feasibility of pedal-to-generator-to-motor chainless bike?

Hey guys, ballpark question here:

If I were to hook up a bicycle crankset to a generator, and connect the generator directly to a hub motor such that human pedaling would (almost) directly power the hub, what kind of efficiency loss am I looking at? I realize that I would be paying an efficiency penalty twice (once at the generator, another at the motor), I'm just unclear on how efficient a typical hub motor actually is.

Basically the idea is to drive the bike with human power only, but with no chain, and no perpendicular drive shafts.

BUT WHY?

Here is the pie-in-the-sky potential application for such a silly idea:

I ride a "fat" bike (a la Surly Pugsley) in the winter, with tires that are almost 4" wide, so you can float on snow better and steering is somewhat sane.

There are two common problems with snow bikes as I see it:

1) The bottom bracket shells are 100mm wide, and the increased Q factor hurts some riders' knees.
2) Sometimes I wish for even more flotation.

I had a silly idea that removing the chainstays (yes, the rest of the rear half of the frame would need to be built to compensate) would allow me to either put the cranks closer to the sides of the rear tire, or increase the overall width of the tire... but the chain would run into the tire. So, use some alternative to a chain drive.

I know it sounds crazy, I'm just looking for numbers. There is also the idea of long chainstays and a transaxle, which has been realized already in the form of a Hanebrink bicycle.

Actual numbers would depend on your specific generator and motor, etc.

I think there is a thread about doing something like this here on ES already, but I might be remembering one on DIY Electric Car forum instead; it was from a couple years ago.

It is possible to do, however.

In theory, you could even make a synchronous setup by using two identical motors, one as generator and one as motor, directly connecting their phase wires so that turning one would cause the other to turn at the same time. Might not work under all conditions as expected, but would not require a controller. Have to gear the pedals to gen-motor low enough that it spins the cranks a decent speed to make it much easier to turn the rest of it, though.


There is a project (by I think ECG) here on ES about a pedal generator that does help charge a battery, that then runs an inverter that runs a motor.

There's at least one other that either does or was going to run something like you want, but I can't remember whose it was.


Another solution is a tophat adapter, to offset a chainring out past the fat tire.

I would think it would be sadly inefficient. It would be the product of:

Generator Eff X Battery Charge/Discharge Eff X Motor Eff

To guess, the numbers for even a good setup might be in the order of:

85% X 97% X 85% = 70%

The battery in this example is the buffer that makes it work without special components that may not exist It adds a bit of weight but doesn't affect efficiency much...

Right on. Thanks, guys!

IIRC, ES member Fitek made a serial, and LOWRACER's bike may be serial.

Fitek uses a serial drive, but...he also runs down a battery as he rides. By that I mean, he can ride his bike by battery alone without moving the pedals, and he can also pedal while he is at a stop to add some charge to the pack. Since humans output 100W-200W, the entire time you are drawing 500W-1000W from the system, the battery pack is slowly draining even if you pedal continuously.

With electrical generation, RPMs count. if your generators (legs) are running 80-RPM, rear wheel performance will be a serious let-down for you (without a battery as a storage). Even if you gear it up (no chain to the rear wheel, but a short chain from BB to nearby gen) you will quickly reach a point where the gen is spinning faster than 80-RPMs but its very hard to turn. There is no "coasting" with generation, if you are not pedaling, you are not generating.

I think your best bet would be a serial hybrid like the first hybrid car the 1899 Lohner-Porsche.

Primary power provided by batteries with the engine (you) driving a generator to charge them or provide a power boost.

I think the new Fisker Karma is a serial hybrid. The first mass-produced serial hybrid since the Owens Magnetic.

I've thought about a series hybrid for a long time. Years ago I read an article about a proposal for recumbents being that you could eliminate the mechanical transmission for an electric transmission.
They've been built:
http://www.hupi.org/HPeJ/0015/0015.html

They state that technology has progressed far enough that now you are able to replace the mechanical tranny with electric and it will weigh less?

If your translation skills are good enough... I'm pretty sure there's a series hybrid being discussed AND being ridden at the first of this video.
I'd sure like to know some of the details of the bike:
[youtube]UzGQ8mPUN-A[/youtube]

the efficancy compared to a chain drive would be crap. 50-60% maybe, compared to 95-97% for a chain drive. The most efficent use of your leg power is to drive the wheels directly. converting your mechanicle energy to electrical then back to mechanicle is never going to be as efficent.

The article by Dr. Fuchs makes some good points, and seems to be well researched. Of course I'm the type who almost falls for all that overunity crap, so who am I to judge?

I think Serias Hybrid human/electric can be very useful - it would address your problem, and might make drive trains on electric assist recumbent bikes & folding bikes easier to design. Who here hasn't at some point drawn up complex assemblies with double freewheels on a jack shaft just to combine human plus e-power. As someone who commutes on a train with a folding bike, I'd definitely consider a reasonable ncrease in cost, weight, and inefficiency if it meant I could ditch the chain.

Not long ago the standard advice regarding regeneration was "forget about it - not worth the cost and complexity. Just add another battery if you want more range". But nowadays that seems to be changing. Technology has improved. Series Hybrid drive trains probably fall into a similar class: "It will never be as efficient as a chain. If you want to do it, plan on building in more room/cost/weight of extra batteries to make up for the efficiency loss". For some, that's not adeal breaker.

What keeps me from considering a series hybrid drive more seriously is the generator part of the equation. There's very little information out there about Series Hybrid drive trains, so I think if someone wanted to build one, they would be pretty much on their own to design the generator and generator controller.

Actually I think series hybrids are far simpler to design than the current split-hybrid stuff that's out there now. No funky electric/mechanical transmission linkages to worry about.

Take an electric car add a generator, simple!

Or

Take an e-Bike add a generator!

The hard part is figuring how to connect a generator or alternator to the crank.

Earlier thread:
http://endless-sphere.com/forums/viewtopic.php?f=2&t=4711&start=0

Even if your motor eff is >95%, the generator isn't going to be much greater than 50%. This is why commercial production of power (even nuclear) is about 35%. Steam turbines aren't very eff either, but consider the most eff commercial power generators are co generation of Natural Gas. Where the NG is burned through a gas turbine, then heat from the discharge is used to run boilers that in turn run steam turbines.
Weak point is going to be an eff generator. Use of hydraulics can be as high as 99%, much better avenue IMO.

zap said:

I'm pretty sure there's a series hybrid being discussed AND being ridden at the first of this video.
I'd sure like to know some of the details of the bike:

Click to expand...

That is interesting...

MattyCiii said:

What keeps me from considering a series hybrid drive more seriously is the generator part of the equation. There's very little information out there about Series Hybrid drive trains

Click to expand...

I agree.

I'm sure mat h physics is right and the efficiency will be low but if the right generator could be found then it would seem to me that folders, recumbents, velomobiles, etc. might be simpler.

Lock, that doesn't look like a video capture, I looked at both sites referenced on the youtube video but could never find anything. Where did you find the pic?

zap said:

Lock, that doesn't look like a video capture, I looked at both sites referenced on the youtube video but could never find anything. Where did you find the pic?

Click to expand...

Sorry. Here:
http://www.flickr.com/photos/24717883@N00/sets/72157600301113889/

Not seeing any markings:

I don't know, but looking at "img_0251" from that flickr site, it looks more like just one long-ass shaft drive than a generator/motor setup.

I can imagine building a Series Hybrid bike using an air core axial flux alternator like the wind power guys build DIY style. Completely enclose it in a big honkin' bottom bracket (like 9" diameter big). Lots of magnets around a large diameter would mean better power at human capable cadence, yes? Rectify the 3 phase output to be DC input paralleled with the main battery as input to the controller. Or even design a generator with more phases - some windmills based on Fisher & Paykel motors rewire their stators as 7 phase (http://www.thebackshed.com/windmill/articles/coglessFP.asp), reducing cogging and upping the voltage.

MattyCiii said:

I don't know, but looking at "img_0251" from that flickr site, it looks more like just one long-ass shaft drive than a generator/motor setup.

Click to expand...

I think you nailed it Matty!
Look at this picture of a shaft drive from wikipedia.


Might be a long flexible shaft???

I was thinking air core axial flux also... it would still need to be geared up but it would be smooth compared to the F&P. They take a lot of work to ease the cogging but do pack a punch.

I think if U really want to make a human input system, look in to the ZTR riding lawnmowers. Honda also made an infinite ratio motorcycle which used six pistons on a swash plate for the hydraulic pump and motor.
Been trying to come up w/ a DD hydraulic hub motor design n my head. One side of the axle would be input, the other output. Use of an R/C motor to drive the hydraulic pump would be eff. An accumulator would act like a battery, by storing power (the same thing is used on home well systems, just larger). Eff of such a system would be much greater than electricity alone. Regen braking would charge accumulator w/ pressure and be 95%+ eff.

PS Why do they use a double geardown (jackshaft) when applying R/C motors?? They should just be using a worm gear, IMO(and mech eng exp).

http://forums.hydraulicspneumatics.com/eve/forums/a/tpc/f/8621030121/m/3841006821

seeking the same answers as the original author, I found this thread...

To those arguing it can't work. see the Mondo Footloose
http://www.mandofootloose.com/en/

Wish I could find more information on that generator in the bottom bracket on that recumbent boom...it just what I'm looking for...still googling

fyi future googlers