Neumotor 8057/75 middrive project, belt drive, adding halls, 10kW+

madin88

10 MW
Joined
May 27, 2013
Messages
3,120
Location
Austria
I would like to convert my Votec bike, which is currently powered by an MXUS 3k hub, to a middrive.

The main reason i want to give away the hub is the poor handling at uneven roads. The whole bike is shaking like hell if the road is bumpy and uneven. There are no issues with the fork or the damper, the only thing which might be not optimal in terms of suspension physics is the low weight of the frame.
It is about 25kg (frame with all parts + battery) fighting against 13-14kg of unsprung mass (motor + rear wheel). Aside from my personal experience this is also in theory anything but optimal.
When i am driving over potholes it comes through quite hard and rough, and the quick road irregularities are not filtered well. the wheel simply cannot follow the surface quick which leads to a very uncomfortable ride than.

My plan is to install the motor on a bracket which will be welded to the swingarm. The problem is there is limited space, so motors like the BigBlock's won't fit, and because i do not like to power it over the chainring on the pedals, units like BBSHD or Lightningrods are no option.

I am thinking of going with a belt drive with about 7:1 reduction (or a bit more even). Belt drive because of the possibility for regen :)
If the diameter of the motor is below 110mm, it should fit without touching the frame if the damper is full compressed, so i started to search for such motors.

first candidate:

- NeuMotors 8057/75

https://neumotors.cartloom.com/storefront/product/126891
http://www.leomotion.com/pi.php/LEO-L8057-075.html

It is a 16P motor and somewhere stands "18S16p" so i think it has 18 teeth (but not confirmed).
as the name says the size is 80mm dia x 57mm wide with a kV of 75 (optimal for my 20s battery).
the rpm range is 4000-8000

other specs:

Case Length 92mm
Diameter 99mm
Weight 2660g
Stator diameter 80mm
Stator length 57mm
Shaft Diameter 14mm
Mounting 6x 56xM6mm
Kv 75U/V
Internal Resistance: 24mOhm
Idle Current: 1.6A @ 10V
Limit (15s): 20'000W
poles: 16 (8 pairs)

At the moment the price is 399USD for the motor and about 60USD for the necessary front mount axle:
http://www.leomotion.com/shop/USER_ARTIKEL_HANDLING_AUFRUF.php?Kategorie_ID=1157&Ziel_ID=8566


126891_631d46309778d32f76d62068d3bf3758_orig.jpg


126891_79603b563f395a9b5f534a9decab3c42_orig.jpg


the 8057 is on the far right:

126891_fea43c66070162ac5201c5fdfef7aa0a_orig.jpg


things to find out for the NeuMotor:

- how is the cogging of 18S16p design (or torque ripple?)?
- would it be possible to install hall sensors?
 
more pics:

PC060967.JPG
 

Attachments

  • neu 1.jpg
    neu 1.jpg
    9 KB · Views: 13,217
  • neu2.jpg
    neu2.jpg
    4.2 KB · Views: 13,217
wineboyrider said:
Controller?

It would need hall sensors, than it should "work" with most e-bike controllers.
Things like back-EMF waveform or inductance are also important to know for choosing the optimal controller.

the 16pole design would require "only" 64000 ERPM for maximum specified RPM which is a big advantage compared to other large RC motors.
 
I don't own the motor yet, but i don't see a problem adding halls if they can be installed between the teeth. Need to find out if they can.
I asked Leomotion (NeuMotor dealer) about the coggin and motor data.

here ya go:

805775.png

the data is "calculated", whatever that means, but it looks promising.
The cogging is quite high they reckon which prevent ma a bit of buying one aside from the missing halls.

This motor should play in a different league than Hobbyking, Alienpower stuff or other china RC motors, and the price is still half of a Hacker Q100 motor.
Also the large 14mm axle with rear mount option makes it a fantastic choice for a mid-drive.
 
When I first got into R/C helicopters Neu motors was the top dog. I ran a 1917 1Y when all the outrunners were still baking themselves. It had a tramendous amount of cogging which never bothered me a bit on a helicopter, plane or car/truck. Also back when controllers didn't have a great slow start and it would spin the Heli almost a full turn which would wake you up in a moment. :shock: I had to send mine in to get repaired because the shaft was only epoxied in......came loose on a hard climb out once....never did it again.

I seem to remember when Steve Neu visited the r/c forums he/they would do custom work as far a winds and custome mechanical changes such as shafts and longer can/stators. Maybe they would add the halls for you?

4400 series looks good as well although has more slots.....has 10K continous and peaks of 20k watts.

Looking forward to your findings.

Tom
 
litespeed said:
4400 series looks good as well although has more slots.....has 10K continous and peaks of 20k watts.
yeah i have seen it, but i think the RPM is to high for a 1-stage reduction and it costs twice as much. Also the bearing looks much smaller compared to the L80xx series.
The design looks similar to Astroflight motors, but much larger and more powerful.

I came across the AWESOME eCalc torquecalc Motor Finder:
https://www.ecalc.ch/torquecalc.php

Choose a motor, type in Torque and RPM and you'll see how it performs.

There is even a Joby JM1, Hacker Q100 or A150 at choice.
 
I saw a video on this motor, 20 kw is pulled from thin air. About half that in real world peak power.

Iirc you do machining, and have done axles before? Those motors seems to relatively thin, could you double to power if you run two of those on one axle? One controller to each motor, they will sync automatically. Should give you double the fun :)
 
Hi Madin,

I really can't help with belt drives, but before you get into all the trouble of building one, have you tried a MAC on the Votec? I really love the MXUS 3k on my steel fatbike (pic attached). Road bump filtering with the 4 inch tyre is second to none. Even for light trails, the fatbike rides fantastic and you can barely feel the trail chatter as it plows thru everything. For commuting on a full suspension bike with regular tyre, I would definitely choose a lighter hubmotor like the MAC/BMS/BPM, unless you plan to ride a lot off road. You may lose the crude power that MXUS motor delivers, but will be rewarded by a more natural feeling ride, simplicity and reliability.

Viktor
 

Attachments

  • fatbike_s.jpg
    fatbike_s.jpg
    450.6 KB · Views: 12,991
macribs said:
I saw a video on this motor, 20 kw is pulled from thin air. About half that in real world peak power.

Iirc you do machining, and have done axles before? Those motors seems to relatively thin, could you double to power if you run two of those on one axle? One controller to each motor, they will sync automatically. Should give you double the fun :)

do you have a link to the vid?

the peak power is always "a thing like that"
Anyway, i already would be happy with 10-12kw peak power.
This is no typical RC motor, it was made for industrial usage and last year offered to the RC crowd. The axle and bearings looking super robust to me.

btw: regarding eCalc this motor performs better as the Hacker Q100-L (which costs 1100€) and also better as the Joby JM1 (which was mentioned already sometimes here in the sphere).
You can choose the motors in the drop down menus. You just need to type in RPM and Nm and voila you get the result.
I don't know if the numbers are 100% correct, and they may be differnt revisions of Joby motors, but the program at least does tell this.
 
Yeah I will look for the video and post a link. If you do some additional fan/blades to aid w cooling 10-12 kw should be possible. Maybe even a hair more for a few seconds. If these are made for industrial usage then for 20 kw I would assume liquid cooling of some sort? Or possible hi speed fan(s) as nobody would care about noise in an industrial setting, at least noise would be tolerated much more so then for e-bikes.

[strike]What is the price of this motor?[/strike]

Missed the price earlier. Yeah really not a good choice to use for dual motor at that price.
 
macribs said:
Missed the price earlier. Yeah really not a good choice to use for dual motor at that price.

why dual motor?
i guess this idea came in your mind because some pics are showing the smallest version of the L80 with 19mm stator width, but this thread is about the large one which has a 3 times wider stator.
When searching the net i could not find much about those motors at all, so i added pics of what i could find.
The Price?
Man, if it can do at least close to what eCalc says, than it would be worth every cent.

I was told that data from Neumotors is often without controller, and other RC motors are usually bench tested incl. controller losses which might explain the higher peak ETA compared to other motors in the calculator.
 
miuan said:
Hi Madin,

I really can't help with belt drives, but before you get into all the trouble of building one, have you tried a MAC on the Votec? I really love the MXUS 3k on my steel fatbike (pic attached). Road bump filtering with the 4 inch tyre is second to none. Even for light trails, the fatbike rides fantastic and you can barely feel the trail chatter as it plows thru everything. For commuting on a full suspension bike with regular tyre, I would definitely choose a lighter hubmotor like the MAC/BMS/BPM, unless you plan to ride a lot off road. You may lose the crude power that MXUS motor delivers, but will be rewarded by a more natural feeling ride, simplicity and reliability.

Viktor

The MAC definitely would not suit my needs in terms of power and furthermore regenerative braking is not possible with those geared motors.
Regen is a must have.

A short while ago i pushed a bionx D-motor to the limits. It is aside from the grin tech all axle hub the most optimized direct drive hubmotor you can get these days with only 4kg. Unfortunately this D-motor makes an ugly noise above a certain RPM probabaly due to its high pole design, and it seems to have very limited torque output.

I have rode a fat bike similar to yours also with MXUS motor. You have right that those large tires are quite good in filtering bumps, but from my experience it works only for VERY SMALL road irregulations.
When i was riding over large bumps or corrugations, the mass of the motor is still very noticeable.

It goes off like this:
when you riding over a bump, the tire compresses like a spring exactly at this moment because the heavy hub motor cannot follow the road surface. than, after you rode over the bump, the tire decompresses and is pushing the hub away from the bottom. In this moment you have almost no traction.
It isn't a problem if you riding straight forward, but there is high risk of loosing traction when cornering (don't ask why i know :lol: )

As you see in my pic i am actually a hub motor guy, but now i like to try something different with the expectation of much more riding comfort with similar performance to a QSV3 or MXUS 3k.
I think the Neu 8057 can make it possible (of course with proper reduction) with the weight of only 2,7kg which is even than no more located in the rear wheel.

I am still looking for a belt drive.
recumpence told me that for his Astro davinci drives he got the pulleys custom made my pfeiffer industries, but they are $$$$
here is a pic of his trike:
http://www.pfeiferindustries.com/timing-belt-pulley-applications-gallery-i-5-l-en.html

there would be also a complete belt drive available from zero motorcycles (but probabaly a bit overkill for my app, and it woud need a differnt motor pulley and adapter mouting the rear pulley to the hub):

http://www.zeromotorcycles.com/shop/index.php?main_page=product_info&cPath=15_6&products_id=181&zenid=b9e13eifqvvi0vv2brelhg6sg4

10-05731.jpg
 
Would timing belt and pulleys from a car engine work? Those timing belts usually good for at least 100K km, and they are routed in the most peculiar ways and still keep on spinning. In the auto industry there are various manufacturers that makes unofficial parts that often comes close in regards to quality and life span, but that a whole lot less expensive then buying big brand name parts or worst of all "original parts" that no auto maker actually manufacturer themselves they just buy bulk and use their own branding.

Disregard my post about real world power, I've not been able to locate the video and my brain is working slowly these days. Could have been I saw a video of a different model or even a different make. Searching for video by model number turned up nothing. Hey if we ain't got a video it didn't happen ;)
 
litespeed said:
What controller would you run madin88?

Tom

I would try it with what i have after adding halls, which would be Adaptto Max-E and a Powervelocity 12F.
A Kelly KBSX with high speed firmware should work well but it has only speed based throttle.

The coming controller from VasiliSk doesn't look bad and it can do 90kERPM which would be more than enough for this 16p motor.
If FOC is ready i will definitely buy one.

https://endless-sphere.com/forums/viewtopic.php?p=1332074#p1332074
 
Hi Madin88

If you go the beltdrive route, you can manufacture the large pulley from a metallic hub and spokes with a 3dprinted belttrack in segments depending on printersize. As there are so many teeth gripping on the large pulley, a properly made 3dprinted one out of abs/petg/ or even pla should work well. I have good experiences with printed pulleys, and there are examples on the tube as well.

The upside is ofcourse cheap price and customizability.
 
Wheazel said:
If you go the beltdrive route, you can manufacture the large pulley from a metallic hub and spokes with a 3dprinted belttrack in segments depending on printersize. As there are so many teeth gripping on the large pulley, a properly made 3dprinted one out of abs/petg/ or even pla should work well. I have good experiences with printed pulleys, and there are examples on the tube as well.

Thanks for bring up this subject. Yes i've seen 3D printed pulleys and i even would have access to a 3D printer and CAD software, but i am afraid it would be to weak or prone to failure for 10kW+ and 1000rpm. What i like about this idea is that existing parts (sproket, disc brake rotor etc) could be used on which the 3D printed pulley ( or segments) can be attached. But i don't have the time to figure everything out and design the parts in CAD..

So i stay in contact with Pfeiffer Industries (thanks to recumpence for the infos) to get the parts fabricated by them.

As belt i decided to use Gates Poly Chain GT Carbon "8MGT-1280-12" which menas it has 8mm tooth gap and is 12mm wide. The standard length of 1280mm fits exactly my frame :)
The torque and power this small belt with carbon cords can transfer is AWESOME.

the driver pulley will have 18T (46mm dia) and the large pulley will be 108T (275mm dia) which gives a reduction of 1:6

btw: the 8057-75 motor is ordered and on the way!!

Now i am looking for a dual disk style hub for 150mm dropouts (optioal would be 10mm axle but this could be easy adapted).
The motoped hub looks nice, but unfortunately it is no more available..
I am still not sure if a usual 44mm bcd with 6xM5 bolts (like used for disc brake rotors) will be up to the task, or too weak for 200Nm +
 
So excited to see your results with this motor. I have been wanting one for months ever since I stumbled upon the website. Is the intro pricing still available? If these even make half of the power the site claims they would be awesome on anything short of a full size motorcycle.

Are you mounting Halls in the NEU motor? Did you talk NEU into adding them? Did you tell them what you were doing with it or get any feedback or tech support from them? I'm curious if they are bike friendly or if they said "airflow, prop, blah blah ..... no warranty" etc. As we've all seen so many times the power ratings seem to change when its from a dead stop on a loaded wheel. I'm skeptical of the 30kw max rating but the 8057 looks like it would eat my 80100.

Controller options have definitely been a concern. I'm having pretty good luck with an old 12Fet Lyen controller and external hall sensor board on an 80kv 80100 at 20s. Not the most elegant or high tech solution but Lyen said he's tested them to 90-100k eRPM so my 50k eRPM hasn't been a problem so far. The speed based throttle is a big concern for me with the small RC motor's low thermal mass. Seems like the speed based throttle makes it more prone to amp spikes with the lyen which has no reservations about pouring on all of the amps from a stop whether I'm hanging on or not.

The 8kw controller from ASI looks super promising as well but I'm not sure when they will be in stock. 320Amp phase current rating for 1 min but 130 amp continuous. I believe its true FOC. If its as high quality as Justin's phase runners but with that high amp rating it will be really nice option.
http://accelerated-systems.com/wp-content/uploads/2016/07/BAC-8000-Product-Sheet.pdf

VasiliSK's controller looks like a solution a lot of us have been waiting for. Hope he brings it to market soon.

I don't recall seeing any of the powervelocity controllers in a high eRPM application. Anybody know how high they have been pushed?

Also, I think the 6x44 bolt circle on a brake flange is more than enough to hold the power. There is a thread where someone breaks it down with the holding force based on thread pitch bolt torque and surface area. its a crazy amount of strength. Not to mention even if it weren't strong enough you are still not going to overcome the friction and shear the bolts at the same time. The seat of the pants engineering tells me that if the brake flange is strong enough to lock up a wheel from a roll than it would be strong enough to handle the same amount of torque in the form of acceleration. So unless you expect your bike to do standing burnouts with your full weight on the bike then it will probably be fine. I've never ridden a bike that can pull more Gs of acceleration than it can braking. That's my unscientific analysis.

Please let us know if you find any nice dual brake flange hubs. I can't find any. Seem to remember a few places were selling them with the DNM front forks but now I cant find them. I was thinking that would be a great solution for mid drives. I could make some 20mm axles in common rear dropout lengths and use spacers to offset the 110mm front hub as needed in a 135 142 or 150mm dropout. There would even be enough room left over for a 6 bolt brake flange to Freewheel adapter to bolt on the right side I think.

Anyway, thanks for taking the plunge on this motor!
 
DanGT86 said:
Is the intro pricing still available?
yes
Are you mounting Halls in the NEU motor? Did you talk NEU into adding them? Did you tell them what you were doing with it or get any feedback or tech support from them? I'm curious if they are bike friendly or if they said "airflow, prop, blah blah ..... no warranty" etc. As we've all seen so many times the power ratings seem to change when its from a dead stop on a loaded wheel. I'm skeptical of the 30kw max rating but the 8057 looks like it would eat my 80100.
yes, yes and yes :)
- I need to figure out first, but if possible i will install the sensors between the stator teeth. Mounting them outside or further away from the coils would be better for signals at really really high magnetic fields (thus phase current's). But furst i want to find out how it works with the usual installation.

- Have askded them. they don't offer this motor with hall sensors.
- i mentioned that it will be for an e-bike but there was no other discussion about that. Leomotion told me they sold already motors to e-bike builders and they send me a pic of the bike (i believe it was a L90 inrunner motor..)

About the 30kW: it makes a huge difference at which RPM.
Maybe it can do it for a short burst at really high RPM, but i am skeptical too. It is probably overrated as most RC motors are.

Please let us know if you find any nice dual brake flange hubs. I can't find any. Seem to remember a few places were selling them with the DNM front forks but now I cant find them. I was thinking that would be a great solution for mid drives.

I have a DNM dual disc hub in "my stock". The distance between the spokeflanges is quite small and it would need 20mm spacer discs on each side.

There is a dual disc hub made by Risse Racing. Check it out:

https://risseracing.com/store/produ...d/620/osCsid/ac693d0d5ed0305722724f7d5cdcf3b2

s-l640.jpg


hmm but the area where the BC threads are looks quite trimmed down or fragile.. what do you think?

BTW: the stopping power of the best bicycle disc braking systems is around 100-150nm.
 
That risse hub does look a bit thin radially around the bolt holes but the pic on their site looks a little better. The way the hub tapers back towards the spoke flange appears to give it a little more support. What I dont know right off hand is the axial spacing between a standard front disk flange to the dropout. If that is too far outboard on that hub it might put you too close to the frame with a large dia sprocket. I like the idea of starting with a hub thats too narrow and using that extra space as adjustment. Its easier to space out the sprocket or use washers inside the dropout than it is to make it narrower. Then again I'm always trying to fit the sprocket and brake on the same side.

How narrow is the spoke flange distance on that dnm hub?

Have you looked at the LMX dual flange hub? Its got a heftier bolt pattern on the non brake side. Its only 142mm wide but thats easier to space than 135 or 110 front.

http://www.lmxbikes.com/en/spare-parts/47-lmx-rear-hub.html

C21CCFE0-B13D-4646-93F1-E63BF348218F.jpeg
 
@ DanGT
yes, the LMX hub would be a goo choice if the axle can be swapped for a longer one (my dropouts have quite much wall thickness).
The problem is i would like to use the 17x1,4" prowheel rim i have left which has 32 spoke holes, but the LMX hub has 36h.

The motor is there!
It took only 3 days from LA to Austria :)
The quality is on a high level and i have confidence that it will work like a charm as an e-bike drive.

measured values:

23mOhm resistance
78µh inductance (with rotor about 40µH +-5)
0,2mm lams
air gap: 0,8mm

rOQaCX.jpg


eejY38.jpg


dNmmuc.jpg


P4YSiV.jpg


FmcowP.jpg


lCJDl1.jpg


0cDm2a.jpg


Td5PEh.jpg
 
Back
Top