Why shouldn't i build a Neumotor RC + planetary gear reduction drive?

neptronix

neptronix

Administrator
Staff member
Joined
Jun 15, 2010
Messages
20,240
Location
Utah, USA
I'm looking at these parts and wondering why someone hasn't done this before. It looks like a slam dunk

Take a neumotor 3210 ( 2kw cont rated )
NeuMotors 3200 Series BLDC Motors: 1,000 to 4,000 watt class - NeuMotors Brushless Motors

Mated to a 1lbs 6.7:1 reduction ( P42 )
Gearboxes - NeuMotors Brushless Motors

Using a 3210 with the 165kv winding, the RPMs work out like..
52v x 165kv * 0.8 = 6864rpm / 6.7 = 1024 RPM

To gear down to 400rpm from there only requires a 2.56:1 reduction ratio which is easy with a rear disc sprocket..
You could run higher kV probably to extract more power..

Finding a 8mm sprocket for the motor is no big deal an you can have a number of companies make you a disc sprocket that bolts onto the brake.
The only remotely hard part is mounting the assembly, but it doesn't look hard to work with.

Results in 3lbs of motor + gearing + ~3lbs of sprocket, chain, and mount... so 6lbs for 2kw of power.. kinda blows away even the best mid drive in power density.. why shouldn't i do it? too loud?
 
Motor plus gearbox looks like it might be sort of long, but I didn't look up the dimensions. It will probably sound like an electric drill.
 
Would the planetary reduction be responsible for the sound itself if i were using a FOC controller?
I've sent them an email to ask about the noise factor.
 
Last edited:
neptronix said:
Would the planetary reduction be responsible for the sound itself if i were using a FOC controller?
Click to expand...
Probably, if they're using straight-cut gears. That's what makes the awful racket inside a GMAC in a small wheel like the 20" I have, when I'm riding at my 20mph cruise speed.

The quieter geared motors use helical gears, but those then need thrust bearings to counter the sideways forces, increasing the thickness of the gearing assembly.

But all the gearboxes will make *some* noise, and the faster the motor spins the more noise there will be from that, too.
 
20mph and a gmac ( really designed for 26" and above ) is, i imagine, a huge racket. Are you using a FOC/sine controller; therefore you're mostly hearing gear noise?

Supposedly, these neumotors and their transmissions are high quality units, basically one rung below astro motor quality, so i'm willing to believe that maybe the gearing isn't more loud than average at least.

I could buy an very big one and run it as a slovenly 4000rpm to reduce the noise.. however.. a MAC is spinning at around 2000RPM and if it makes enough racket already.. yeah.. we may need to rethink the drive unit.

I just thought man, this is a quick shot to a really good rear RC drive, so little fabrication needed.
 
I am skeptical about the torque handling capabilities of a sub-2 ounce gearbox. Sounds like it would have just enough beef to turn against, like maybe... air?
 
Agree with Chalo. The 32xx series motors are not to be combined with the p42 reduction (see chart).
The P62 is suitable for the 32 series motors and might have a chance to survive in an ebike application. But is longer and heavier.
 
The website contradicts itself on that.

NeuMotors 3200 Series BLDC Motors: 1,000 to 4,000 watt class - NeuMotors Brushless Motors
May be mated with P42 or P62 gearboxes for very high power-to-weight drives.
Click to expand...

Gearboxes - NeuMotors Brushless Motors
P42 drive is best suited for systems in the 3,000 – 5,000 watt range turning props up to 30 inches in diameter.
P62 is suitable for well over 10,000 watts. The 12mm shaft easily handles props over 30 inches in diameter.
Click to expand...

Using a P42 with 3kw peak power, 1.5kw cont doesn't sound like too much of a stretch.
 
They mention turning a prop against, indeed, air. Steady torque requirements for an aircraft are nothing compared to transient load a gearbox might encounter with a bicycle over a bumpy road, I'd assume.

If it's not too expensive and parts are available, though, maybe rebuilding it every now-and-then might not come out too bad, since it's easier to reach than a geared hub anyway.
 
Called them.

They say the reductions are not designed with noise minimization in mind.
They say people use these reduction drives with pumps and other types of applications and the ratings are based on industrial + air prop use.
They seemed offended that i referred to it as a RC motor. ( oops )
They say to minimize initial torque and maximize RPM and watch the heat on the motor ( makes sense )
Output torque is rated in the 10's of newton meters, input torque is rated in the 1's of newton meters, so this explains the gear reduction's size.

The rational thing to do for the longevity of the gearbox is to tune the motor with a more gasoline engine like torque curve IE phase to battery current ratio around 1:1, initial torque being kind of weak.

Some thoughts on reducing the noise..
Wrapping the reduction with dynamat or similar materials to consume the sound energy is said to be effective. Since we are reducing the heat shedding ability of the transmission we must make up for it though..
I need to look at the maximum amount of noise reduction you could get from adding mass before saying yes, this is a good idea.


We could go a half RPM route and just throw away most of the power potential with a 4410 and chose the much larger p62 transmission. Then i get the ability to goose this motor.

1741885293011.png

Gearing math:
4410 @ 78kv
52 x 78 * 0.8 = 3244.8 RPM loaded
/ 6.7 = 484RPM = basically wheel speed

Extra ~600$


Or go in the opposite direction and build a small scale model before buying the big one

IE
2207 motor $362 + p42 gearbox $295
= $657 drive unit

2207 should have a stubby form factor instead of a long one.
With 276kv winding we end up at 11481.6rpm loaded
/ 6.7 = 1713rpm
Now we need a 4.2:1 minimum reduction to the rear wheel.. on the difficult end of doable
..but this would have bad hill climbing power. on a bike with a 29" wheel.. we probably want more like 7:1, which is too high..

However if we use a 20" wheel ( my recumbent ), then it's no problem... so it would work for the poor man's dual stage RC drive.

Should sound like a blender, but it's a $657 experiment instead of a $1600 experiment.

I think i need to spend some time researching the degree to which you can externally consume gear motor noise at high frequency.. if the maximum noise reduction achievable is low then we might want to bail.

Possible that you could wrap half a pound of dynamat around the transmission.. with such good power density, adding half a pound is ok. I probably need to add 0.25lb of aluminum heatsink to the motor too, to make up for the heat that the transmission cannot expel now. Or i need a bitchin heatsink on the other side of the transmission IDK
 
Last edited:
Deepseek says:

2. Lubrication Optimization​

  • High-Performance Lubricants: Use synthetic oils with additives to reduce friction and dampen noise. Adjust viscosity to balance damping and efficiency.
  • Efficient Lubrication Methods: Implement oil jets or mist systems to ensure consistent coverage at high RPMs, avoiding churning losses.

3. Vibration Damping and Isolation​

  • Isolation Mounts: Use elastomeric or active damping mounts to decouple the transmission from the structure.
  • Housing Stiffening and Damping: Reinforce the housing with ribs or use composite/metal materials with high damping capacity. Apply constrained-layer damping treatments to surfaces.

4. Acoustic Enclosures and Absorption​

  • Sound-Absorbing Enclosures: Design enclosures with acoustic foam or mass-loaded vinyl, ensuring adequate ventilation to manage heat.
  • Barrier Materials: Line housing interiors with noise-reducing materials like rubber or thermoplastic layers.

The idea of using rubber somewhere in the mounts sounds about right, this is critical on a car
 
Possibly Dynomat (and more layers of Dynomat) may not be the best solution. How about a measurement of the offending frequency range and a study to identify which material is best for damping it?
 
I called dynamat and a few other companies.
No good answer about high frequency sounds, which i assume there are some of in this fast spinning transmission.

May just have to buy and measure, yeah.
 
Remember that acoustic dampening materials are also great thermal insulators. The gearbox is going to not only make heat, it's going to be heated by the motor attached to it.

I don't know how hot it might get, or if that might affect it's materials or lubrication, but it might be something to check on.

When I asked Grin Tech about the GMAC gear noise, they said that there can be a little noise reduction by packing a gearbox with grease, but it likely wouldn't help much (and it didn't). Based on that, oil-filling all the airspace in a gearbox might help too, but also not much.

From the various gearboxes I've seen inside of on stuff I've opened up, I'd guess the best way to reduce gear noise is use ones with more contact surface to transfer the load (which usually means helical vs straight cut (I think theyre also called spur?)).

Higher speed is going to make more noise either way.



The GMAC noise was with the Phaserunner v6, so it's the gears.

The two different kinds of Fusin geared hubs I had were also noisy gears, but not as bad because they were in bigger (26") wheels. I don't recall how noisy the Ezee was as taht was only used on my brother's trike and he only rode it the once, IIRC. All of these are straight-cut gears.
 
Yup that's why i gotta add extra heatsinks and overspec the motor - the transmission is no longer a heat bank.
That's why i thought of a 32xx first.
 
I have to say those motors look really interesting when you look at their weight... 1:7 reduction on the final drive is definitely doable, with smaller bike chains you can run something like 10:70 no problem. On a motorbike I'd expect two stage gearbox before the final drive I guess.
 
Another possible issue is shock loading. Those gearboxes are really made for big propellers that generally won't see any shock loading. On a bike, if you are giving full power and go over a big bump, the motor can spin up while the wheel is off the ground and deliver a huge shock load when the tire regains traction. This is what can break things. Some kind of shock buffer (like motorcycles) would reduce the chances of breakage.
 
Oh, very good point! If you use a motorcycle hub, you can get one from a smallish road bike that will come with a rubber damped (cush drive) rear sprocket, like so:

1741964704486.png

The unfortunate consequence is that they will often add a bit of slop, especially as they wear out.

KTM puts a similar mechanism in their new clutches, but that would be harder to fit near a gearbox, i think.
 
Not to protect a gearbox but to protect the freehub I built a chain guide that goes on the driven side and is sprung with some damping to act similar to a cush drive, I called it a cush guide.

I still kinda think a belt reduction is a better choice over a gearbox, would be cool if there was a standard belt reducer box with input and output shafts for such a purpose. I guess there are kinda things like that used on e-skateboards.

I have at this point a lot of experience in sound damping, although in fairly niche cases. First you want to isolate the gearbox from anything it can transfer sound to, which will be a pain from the motor if it wasn't designed that way in the first place. Higher frequencies are easier to damp and most materials aren't exactly dramatically better or worse at certain frequencies based on their properties but moreso their construction, thickness for instance but that's a pretty big generalization. Something with viscoelastic properties may be the best option, that is you can't use anything porous, something slows sound propagation by making the sound go between high and low impedance materials over and over again like in a closed cell foam or in the case of an open cell foam by doing a little of that but mostly causing very high air flow resistance inside the material that eats acoustic energy. Those materials will be very thermally insulative. But a visco elastic material will be solid so still poor but some thermal transfer and damp sound by converting it to heat. So like sorbothane is the famous one and butyl rubber is the other which is what constrained layer dampers (CLD) like dyna mat use. So a thinish layer of one of those with a heatsink on the other side, the heatsink will act as the mass in a CLD system. May even be possible to mix a thermally conductive material like copper powder into the material if heat was an issue.

But before doing any of that I would would just wrap the gearbox up and see if it even does overheat.

I've done a fair amount of research into sound damping lubricants and boy there is not a whole lot of information on it, it's very complicated.
 
fechter said:
Another possible issue is shock loading. Those gearboxes are really made for big propellers that generally won't see any shock loading. On a bike, if you are giving full power and go over a big bump, the motor can spin up while the wheel is off the ground and deliver a huge shock load when the tire regains traction. This is what can break things. Some kind of shock buffer (like motorcycles) would reduce the chances of breakage.
Click to expand...

Damn, i didn't think about that since this i my first drive.
Would a freewheel fix this?
 
bananu7 said:
Oh, very good point! If you use a motorcycle hub
Click to expand...

Interesting idea, but can't do that, the bike needs to be pedal-able :/

@scianiac thanks for the feedback, that kinda lines up with my research so far, i know a bit about acoustics from being a sound engineer. But when it comes to dampening motor sounds...
 
Last edited:
neptronix said:
Damn, i didn't think about that since this i my first drive.
Would a freewheel fix this?
Click to expand...
A freewheel would help partially (in one direction), but if the wheel was jerked backwards (which will also happen during suspension movement if you don't build a constant chain tension system) it'd still pull on the gears with a lot of force.
neptronix said:
Interesting idea, but can't do that, the bike needs to be pedal-able :/
Click to expand...
Well then...
DSC_3092-lncz8sno.jpg

(credit: Texugo.pl)
I've been eyeing this one for a while. It's a really sweet machine, i hope they get the price down at some point and make larger series runs. Coming back to it I see more and more details that I believe were very good strategic choices (e.g. tiny chain and massive reduction on the motor - sounds familiar?). What's not clear from the pic is that the bike uses a planetary gearbox hub rigidly mounted in the middle as the pedal gearbox, so that you can keep up with the motor at all speed ranges.
 
Last edited:
Let's say we want to use a freewheel and create a mid drive..
You could run 36v on this thing through the reducer and the loaded RPM would be 257rpm... 3:1 stage reduction to 85rpm crank speed. This motor is about 5lbs.

1742071482425.png

In this case we are probably getting 2000w continuous since we're running a super low RPM.
We could probably get a 4410 wound to the same KV and have a 1500w... that motor would be 4lbs but.. that's still a pretty good power to weight ratio versus something like a BSSHD!
 
mm no way would i buy that bike instead of build an addon mid drive i could use on anything..
I want a normal bike that just happens to have a motor on it.

What if i uses a freewheel and have a hardtail bike?

Have i defeated the shock loading thing well enough?

Do i need to go mid drive to avoid adding other complex systems?
 
neptronix said:
mm no way would i buy that bike instead of build an addon mid drive i could use on anything..
I want a normal bike that just happens to have a motor on it.

What if i uses a freewheel and have a hardtail bike?

Have i defeated the shock loading thing well enough?

Do i need to go mid drive to avoid adding other complex systems?
Click to expand...
I wasn't implying you buy one, it was just to illustrate the drive idea. If you combine the motor and motor reduction with an intermediate pedal gearbox, you can still achieve a transferrable, modular system. I mean, kind of like CYC, but I guess building your own is the point.

Hard tail might help a bit but it's a wasted opportunity to not have rear suspension, and the wheel will still likely cause high transient loads like I mentioned a couple posts above. You just need a gearbox that's essentially "overbuilt" from a perspective of an airplane gearbox.
 
Last edited:
You must log in or register to reply here.
Back
Top