Still searching for the perfect ultra-light ebike kit for road bikes — here’s the idea in my head

[Katrina92]

I've done quite a bit of research on e-bike kits and have some thoughts.

As I mentioned in my previous post, I’m looking for a kit that is compatible with my lightweight road bike (Scott Addict).
A few days ago, I tried out the Scott Addict eRIDE 30 and confirmed that the torque from the Mahle X20 is enough for my needs.
After looking around, I realized most conversion kits on the market weigh around 4 kg.

I consulted with my engineer friend, who's a seasoned motor expert. He told me that the size of the motor has a lot to do with torque. I asked him if it's possible to have a motor with the torque of the Mahle X20, but smaller in size.
To my surprise, he told me it could be at least 30% smaller than the X20.

So, based on our conversation and my ongoing research, I've come up with an idea:
Is it possible to create an ultra-small, ultra-light hub motor? (Of course, it needs to ensure it has enough torque for hill climbing.)

 

[Stefan Mikes]

It is possible to buy a good road e-bike before the Trump tariffs take the full effect. Like a Scott eride or Trek Domane+
I'm afraid you are looking for a unicorn Katrina.

 

[Gionnirocket]

How do you plan on getting the idea from your head to your bike?
That said.. Anything is possible if you want to pay for it.

 

[Bicyclista]

As I mentioned in my previous post, I’m looking for a kit that is compatible with my lightweight road bike (Scott Addict).

Is it possible to create an ultra-small, ultra-light hub motor? (Of course, it needs to ensure it has enough torque for hill climbing.)

In my limited experience, the smallest hub motor I've seen is the front-hub motor (about 2.2 lbs) on the Gocycle. I doubt Gocycle will sell you one separately, but used Gocycle motors occasionally show up on eBay, such as this one. According to Richard Thorpe, the former McLaren F1 engineer who owns Gocycle, his motor has an "equivalent" torque of 60Nm, which is enough for hill climbing. (I have a Gocycle GS, and I find I can climb moderate hills easily.) But the 60Nm torque figure is only good for Gocycle's 20-inch wheels. It is likely to be much lower when installed on your Scott Addict's larger wheels, but I don't know the conversion factor. Still, if you have strong legs, it's probably enough to assist hill climbing effectively.

The Gocycle G3 motor in the eBay listing is a 22-volt motor. The newer G4 models use 36 volts. In any case, you would have to match your electric system to the motor's electrical requirements. A battery, depending on size, would add several more pounds to the system, assuming you find a good way to attach it. The whole project is daunting.

Me, I would just buy a ready-made Scott Addict ebike!

 

[Biplaneguy]

Make the motor smaller and the internal stresses go up for the same torque, so longevity might become an issue. Also making it smaller means more issues with heat rejection.

 

[BlackHand]

Maybe you should get an exoskeleton, then you don't have to mess with your bike at all. 2.2kg for the system, 2.6kg with a spare battery which you would almost certainly need.

Buy Hypershell X Series: Advanced Exoskeleton Technology

Discover Hypershell's wearable robot exoskeletons designed for heavy loads, improved walking speed, and enhanced quality of life.

hypershell.tech

 

[Stefan Mikes]

Maybe you should get an exoskeleton, then you don't have to mess with your bike at all. 2.2kg for the system, 2.6kg with a spare battery which you would almost certainly need.

Buy Hypershell X Series: Advanced Exoskeleton Technology

Discover Hypershell's wearable robot exoskeletons designed for heavy loads, improved walking speed, and enhanced quality of life.

hypershell.tech

I'm sorry Steve for laughing. It even beats the "airbag bike helmet"!

 

[jabberwocky]

The ebike market is hyper competitive, so I suspect that if a significantly smaller, lighter motor was easily achievable someone would be making it. There are very likely aspects your engineer friend isn't considering that make it a more difficult problem than you think.

 

[Jeremy McCreary]

Make the motor smaller and the internal stresses go up for the same torque, so longevity might become an issue. Also making it smaller means more issues with heat rejection.

Also, going smaller might require more expensive materials.

Heat rejection's especially important for a hub-drive bogged down on a hill. Hub motor shaft speed is tied to drive wheel speed, not cadence (as it would be with a mid-drive).

So the more you slow on a hill, the less efficient the hub motor becomes, and the greater the fraction of input electrical power lost to motor heating. That also means less power for forward propulsion.

If your legs can't make up the shortfall in assist, you'll slow even more. If the hill's long enough and steep enough, you can end up triggering the motor's thermal protection before you ever get to the top.

Clearly, the faster a hub motor sheds heat, the better.

 

[Catalyzt]

My gut is telling me: Don't do this.

The thing about CF-- as I understand it-- is that it's extremely lightweight and durable and very safe when subjected to stress from the vectors it was designed for.

At its simplest level, this is why CF works for airplane hulls and not for submersibles, or why earlier CF bikes were often great on the road, but experienced structural problems when dropped, or after relatively trivial accidents. And just at a glance, the Addict's frame seems kinda exotic; it's not JUST a superlight CF frame, it's a hyperlight, with some very thin spots. I'd be nervous putting any kind of motor on any part of that bike. Sure, the motor on a hub drive is on the wheel, but...

Is the rear triangle designed for the added force? Would that really be the same as if you were just, like, twice as strong as you are?

I don't think anyone can give you a fair and accurate answer to that question.

Ebike motors are, basically, a little weird in terms of how they place stress on a materials. Rotational energy is always weird, right? I split a TSDZ2B motor housing-- and that's steel, I think-- doing almost nothing, just puttering along on level terrain in power assist level 1. It was an unusual build, but not radical, and built by someone who really knew what they were doing. The only things that were out of spec were that the bottom bracket was wider than recommended, and the chainline was not great. It's possible I just got a bad one, that the metal in the housing was no good, but I think a lot of little factors just stressed the motor housing in ways it wasn't designed for.

I also blew up a 250W front hub motor, probably from climbing hills that were too long and too steep. It was probably all of about 40nm, and I found it perfectly serviceable for grades of 20% or so in granny gear, plenty of power. Like the mid drive, it took about 700 miles to fail. It showed no signs of problems until the last 50 miles or so. My buddy weighs about 40 pounds more than I do and he's very strong; we were just asking it to do something it was not designed to do.

I will rebuild the mid-drive eventually, probably with a DM02. The front hub motor on the other bike cost me $450 or so, I had a lot of fun with it, money well spent, lots of good memories.

* * * * *

I mean, I get it. I'm a weight weenie, too. When I dropped 2.5 pounds from my 49 pound, 40nm, 250 watt eMTB, it was like a different bike. I took it up mountains, got it up to 40+ MPH downhill speeds, I can carry it down stairs more easily. I'm still crazy about it. Of course I'm tempted to drop more. I looked at a new fork today; at 44 pounds, my bike would be such a monster...

That would cost me about $1,500. And it might be worth it. CF wheels, going with tires that are .1 or .2 inches thinner... that might get me to 42. That's another two grand. At that point, having already spent $2,700, I'm up to what... a $6,000 42 pound eMTB... and I could have bought a 42 pound eMTB off the shelf for...

About $6,000.

If I were you? I'd get a second bike, or build a second aluminum bike, definitely with a steel fork if you want a front hub drive. (And by the way, I *loved* the way this particular relatively light-weigh, low power front drive handled. Lean it way the hell over on a decreasing radius curve, hammer down on the pedals and punch the throttle, and you can exit the turn at astonishing speed. IMHO, if you go front hub, and go too heavy and powerful, that's when you have the weird handling issues.) Having two bikes is great, particularly when a friend comes to visit from out of town, when one of them is in the shop, etc.

Not what you want to hear, I know.

 

[JRA]

The leader in hub motor knowledge and components is at ebikes.ca. Justin at Grin has done more to develop hub motors than anyone for the DIY sector and is very free with his conclusions that he posts on his site. Check there to at least see what is perhaps possible?

 

[mschwett]

The leader in hub motor knowledge and components is at ebikes.ca. Justin at Grin has done more to develop hub motors than anyone for the DIY sector and is very free with his conclusions that he posts on his site. Check there to at least see what is perhaps possible?

they’re serving a very different market, with nothing even remotely as light as the X20 that @Katrina92 is trying to improve on.

The X20 weighs 1.39kg. a typical hub might weigh 200g, so the delta is more like 1200g. I am not aware of anything else that light, and while it’s bigger than a normal hub it’s smaller than either a disk rotor or most any modern cassette.

I don’t think there’s much need to make it smaller. lighter is always better but you can bet that if it was easy to do so, mahle would have done so given that the entire purpose of the x20 is for stealthy, light e-bikes.

to the OPl as mentioned in one of your other threads on this, mahle now makes a complete system around the X20, the XS, which incorporates everything into just the battery and motor. total weight is 2.5kg. you are not going to get it lighter than that with readily available current technology!

 

[mschwett]

also.… the addict e-ride is a very nice bike, maybe pick up a used one? it’s essentially the bike you are trying to build yourself.

notice that 1) you can’t even see the motor and 2) the chain stays are substantially reinforced relative to the regular addict. the additional dimension also integrates the electrical dropouts, so the power to the moto or is completely invisible and no wires need plugging in. this is as clean as it gets in 2025.

 

[Katrina92]

It is possible to buy a good road e-bike before the Trump tariffs take the full effect. Like a Scott eride or Trek Domane+
I'm afraid you are looking for a unicorn Katrina.For now, I’m not really looking to buy a whole new bike — I’m more interested in exploring what’s possible with a conversion kit. I don’t necessarily think what I’m hoping for is impossible or doesn’t exist.

For now, I’m not really looking to buy a whole new bike — I’m more interested in exploring what’s possible with a conversion kit. I don’t necessarily think what I’m hoping for is impossible or doesn’t exist.

 

[Katrina92]

In my limited experience, the smallest hub motor I've seen is the front-hub motor (about 2.2 lbs) on the Gocycle. I doubt Gocycle will sell you one separately, but used Gocycle motors occasionally show up on eBay, such as this one. According to Richard Thorpe, the former McLaren F1 engineer who owns Gocycle, his motor has an "equivalent" torque of 60Nm, which is enough for hill climbing. (I have a Gocycle GS, and I find I can climb moderate hills easily.) But the 60Nm torque figure is only good for Gocycle's 20-inch wheels. It is likely to be much lower when installed on your Scott Addict's larger wheels, but I don't know the conversion factor. Still, if you have strong legs, it's probably enough to assist hill climbing effectively.

The Gocycle G3 motor in the eBay listing is a 22-volt motor. The newer G4 models use 36 volts. In any case, you would have to match your electric system to the motor's electrical requirements. A battery, depending on size, would add several more pounds to the system, assuming you find a good way to attach it. The whole project is daunting.

Me, I would just buy a ready-made Scott Addict ebike!

I hadn’t looked into the Gocycle motor before — I’ll check it out, thanks!

 

[Katrina92]

The ebike market is hyper competitive, so I suspect that if a significantly smaller, lighter motor was easily achievable someone would be making it. There are very likely aspects your engineer friend isn't considering that make it a more difficult problem than you think.

I agree... there are definitely a lot of challenges. But I also wonder if it’s partly because the demand for small-torque motors is pretty niche in the overall ebike market.

 

[Katrina92]

How do you plan on getting the idea from your head to your bike?
That said.. Anything is possible if you want to pay for it.

My engineer buddy and I have been talking a lot about this stuff. Maybe we can dive into some technical details later.

 

[Katrina92]

Make the motor smaller and the internal stresses go up for the same torque, so longevity might become an issue. Also making it smaller means more issues with heat rejection.

My friend and I also talked about the issues of heat dissipation and durability. He said those can all be figured out.

 

[Katrina92]

My gut is telling me: Don't do this.

The thing about CF-- as I understand it-- is that it's extremely lightweight and durable and very safe when subjected to stress from the vectors it was designed for.

At its simplest level, this is why CF works for airplane hulls and not for submersibles, or why earlier CF bikes were often great on the road, but experienced structural problems when dropped, or after relatively trivial accidents. And just at a glance, the Addict's frame seems kinda exotic; it's not JUST a superlight CF frame, it's a hyperlight, with some very thin spots. I'd be nervous putting any kind of motor on any part of that bike. Sure, the motor on a hub drive is on the wheel, but...

Is the rear triangle designed for the added force? Would that really be the same as if you were just, like, twice as strong as you are?

I don't think anyone can give you a fair and accurate answer to that question.

Ebike motors are, basically, a little weird in terms of how they place stress on a materials. Rotational energy is always weird, right? I split a TSDZ2B motor housing-- and that's steel, I think-- doing almost nothing, just puttering along on level terrain in power assist level 1. It was an unusual build, but not radical, and built by someone who really knew what they were doing. The only things that were out of spec were that the bottom bracket was wider than recommended, and the chainline was not great. It's possible I just got a bad one, that the metal in the housing was no good, but I think a lot of little factors just stressed the motor housing in ways it wasn't designed for.

I also blew up a 250W front hub motor, probably from climbing hills that were too long and too steep. It was probably all of about 40nm, and I found it perfectly serviceable for grades of 20% or so in granny gear, plenty of power. Like the mid drive, it took about 700 miles to fail. It showed no signs of problems until the last 50 miles or so. My buddy weighs about 40 pounds more than I do and he's very strong; we were just asking it to do something it was not designed to do.

I will rebuild the mid-drive eventually, probably with a DM02. The front hub motor on the other bike cost me $450 or so, I had a lot of fun with it, money well spent, lots of good memories.

* * * * *

I mean, I get it. I'm a weight weenie, too. When I dropped 2.5 pounds from my 49 pound, 40nm, 250 watt eMTB, it was like a different bike. I took it up mountains, got it up to 40+ MPH downhill speeds, I can carry it down stairs more easily. I'm still crazy about it. Of course I'm tempted to drop more. I looked at a new fork today; at 44 pounds, my bike would be such a monster...

That would cost me about $1,500. And it might be worth it. CF wheels, going with tires that are .1 or .2 inches thinner... that might get me to 42. That's another two grand. At that point, having already spent $2,700, I'm up to what... a $6,000 42 pound eMTB... and I could have bought a 42 pound eMTB off the shelf for...

About $6,000.

If I were you? I'd get a second bike, or build a second aluminum bike, definitely with a steel fork if you want a front hub drive. (And by the way, I *loved* the way this particular relatively light-weigh, low power front drive handled. Lean it way the hell over on a decreasing radius curve, hammer down on the pedals and punch the throttle, and you can exit the turn at astonishing speed. IMHO, if you go front hub, and go too heavy and powerful, that's when you have the weird handling issues.) Having two bikes is great, particularly when a friend comes to visit from out of town, when one of them is in the shop, etc.

Not what you want to hear, I know.

Wow, you have a lot of experience with this! When you say ‘we were just asking it to do something it wasn’t designed for,’ apart from all the climbing and long hours, what else do you think was pushing it beyond its limits?

 

[Base]

For now, I’m not really looking to buy a whole new bike — I’m more interested in exploring what’s possible with a conversion kit. I don’t necessarily think what I’m hoping for is impossible or doesn’t exist.

Long time a go there was the Gruber Assist

https://www.bikeradar.com/reviews/bikes/electric-bikes/gruber-assist-kit-review

also know in some mechanical doping scandals in UCI competition

Femke Van den Driessche - Wikipedia

en.wikipedia.org