Would a 250W mid drive be worthwhile for me - a 'reasonably' fit cyclist?

[JES2020]

I think with a hub motor you have limits, but for different reasons. With a mid drive, the mid pulls like hell on the chain, which pulls the chain and seatstays with it unless they are strong enough to resist. Duck soup for an emtb rear triangle. A road bike with Columbus tubing? Too big of an ask as we described above.

A hub motor is a different animal. It powers the bike thru the axle. That means in addition to it being a single-speed assist independent of the drivetrain (with the associated implications on hill climbing) it is putting rotational force on the dropouts in order for it to be able to spin forward. Just like the lightweight stays were not designed for a kilowatt of oomph, neither are the dropouts expecting to get enormous twisting forces applied to them, and if they are steel they will spread so they are no longer parallel, at which point the frame is literally ruined. If its an alloy frame, then the dropouts just snap instead of bending/spreading because aluminum.

The above is the worst case scenario. You counteract that unintended force with something known as a torque arm. And there are good quality ones and crappy ones. But either way a donor ebike is a more suitable donor if it has beefy parts on it. A proper old school lightweight road bike is not that, and as such you should limit it to a lower-power solution. And still do torque arms because there is (unless you have a frame building shop and jigs+brazing tools+spare dropouts) *no* coming back from dropouts that have spread past parallel and let the axle spin.

So short answer is yes I would go low power on a hub motor too if the donor is a quality road bike. Given my druthers I would do a dialed-back BBS02. Like I said I know of two hard-core cyclists who have done it and they were in love when they had it all dialed in.

Those torque arms are made of stainless steel ..enough said!

 

[PedalUma]

If 2000w is marginal on such a heavy bike (with hub-motor), it must be limiting for steep climbing / sustained climbing even more than it limits top speed or range(?).

I regularly do 350 to 750W conversions of older steel frame bikes with a mid-drive. @m@Robertson is correct. Just dial back the amps. I do not like lower powered hub-drives or the balance of them. There is also the issue of rotational weight with them and the motor only having one speed. My go to bike is a lugged steel three speed mid-drive with 350W. Zoom in on the motor between the pedals. The 'water bottle' is the battery. Two Dutch aluminum Lekker bikes just came in with Nexus 8 speeds. https://www.lekkerbikes.com/us/product/amsterdam-8-speed-commuter-bike/ But nothing rides like a steel frame.

 

[m@Robertson]

Those torque arms are made of stainless steel ..enough said!

"those" torque arms?

 

[JES2020]

"those" torque arms?

Dem torque arms?

 

[m@Robertson]

Dem torque arms?

Dose torque arms. That we never said anything specific about? I mentioned there were kaka ones and good ones and never took it further.

 

[stw]

I think with a hub motor you have limits, but for different reasons. With a mid drive, the mid pulls like hell on the chain, which pulls the chain and seatstays with it unless they are strong enough to resist. Duck soup for an emtb rear triangle. A road bike with Columbus tubing? Too big of an ask as we described above.

A hub motor is a different animal. It powers the bike thru the axle. That means in addition to it being a single-speed assist independent of the drivetrain (with the associated implications on hill climbing) it is putting rotational force on the dropouts in order for it to be able to spin forward. Just like the lightweight stays were not designed for a kilowatt of oomph, neither are the dropouts expecting to get enormous twisting forces applied to them, and if they are steel they will spread so they are no longer parallel, at which point the frame is literally ruined. If its an alloy frame, then the dropouts just snap instead of bending/spreading because aluminum.

The above is the worst case scenario. You counteract that unintended force with something known as a torque arm. And there are good quality ones and crappy ones. But either way a donor ebike is a more suitable donor if it has beefy parts on it. A proper old school lightweight road bike is not that, and as such you should limit it to a lower-power solution. And still do torque arms because there is (unless you have a frame building shop and jigs+brazing tools+spare dropouts) *no* coming back from dropouts that have spread past parallel and let the axle spin.

So short answer is yes I would go low power on a hub motor too if the donor is a quality road bike. Given my druthers I would do a dialed-back BBS02. Like I said I know of two hard-core cyclists who have done it and they were in love when they had it all dialed in.

Will soon be building a geared-hub conversion (GMAC) so I'll know more about what to expect for climbing and sustained climbing on a hub drive. I'll also know more about how it feels to use PAS with hub-drive since my only ebike experience so far is a torque-sensing Bosch mid-drive. The future mid-drive conversion I've contemplated is for a 26x4" tire bike that would do longer climbs on mountain roads (not eMTB) which is where mid-drives shine.

My hub drive conversion will use the Grin Tech torque arm for the reasons you mention.

To the original post question, my experience with a 250w mid-drive (Bosch) suggests it would offer a lot for a fit cyclist who likes to pedal.

 

[m@Robertson]

Will soon be building a geared-hub conversion (GMAC) so I'll know more about what to expect for climbing and sustained climbing on a hub drive. I'll also know more about how it feels to use PAS with hub-drive since my only ebike experience so far is a torque-sensing Bosch mid-drive.

Just bear in mind the pedal-assist experience on a hub motor can vary *dramatically* based on what you do with the controller and the bike's gearing. Your typical stock ebike is geared too low so that makes it easy to 'ghost pedal' whether you want to or not, and your typical factory settings of pedal assist over power the motor out of the gate rather than starting out low and ratcheting up. Gear the bike so it can be pedaled just a little faster than the motor can take it - and dial back the pedal assist (you can always use the throttle if you need an unfiltered boost) and suddenly you have a whole different kind of ride.

If you are using a GMAC then I assume the controller is one of the Phaserunner flavors, with a CA? Should give you all the flexibility you could possibly ever want... you just have to dig into it.

 

[PedalUma]

@m@Robertson, Here is the bike that needs a new controller. The original is out of stock and has not been made for a decade. I could use some help. I sent a private message but cannot post photos in it. Thank you.

 

[stw]

Just bear in mind the pedal-assist experience on a hub motor can vary *dramatically* based on what you do with the controller and the bike's gearing. Your typical stock ebike is geared too low so that makes it easy to 'ghost pedal' whether you want to or not, and your typical factory settings of pedal assist over power the motor out of the gate rather than starting out low and ratcheting up. Gear the bike so it can be pedaled just a little faster than the motor can take it - and dial back the pedal assist (you can always use the throttle if you need an unfiltered boost) and suddenly you have a whole different kind of ride.

If you are using a GMAC then I assume the controller is one of the Phaserunner flavors, with a CA? Should give you all the flexibility you could possibly ever want... you just have to dig into it.

thanks for the info. yes, using one of the Phaserunner controllers and CA3. Interesting to learn how you have to set up the controller to give you the chance to apply some power to the pedals. If I'm understanding it correctly, it's easy for a strong hub motor to be set up out of the box to give you enough power/speed in PAS settings to outrun your pedaling so you dial it back and program the PAS settings to give you less/right amount so you still feel yourself pedaling (not ghosting). In other words, you program it to feel a bit like a torque sensing response where it responds because you apply power. I'll be learning the CA3 programming for that.
I've chosen higher gearing than I'd use on a pushbike so I'm headed in the right direction I think, but I'm not going for quite the mile-eater your bikes are ;-)

 

[m@Robertson]

thanks for the info. yes, using one of the Phaserunner controllers and CA3. Interesting to learn how you have to set up the controller to give you the chance to apply some power to the pedals. If I'm understanding it correctly, it's easy for a strong hub motor to be set up out of the box to give you enough power/speed in PAS settings to outrun your pedaling so you dial it back and program the PAS settings to give you less/right amount so you still feel yourself pedaling (not ghosting). In other words, you program it to feel a bit like a torque sensing response where it responds because you apply power. I'll be learning the CA3 programming for that.
I've chosen higher gearing than I'd use on a pushbike so I'm headed in the right direction I think, but I'm not going for quite the mile-eater your bikes are ;-)

CA's are a whole different world. Tons of flexibility but Grin only recently published a manual after years of it being on the market. Its always been a sort of product you had to puzzle out for yourself and doing that with some community help was part of your rite of passage into the DIY temple.

I have personally stayed away from them. Partly because I'm not enamored of the display quality, but mostly because I have gotten what I have needed from KT controllers and I have never felt a need to branch out. If T Tangent Motors had not gone dark right about at the time I was going to buy one, I would have bitten the bullet for my Guerrilla Gravity Smash build. But I had to go with the Cyc and an ASI BAC instead... totally different world and frankly it hasn't impressed me. For hands-on, granular control I think you are probably going the best route.

 

[m@Robertson]

@m@Robertson, Here is the bike that needs a new controller. The original is out of stock and has not been made for a decade. I could use some help. I sent a private message but cannot post photos in it. Thank you.

Boy I wish I could be of some use here, but for my hub motors, I have always stuck with KT controllers and have never branched out into anything else, other than what I just posted above. My gut feeling is your most likely bet is a Grin controller. But this is something where I have willfully stayed ignorant by picking one lane and staying in it. I can say that from what I have seen behind the scenes from folks who are programming the ASI controllers (not speaking of the Cyc guys here) that there is a ton of nuance and possibility there, but you really need to commit a lot of time and effort to learning that platform to get results. I am thinking the best option for a reasonably swift solution - absent finding some basic controller that can be spliced in - is a Grin option. Maybe talk to them via email and see what they say? They have lots of time in the field and may know something less nebulous than what I am saying here.