• Howdy! we're looking for donations to finish custom knowledgebase software for this forum. Please see our Funding drive thread

My RC Mid Drive with Single Stage Reduction

FZBob said:
cheapcookie said:
Interesting, do some of these outrunners come with temp sensors? whats the motor heating rate 4svs6s?
All six of my motors have sensor cables. Five of them have six wires in the sensor cable, which I believe indicates they have temperature sensors. As I run sensorless FOC, my sensor cable is not connected...

My big motors don't get warm to the touch in my normal riding (~150 watts mostly), but I'm not sure about high power. If I get a chance I will plug my cable into the VESC, and see if I get motor temperature readings. Not sure when, as my wire routing is a pain...

Yeah have you seen the connectors ebikes.ca has recently "pushed forward"?
 
Hi all. I'm new to the forum and have been attempting to read as much as possible before asking any questions. Some great ideas here :shock: Once I get something worth posting I'll put up a couple of pics.

Hey c_saayman, whereabouts in Aus are you? I'm in Brissy.
 
I found another advantage to a light, low power E-Bike. Last week I traveled from California to Arkansas. There are great trails in Bentonville, so I shipped my bike. https://www.bikeflights.com/

Since the bike is relatively light, the shipping rates were really good. Another 2 pounds and shipping cost would have jumped by about 40%.

I was able to pack two Li batteries in my carry on, as they were under 100 watt hours each. Funny story - I was worried about the batteries going through security. The TSA inspector said he didn't care about the batteries, but he needed to carefully inspect my PBJ sandwitch, as they recently found a "Banana Bomb"!

A heavy E-bike with a big battery would be a big hassle/expense to ship. Once the battery goes over 100-150 watt hours, things get very complex.

Bike Box.jpg
 
That box for the bike is really neat. Do they allow max 100Wh per battery or per 'luggage' ? If it's per battery then a lot of trick can be made by splitting the battery.

I made some improvements for my motor bracket. The plastic was not dissipating heat and i was constantly overheating. Now im using 6mm aluminium with some 3d printed mold on the frame. Cooling has been improved significantly.

DSC8371.jpg


DSC8373.jpg


I tried 3 sprocket: 9t,10t and11t.
9t is really rough. I don't really like it.
10t is smoother with less torque. A good compromise.
11t is again smoother but not smooth enough, torque is pretty low.

I feel that #25 drive-train cannot be absolutely smooth like a bicycle chain. The 11t was ticking .. maybe a 13t-15t would be satisfactory.

I am feeding the motor with 40A, any more and the motor cogs and skips at full throttle, not a nice feeling. With 14-15v from the battery and 25-27A it can reach ~370W. I cannot really achieve more than this, the 9:1 reduction is limiting the power.

A small planetary gearbox on the motor shaft would be perfect.. and maybe a #219 chain or bicycle chain for improved smoothness.
 
aCeMadMod said:
But the #25h chian
Takes 40 amp at 50 volts. And looks like it can more. Before it cracked.
I know from real world data .

That's a serious 2kw of power but was it first stage or second stage reduction ?
If it was second stage that's really good, my only concern is that it's really rough. When the motor develops torque i feel the whole crank is vibrating like hell. What was the size of the small sprocket which you used?
 
Hi it on the 2 stage . My sockets are bigers 56/ 46t
Cyclone motor
It help load teeth better.

I feel it could take 60amp. 75amp before breaking in half

I test it soon.

My chians are cheap. $7

25h# can take it...this my shoping ebike

Good luck
 
FZBob

I have an ideea which you also had in the past. But it seems feasible to me now. What if i put another alu plate in front of the motor plate and mount 2 spur gears for a primary reduction of 3:1 (45t-15t), will a mod 1 spur gear survive ? (let's forget the freewheel for now)

These are Mod1 spur gears for cheap:
https://www.aliexpress.com/item/32946542536.html?spm=a2g0o.detail.1000060.1.37f1620bXfwKpT&gps-id=pcDetailBottomMoreThisSeller&scm=1007.13339.146401.0&scm_id=1007.13339.146401.0&scm-url=1007.13339.146401.0&pvid=0369aefd-29a5-4734-b9d4-06521905ebf5

1-pc-1-mod-Spur-gear-1-M-15-20-16-18-35-50-60-25.jpg


If not, wil Mod 1.5 do it ? (or Mod 2)
 
I think you'll need to run those in some sort of oil bath/enclosure. Since they are metal and straight cut you'd be getting a lot of noise.
I'd look at either a belt, another chain primary or preferably a helical planetary. All of which kind of fall outside the scope of this project.
 
fixvid said:
That box for the bike is really neat. Do they allow max 100Wh per battery or per 'luggage' ? If it's per battery then a lot of trick can be made by splitting the battery.

I made some improvements for my motor bracket. The plastic was not dissipating heat and i was constantly overheating. Now im using 6mm aluminium with some 3d printed mold on the frame. Cooling has been improved significantly.

I tried 3 sprocket: 9t,10t and11t.
9t is really rough. I don't really like it.
10t is smoother with less torque. A good compromise.
11t is again smoother but not smooth enough, torque is pretty low.

I feel that #25 drive-train cannot be absolutely smooth like a bicycle chain. The 11t was ticking .. maybe a 13t-15t would be satisfactory.

I am feeding the motor with 40A, any more and the motor cogs and skips at full throttle, not a nice feeling. With 14-15v from the battery and 25-27A it can reach ~370W. I cannot really achieve more than this, the 9:1 reduction is limiting the power.

A small planetary gearbox on the motor shaft would be perfect.. and maybe a #219 chain or bicycle chain for improved smoothness.
They allow up to 100 Watt Hours per battery, and I couldn't find any limit on how many you could take. Carry on only. If you get special permission from the airline, you "may" be able to carry on two 150 watt hour batteries. I didn't want to deal with that...

I don't remember if you mentioned it before, but what bike gearing do you run? I have 40T and 22T front chainrings, and an 11-40 cog set on the back. Most of my serious climbing is in the 22 front ring, and the 36 or 40 on the back. So I climb at between 2.5 and maybe 4 mph on steep climbs. Until very recently, I used about 125 to 150 watts on these climbs. However, my knee is getting worse, so recently I have been using around 250 watts. If your bicycle gearing is not low enough to climb very slowly, then this drive will probably struggle.

Also, how many Amp Hours is your battery? Do you know the C rating? My 5.8 Ah batteries sag sooner than my 8 Ah battery. And those are all 30C Zippy RC batteries.

If you take a look on McMaster Carr, they have a bunch of gears, bearings, spacers, clips, keys, etc. It's the mechanical engineers favorite website. Also, they have Autocad (2D) and 3D drawings for everything which you can download.
 
Grantmac said:
I think you'll need to run those in some sort of oil bath/enclosure. Since they are metal and straight cut you'd be getting a lot of noise.
I'd look at either a belt, another chain primary or preferably a helical planetary. All of which kind of fall outside the scope of this project.

You're right that gears needs lubrication and probably belt would be the most simple solution.
I was struggling to find a planetary gearbox for rc motors, do you have any example?


FZBob

My gearing is 46t - 28t, I cannot go lower then this. I have the same problem as you, bad knees, and I'm struggling on hills. It would be preferable not to pedal at all and that's what I'm aiming for.

I really feel that there is more power in that motor but rpm is not enough. Dumping a lot of amps is not very efficient and a lot of that is wasted as heat. I would preferably set a limit of 20-25A at a higher voltage and make a smoother drive train with a larger drive sprocket.

My battery is 4s6p 250wh and it does sag under load. I saw max battery consumption around 28A(voltage drop around 2v) with 50A motor current but at this level the motor is cogging (skipping) sometimes at hard acceleration. 40A motor current limit is a safe bet with no cogging. Do you experience the same?

Screenshot_20191101-104347_Photosc1dba17dc7465421.png
 
Fixvid I think with your requirements you should be looking at a low voltage CYC build. A small RC setup will always be at best a mild assist rather than a leg replacement.
Guys have run them with Castle Creations ESCs and VESCs using throttle only.
 
Grantmac you mean the CYC x1 pro? That's a nice drive but it costs big bucks. I don't really want to spend 1k, I'd rather improve my skills at building a drive myself.
The goal is to be able to climb the steepest hill without pedaling and light enough for me to lift it on a roof rack. My scalpel has 15.2kg and I'm not comfortable lifting more weight. That's my problem with commercial ebikes, they are so heavy. I'm eyeing a new Cannondale habit for a new build.
 
By the time you add up component costs I think you'll find the CYC less expensive then engineering something similar. You could probably get by with 500w and ~25-30:1 total reduction provided you didn't mind climbing at a snails pace, but it's still going to be a +40# bike. A single stage build isn't going to happen unless you crank ~5kw through it and aren't interested in pedaling, at which point it's going to be +65#.
There is an old saying:
Fast/Cheap/Light, you get 2 at most.

PaulD has come as close to what you're after as anyone else:
https://endless-sphere.com/forums/viewtopic.php?f=28&t=87748
Again, hardly cheap.
 
Paul's drive is really slick, but I would still prefer to develop mine. I'm going today after the new bike and begin doing the CAD stuff. In simple terms 3:1 primary belt reduction and 6:1 secondary 25h chain reduction. Compact as possible. I'm aiming at around 800w. Speed doesn't really matter just to be able to climb the steepest hill reliably. I feel there is huge efficiency loss pushing a bldc motor at low rpm. BTW, what do you mean with '40#' is that speed in mph?
 
fixvid said:
My gearing is 46t - 28t, I cannot go lower then this. I have the same problem as you, bad knees, and I'm struggling on hills. It would be preferable not to pedal at all and that's what I'm aiming for.

I really feel that there is more power in that motor but rpm is not enough. Dumping a lot of amps is not very efficient and a lot of that is wasted as heat. I would preferably set a limit of 20-25A at a higher voltage and make a smoother drive train with a larger drive sprocket.

My battery is 4s6p 250wh and it does sag under load. I saw max battery consumption around 28A(voltage drop around 2v) with 50A motor current but at this level the motor is cogging (skipping) sometimes at hard acceleration. 40A motor current limit is a safe bet with no cogging. Do you experience the same?
Your lowest gear is close to my 2nd gear. I can usually climb most hills in second, although lately I'm relying on first gear more often. I climb around 100-200 watts. HOWEVER, if I take my feet off the pedals and try motor only, it falls on its face, even at full throttle. Barely climbs shallow hills in the lower gears. I'm not sure why this is. I agree that if your goal is not to pedal, this style of build probably won't satisfy.

There is definitely much more power in these motors! POWER is proportional to TORQUE (amps) x RPM. If you triple the RPM, you will triple the POWER. Of course then you need to gear it back down. It all depends on your needs. For my needs, (around 200 watts continuous and VERY short bursts to 500 or so, light weight and cheap), it works great and is very efficient.

If you need to pull 500 watts at very low RPM, you may run into cogging. I just looked at the video you linked back on July 11, and my bike does exactly the same thing. I ran my orange motor at 48 amp motor limit for a year, and briefly at 53 amp. I'm sure I could get it to cog, but I just don't run into that in normal riding.

Just curious, are you trying to accelerate hard from very low crank RPM? If I need a bit of assist and the crank is spinning less than say 40 RPM, I try to bring in the throttle gently. A few times I have been caught in the wrong gear, with the crank barely turning, and tried to catch it with a handful of throttle, and it cogged. But that's my fault...
 
Fixvid -

Grantmac makes a lot of good points. Designing and developing a proper gear reduction will probably be about an order of magnitude more work/difficulty than a single reduction system. I learned long ago not to build anything that I could just buy. I just took a quick look at the CYC, and it looks pretty nice. On sale now for $810. I'm almost tempted, and I'm not even in the market...

Looking at your requirements, I have a few questions.
Climb the steepest hill without pedaling -
How steep?
20% (Tough, but I used to do long slogs without assist)
25% (May need to balance between rear tire traction and looping out)
30% (I'm not sure I could climb that on any bike)
Almost Vertical? (Real Moto Hillclimbers - fun to watch on Youtube!)
How fast?
2.5 MPH? (barely controllable)
5 MPH?
How much do you weigh?
With this information we can calculate theoretical power requirements. Then I would double that... (I'm betting you would want at least 1500 watts)

What range would you like without pedaling?
How much total climbing would you like without pedaling?
That will give some idea of battery requirements and weight.

Weight limit 15.2 Kg (33.4 lb)
How light (expensive) is the donor bike?

Just curious - has anyone on the ES forum or elsewhere come close to these goals? I would love to see a table surveying DIY E-Bikes listing Power, Total Weight, Conversion Weight and Battery Capacity/Weight. Power vs Conversion Weight (with & without batteries) would be an interesting graph.
 
I just wanted to chip in and say that I am glad to see this experiment still ongoing. I've never seen anything like it on here and dig the pioneering spirit.

I wanna do this with a super tiny 3lb geared motor. The end result may be a little heavier and less efficient than what you're up to, but ultra light, easily serviceable and simple mid drives are rad, and I think they are way cooler than these other bolt on mid drive kits out there.

Keep up the great work on proving this!
 
I have the new bike, its Cannondale Habit Carbon 2 SE. Currently has 13kg but i will put it on a weight loss diet until it reaches 12kg. Maximum weight without batteries should not exceed 15kg because i have to put it on a roof rack and i'm not the strongest.




We have 3kg to play with for the motor, bracket, transmission, pedal freewheel and controller. Battery will be removable and im thinking of using 10s2p configuration with 21700 5000mAh cells capable of delivering 15A.

Steepest gradient which it should climb is 25% at low speed around 3-4mph. I have 63 kg so i'm fairly light and won't need a ton of power to climb a steep mountain. My other bike with Shimano E8000 mid drive can climb anything with a little effort and it has a 3 stage gear reduction around 25:1 - 20:1.

My ideea is using a small RC motor like Turnigy D5035 or D6374 or this cheap chinese outrunner:
https://www.aliexpress.com/item/33011400738.html?spm=a2g0o.detail.1000014.23.4c7d6d4bO5mMcK&gps-id=pcDetailBottomMoreOtherSeller&scm=1007.13338.128125.0&scm_id=1007.13338.128125.0&scm-url=1007.13338.128125.0&pvid=ab9d1249-6d2e-4171-bdc2-6c7b14c5d059

Primary reduction will use htd5 belt with 20t drive pulley and im thinking of printing the 80t driven pulley .. I would save 25bucks and will be easier to integrate a one way bearing into the big pulley:
https://www.aliexpress.com/item/32833380128.html?spm=a2g0s.8937460.0.0.32722e0eHUgWd4
Is the 20t pulley big enough to prevent any slippage (15mm belt)?

Secondary reduction will be driven by a 12mm jackshaft and 15t 25h sprocket to reduce the vibration of really small sprockets. Driven sprocket will be 90t and i'm planning to convert to ISIS bb with trial cranks + freewheel and adapter for 104bcd chainrings or 76bcd because i have eagle and those chainrings come with 76bcd hole spacing.

Total reduction: 24:1 with around 700W - 1000w available. I think this drive will surpass the Shimano one which is pretty good benchmark.
 
You might be surprised at how much power is actually being produced by the Steps E8000 "250w" isn't the actual output.

That said I'd look at GNG for various belt parts since many of their early drive systems had a belt primary. They have since gone plentary though and personally thats the direction I'd look. Specifically surplus/recycled drill drives can be useful.
 
fixvid said:
I have the new bike, its Cannondale Habit Carbon 2 SE. Currently has 13kg but i will put it on a weight loss diet until it reaches 12kg. Maximum weight without batteries should not exceed 15kg because i have to put it on a roof rack and i'm not the strongest.


We have 3kg to play with for the motor, bracket, transmission, pedal freewheel and controller. Battery will be removable and im thinking of using 10s2p configuration with 21700 5000mAh cells capable of delivering 15A.

Steepest gradient which it should climb is 25% at low speed around 3-4mph. I have 63 kg so i'm fairly light and won't need a ton of power to climb a steep mountain. My other bike with Shimano E8000 mid drive can climb anything with a little effort and it has a 3 stage gear reduction around 25:1 - 20:1.

My ideea is using a small RC motor like Turnigy D5035 or D6374 or this cheap chinese outrunner:
https://www.aliexpress.com/item/33011400738.html?spm=a2g0o.detail.1000014.23.4c7d6d4bO5mMcK&gps-id=pcDetailBottomMoreOtherSeller&scm=1007.13338.128125.0&scm_id=1007.13338.128125.0&scm-url=1007.13338.128125.0&pvid=ab9d1249-6d2e-4171-bdc2-6c7b14c5d059

Primary reduction will use htd5 belt with 20t drive pulley and im thinking of printing the 80t driven pulley .. I would save 25bucks and will be easier to integrate a one way bearing into the big pulley:
https://www.aliexpress.com/item/32833380128.html?spm=a2g0s.8937460.0.0.32722e0eHUgWd4
Is the 20t pulley big enough to prevent any slippage (15mm belt)?

Secondary reduction will be driven by a 12mm jackshaft and 15t 25h sprocket to reduce the vibration of really small sprockets. Driven sprocket will be 90t and i'm planning to convert to ISIS bb with trial cranks + freewheel and adapter for 104bcd chainrings or 76bcd because i have eagle and those chainrings come with 76bcd hole spacing.

Total reduction: 24:1 with around 700W - 1000w available. I think this drive will surpass the Shimano one which is pretty good benchmark.

Just for fun, I went here to take a quick look at power requirements.
http://www.u.arizona.edu/~sandiway/bike/climb.html
That says you would need about 360 watts. However, I believe that is average power for a climb, and assumes perfectly graded and compacted dirt. I also believe that is watts applied to the tire contact patch, and does not account for efficiency. I would want 2-3X that amount. So your 700-1000 Watts is probably a good estimate.

At a guess, this should be do-able in your 3 KG.

If I was going to build a 2 stage, I would take a look at a gearbox with 5-7x reduction in one stage. Possibly using a 60-80T plastic spur and 12 T Pinion. I would plan on spending some time calculating tooth loading, and discussing technical issues with gear suppliers. Then I would take a look at #35 Go-Kart chain or a similar size for the second stage. Probably a 12-15 T on the gearbox, mounted to a freewheel or sprag clutch, and a 60T on the crank (same dia as my 90T #25 chainring). This would require proper analysis of the gears, and some machining.

If I wanted to do something much simpler, I would do a first stage using #25 chain, 12T and 60T, for 5x reduction. Second stage #35 chain as above. And a bit of steam punk look with the extra sprocket.

When I tried a belt drive on my build, I found it to be far less forgiving than the chain, in addition to being bulkier, and handling less power in a given space. But they are quiet and sexy...

Since this sounds like a really interesting build, I would suggest starting a dedicated thread for it, so it gets the attention it deserves. Throw up a link here, and I will follow.
 
Grantmac said:
You might be surprised at how much power is actually being produced by the Steps E8000 "250w" isn't the actual output.

That said I'd look at GNG for various belt parts since many of their early drive systems had a belt primary. They have since gone plentary though and personally thats the direction I'd look. Specifically surplus/recycled drill drives can be useful.

I like the drill planetary. Just curious - can you recommend any specific drill models? Do you know if anyone has taken a drill planetary and mated it to a bike friendly case? What is the typica reduction per stage?

Might actually be worth buying a new high end drill just for the motor and reduction unit, as I believe they use BLDC motors.
 
The robotics guys use Dewalt gearboxes. Unsure on maximum input RPM or durability in a traction application. They can be very high reduction.
There was someone doing a two stage planetary build on here not that long ago:
https://endless-sphere.com/forums/viewtopic.php?f=28&t=98618&hilit=Planetary
 
I'm not sure if he passed the planning phase but it is sure a nice ideea. In the meantime I found this planetary which resembles to the cyclone one and looks like a though one:
1Set-36T-Gear-Diameter-38mm-Thickness-12mm-Electric-Vehicle-Motor-Steel-Gear-Gear-Ring-Clutch.jpg_640x640q70.jpg_.webp

https://s.click.aliexpress.com/e/BWrqh4GY

I think a 3d printed enclosure / motor adapter wouldn't be hard to make. It just needs a shorter and fatter motor. My only concern is that it looks like the input is a flange and the output will drive a shaft with keyway , shouldn't be it the other way around?

According to my calculation the gearbox is 6:1 ratio and it's not big, ring gear having an outer diameter of ~10cm.
 
Back
Top