Freehub vs Freewheel axles

One of the articles of faith that we see repeated over and over is the statement that (other things
being the same) the axle is less likely to break in a freehub because the bearings are further
outboard thus reducing the length of unsupported axle. I submit that the length of unsupported axle
is reduced ONLY ON THE DRIVE SIDE. There is a bunch of unsupported axle on the non-drive side
especially in the case of really wide OLN tandem hubs (145mm to 160mm). I think the optimm tandem
rear hub would be constructed to provide zero dish which would most likely mean that there would be
as much unsuported axle on the non-drive side of a cassette hub as there would be on the drive side
of a freewheel hub. Granted the freehub would have less unsupported axle on the drive side.

Hubs could be made to reduce the unsupported axle length on the non-drive side by moving the bearing
outside the spoke flange on the non-drive side. Since to my knowledge no manufacturer bothers to do
this even on tandem hubs of extreme width it leads me to think that perhaps the length of
unsupported axle is of less importance than quoted wisdom would suggest.

My 1985 Santana tandem has a dishless freewheel rear wheel and in all that time it has had no
broken axles.

Bob Taylor

 

Robert Taylor wrote:

> One of the articles of faith that we see repeated over and over is the statement that (other
> things being the same) the axle is less likely to=

> break in a freehub because the bearings are further outboard thus reducing the length of
> unsupported axle. I submit that the length of unsupported axle is reduced ONLY ON THE DRIVE SIDE.
> There is a bunch o=
f
> unsupported axle on the non-drive side especially in the case of really=

> wide OLN tandem hubs (145mm to 160mm).=20

That's true, but the chain pull, which is the largest force acting on=20 the axle, is basically only
stressing the right side.

> I think the optimm tandem rear hub would be constructed to provide zero dish which would most
> likely mean that there would be as much unsuported axle on the non-drive side of a cassette hub as
> there would be on the drive side of a freewheel hub. Granted the freehub would have less
> unsupported axle on the drive=

> side. =20
>=20
> Hubs could be made to reduce the unsupported axle length on the non-drive side by moving the
> bearing outside the spoke flange on the non-drive side. =20

They could, but this would be a "solution in search of a problem" since=20 the left side never
bends/breaks.

> Since to my knowledge no manufacturer bothers to do this even on tandem hubs of extreme width it
> leads me to think that perhaps the length of unsupported axle is of less importance than quote=
d
> wisdom would suggest.

That is true as far as the left side is concerned, because of the much=20 lower stresses experienced
by the left side.

Sheldon "It's The Chain Pull, Not The Weight Load" Brown
+----------------------------------------------------+
| I=92m not convinced that this was the right time | to attack Iraq, but I would like to express
| my | support for, and gratitude to U.S. and allied | fighting forces, and to wish them all
| success. |
+----------------------------------------------------+ Harris Cyclery, West Newton, Massachusetts
Phone 617-244-9772 FAX 617-244-1041 http://harriscyclery.com Hard-to-find parts shipped Worldwide
http://captainbike.com http://sheldonbrown.com

 

I've just realized that I didn't word something too well when I was talking about moving the
non-drive side bearing outside the spoke flange. That bearing is of course outside the flange but
only just. The point I wanted to make was that it could be moved as far outboard of the spoke flange
as the bearing on the drive side but no manufacturer bothers to do it so maybe that length of
unsupported axle isn't as deadly as it's so often said to be.

Also in anticipation, yes it is true that a freehub will still have less usupported axle width in
total (drive side+non drive side) which is almost certainly a good thing. I'm just not convinced
that it is a critical thing.

Bob Taylor

 

In article <[email protected]>, Robert Taylor
<[email protected]> wrote:
>One of the articles of faith that we see repeated over and over is the statement that (other things
>being the same) the axle is less likely to break in a freehub because the bearings are further
>outboard thus reducing the length of unsupported axle.

I don't think it is an article of faith - it is an empirical observation you tend to hear from
people who have fixed lots of broken axles.

> I submit that the length of unsupported axle is reduced ONLY ON THE DRIVE SIDE.

But why should I care about the unsupported axle on the left side when people don't break it there?
Nothing to fix! Say, maybe that means we should make the axle a little thinner on the left side...

I can't name a single axle I have ever seen broken on the left side. I'm sure it's possible but it
is not a normal occurrance. Broken axle on the drive side however is quite a common thing,
especially for low quality mild steel axles which break at the drive side cone, and respaced hubs
that have extra dish to accommodate 8-speed freewheel width.

--Paul

 

>The point I wanted to make was that it could be moved as far outboard of the spoke flange as the
>bearing on the drive side but no manufacturer bothers to do it so maybe that length of unsupported
>axle isn't as deadly as it's so often said to be.

>
>Also in anticipation, yes it is true that a freehub will still have less usupported axle width in
>total (drive side+non drive side) which is almost certainly a good thing. I'm just not convinced
>that it is a critical thing.

For us big fellows who rode 6 and 7 speed freewheel designs, it is a critical thing. Those old hubs
bent plenty of axles.

While it may be possible to get by with more overhang than there is with Shimano freehub, there is
no reason to.

jon isaacs

 

[email protected] (Robert Taylor) wrote:

> One of the articles of faith that we see repeated over and over is the statement that (other
> things being the same) the axle is less likely to break in a freehub because the bearings are
> further outboard thus reducing the length of unsupported axle. I submit that the length of
> unsupported axle is reduced ONLY ON THE DRIVE SIDE. There is a bunch of unsupported axle on the
> non-drive side especially in the case of really wide OLN tandem hubs (145mm to 160mm).

Another fact conveniently overlooked by those who praise cassette hubs is that such hubs are limited
to the use of only externally threaded axles with a maximum diameter of 10mm. That is to say, the
weakest type of rear axle is the only type of rear axle you can use with a cassette hub.

Freewheel hubs are able to use axles as large as 19mm, the limitation being the inside diameter of a
splined freewheel removing tool. Since freewheel hubs can be made to use sealed cartridge bearings,
there is no need to provide preload adjustment, and the axle can be smooth on the outside. The
absence of external threads allows the axle to be much stronger in any given size and material.

Chalo Colina

 

Hey Bluto,

Can you recommend a 14mm BMX hub that will accept road freewheels, and take slotted axles (to fit
road dropouts)?

I bought an Odyssey Hazard with this idea in mind, but it's large locknut gets in the way of a
multi-speed freewheel going on there. Do certain brands of freewheels have more clearance? Also, I'm
not sure which sealed bearing hubs make an accompanying slotted axle (that will be not-threaded in
the right places).

And hey, don't Phil cassette hubs have a 15mm axle? Not exactly affordable, but there.

Thanks,

Bruce

> Another fact conveniently overlooked by those who praise cassette hubs is that such hubs are
> limited to the use of only externally threaded axles with a maximum diameter of 10mm. That is to
> say, the weakest type of rear axle is the only type of rear axle you can use with a cassette hub.
>
> Freewheel hubs are able to use axles as large as 19mm, the limitation being the inside diameter of
> a splined freewheel removing tool. Since freewheel hubs can be made to use sealed cartridge
> bearings, there is no need to provide preload adjustment, and the axle can be smooth on the
> outside. The absence of external threads allows the axle to be much stronger in any given size and
> material.
>
> Chalo Colina

 

Chalo Colina wrote:

> Another fact conveniently overlooked by those who praise cassette hubs is that such hubs are
> limited to the use of only externally threaded axles with a maximum diameter of 10mm. That is to
> say, the weakest type of rear axle is the only type of rear axle you can use with a cassette hub.

That's as may be, but in practice, axle failures in cassette Freehubs are virtually unknown, so
there's no need to make them thicker!

If I replaced the wooden legs of my kitchen table with a solid reinforced concrete pedestal, the
table would be much stronger, probably able to sustain loads of several tons. However, since I went
on my diet, I rarely load the table with more than half a ton of food, so the wooden legs are more
than adequate.

Sheldon "Enough Is Enough" Brown +---------------------------------------+
| Whatever became of eternal truth? |
+---------------------------------------+ Harris Cyclery, West Newton, Massachusetts Phone
617-244-9772 FAX 617-244-1041 http://harriscyclery.com Hard-to-find parts shipped Worldwide
http://captainbike.com http://sheldonbrown.com

 

On 24 Mar 2003 17:48:30 -0800, [email protected] (Bluto) wrote:

>[email protected] (Robert Taylor) wrote:
>
>> One of the articles of faith that we see repeated over and over is the statement that (other
>> things being the same) the axle is less likely to break in a freehub because the bearings are
>> further outboard thus reducing the length of unsupported axle. I submit that the length of
>> unsupported axle is reduced ONLY ON THE DRIVE SIDE. There is a bunch of unsupported axle on the
>> non-drive side especially in the case of really wide OLN tandem hubs (145mm to 160mm).
>
>Another fact conveniently overlooked by those who praise cassette hubs is that such hubs are
>limited to the use of only externally threaded axles with a maximum diameter of 10mm. That is to
>say, the weakest type of rear axle is the only type of rear axle you can use with a cassette hub.
>
>Freewheel hubs are able to use axles as large as 19mm, the limitation being the inside diameter of
>a splined freewheel removing tool. Since freewheel hubs can be made to use sealed cartridge
>bearings, there is no need to provide preload adjustment, and the axle can be smooth on the
>outside. The absence of external threads allows the axle to be much stronger in any given size and
>material.
>
>Chalo Colina

Shimano Freehubs limit one to a 10mm axle because they bolt the freewheel assembly to the hub shell
with a 10mm a/f hex key. The 10mm restriction is not inherent in the freehub idea, and lots of
manufacturers use bigger axles, eg 19.5mm Chris King, any 12mm bolt through DH hub. It is also
possible to put the bearings right next to the dropouts, and even to put another bearing at the
flange position too, there are 3 and 4 bearing hubs out there; my Joytech singlespeed disc hub is
just one example. But, as everybody else has already pointed out, a 2 bearing freeehub with the
driveside bearing under the 3rd sprocket and the non drive side bearing just outside the flange, and
with a 10mm threaded axle, has proven to be adequate for solo road and XC bikes, which are the high
volume classes Shimano is interested
in. Freeriders, downhillers, dirt jumpers and tandemists who find 10mm axles a problem are well
catered for by specialist manufacturers.

Kinky Cowboy

*Your milage may vary Batteries not included May contain traces of nuts.

 

"Bruce Lange" <[email protected]> wrote:

> Hey Bluto,
>
> Can you recommend a 14mm BMX hub that will accept road freewheels, and take slotted axles (to fit
> road dropouts)?
>
> I bought an Odyssey Hazard with this idea in mind, but it's large locknut gets in the way of a
> multi-speed freewheel going on there. Do certain brands of freewheels have more clearance? Also,
> I'm not sure which sealed bearing hubs make an accompanying slotted axle (that will be
> not-threaded in the right places).

I have run into this issue before, and resolved it in one of two ways:

1) Turn or grind the too-large locknut to an acceptable size.

2) Make my own axle assembly (still easier than making my own hub).

The quick-n-dirty way to get that locknut down to size is to affix it to an axle, affix the axle
to a drill, and spin the sucker while holding it against a bench grinder. Since many hand drills
max out at
3.7mm,using a 1/2" flat head bolt with a 1/2" nut to bind the locknut up against the tapered
underside of the head might be a better approach.

The tasty way to solve the problem is to make a stepped shaft
3i(1/4") on the freewheel side end and 15mm through both bearings. Put a M14x1 thread on the left
end only as far as the locknut protrudes and no farther. Drill and tap both ends M8x1.25. Make
a tubular sleeve that corresponds to the width separating the inner bearings. Install this axle
and use 8mm bolts and washers to fix the axle in the frame.

Chalo Colina

 

"KinkyCowboy" <[email protected]> wrote in message
news:[email protected]...
> Shimano Freehubs limit one to a 10mm axle because they bolt the freewheel assembly to the hub
> shell with a 10mm a/f hex key. The 10mm restriction is not inherent in the freehub idea, and lots
> of manufacturers use bigger axles, eg 19.5mm Chris King, any 12mm bolt through DH hub. It is also
> possible to put the bearings right next to the dropouts, and even to put another bearing at the
> flange position too, there are 3 and 4 bearing hubs out there; my Joytech singlespeed disc hub is
> just one example. But, as everybody else has already pointed out, a 2 bearing freeehub with the
> driveside bearing under the 3rd sprocket and the non drive side bearing just outside the flange,
> and with a 10mm threaded axle, has proven to be adequate for solo road and XC bikes, which are the
> high volume classes Shimano is interested
> in. Freeriders, downhillers, dirt jumpers and tandemists who find 10mm axles a problem are well
> catered for by specialist manufacturers.

The bearings are in the wrong place to start with. They should be mounted in the frame.