Vehicle: Car: Transmission: Continuously Variable
bike hub with two motors   (+5, -1)  [vote for, against]
Coward's preface: It's late and well out of my area of expertise, so there's a high probability of this being a crap idea.

Hub motors are a simple way to add electric power assistance to a bike. A hub motor fits unobtrusively inside the wheel. The problem with hub motors is the lack of gearing; they deliver power optimally only when the bike is moving in a particular range of speeds. Thus hub motors are very inefficient on hills and on initial accelaration.

What is needed is a hub motor with internal gearing, and more preferably CVT (continuously variable transmission). The problem with most CVTs are that they don't fit naturally into a hub which is thin and radially symetrical.

My idea is to make a hub with two motors and two concentric planetary gears (also known as epicyclic gearing). see illustration. Although planetary gears are not technically CVT, they perform the same function.

The centre gear (blue) is connected to the wheel axle which is held stationary relative to the bike frame. The first motor drives the inner planet gears (orange), the second motor drives the outer planet gears (green). The outer ring (purple) is connected to the wheel.

The speed of the two motors essentially creates a continuously variable gearing ratio. The speed of the motors is chosen to minimise the torque on each motor, thus maximise the power transmitted to the wheel. This is a similar to how a Prius works. So at one moment the inner and outer gears are supplied with 150W and 50W respectively, while at another moment the inner and outer gears are both supplied with 100W.

Because the motors are always operating at low torque, the motors can be relatively small. Hopefully two small motors is not too much heavier or larger than one large motor.
-- xaviergisz, Jan 30 2009

illustration http://imgur.com/a/4o9Yo
[xaviergisz, Jan 30 2009, last modified Dec 12 2011]

Toyota Prius - Power Split Device (PSD) http://eahart.com/prius/psd/
[xaviergisz, Jan 30 2009]

Nohassel motor drive http://www.endless-...7358&st=0&sk=t&sd=a
An interesting open-design project. Two independently controlled motors, coupled by an epicyclic differential - this gives the advantages of a CVT and allows both freewheeling and regeneration, all within a single unit. [xaviergisz, Jan 31 2009]

NexxtDrive electric propulsion system http://www.bike-eu....e-transmission.html
Its patented “Power Combining” system consists of 2 interacting electric motors, combining the mechanical and electrical inputs through a compact CVT. [xaviergisz, Dec 23 2010]

Recent CVT http://machinedesig...d-automotive-transm
Supposedly even includes a fully gears-running neutral-motion setting [Vernon, Apr 01 2012]

2 electric motors in 1, sort-of http://books.google...v=onepage&q&f=false
Don't call the parts "rotor" and "stator"; call both of them "interactors" [Vernon, Apr 01 2012]

An electric assist bicycle drive with automatic continuously variable transmission http://ieeexplore.i...sp?arnumber=4771266
A trial design is given for a drive combining two planetary gears and two motor/generators in the rear hub. [xaviergisz, Apr 04 2012]

US8297384 http://www.google.c...u/patents/US8297384
Although this patent has very little to do with this idea, it cites this idea as prior art (my first HB idea as a patent citation) [xaviergisz, Jun 17 2014]

+ Wow.
-- Zimmy, Jan 30 2009


compared to a combustion engine force can be utilized without movement. I don't think any gears are required. I am not sure if it is safe to say that the amount of force( or torque?) for accelleration is constant at any speed. I think it causes a problem if you can't accelerate and you run one magnet like I am sure you are thinking about that already. You could push off with your feet or have a self regulating motor that doesn't allow you to fry it.

I think what we are dealing with is baked. It is the future and Kaneda had it on his bike in the movie Akira. I am pretty sure.
-- MercuryNotMars, Jan 30 2009


Sorry, your missing something important. If the two motors do not have equal torque, then one will drive the other backwards. You might as well put both motors on the same shaft, or better just have one motor and a three speed hub gearbox, or create an automatic trans with a couple of clutches and that planetary gear set. The planetary gearbox in the Prius works like a differential and works because the electric motor always has more available torque than the gas motor or works as a generator when the gas motor over torques the electric.
-- MisterQED, Jan 31 2009


//If the two motors do not have equal torque, then one will drive the other backwards//

So could you run the motors so their torques are always matched?
-- xaviergisz, Jan 31 2009


Electric motors have basically a flat torque curve all across their RPM range. You can change that with gearing, but it trades one for the other. So say we have an application where we have a wheel that goes from 0 to 1000 rpm and we have electric motors whose range is 0 to 1000 rpm. To maximize acceleration we need to maximize torque on the wheel. So we could use a single motor whose torque is 100 (units removed for simplicity, I'll just put T for torque). It works but not as well as one with a three speed (say 4:1/2:1/1:1) gear box whose starting torque is 400T. Your idea could use two 50T motors linked by a 1:1 diff but that would work just like a single 100T as the motors would just add like they were on a single shaft. You could use a 75T and a 25T motor linked by a 3:1 diff, but the 25T would rev limit at 333rpm wheel speed. So guess up to that point you'd get 150T at the wheel 3x25+75, but I think it cuts the range down at the top end.
-- MisterQED, Jan 31 2009


I think driving the other moter backwards is not a problem. if this is a gas-electric hybrid. you can think of this as a fancy drive trane you can run the engine at full tilt and run the electric backwards generating electricity even with the brakes on you should be able to control this with a dimension of the electricity stopping the motor you shouldn't really need any different gears. over the long term the top average speed would be dependant on the motor at full thinking of the electric as a power boost. if you can get an engine that will get you up to 65 in its peak then you have power for passing in your motor. you can shut off your engine part time if you don't use all the power. You save a little on engine size and the gears and clutch and Electric is the best thing for acceleration with the infinite something you can call infinite gears. the size of battery probably isn't that big with a gas engine handy.

I am not sure I understand the complicated power split divice. I might be imagining prius design on a bike. I think the complication is with tieing in what looks like three driving motors/engine
-- MercuryNotMars, Feb 01 2009


I was thinking about my suggestion and I think to make my suggestion work exactly how you want you would want two motors and an engine. you have the engine and the motor together. You can think if it as a starter and an alternator too. The Engine wants to run full power or not at all. No change in speed no shifting gears. It is all controlable and regenerative you don't need slipage of a clutch you can connect disengage any part by speed matching.

Thought experiment accelleration from 0-65 MPH. The engine puts out enough power to get you to 65 and overcome wind resistance. engine speed and gears and motor armatures are brought up to speed thanks to the starter motor and the bike is standing still. Engine powers this system and there is nothing about this that makes the bike move forward. The hub engine starts generating power and for the purpose of this thought experiment. offering resistance torque resistance and power generation. for the purpose of the thought experiment lets tie the power generated and expend it in the starter moter helping the engine. we have thus tied the two motors together with a constant amount of torque and that comes out of the system in the wheel. Minus wasted energy all of the engine power goes into acceleration. No shifting, no changing engine speed. at no time is the engine at anything less that peak performance speed. Anything you do not use in acceleration can be stored in the battery and any more acceleration you want can be taken from the battery. You can generate power from the engine and regenerative braking at the same time. If you want to cut out the motors while at cruzeing speed you can simply stop the hub motor and lock the hub motor. You can maintain speed with addition or subtraction of torque with the starter. If you don't have to build an engine for throtteling up and down I imagine there are unique benefits.

There would be no throttle just power management systems. You might have a GPS you probably have a display of how long you can maintain a speed in a graph. I imagine it will be a little more like your cruze control. setting the speed you want.

Sorry if I am running with your idea and it is nothing alike but you got me thinking.
-- MercuryNotMars, Feb 02 2009


I've made a small variation to this idea. The pink ring (in the illustrated idea) acts like a flywheel. However, a flywheel that is only 10cm radius is not going to store much energy (appox. 10x less) compared to a flywheel with a radius almost as large as the rim of the wheel (30cm).

The added illustrations (the last two illustrations) show a large pink ring (almost extending to the rim of the wheel) connected to the small pink ring of the previous illustrations. If the large pink ring weighed 1kg and rotated at 20rev/s it could store 720J.

Edit: The energy to accelerate 80kg to 20km/h is 1000J, so a 2kg flywheel could store enough energy 'regeneratively' to stop/start the cyclist without needing to use the friction brakes.
-- xaviergisz, Apr 01 2012


// The first motor drives the inner planet gears (orange), the second motor drives the outer planet gears (green) //

How?
-- BunsenHoneydew, Apr 07 2012


//How?//

I haven't illustrated them, but the motors would be outside the gears, i.e. one above and one below the gears in the illustration. The motors would be attached to the rings (grey) that are connected to the planet gears (green, orange).

Now I think about it, the flywheel embodiment of the invention would require 3 motors. The two as before, and an additional motor driving the central sun gear (blue).
-- xaviergisz, Apr 08 2012


What [Zimmy] said.
-- 2 fries shy of a happy meal, Oct 01 2013


I thought the point of electric motors is that they are good over a very wide range of speeds, so that gears are not needed; is this not so?
-- MaxwellBuchanan, Jun 17 2014


The problem is both torque than speed. My intuition is that small/light motors are not good at high torque conditions (e.g. accelerating from stationary). Even if you could make a small motor with high torque, the drain on the battery would probably not be good for it.

The idea is that you can have two motors (or three motors in the flywheel embodiment of the idea), all running within their optimal range of torque and speed, to efficiently deliver a broader range of torque and speed.
-- xaviergisz, Jun 17 2014


I was not a bike rider when you wrote this originally, now I am an avid one. Or rabid one, whichever. I like using my leg muscles to power me and that is kind of the whole point of riding. But you would use this more of a form of transportation, more like a motorcycle, I guess. Anyway, I love my bike. Her name is sunny.
-- blissmiss, Jun 17 2014


// ... the point of electric motors ... very wide range of speeds, so that gears are not needed ...//

Yes and no.

In almost all cases, if there is a transmission, a smaller motor can be used to get similar overall performance. After switching to a higher gear, max torque to the wheels is reduced, but at higher speeds usually the battery and motor controller are limiting power output, not the motor. The exception is if you built a vehicle that has excessive battery/controller power, and the max motor torque is set based on the torque required to spin the tires. In that case there is no advantage to increasing the available torque at low speeds.

On the other hand, while a transmission allows using a smaller motor, it is usually better to use a larger motor (and motor controller) so that the cost and complexity of a multi-speed transmission is not needed.

On the other other hand, in most transportation applications a single speed gear reduction is still used. If max speed is not equal to max motor RPM then you could have built it with a smaller motor and a gear reduction, and a single speed gear reduction is generally cheaper than an oversized motor. It's possible to trade off RPM limit for torque in a motor design by increasing the diameter (keeping the overall size the same). That's why pancake style motors are sometimes seen on e-bikes. Unfortunately that approach has only been taken so far, so even a pancake motor will tend to have a gear reduction of some kind. I've often wondered if it would be possible to take that concept much farther, resulting in a toroidal shaped motor with spokes or a thin disk transmitting the force to the hub, but then you start running into practical considerations like bearings, gyroscopic effects, etc.
-- scad mientist, Jun 17 2014



random, halfbakery