bike going instead of the mid drive beating on the drivetrain from a standstill. On my first mid+hub bike I found I didn't need to downshift at a stop, and I could leave the bike permanently in an 11T cog (flat ground). Doing that, that $7 cog still lived a hard life but it lasts for 1000-1100 miles. So I knew the dual-motor thing worked to spare the drivetrain. On the Bullitt, I have set the 160Nm BBSHD to not start up until 6 mph+. From zero to 5.99 mph, its my legs and the front hub. Also of course the BBSHD rolls on the power gently at the start.

The result is a bike that I can comfortably pedal loaded (no throttle) from a standstill. I think I am just over 1500 miles on it and still using original drivetrain components with no signs of wear (I keep it 3 cogs up in back and the cluster is steel). If I do use throttle its a one-one-thousand short bump on the front motor, and sometimes I hit both of them to get up to balance under load. Never more than about 2 seconds. If I build another cargo bike (I am looking at a Bullitt X later this year) it will absolutely be 2wd; this works so well.

Dual Motor AWD Electric Bikes (the good and bad)

I like to build top-quality-component ebikes from the frame up. Quite a few of them are dual motor or AWD or 2WD or whatever you want to call them. Why would you build an AWD ebike?

talesontwowheels.com

The fact that the Bafang motors do not deliver peak power like a light switch is why 300Nm is in fact not possible. So I'm changing what I said earlier as I forgot about the ramp-up they go thru.

BBSHD Programming For The Pedaling Cyclist

(But Not For The Throttler) There is a follow-on to this article that tinkers with these settings a bit further. In August 2023 I published another update that passed along what I’m doing dif…

talesontwowheels.com

 

[Ken M]

You are stating the obvious, but its only a part of the story. Here's reality: When starting from a stop, the motor either lugs in a cog too small for it, or spins fast in a gear properly sized for it. (Fix the lugging in the small cog so it doesn't need to shift? Use more torque, which tears apart the drivetrain). If you do it right with a granny gear, you need to shift to continue accelerating or the motor reaches peak rpm and you are riding a bike at 15 mph at full rpms. Which is stupid.

At a guess you're unaware of how low the peak rpms are on most of these motors. There's the big limiting factor.

I did say there is a level where gears are not necessary. Bigass 3-6kw hubs like a QSv3 are how that is usually done. But we are talking mids here. I have an enduro bike with a Cyc X1 Pro. That motor as-configured peaks at 3000w, its controller (a BAC800 done for me by Electric Race Technologies) has shown 60a output and climbing before I chickened out and lifted. Oh and the motor can sustain 10000 rpms. Ten Thousand. It outputs 180 Nm. My front ring is 32T (there's your small ring), and my EX1 tool steel chainring in the back goes as high as I think 46, although I don't go past the one just under that to keep chain alignment safe on a motor whizzing that fast.

THAT motor can skip the gear changes. But its not practical. Think about what I am listening to as a motor runs up at rpms like that. Consider for a moment what you are up against with respect to the chain, chainrings, cogs. Can a belt do better? Maybe. I only know of a Lightning Rods Big Block that has used them though. And every one of those are effectively custom builds.


Yeah but then its no longer a bicycle. Doing that with a motor is what is commonly called a jackshaft, if I am reading you right. They are some really high power builds, typically, but mostly eMTBs. I have seen a Big Fat Dummy that used a LR big block ... but it was a train wreck that never worked right. Look to E-S for the build saga and aftermath on it.

Nobody does this with cargo bikes, I assume, because of the lack of reliability. My Cyclone and Cyc builds are a kick in the ass, but they have too much spinning steel, extra freewheels and bearings to make them practical daily drivers.

Big surprise... the manufacturer would not warranty it. If you build smart, and you ride smart, it will work fine. This is a thing I saw when I was building hot rod cars. Some guys wanted guarantees and warranties. But if you are going to play in the big leagues, you have to acknowledge that your wallet is the warranty and your brain is the guarantee.

Its been figured out and done before with success. You are far from the first person to think on this and won't be the first build of one, by any stretch. Should it be done on a cargo application? Only field experience will tell you this. So put the keyboard down, wipe off the donut crumbs and get out into the garage.

I'm actually working with someone on an IGH design that we want to spec with a torque rating over 300nM in any gear (3 and 5 speed versions planned and mainly targeting the Bafang M600 and M620 but would be even attractive on the EU spec mid drives for reliability and long life. We are just trying to assess if the needs of an ebike with these motors is served by any of the current IGHs without being ridden gently.

 

[Ken M]

I have done this on all three of mine, and its critical to their survival. Its worth noting the older UART M620's had a very similar settings interface that allowed the same thing: Slow application of initial power that ramps up gently. Also heavy reduction of power as cadence increases past different points at different assist levels. If you can spin the crankarms you don't need heavy electrical assist.

I did take an entirely different approach on my Bullitt, that was a follow-on to earlier efforts that had great success: A front hub motor can be used to get the bike going instead of the mid drive beating on the drivetrain from a standstill. On my first mid+hub bike I found I didn't need to downshift at a stop, and I could leave the bike permanently in an 11T cog (flat ground). Doing that, that $7 cog still lived a hard life but it lasts for 1000-1100 miles. So I knew the dual-motor thing worked to spare the drivetrain. On the Bullitt, I have set the 160Nm BBSHD to not start up until 6 mph+. From zero to 5.99 mph, its my legs and the front hub. Also of course the BBSHD rolls on the power gently at the start.

The result is a bike that I can comfortably pedal loaded (no throttle) from a standstill. I think I am just over 1500 miles on it and still using original drivetrain components with no signs of wear (I keep it 3 cogs up in back and the cluster is steel). If I do use throttle its a one-one-thousand short bump on the front motor, and sometimes I hit both of them to get up to balance under load. Never more than about 2 seconds. If I build another cargo bike (I am looking at a Bullitt X later this year) it will absolutely be 2wd; this works so well.

Dual Motor AWD Electric Bikes (the good and bad)

I like to build top-quality-component ebikes from the frame up. Quite a few of them are dual motor or AWD or 2WD or whatever you want to call them. Why would you build an AWD ebike?

talesontwowheels.com

The fact that the Bafang motors do not deliver peak power like a light switch is why 300Nm is in fact not possible. So I'm changing what I said earlier as I forgot about the ramp-up they go thru.

BBSHD Programming For The Pedaling Cyclist

(But Not For The Throttler) There is a follow-on to this article that tinkers with these settings a bit further. In August 2023 I published another update that passed along what I’m doing dif…

talesontwowheels.com

Long term I would like to see a Semi-recumbent with a front axle transmission (Like Kervelo showed before going bankrupt) and a rear in-runner motor powering the rear wheel via a belt. That is a product that would take time to find wide acceptance but for urban mobility it just seems conceptually sound.
.
It seems some keep insisting the ebike motors actually need the typical drive train gearing the rider benefits from but both DD and geared hub motors do not need gears to support the full range of bike speeds (they drive the rear wheel directly so even though the geared hub motors have one reduction planet set (usually 5:1) they are not utilizing the drive train gearing and perform fine over the travel speeds of ebikes)

 

[Ken M]

I have done this on all three of mine, and its critical to their survival. Its worth noting the older UART M620's had a very similar settings interface that allowed the same thing: Slow application of initial power that ramps up gently. Also heavy reduction of power as cadence increases past different points at different assist levels. If you can spin the crankarms you don't need heavy electrical assist.

I did take an entirely different approach on my Bullitt, that was a follow-on to earlier efforts that had great success: A front hub motor can be used to get the bike going instead of the mid drive beating on the drivetrain from a standstill. On my first mid+hub bike I found I didn't need to downshift at a stop, and I could leave the bike permanently in an 11T cog (flat ground). Doing that, that $7 cog still lived a hard life but it lasts for 1000-1100 miles. So I knew the dual-motor thing worked to spare the drivetrain. On the Bullitt, I have set the 160Nm BBSHD to not start up until 6 mph+. From zero to 5.99 mph, its my legs and the front hub. Also of course the BBSHD rolls on the power gently at the start.

The result is a bike that I can comfortably pedal loaded (no throttle) from a standstill. I think I am just over 1500 miles on it and still using original drivetrain components with no signs of wear (I keep it 3 cogs up in back and the cluster is steel). If I do use throttle its a one-one-thousand short bump on the front motor, and sometimes I hit both of them to get up to balance under load. Never more than about 2 seconds. If I build another cargo bike (I am looking at a Bullitt X later this year) it will absolutely be 2wd; this works so well.

Dual Motor AWD Electric Bikes (the good and bad)

I like to build top-quality-component ebikes from the frame up. Quite a few of them are dual motor or AWD or 2WD or whatever you want to call them. Why would you build an AWD ebike?

talesontwowheels.com

The fact that the Bafang motors do not deliver peak power like a light switch is why 300Nm is in fact not possible. So I'm changing what I said earlier as I forgot about the ramp-up they go thru.

BBSHD Programming For The Pedaling Cyclist

(But Not For The Throttler) There is a follow-on to this article that tinkers with these settings a bit further. In August 2023 I published another update that passed along what I’m doing dif…

talesontwowheels.com

It's my understanding the the max motor torque is at stall / stop so it seems very possible a combined 300NM is possible. It drops down dynamically but remains high at slower speeds. It's almost natural for a rider to stand on the cranks from a stop to get going and this torque typically will max out the motor assist on most pedelecs.

If a pedelec with a Bafang M620 was being held in place and say a 200lb then put their full weight on one horizontal pedal which is torque sensed to deliver the full motor assist, I'm wondering why that would not result in a potential torque to the rear axle / IGH of 300nM (assuming the drive ration is 1:1 front to rear). It just seems to me this is possible on both cargo and regular ebikes that have the Bafang M620 or BBS2HD and no one really openly discusses it. I'm not saying these ebikes couldn't be ridden such that the max rating of the existing IGHs is not exceeded but the fact is these motors sure seem to have the potential to damage some expensive IGHs if people just assume they are designed to handle the full performance of the motors.

I've seen advice in articles to ride Bafang Ultra ebikes conservatively if they have an rear IGH but how many buyers are being instructed about this caution?

 

[mschwett]

people aren’t “openly discussing it” because the vast majority of bikes like this have a roughly 1:2 drive ratio front to rear before the IGH, as discussed over and over! you’re applying an artificial constraint which doesn’t seem to have any upside other than the idea that a 1:1 size of the front and rear cog is somehow efficient.

once you decouple the pedals and motor from the driven wheel, you have to transmit the power and some losses will occur from moving the belt/chain, spinning the cogs, friction within the belt/chain, and friction in the various interfaces. i can’t imagine there’s much difference in efficiency with varying sizes of cogs front to rear. it’s free gearing, no?

an amateur rider also typically has their hands on the bars when they stand up at a start, making that full torque on the end of the crank unlikely. and really strong / athletic riders who do put that kind of power down don’t use super powerful mid drive motors or cargo bikes….

 

[Ken M]

people aren’t “openly discussing it” because the vast majority of bikes like this have a roughly 1:2 drive ratio front to rear before the IGH, as discussed over and over! you’re applying an artificial constraint which doesn’t seem to have any upside other than the idea that a 1:1 size of the front and rear cog is somehow efficient.

once you decouple the pedals and motor from the driven wheel, you have to transmit the power and some losses will occur from moving the belt/chain, spinning the cogs, friction within the belt/chain, and friction in the various interfaces. i can’t imagine there’s much difference in efficiency with varying sizes of cogs front to rear. it’s free gearing, no?

an amateur rider also typically has their hands on the bars when they stand up at a start, making that full torque on the end of the crank unlikely. and really strong / athletic riders who do put that kind of power down don’t use super powerful mid drive motors or cargo bikes….

Here's what I question. I have read that the gear reduction inside the M620 motor itself is something from 18:1 to 28:1. The design uses 3 levels of gear reduction ... it makes no sense to do that and then give 50% of the torque gained via gear reduction with the belt / chain ratio unless there was too much reduction to allow the top speed / cadence desired when using an IGH. I don't think the Kv of the Bafang motor is such that it would need this. You get some of that back via the underdrive gears on the IGHs but even then I just don't it makes sense unless not wanting the full torque from rider and motor to impact the IGH.

 

[Ken M]

You are stating the obvious, but its only a part of the story. Here's reality: When starting from a stop, the motor either lugs in a cog too small for it, or spins fast in a gear properly sized for it. (Fix the lugging in the small cog so it doesn't need to shift? Use more torque, which tears apart the drivetrain). If you do it right with a granny gear, you need to shift to continue accelerating or the motor reaches peak rpm and you are riding a bike at 15 mph at full rpms. Which is stupid.

At a guess you're unaware of how low the peak rpms are on most of these motors. There's the big limiting factor.

I did say there is a level where gears are not necessary. Bigass 3-6kw hubs like a QSv3 are how that is usually done. But we are talking mids here. I have an enduro bike with a Cyc X1 Pro. That motor as-configured peaks at 3000w, its controller (a BAC800 done for me by Electric Race Technologies) has shown 60a output and climbing before I chickened out and lifted. Oh and the motor can sustain 10000 rpms. Ten Thousand. It outputs 180 Nm. My front ring is 32T (there's your small ring), and my EX1 tool steel chainring in the back goes as high as I think 46, although I don't go past the one just under that to keep chain alignment safe on a motor whizzing that fast.

THAT motor can skip the gear changes. But its not practical. Think about what I am listening to as a motor runs up at rpms like that. Consider for a moment what you are up against with respect to the chain, chainrings, cogs. Can a belt do better? Maybe. I only know of a Lightning Rods Big Block that has used them though. And every one of those are effectively custom builds.


Yeah but then its no longer a bicycle. Doing that with a motor is what is commonly called a jackshaft, if I am reading you right. They are some really high power builds, typically, but mostly eMTBs. I have seen a Big Fat Dummy that used a LR big block ... but it was a train wreck that never worked right. Look to E-S for the build saga and aftermath on it.

Nobody does this with cargo bikes, I assume, because of the lack of reliability. My Cyclone and Cyc builds are a kick in the ass, but they have too much spinning steel, extra freewheels and bearings to make them practical daily drivers.

Big surprise... the manufacturer would not warranty it. If you build smart, and you ride smart, it will work fine. This is a thing I saw when I was building hot rod cars. Some guys wanted guarantees and warranties. But if you are going to play in the big leagues, you have to acknowledge that your wallet is the warranty and your brain is the guarantee.

Its been figured out and done before with success. You are far from the first person to think on this and won't be the first build of one, by any stretch. Should it be done on a cargo application? Only field experience will tell you this. So put the keyboard down, wipe off the donut crumbs and get out into the garage.

I'm actually not looking for hot rod level performance but I would like to know that there is a production IGH that is truly capable of reliable performance if used with a Bafang M620. I don't think the Kv of the motor is what is making the ebike brands put 2:1 reduction front to rear....I just think they do it to reduce the torque transferred.

By the way there is no ebike regulation that states that the motor power must go thru the same drive system as the rider. Look up the Izip Express as it had separate rider and motor drive systems to the rear wheel.

As for belts there are number of Bafang M620 ebikes using the Gates belt so I think it's fully capable and probably still last many times as long as a chain.

 

[scrambler]

I would like to know that there is a production IGH that is truly capable of reliable performance if used with a Bafang M620.

The best ones we know of are the Kindernay and the Rohloff.
The new Kindernay VII has the edge with a bit more strength and the right number of gears

 

[Ken M]

The best ones we know of are the Kindernay and the Rohloff.
The new Kindernay VII has the edge with a bit more strength and the right number of gears

I checked with Kindernay and they said they do not consider an ebike with the Bafang M620 to be a bike and will not warranty their IGH on those models (probably why you don't see them on Bafang M620 ebikes from any brand).

 

[scrambler]

WattWagons has been configuring bikes with them for a while now
You can go browse the WattWagons forum and ask users of the Hydra with Kindernay.