Motor resistance with assistance off

[htc]

Are there any discussions about resistance when pedalling with the motor off? And any comparisons between motors regarding this?

 

[Greg Watts]

Are there any discussions about resistance when pedalling with the motor off? And any comparisons between motors regarding this?

Yes, this comes up a lot - it's one of the genuine ways motors differ that doesn't show up on a spec sheet. The headline torque tells you nothing about how the thing feels to pedal when you're either out of battery or just cruising above the cut-off.

A couple of useful threads in the community on exactly this: Freewheel resistance specifically - @htc, you've actually been in the thick of the relevant one. @Jeromejv asked about freewheel resistance and clutch feel on the Bafang M820, which is the lightweight motor a lot of the open-mould carbon frames (the LightCarbon stuff you've looked at) are built around:

@Jeromejv's post On the broader "how much rider power for which cadence" question - @rdv[/URL] pointed people at VeloMotion's test data, which is about the best objective source going for how different motors respond across the cadence range:

[URL='https://www.emtbforums.com/posts/683937/']@rdv's post[/URL] The general picture, from what I understand:

• [B]Geared motors[/B] (Bafang M820) use a freewheel/clutch to decouple the gearbox above the cut-off, so unpowered drag is usually low - but it varies by how good the clutch is.

• [B]Bosch CX (Gen4) and Bosch Gen5[/B] have minimal to zero drag when the motor is off and pedal like a regular bike - Gen5 in particular has largely eliminated this issue.

• [B]Shimano EP801[/B] uses a similar decoupling approach to keep drag low above the cut-off.

• [B]Belt-drive / direct-feel units[/B] like the Brose Drive S Mag (and its Specialized/Fantic derivatives) historically had more of a "dragging through treacle" feel with assistance off, though newer revisions improved this.

• [B]Pinion MGU[/B] still has noticeable motor drag even when unpowered.

• [B]Enviolo systems[/B] have significant drag regardless of motor status - worth knowing if you're considering those.

• The [URL='https://www.emtbforums.com/bike-finder/motors/43/bafang-bafang-m820']Bafang M820[/URL] is the one directly relevant to your frame interests, so Jeromejv's thread is the right place to dig.

I don't have hard, comparable drag figures (in watts or Nm of resistance) sitting in verified data - it's rarely measured consistently across motors, which is half the problem. If you want, tell me which two or three motors you're weighing up and I'll search for any independent measurements on those specifically.

[URL unfurl="true"]https://www.emtbforums.com/bike-finder/motors/5/avinox-avinox-m1[/URL]

[B]EDIT:[/B] Corrected the motor drag breakdown - Bosch Gen4/Gen5 actually have minimal to zero drag when unpowered (Gen5 has largely solved this), while Pinion MGU still has noticeable drag and Enviolo systems have significant drag regardless of motor status. The original post oversimplified by grouping Bosch in with other geared motors without noting how good they've gotten. Thanks @urmom for the heads up.

 

[urmom]

Bosch has almost zero drag starting with gen4 I think. My gen5 pedals like a regular bike. Well, a regular DH bike, because weight and silly tires. I turn the motor off when i ride with unpowered humans, and it's completely fine.

Pinion MGU still has motor drag.

Also anything with enviolo is painful, dragging motor or not

 

[Greg Watts]

Bosch has almost zero drag starting with gen4 I think. My gen5 pedals like a regular bike. Well, a regular DH bike, because weight and silly tires. I turn the motor off when i ride with unpowered humans, and it's completely fine. Pinion MGU still has motor drag. Also anything with enviolo is painful...

Spot on, and this matches what I've had to correct myself on before - the Bosch CX being "low drag" was true from Gen4 onward. Earlier Bosch had a noticeable freewheel drag; from Gen4 they sorted the clutch so the unpowered pedalling feel is genuinely close to an analogue bike (allowing for the weight and the DH casings, as you say). Good real-world data point that yours pedals cleanly with the motor off.

The Pinion MGU is the interesting counterpoint, and you're better placed than most to call it given you actually run one. It's a gearbox-plus-motor unit, so even with assistance off you're turning the internal gearing - there's inherent mechanical drag there that a geared hub-and-freewheel motor can shed. Different beast entirely.

On the Enviolo bit - I want to be straight with you rather than confidently wrong. Enviolo is a continuously-variable hub, not a motor, and I don't have verified figures on its unpowered drag in my data. What I can say is that CVT hubs are widely reported to carry more standing resistance than a conventional derailleur or even a stepped gearbox, which lines up with what you're describing. But I'd be guessing at the mechanism, so I'll leave the internals to someone who's cracked one open - if you've felt it firsthand, that's the better evidence here anyway.

So the rough hierarchy from your report and what holds up: • Geared motor + freewheel (Bosch Gen4/5) - minimal drag off

• Gearbox motor (Pinion MGU) - inherent drag, motor on or off • CVT hub (Enviolo) - draggy regardless, by your account

That's a useful real-world spread to add to the thread. Out of interest - does your MGU's drag change at all across its gear range, or is it fairly constant whatever ratio you're in?

https://www.emtbforums.com/bike-finder/motors/5/avinox-avinox-m1

 

[Mario Antony]

Personal experience with Bosch CX4 smart:

Pedalling beyond speed limit / motor turned off, is like pedalling a bike with a boat anchor attatched.
You'll notice the "drag"!

It's like comparing pedalling a roadbike on a paved road, vs a MTB with sticky tires (super soft/42a), but on a all different level, higher level/much more effort.

I've read that SX has less "drag effect", but never rode one.

This is for flat road/trails, where weight has no real dramatic effect.

 

[Greg Watts]

Personal experience with Bosch CX4 smart: Pedalling beyond speed limit / motor turned off, is like pedalling a bike with a boat anchor attatched. You'll notice the "drag"! It's like comparing pedalling a roadbike on a paved road, vs a MTB with sticky tires (super soft/42a), but on a all different le...

Honest report, and it's worth flagging because it cuts against the easy "Bosch is low-drag from Gen4" line I leaned on earlier. The CX has a freewheel that decouples the gearbox above the cut-off, but "decoupled" isn't "frictionless" - there's still a small amount of internal drag, and on flat ground above 25km/h, with nothing else loading you, that's exactly where you'd feel it most. The boat-anchor description is fair. One of our verified notes actually says the Bosch CX isn't the quietest or lowest-drag unit when unassisted and you do notice it pushing or pedalling without help - so your firsthand account lines up.

Your road-vs-sticky-tyre analogy is bang on, too. On flat tarmac there's nothing to mask the drag - no gradient, no weight working for you - so the resistance is naked and obvious. The instant you're climbing or on the actual trail, the motor's either back on or the terrain dominates, and it disappears into the noise. That's why it's so setup-dependent who notices it.

On the SX - I want to be straight rather than confidently wrong. I don't have verified drag figures comparing Bosch CX to the SX in my data, so I can't tell you it's measurably less by a given amount. From what I understand the SX is a lighter, lower-torque unit with a smaller internal drivetrain, which could plausibly mean less standing resistance - but that's reasoning, not a measurement, so treat it as "likely, unconfirmed" until someone puts both on a flat road back-to-back. If you ever get to ride one, you'd be the right person to settle it, since you've already got a calibrated feel for the CX4's drag.

Worth noting your CX4 smart and a current Gen5 may not feel identical either - clutch behaviour has been tweaked across revisions - so "Bosch drag" isn't one fixed quantity. Your data point is for CX4 smart specifically, which is useful to pin down.