Most quiet e-bike motor?

watchdog_timer · Oct 9, 2019

I am making a pedal-powered generator and would like to make it as quiet as possible. I have used a Conhis direct-drive hub motor and a Bofeili mid-drive motor on past units. Both motors made more noise than I would like. Is there anything that would be more quiet, excluding those that can't be readily purchased individually (e.g., Bosch and Yamaha)? Among the BBS01, BSS02, TSDZ2, and Ling Bei mid-drives, which is the most quiet?

amberwolf · Oct 9, 2019

Anything with internal or external gears, chains, sprockets, etc will be louder than any DD hubmotor.

The DD hubmotor in your generator is probably not making much noise, but whatever you're driving it with probably does.

Alternately, if you are using a trapezoidal controller in regen braking mode to rectify the electricity from the generator-motor, it's probably making significant buzzing in the process, in the motor and possibly the controller.

Using a sinewave controller instead would probably be much quieter.

So would just using a 3-segment rectifier bridge on the motor phase wires directly.

markz · Oct 9, 2019

Convert to belts instead of using chains.

watchdog_timer · Oct 10, 2019

I am simply rectifying the output using a full-wave bridge rectifier, then buffering it with a 10 Ah LiFePO4 battery and some capacitors.

I've tried driving the hub motor with both a chain and a V-belt. Both were fairly loud.

It was worse when the axle nuts of the hub motor were tightened down, so I left it semi-loose in the dropouts and tightened it by moving the bottom bracket instead. Is it possible I just got a faulty motor?

E-HP · Oct 10, 2019

watchdog_timer said:
I am simply rectifying the output using a full-wave bridge rectifier, then buffering it with a 10 Ah LiFePO4 battery and some capacitors.

I've tried driving the hub motor with both a chain and a V-belt. Both were fairly loud.

It was worse when the axle nuts of the hub motor were tightened down, so I left it semi-loose in the dropouts and tightened it by moving the bottom bracket instead. Is it possible I just got a faulty motor?

Don't know if this is similar, but my DD motor howls under regen (sounds like poorly adjusted rim brakes) depending on the regen level, silent otherwise. Might be resonance, based on feedback when I posted the question. That would be harder to predict, but I'm guessing the motors that are silent with regen (some members replied saying their regen is silent) would be ones that would work for you.

amberwolf · Oct 10, 2019

watchdog_timer said:
I am simply rectifying the output using a full-wave bridge rectifier, then buffering it with a 10 Ah LiFePO4 battery and some capacitors.

Then that eliminates PWM/etc as a factor.

I've tried driving the hub motor with both a chain and a V-belt. Both were fairly loud.

When you hand spin the wheel the motor is in, no chain attached, no belt, does it still make the same noise? (yeah, it'll be hard to spin it fast enough, but it's worth a try)

If it does, then my guess is that motor design just happens to be noisy at the RPM it's being run at. If you run it at a significantly higher or lower RPM, is the noise the same?

It was worse when the axle nuts of the hub motor were tightened down, so I left it semi-loose in the dropouts and tightened it by moving the bottom bracket instead. Is it possible I just got a faulty motor?

Most likely your frame happens to be just the right design to be resonant with the motor under those conditions. That has happened to a few bike builders here on ES, and various solutions have been tried, with some success, including using sprayfoam to fill the bike frame tubing, adding rubber damping blocks between stays, wrapping the tubing with old innertubes, etc.

watchdog_timer · Oct 11, 2019

Thanks for your advice. Yes, the hub motor only made noise under load. When it wasn't connected to anything, it was much more quiet.

The motor was mounted in a custom mount I made from 2"x3/16" aluminum flat bar which clamped onto the end of a piece of 1"x2"x1/8" rectangular aluminum tube:

I tried a couple of other mounts made from the same-size flat bar, as well as some thick rubber between the motor mount and the rectangular tube, but it only marginally helped.

This was a 36V hub motor. A similar motor is available as a 24V motor. I'm wondering if the 24V motor operated at a higher RPM might solve the problem...?

E-HP · Oct 11, 2019

watchdog_timer said:
Thanks for your advice. Yes, the hub motor only made noise under load. When it wasn't connected to anything, it was much more quiet.

The motor was mounted in a custom mount I made from 2"x3/16" aluminum flat bar which clamped onto the end of a piece of 1"x2"x1/8" rectangular aluminum tube:

I tried a couple of other mounts made from the same-size flat bar, as well as some thick rubber between the motor mount and the rectangular tube, but it only marginally helped.

This was a 36V hub motor. A similar motor is available as a 24V motor. I'm wondering if the 24V motor operated at a higher RPM might solve the problem...?

Is there anywhere in the mount that you can modify and add some rubber pads to eliminate the resonance, like motor mounts on a car engine? It's hard to see how those plates are attached, but if they're welded, you might be able to make some extension plates and rubber mount those the existing ones.

watchdog_timer · Oct 11, 2019

Is there anywhere in the mount that you can modify and add some rubber pads to eliminate the resonance, like motor mounts on a car engine? It's hard to see how those plates are attached, but if they're welded, you might be able to make some extension plates and rubber mount those the existing ones.

Yes, I tried putting various thickness of rubber underneath the motor mount, before bolting it to the tubing. It made little difference.

Aquakitty · Oct 11, 2019

watchdog_timer said:
Is there anywhere in the mount that you can modify and add some rubber pads to eliminate the resonance, like motor mounts on a car engine? It's hard to see how those plates are attached, but if they're welded, you might be able to make some extension plates and rubber mount those the existing ones.

Click to expand...

Yes, I tried putting various thickness of rubber underneath the motor mount, before bolting it to the tubing. It made little difference.

I don't know the technicalities, but are you getting a sort of high-pitched whiny buzz noise? My DD motor does this, but the TSDZ2's do not. The TDSZ2 also has space in the motor casing where you could put heat pads or just sound dampening.

That's an interesting setup. What are you using it for? I made a similar looking bike desk from a semi-recumbent exercise bike.

watchdog_timer · Oct 11, 2019

I don't know the technicalities, but are you getting a sort of high-pitched whiny buzz noise? My DD motor does this, but the TSDZ2's do not. The TDSZ2 also has space in the motor casing where you could put heat pads or just sound dampening.

Yes, that was the sound I was getting. (I say 'was' because I'm currently using a Bofeili mid-drive motor.)

Thanks for letting me know about the TSDZ2. How does its noise level compare to a Bafang or Ling Bei mid-drive motor?

I prefer using a mid-drive motor in a project like this because it's simpler to install and maintain, more compact, and doesn't need a chainguard. I would consider using a DD again if I was certain it would be significantly more quiet. In theory, it should be, but now I'm beginning to wonder how often that is true...

That's an interesting setup. What are you using it for? I made a similar looking bike desk from a semi-recumbent exercise bike.

It was my home office desk. Here's my current one:

It has six 12V DC power outlets plus a fan, all individually monitored and controllable from a web interface run off an Orange Pi mounted on the back. It powers my computer, monitor, printer, modem, and a AA/AAA charger we use in LED lights around our house. It also recharges the phones in our house. I use it 4-5 hours a day while working on my computer and generate between 200 to 250 Whr a day.

I've been asked to make something similar for someone else in town, but it looks like Bofeili is out of business. That's what prompted my original question.

I'd be curious to hear more about the bike desk you made and how you're use it.

qwerkus · Oct 12, 2019

Love the idea, but I think it would be a whole lot more efficient if you put a flywheel in the system... This would even out the generators power input, and make it easier to setup a speed at which it runs silent. DD is the way to go; IIRC there is a small 9c hub that would be just perfect.

watchdog_timer · Oct 13, 2019

Love the idea, but I think it would be a whole lot more efficient if you put a flywheel in the system... This would even out the generators power input, and make it easier to setup a speed at which it runs silent.

I don't think you gain much in efficiency with a flywheel.

I generate about 45-60 W, depending on my load, and pedal at a cadence of 80 RPM while I work. The power output of people using a (non-electricity-generating) bike desk in a 2014 study averaged 59-86 W, depending on the resistance level, pedaling at the same cadence.

If those study participants were driving a generator with 80% efficiency (typical of most bike motors), their power output would be 47-69 W. Not that much different.

But you're right--one advantage of using a hub motor is you could change the speed at which it turns for a given pedaling cadence. That would allow you to adjust it to whatever is most comfortable cadence for the user, as well as possibly improve the efficiency.

The downside is the resulting desk would be heavier and larger, making it more difficult to move, and require making an enclosure over the chain and hub to keep cables from being caught in them.

Aquakitty · Oct 13, 2019

That's really cool. I just made mine using one of these: https://www.amazon.ca/Exerpeutic-1110-300SR-Foldable-Recumbent/dp/B004O6TXVS/ref=sr_1_4?dchild=1&keywords=Exerpeutic&psc=1&qid=1570999764&sr=8-4 and an old spring-loaded CRT monitor arm which happens to fit my laptop nicely. Not very exciting. I mostly use mine for working and Zwift cheating.

I really like your seat. My bike desperately needs a different seat because it puts you too far forward which brutalizes the butt. What did you make the seat base from?

billvon · Oct 13, 2019

watchdog_timer said:
I am simply rectifying the output using a full-wave bridge rectifier

That's your problem. The rectifiers provide a square wave "drive" (load) and cause harmonic noise. The noise is then amplified by the sides of the motor.

watchdog_timer · Oct 14, 2019

Aquakitty said:
That's really cool. I just made mine using one of these: https://www.amazon.ca/Exerpeutic-1110-3 ... 764&sr=8-4 and an old spring-loaded CRT monitor arm which happens to fit my laptop nicely. Not very exciting. I mostly use mine for working and Zwift cheating.

I really like your seat. My bike desperately needs a different seat because it puts you too far forward which brutalizes the butt. What did you make the seat base from?

My mother-in-law has one of those, too. Are you generating electricity with it, or just use it for exercise?

The seat base is 1-1/2" of regrind foam rubber topped with 1/2" of medium-density foam, which is wrapped in batting and mounted on a 1/2" plywood base. The seat is covered with black upholstery vinyl. (In keeping with the spirit of the project, I sewed the cover using a treadle-powered sewing machine I bought at an auction 30 years ago

.)

watchdog_timer · Oct 14, 2019

billvon said:
That's your problem. The rectifiers provide a square wave "drive" (load) and cause harmonic noise. The noise is then amplified by the sides of the motor.

I'm a bit surprised the sides of my motor would amplify the harmonics that much, because they aren't that large. I was using a small DD 250 W mini hub motor that's about 145 mm in diameter, not one of the larger DD motors thatseem more popular on this forum.

Here's the rectifier portion of my schematic:

What would you recommend doing instead?

billvon · Oct 14, 2019

watchdog_timer said:
I'm a bit surprised the sides of my motor would amplify the harmonics that much, because they aren't that large.

Right, but keep in mind that any membrane (even the side of a motor) is going to have a frequency it works best at. So you are hearing the frequencies that the structure is good at creating.

What would you recommend doing instead?

That's not an easy problem to solve. One thing that might work is to add a boost converter with a largish inductor after the bridge rectifier, so the current drawn is more sinusoidal than square wavey. But the voltages/motor constant have to be right for that to work. Another way that _might_ work is to use a good motor controller that implements regen, then use regen to capture the energy. Most good controllers use a sort of boost converter (using the inductance of the winding) to get energy out of the motor.

watchdog_timer · Oct 14, 2019

billvon said:
That's not an easy problem to solve. One thing that might work is to add a boost converter with a largish inductor after the bridge rectifier, so the current drawn is more sinusoidal than square wavey. But the voltages/motor constant have to be right for that to work. Another way that _might_ work is to use a good motor controller that implements regen, then use regen to capture the energy. Most good controllers use a sort of boost converter (using the inductance of the winding) to get energy out of the motor.

Let me see if I understand what's happening.

Each of the three phases of the generator produces a sine wave voltage. When connected to a full-wave rectifier, current will flow only when the phase voltage exceeds the forward voltage of the rectifier diodes. When the phase voltage is less than the rectifier diode's forward voltage, no current will flow.

However, with a 12 V battery wired in parallel with the rectifier, the phase voltage now has to exceed the diodes' forward voltage drop plus the battery voltage before it will conduct current, since the diode's cathodes are at +12 V. This prevents current flow during a longer period of the generator's period, causing an output more like a square wave than a sine wave.

And, since a DD hub motor turns at a relatively low speed, the frequency of this square wave is closer to the range of human hearing and/or the natural harmonics of the case, causing more noise.

By using a boost converter after the bridge rectifier, you're essentially lowering the forward voltage needed to get current flow through the rectifier diodes and thereby increasing the proportion of the generator's period during which current will flow.

Do I have that correct?

fechter · Oct 14, 2019

watchdog_timer said:
Do I have that correct?

Yep.

A dc-dc converter will have some losses though, so making it quieter will have a drawback. Pole count in the motor will make a difference too, but mainly you want to avoid any resonance in the parts.

billvon · Oct 14, 2019

watchdog_timer said:
By using a boost converter after the bridge rectifier, you're essentially lowering the forward voltage needed to get current flow through the rectifier diodes and thereby increasing the proportion of the generator's period during which current will flow.

Yes. And I would add that since a boost converter has the characteristic that it is a current sink when running open loop, current will flow during almost all of the cycle.

watchdog_timer · Oct 17, 2019

billvon said:
Yes. And I would add that since a boost converter has the characteristic that it is a current sink when running open loop, current will flow during almost all of the cycle.

That makes sense.

Thanks for all your help. I have learned a lot from this discussion.

Most quiet e-bike motor?

1 mW

Administrator

100 TW

1 mW

10 GW

Administrator

1 mW

10 GW

1 mW

100 W

1 mW

10 kW

1 mW

100 W

1 MW

1 mW

1 mW

1 MW

1 mW

Administrator

1 MW

1 mW

Similar threads