hybrid bicycle with electric CVT?

[NiHaoMike]

Has anyone thought about building a hybrid bicycle with an electric CVT? (Basically, it would work like a scaled down version of the CVT in the Prius, with the pedals instead of the engine.) One motor is directly coupled to a wheel, and the other motor has the shaft coupled to the rear wheel, the casing to the pedals, and slip rings used for the electrical connections. Two bidirectional DC/DC converters connect the motors to the batteries, with a microcontroller managing power flow.

To ensure it rides like a normal bicycle, the "gear shifting knob" will control the proportion of pedal speed to wheel speed. In other words, the microcontroller will measure the wheel speed and calculate a pedal speed based on the "gear ratio" selected by the knob.
* If the pedals are rotating faster than that, one DC/DC converter will pull power from the split motor to load down the pedals to slow them down to that speed while the other DC/DC converter will supply power to the wheel motor to (by default) net zero battery current.
* If the pedals are slower than the calculated speed, the motors will be inactive and the bike will "glide".
* If the pedals are going backwards, the bike will regen at a rate proportional to the speed of the pedals going backwards. (In other words, pedal backwards to regen!)
* Optionally, if the pedals are moving while the brake is engaged, use that to recharge the batteries.

To make use of electric boost, a twist handle will be used to control battery power used for boost just like a regular electric bike. (Optionally, the handle can twist backwards to engage regen, as an alternative to pedaling backwards. It can also allow for regen while pedaling forwards.)

The display panel will be based on a small video LCD, so the display layout and data will be defined by software. (A cheaper alternative can be a plain text LCD, although small video LCDs are cheap nowadays, often cheaper than some text LCDs! However, text LCDs are orders of magnitude easier to interface than video LCDs.) It can be programmed to look like a panel of meters (either digital or "analog"), a pictorial diagram, a combination, or whatever else you like! What I think would be interesting would be an animated diagram with icons of the wheel, batteries, and pedals arranged in a triangle. Arrows of varying sizes would show the flow of power when in operation (pedaling, boost, regen, and pedal recharge). For additional information, a few meter displays would be used. The diagram can also include the motors to be more accurate, but that would probably clutter it up too much.

The microcontroller would have to be advanced to accommodate the sophisticated system, so it would likely be something like the Parallax Propeller as it would be well suited for the many threads needed.

I don't think I'll actually build such a bike as bikes are of very limited use in my current neighborhood, but I will be happy to help with the development if someone else wants to build one.

Check hybrid bicycle with electric CVT? - CleanMPG Forums for an existing discussion about it.

[captainslug]

Sounds more complicated than is needed.
Many E-Bike hub motor kits come with pedal-elec options that will shut the throttle off if the pedals are not moving. You could simply short out the throttle so that any amount of pedaling will trigger the throttle, or build a simple circuit that will alter the throttle (resistance) based on how quickly the pedals are spinning.

[NiHaoMike]

Those work somewhat like the IMA system in Hondas, not the HSD system in Toyotas.

The advantages I see with a HSD-like bike are:
*"infinite" gear ratios with a simple mechanical system (IMA-like bike would need a mechanical CVT to do the same)
*operation is mostly defined in software, allowing extensive customization if open source
*since power flow is managed electronically, overall energy usage can be accurately measured and recorded (think Scangauge for your bike)

Also note that HSD is superior to IMA at low speeds, exactly where bikes operate at!

[blueflame]

You kinda lost me there but it seems you have given it lotsa thought, congratulations for working on improving on the perennially wonderful bicycle.

I think the secret to the bicycles success is simplicity. Like the Volkswagen, a machine for the people.

LCD screens and high tech componentry will make a cheap reliable form of transport into a rich mans toy, and rob the direct applicability of such a design. Sort of like fitting a FM radio in your underwear.

Constantly variable transmission scooters (of which I have many) are way superior to fixed gearing but add considerable weight and generally require a more powerful engine. A couple of early honda scooters I have with fixed ratio drive weigh only 45kg( i can pick them up and swing them round just holding a handlebar grip and the rear carrier!) instead of the 75kg of most lightweight cvt scoots. Once you improve power, acceleration and top speed you will also need to improve suspension, brakes and beef up the frame. However, as the weight is less, they are still worthwhile designs.

Regen need only be connected to the brake levers as like a brake light sensor, so that as you pull more on the levers, braking and regen occur together.

As commented above, I think its easy to get lost in possibilities and forget about simplicity, of which most beautiful and practical things involve.

Like Grasshopper walking across the rice paper, not leaving a wrinkle....( just lots of rubber smoke)!!

Hahaha, my bicycle project is a 20kg chopper style bicycle with a fat back slick tire, lay down drag bike/racing bike seating and bars(to keep weight over the front wheel), and a 5-7kw dc motor direct drive to big rear tyre via a ribbed belt, no controller just 100% power with a relay and switch.

0-30mph in 1 or 2 secs? And replace the pedals with a beer keg for when i get thirsty

[elhigh]

It sounds awfully complex. And I worry that adding yet another step into the drivetrain is just one more little drain on what is a very small resource. When all you've got is humanpower for motive effort, I say the most direct transfer of power is the best option.

Have you read about the NuVinci CVT? I don't remember what the efficiency numbers are but they're pretty good, and it's small enough to be a relatively unobtrusive in-hub installation.

[MetroMPG]

Originally Posted by captainslug

Sounds more complicated than is needed.

Originally Posted by elhigh

It sounds awfully complex.

Definitely, and absolutely.

But as an engineering project, it would be fantastically interesting!

I've never heard of a "Prius" bicycle before, though I have had the chance to examine a home built "Chevy Volt" style bicycle:

- 100% electric drive, with regen
- small battery pack
- pedals connected solely to a generator

Less efficient than a regular bike, but with vastly different performance characteristics (easier to climb hills, accelerate rapidly, etc).

[blueflame]

NihaoMike: dont give up yet! The most efficient mode is gliding. An aero shell on the front(which the rider can tuck into) would yeild a considerable efficiency gain.

I heard here that someone went 82mph on a human only powered bicycle with aero's. Add electric and you've got a world beater.

My idea for lightweight cvt is 2 tapering urethane rollers on separate shafts that move together and apart, one driving the other, the second roller shaft speed changing as the mating surface move across the different circumferences of the taper...

The electric cycle IS the future. I see your in Texas. Big oil so close to home! Sweet! Perfect!

[NiHaoMike]

Originally Posted by blueflame

LCD screens and high tech componentry will make a cheap reliable form of transport into a rich mans toy, and rob the direct applicability of such a design. Sort of like fitting a FM radio in your underwear

Electronics are surprisingly cheap nowadays. The Propeller CPU is only about $12 or so even in small quantities. A text-only LCD costs as little as $8, a small video LCD maybe $25. The power electronics and motors should cost about as much as those in a regular electric bicycle. The batteries should be much cheaper as it is a hybrid so they can be smaller.

You can use the same argument as for why buy a really feature-rich cell phone when the most basic ones work just as well for making calls. Or why buy a Prius when a used Civic is much cheaper and can get just as good a MPG. Or why buy a 67" 240Hz 1080p TV when a no-frills 32" will work just fine for watching the news. Some are willing to pay for some extra features.

And of course, a hybrid bicycle is the most affordable hybrid vehicle.

[blueflame]

True. My apologies.

But I dont fully understand your written design description.

I guess a diagram would help to understand properly. Maybe that why everyone poo poohed at first.

Dont give up! Have you seen the optibike? Optibike - High Performance Electric Bikes, Finest Electrical Motorized Bike, Fastest Electric Powered Bicycle - Home

[NiHaoMike]

Originally Posted by blueflame

True. My apologies.

But I dont fully understand your written design description.

I guess a diagram would help to understand properly. Maybe that why everyone poo poohed at first.

Dont give up! Have you seen the optibike? Optibike - High Performance Electric Bikes, Finest Electrical Motorized Bike, Fastest Electric Powered Bicycle - Home

The bikes in the link are based on an IMA-like system. And they're really overpriced.

To better understand my idea, first study the HSD system used in Toyotas. Then just scale it down and replace the engine with the pedals.