Lightest motor for hill climbing and 135mm dropout with disk brake?

the3rd8

1 mW
Joined
Sep 25, 2011
Messages
13
I am trying to figure out what motor to get for my hill climber project and I thought I could use some advice.

Here are the requirements:
- Rear hub
- Geared
- Dropout width 135mm
- Disk brake
- Estimated max load including bike+rider 300lbs
- Able to climb steep hills at slow speeds (5 mph) without pedaling for at least 5 min
- Wheel size 20"
- Max speed desired on flats without pedaling 10mph
- Freewheel or cassette - only one cog will be installed
- Throttle only

Nice to have:
- Under 5-6 lbs
- Low diameter
- Quiet
- Rainproof (I'm in the PNW)

Any suggestions would be appreciated.
 
I'd say a bafang g310 would be perfect, except for the steep hills climbing without pedaling part. You have 5-10 minutes before that thing melts if you don't pedal. Other small motors in it's class will do way worse than that. Hell, they'll probably just stall.

But the bafang G310 has dual reductions inside so it can produce a crapton of torque per weight, but in a 20 inch wheel you are going to max out the electrical rpm of most controllers except for the phaserunner, VESC, and some RC controllers. That's the downside to the high power per pound.

Motor simulator link:
http://www.ebikes.ca/tools/simulato...&hp=0&grade=0&axis=mph&batt=B4814_EZ&blue=Lbs

As you can see, it overheats rather quickly. Add pedaling power and you greatly extend the time to overheat on a 7% grade.

OK, you might say.. "it's too fast". Yes, but what you need is a motor with an even crazier internal reduction ratio if you want even more low end torque at the expense of speed. Such a thing doesn't exist. Most people wanna do 20mph.

Without requiring an exotic controller... i would go with a standard single reduction small geared hub on a very low speed wind.
The only single reduction geared motor i can personally vouch for is the MAC, but it's probably larger and more powerful than you're looking for. But there's tons of other smaller geared motors you can hook up to the simulator.

Also i can't think of a rain proof motor. Every one has a place to take water in that i've seen. Thermal expansion and contraction of seals and bearing housings is your worst enemy. Big direct drive motors get rusty fast because the water pools up right on the magnets and iron. Geared motors have a space between the motor shell and the stator, so actually they're a lot more water resistant. Mid drives? dunno.
 
Your motor weight target is inconsistent with your implied torque requirement. You'd have to use a sub-6 pound hub motor with reduction to the bike's crank (like Stokemonkey) to meet those two criteria at the same time. And in that case you won't be using just one speed for the pedal drive.

I'd look at the 12T MAC motor driven at its maximum tolerable current, with just enough voltage to meet your speed requirements. 7S of lithium ion would be about right. I'm not sure what the rated maximum current is, but even if it's only 30-40 amps, you're looking at an unusual controller to deliver that combination of voltage and maximum current.
 
The New Mac, ( available for almost 2 years now ) it has a temp sensor inside it.

The first generation Mac , that I have did not have a temp sensor inside the motor.

When bought from a place like em3ev.com you can get the necessary items to make the temp sensor work correctly. You will need a 12 t Mac, a controller that is rated up to at least 40 amps, and the Cycle Analyst V3 .
With all three of those items the Cycle Analyst V3 will sense when the Mac motor is about to over heat, which is the killer of geared hub motors when going up long and steep hills.

The Mac motor before lacing up to a rim is 9.5 pounds, I Have two Q100 c cst motors that I like the lighter weight of better than the Mac, but I now as of last week stopped using them and am only using the Mac because it will run on just a couple of hundred watts , or much higher wattages and twice the amps that a small hub motor will run on.

em3ev.com says to use up to 30 amps on the Mac, however many people use up to 35-40 amps for short distances , like that hill you have to go up.
I do not know how low a voltage you can go on most controllers, I think it is 36 volts.
So you will want a 36 volt battery pack, or two 5s lipo batteries.
I mention lipo because the hill you say you need to go up you will be pulling 35-40 amps out of the battery pack, and very few 18650 packs will do that. You would need a 36volt 18650 pack with 8 or more like 10 parallel strings . Very Expensive for such a pack.

Lipos are cheaper only because you can buy just a few at a time. Good Lipo chargers are expensive, and you would learn how to solder and make your own charging and battery to controller wiring harnesses.
Only get lipo's when you know how to charge and discharge them. and if you live in a house and can make your own battery bunker.
For the high amp capable battery pack you will need to go up that hill, you will want to take the battery pack off the bike each evening for safe charging and storage.
 
He says he is only going to use/install one cog on the rear wheel.

A mid drive would be better , but only if he ran many cogs on the rear with a derailleur.

For what he wants to do , if he were to use a mid-drive I would go with one of the newer systems that uses 11-48 or 11-50 tooth cog set. 8/9/11, or the new Sram NX 12 speed e-bike drive train.

Hillhater said:
Why a hub motor ?
Many mid drives could do this much easier.
 
Hillhater said:
Why a hub motor ?
Many mid drives could do this much easier.
Cannot use a mid drive as it's for a hand driven bike.
It has to be a hub.
 
Justin tells me that operating the G310 at a high RPM in a tiny wheel = magnets being flung off the motor ring.
So you can nix that idea.

A high turn count MAC is the ideal ticket for a lot of torque in the smallest package on steep hills.
You will not find a higher power density motor for the job unless you venture into rear wheel chain drive ( modified mid drive ) or mid drive solutions.
 
MAC 10T for a 20".
Field Oriented Controller like a Phaserunner will tone it down to near silence.

This might give better detail:
http://www.triketech.com/Drivetrain/PowerAssist/HPV-MAC-V2.html
 
One could push more power into any hub with better cooling. If you don't mind getting grease between your nails.
 
Can you post a picture of the bike? I have a friend with a handcycle using a BBS02 which she loves.
A mid drive to IGH would be pretty slick.
 
Triketech, that's an impressive site ya got there, and nice build.

MAC in a 20" wheel is certainly a golden motor configuration. One could sustain 40mph for decent periods of time on an upright bike with that motor + wheel combo. I'm sure it climbs hills like a beast as well.

20" wheels are certainly an under appreciated way of producing A LOT of torque and continuous power with a small motor.
 
I went to bikecalculator.com. 300 lb up a steep (10%) grade at 5mphis 330W. Very doable, if you are at an efficient hub rpm you will draw 415W, 85W will go to heat in the hub, and a hub can shed 85W.

5mph on a 20 inch wheel is 84rpm, that is where it gets hard. To still be in an efficient motor range that means no-load motor rpm about 125rpm (motor efficiency plummets once you go below about 2/3 of no-load rpm).

First choice is the xiongda 2-speed, it's low gear is about 125rpm, it will winch you over that hill, no risk of cooking the motor. But I don't think you get disk brake and 135mm. 135mm without disk or 145mm with disk if I recall.

Second choice would be a Q128C (500W) or Q128H (800W), 201 rpm no-load rpm. To stay in efficient motor rpm you need to climb the hill at 8mph, but they have enough power to do it, especially the Q128H. But overheat risk is there, I doubt you'd want to sustain 8mph up a 10% grade more than 10 minutes.

These calcs are not counting your pedal power. If you are contributing 100W it would be fine.

Bafang G310 is a good motor but not a good choice here. It is geared too high and you'd overheat it quickly.

These are the light options, 3kg. I think heavier motors can do it but I don't know the options.
 
Remember when you're doing your thermal limit estimates that the reduction gearing also suffers some power loss and heat accumulation. I'm guessing on the order of 10%.
 
Chalo said:
Remember when you're doing your thermal limit estimates that the reduction gearing also suffers some power loss and heat accumulation. I'm guessing on the order of 10%.
A common number for a planetary stage is 3%. That doesn't increase for double reductions that are achieved using planet gears that are one piece with two diameters, the number of tooth meshes doesn't change. Double reduction is better thermally because the same teeth mesh once, not twice.

In my hub the gears are are made to close tolerances, and the sun (on the motor) is metal (and helical and ground, a good quality gear). I just don't see much sign the gears fail enough to indicate nylon is a bad design choice.
 
Chalo said:
Remember when you're doing your thermal limit estimates that the reduction gearing also suffers some power loss and heat accumulation. I'm guessing on the order of 10%.

I got off track on plastic vs metal (confused with another thread).

You are totally right gear loss needs to be added to motor efficiency. They both become heat in the hub.
 
tomtom50 said:
Chalo said:
Remember when you're doing your thermal limit estimates that the reduction gearing also suffers some power loss and heat accumulation. I'm guessing on the order of 10%.

I got off track on plastic vs metal (confused with another thread).

You are totally right gear loss needs to be added to motor efficiency. They both become heat in the hub.

I'm not familiar with measurements from geared hub motors, but I'm familiar with the range of gearhub efficiencies ranging from low 90% range (Sturmey Archer 3sp and comparable) to below 80% (NuVinci). I assume these efficiencies include chain losses, so it's not all gear losses. At the same time, they don't include large reduction ratios such as are normal for geared hub motors.
 
Chalo said:
I'm not familiar with measurements from geared hub motors, but I'm familiar with the range of gearhub efficiencies ranging from low 90% range (Sturmey Archer 3sp and comparable) to below 80% (NuVinci). I assume these efficiencies include chain losses, so it's not all gear losses. At the same time, they don't include large reduction ratios such as are normal for geared hub motors.
As a kid I took apart my sturmey archer hub a couple of times. The planet gears were on plain bearings and the gear surface quality was not high.

The planet gears in my hub motor are on ball bearings and the gear finish is better.

Although the ratio in my hub motor is high (10.66) it is with two meshes and the ratios are moderate.

Nuvinci if I recall is not gears but rolling elements transmitting torque by friction. Really different. That concept was worked for years for automotive and was always less efficient than a gearbox.

Rohloff hubs have a lot of stages and pretty good efficiency, I think would fit with the 3% per stage rule of thumb. Good execution makes a big difference.

The execution on my hub motor was pretty Impressive, photos are in
https://endless-sphere.com/forums/viewtopic.php?f=3&t=97415&p=1427659#p1427459
 
I've been using a Mac 8t in a 20" wheel on a front loading cargo bike for about 4 months, and I live on a hill that's 150 ft climb that averages 9% but peaks somewhere around 15%.

In the peak of summer, once or twice I saw the motor reach 90C after I'd reached the top of the hill and stopped at the mailbox (no more airflow), but never above that.

It's a 100lb bike, and I'm 180lb, and I've had a 150lb passenger, my experience there is that if I don't have a fairly full charge, we bog down on the 15% - I kind of wonder if that's where a 10t would do better?

Anyway, that's just to chime in that the Mac with Phaserunner has done well for me in a challenging situation.
 
Back
Top