I never criticized your point of view. Cynikal said "I think shimano has the splines patented and most wheel makers pay a royalty. Don't quote me on that. " to which you replied "I am not sure, but I think you are wrong about the splined interface for cassettes being proprietary."

I said that it was proprietary. Being a de-facto standard and being proprietary are not necessarily mutually exclusive (though usually they are).

I could be wrong, but I'm pretty sure there is no ISO (or similar) standard for Shimano splines.

Case in point
http://ridemonkey.com/forums/showthread.php?t=111155 specifically, http://ridemonkey.com/forums/showpost.php?p=1467124
Single speed cogs designed to fit king splines that need filing to fit some other hubs.

Shimano don't hand out specs so competitors can clone their successful products.
It's the same in practically every industry. The status quo don't help the emerging companies that are out to take their lunch. As a result many emerging companies reverse engineer by default because they usually don't have the language skills. experience or track record to do it any other way.

Since we don't even know what tolerances the spec calls for and it's likely different people are building to their own tolerances how can it be a good idea to use this system in a situation that requires a precise fit?

I like the splined idea, but since no one makes it yet...

Bolt-on disk is my solution

splined
 

been done...


http://www.bikeforums.net/showpost.p...7&postcount=32

http://www.bikeforums.net/showthread.php?t=274373

I should also point out that some trials bikes have fixed-gear cassettes. The freewheel is in the bottom bracket.

Not exactly a "solution" to this "problem" but it is another splined fixed-gear system.

^yes but again, since there is a freewheel in the bottom bracket, the force on the cassette is only in one direction.

aluminum splines will develop back and forth play. that's why this design isn't used.

The eno is the same setup as Miche and we know what the problems are there.

The PMP costs a mere $350 or so, and has all the problems we've been discussing here.

Yeah, I know. Just thought it was interesting, since we are on the topic of non-threaded fixed hubs.

i'm with you now, it is an interesting design. got any idea as to the reason for it?

the one guy i've talked to said that fixed rear wheels are cheaper than freehub rear wheels and since they taco wheels all the time it just makes more sense to have the freewheel in the bottom bracket. i dunno though, that seems like a lot of effort and incompatible design just to save a few bucks.

Both the miche and eno have a type of spline, that is about all that is the same.
PMP, I know all that. Reply was 4 willfcc.

I've heard it's so you can shift without pedaling. You can do that with an internally geared hub, but those would be pretty expensive.

Even more so why I like my Surly version...
Less than $120. ENO splined version will probably be twice that.
Cog comes off with a 6mm wrench. ENO will still need lockring wrench to remove cog.
ENO Cogs will probably be $30+, less than Boone (I've been waiting since April for THAT to arrive).
No proprietary parts.

Durability? I'm just one guy, so testing could take a while.

It's all about control on the rear wheel. It allows your freewheel pawls to be engaged / quicker engagementwhen you're on the rear wheel with the brake locked. Thats is all. Front freewheels on trials bikes are an old trick, have been around for years. I have an old 11t Monty fixed cog somewhere from my trialsin days.

Can you still get the kogswell cogs like you used, though?

An elegant execution.

Does Kogswell still make bolt-on cogs?

-Rob.

The supplier of the un-drilled cogs stopped making them (due to a change in the BMX world, I think). But we've found another supplier and we're back on. This time with 18s and 20s.

Still some 18s left. And lots of 16s.

How do we get them? Nothing on the web site that I can find.

I continue to wonder why this is. I wonder if manufacturers did to see if people would make up reasons for it. I'm not criticising anyone, since I don't know the answer either, but nothing I've heard so far washes:

But BMX freewheels will thread on to the same hub as fixed cogs, and in any case, tacoing the wheel will kill spokes and rims, not hubs.

Someone once did this along with a deraileur so you could shift while coasting, but it didn't catch on because it adds a small amount of drag to solve something nobodoy considered a problem. In any case, we're talking single speed trials bikes in this case.


But there's no difference. If there's a freewheel in the drivetrain anywhere, you can rotate the pedals backward, even with the brake locked. The only difference is whether the chain moves when you do it.

This leads me to grudgingly accept the only explanation I've heard that makes even a little sense: If you've landed on an obstacle such that you are sitting on top of your chain and it can't move, a forward freewheel will still let you rotate the cranks backward.
I can't really get my brain around why that's advantageous enough to make it all worthwhile, so I'm not entirely convinced. But it's the only thing I've heard that relates to the actual difference between the freewheel positions: the cranks can move when the chain can't.

OK, I believe you... but what tool is used to put a square end toward on a cut spline at the flange end of things? Only machine tools I've familiar with would leave some kind of radius at the end of the cut, which would only work with corresponding radius on cogs, not at all the standard cogs OP is talking about. Although if the hub was cut with a shoulder to stop the cog against the hub before the radius end of the spline starts...

Now that I've read through this, the easiest solution would be if someone made a SS specific hub with ISO disc brake mounting. As a stop-gap, any cog mfg. could make a killing simply by including holes to disk-brake standards. Kogswell might be close, but even better if the holes matched those on disk brake rotors exactly...

Aluminum splines should not be used.

Splined solutions can be very robust to indestructible. It all depends on the implementation. As an example, there's a splined interface all you guys use every time you pedal, and that interface takes the strongest torque in the drivetrain. It's the crank-to-axle, which can be square taper (a generalized form of splined interface), Octalink or some ISIS (and more, in the BMX world).

FWIW I've heard of people developing play in ISIS cranks when using them on the street.

True, so have I. But I doubt it's inherent to the ISIS spline interface per se, otherwise all ISIS cranks would show the same problem. That's what follows logically. My totally uninformed and purely speculative guess is that those ISIS cranks were not installed quite right (not inserted correctly during installation).

If you look outside of bicycles for a minute, splines, keyways, shear bolts, and square shafts are all pretty standard design features, but I've never seen a threaded interface being used to transmit torque around the same axis as the threads, especially a reversing torque. It's completely illogical.