It would probably be more aero and I would bet some study of track frames could even tell you how much.

But, internal geared hubs are very inefficient compared a standard drivetrain. Also, they are very heavy.

This is a very interesting idea. I suppose it could be very efficient in a flat course like IM Florida where a bike a bit heavier wouldn't matter.

The questions I'd have are:
1) Is this actually doable? Can you adapt a rear Zipp disc and put this hub in there?
2) Is the gear ratio usable?

Also, here's a DI2 internal hub with 11 speeds - interesting!
http://www.shimano.com/....-type-.fh_road.html

And here's a road bike with it - clean!
http://www.coolhunting.com/...imano-alfine-di2.php

Cool I hadn't seen the DI2 one.

With that I'm sure someone will give this a try.

Power Tap should/could look into this. Add some strain gauges in there and you really do have a neat set up.

It not the weight that kills you, it is the extra friction of an internal hub that gets you. Think at least 5% less efficient. So would you like to loose 5% of your power for a tiny drag reduction?

You could possibly make it work on flat windless course if you were careful though. The middle gear of a internal geared hub is essentially a direct drive, so the efficiency would be the same a conventional drivetrain (maybe even slightly better because of no loss of derailleur pulleys and straight chain line). So you set your gearing up so that middle gear is the for the speed you will hold for the entire course, you would get great efficiency and a small drag reduction. Compared to simply a single speed, you would have some lower gears for starting and some higher gears if there a downhill. But, the efficiency drops when you are not in that middle gear, so you would need to be riding that one gear basically the whole time for it to make any sense.

I wonder if wheelbuilder could make a disc cover for a regular spoked internal gear hub wheel? It'd be a little different than their normal covers since IGH wheels aren't dished.

Ok, I see your point. 5% extra friction is a lot. But indeed, it gets interesting if one can set the gearing as you explained.

Any idea of the drag reduction one would get by using this and getting rid of the front derailleur?

What makes an internal geared hub 5% less eff than a standard external rear D and cassette ?

Is it due to the size and the fact that most have (3) smaller gears versus (1) larger one for each gear ratio ?

According to wikipedia, common internal hub gear systems are about 2% less efficient. But when it comes to "Advanced internal hub gear systems" such as the Alfine they are actually more efficient:


Source: http://en.wikipedia.org/...r#Advanced_hub_gears

Mostly more surface area from the gears meshing and more friction from the meshing it self. When a chain contacts a gear it has a rooler and that roller contacts the gear in almost a parallel manner.

Gears contacting gears don't have rollers and the geometry of the mating surfaces are more complex.

One similar idea that may come about (its used on downhill Mtbs already) is a gear box, which is all chain driven, just sealing in a container and can be within a frame shroud in the front triangle

---

Styrrell

More efficient than a chain drive? Don't think so. The oil bath itself adds drag for instance.

---

---

Kat Hunter reports on the San Dimas Stage Race from inside the GC winning team
Aeroweenie.com -Compendium of Aero Data and Knowledge
Freelance sports & outdoors writer Kathryn Hunter

Love this quote:

"The advent of these options make for a sleeker new school commuter unlike the clunky wire-ridden beasts of yesterday."

I like my clunky wire(cable?)-ridden beast!

---

/kj

http://kjmcawesome.tumblr.com/

Yeah, seems weird to me too. However, most internal gear hubs tend to be tested at lower wattage as they are aimed towards recreational / commuter bikes. Could an oil bath get more efficient at higher wattage as it warms up? Here's what wiki says



Here's the link to the paper they reference (I could only find this one page):
http://www.hupi.org/HuPIannounce.pdf


Thoughts?

Car mfg and motorcycle mfg have worked on geardrives for a long long time with the idea of limiting losses. Still not as efficient as a chain drive.

Maybe someone will come up with a new design to make them more efficent, but their isn't a lot of room to work with. The best chaindrives are very very efficent.

---

Styrrell

I still don't see anywhere claiming a hub transmission is more efficient than a typical chain drive. that paper still seems to say 2% less efficient for the hubs.

am I missing it?

---

---

Kat Hunter reports on the San Dimas Stage Race from inside the GC winning team
Aeroweenie.com -Compendium of Aero Data and Knowledge
Freelance sports & outdoors writer Kathryn Hunter

Let's work with a 2% difference under typical load as claimed in the paper. For an IM bike leg at an average of 300W, that would be 6W lost to transmission.

What kind of aero saving can we expect from getting rid of the front and rear derailleur plus the cassette? Could it be more than that?

I read it more like 1.8% less efficient instead. :)

There are no 90 degree turns made in the load path, so that should keep the efficiency high, so I'm not surprised that internally geared hubs aren't fairly close to chain drives. CVT's would be worse, for sure. Fluid drag from the oil is probably more than overcome by the reduction in friction, plus the bonus of minimizing wear. The bearing/seal drag from an internally geared hub is probably a bit higher, too.

Are belt drives more, same, or less efficient than chain drives? Eliminating the chain lubrication variable from *most* non-tweaker types might actually mean lower net losses.

---

The question of who is right and who is wrong has seemed to me always too small to be worth a moment's thought, while the question of what is right and what is wrong has seemed all-important.

-Albert J. Nock

full article: http://www.bikeradar.com/...ich-is-faster-36074/

---

---

Kat Hunter reports on the San Dimas Stage Race from inside the GC winning team
Aeroweenie.com -Compendium of Aero Data and Knowledge
Freelance sports & outdoors writer Kathryn Hunter

Belt drives have been shown to be less efficient, but they also are single speed unless you also add a chain drive RD or gearbox. Gearsdrives for bikes don't need an oil bath, they are pretty well sealed and don't heat up. The Oil baths in motor driven vehicles are their for cooling (you need some for lube, but the bath is cooling)

---

Styrrell

Thanks! I'm not surprised in the least by the differential.

---

The question of who is right and who is wrong has seemed to me always too small to be worth a moment's thought, while the question of what is right and what is wrong has seemed all-important.

-Albert J. Nock

That makes no sense, since the internal hub drivetrain needs the whole chain and sprocket system anyway, just the same as a derailleur drivetrain! Sure you get rid of the rear derailleur, but that is not even a big component of the losses.

With just a single CR driving a single cog you can put them much closer to the frame and don't have the FD/RD/cogset hanging in the wind. Should be much more aero, with much meaning their isn't a lot of gains to be made these days .

---

Styrrell

Agree. Gaining 6watts via an aero improvement seems plausible.

Does make servicing an interesting problem to think about. Geared hub tucked down inside a solid disc.

I have heard somewhere that the difference between the track and the road version of the Hooker was about 10W. If you substract the contribution of the brakes, it's about 8.5W more due to the drivetrain.

You can hide the chain too

---

http://cds-0.blogspot.com

This is super interesting - thanks for sharing.

So, to wrap it up:
- We have a 8.5w gain by getting rid of the FD, RD, Cassette, 2nd chainrig. Potentially a bit more if we hide the chain
- There's a 2% decreases in drivetrain efficiency dur to increase friction in internal hubs. That's 5w at 250w
- Weight: The Alfine Di2 is 1600g - but we get rid of the RD (270g ultegra Di2), FD (162g), Cassette (208g) - 2nd chain rig (150g). So about a 900g of weight penality, which is a rotational mass.


Interesting...