Dish the rear? Why not offset the frame?

Quoted message said:

Has anyone ever made a frame that has the dropouts off center? If you
moved them 9mm to the right, you could use the same spoke lengths with
an Ultegra rear hub (quick and dirty calculations). The rim wouldn't be
centered on the hub, so I guess that getting it centered in the frame
would be slightly tricky, but I'm sure we could all come up with a
simple enough solution.

Maybe I'm not understanding what you propose quite right, but the right-side
(derailleur-side) spoke position is determined by the location of the
largest cassette/freewheel cog. If you move that further to the right, then
you also have to move the chainring side of the crankset further to the
right as well, creating a rather-massive offset on the right side, compared
to what's on the left. You could offset the seat tube to take care of that,
but this is heading into some very strange territory... a markedly
non-symmetrical bicycle. Since rear wheel dish has been successfully
addressed in a number of ways, I don't think it's a serious issue that's in
need of a frame redesign.

But maybe I'm just not understanding your proposal?

--Mike-- Chain Reaction Bicycles
www.ChainReactionBicycles.com

"Mike Reed" <[email hidden]> wrote in message
news:[email hidden]...

Quoted message said:

Has anyone ever made a frame that has the dropouts off center? If you
moved them 9mm to the right, you could use the same spoke lengths with
an Ultegra rear hub (quick and dirty calculations). The rim wouldn't be
centered on the hub, so I guess that getting it centered in the frame
would be slightly tricky, but I'm sure we could all come up with a
simple enough solution.

Is this a reasonable approach?

Just wondering,
Mike

Mike Reed said:

Has anyone ever made a frame that has the dropouts off center?

Yes.

http://www.surlybikes.com/pugsley.html

I think Mike is talking about building a symmetrical (not dished) rear
wheel, and offsetting the dropouts so that the chainline is
maintained.

However, I think the drive side dropout and flange could not be
relocated, since those are fixed by the cluster width and the
chainline. So to make an undished wheel the flange-flange spacing
would have to decrease, so that the vertical angle difference is
eliminated. That results in a new hub shell that is much narrower
than today's hub. Spoke tensions could be equalized, but what would
be the effect on wheel lateral stability?

The non-drive dropout could move closer to the centerline, resulting
in an assymmetrical frame. The axle could be shortened, improving
axle rigidity and reducing axle stress.

Tony Oliver discussed this idea in his great book "Touring Bikes."

I'm not a framebuilder, but would this really help?

Ken Freeman

"Mike Jacoubowsky" <[email hidden]> wrote in message
news:[email hidden]...

Quoted message said:

Quoted message said:

Has anyone ever made a frame that has the dropouts off center? If
you
moved them 9mm to the right, you could use the same spoke lengths
with
an Ultegra rear hub (quick and dirty calculations). The rim
wouldn't be
centered on the hub, so I guess that getting it centered in the
frame
would be slightly tricky, but I'm sure we could all come up with a
simple enough solution.

Maybe I'm not understanding what you propose quite right, but the
right-side (derailleur-side) spoke position is determined by the
location of the largest cassette/freewheel cog. If you move that
further to the right, then you also have to move the chainring side
of the crankset further to the right as well, creating a
rather-massive offset on the right side, compared to what's on the
left. You could offset the seat tube to take care of that, but this
is heading into some very strange territory... a markedly
non-symmetrical bicycle. Since rear wheel dish has been successfully
addressed in a number of ways, I don't think it's a serious issue
that's in need of a frame redesign.

But maybe I'm just not understanding your proposal?

--Mike-- Chain Reaction Bicycles
www.ChainReactionBicycles.com

"Mike Reed" <[email hidden]> wrote in message
news:[email hidden]...

Quoted message said:

Has anyone ever made a frame that has the dropouts off center? If
you
moved them 9mm to the right, you could use the same spoke lengths
with
an Ultegra rear hub (quick and dirty calculations). The rim
wouldn't be
centered on the hub, so I guess that getting it centered in the
frame
would be slightly tricky, but I'm sure we could all come up with a
simple enough solution.

Is this a reasonable approach?

Just wondering,
Mike

Mike Reed said:

Has anyone ever made a frame that has the dropouts off center? If you
moved them 9mm to the right, you could use the same spoke lengths with
an Ultegra rear hub (quick and dirty calculations). The rim wouldn't be
centered on the hub, so I guess that getting it centered in the frame
would be slightly tricky, but I'm sure we could all come up with a
simple enough solution.

Is this a reasonable approach?

It has disadvantages.

I have a mountain bike from 1988, built by UK framebuilder Dave Yates
for the shop team I ran at the time.

Late 80s thinking was that chainstays on mountain bikes should be a
short as possible, so this bike used an 88mm bottom bracket shell to
allow 15.5in chainstays. While we were getting silly, Dave built the
rear end offset so that the rear wheel required no dish; the rim sits
exactly between the hub flanges. That would make for a nice strong rear
wheel, we thought. It also made for wheels that were a pain to build.

As Mike has pointed out, this required the chainline to be moved further
away from the centre of the frame. No clamp-on front derailleur would
reach so Dave made an extended braze-on fitting, and we used SunTour and
Mavic front derailleurs. An insanely long bottom bracket axle was also
required; something like 127mm if memory serves, and to place even that
far enough to the right we had to mount the cups in an aluminium sleeve
and hold that in the frame with a pinch-bolt arrangement. (Imagine a
tandem eccentric without the offset.) Initially we also had to bodge up
a combination of cups with unusually deep walls and an axle with the
widest possible bearing surfaces we could find - these were the days of
loose balls in bottom brackets.

It was, to say the least, a Rube Goldberg affair.

Didn't someone (Cinelli?) try and make a dishless rear wheel by moving
the sprockets outboard of the frame with a special hub and dropout
arrangement?

Mike Reed said:

Has anyone ever made a frame that has the dropouts off center? If you
moved them 9mm to the right, you could use the same spoke lengths with
an Ultegra rear hub (quick and dirty calculations). The rim wouldn't be
centered on the hub, so I guess that getting it centered in the frame
would be slightly tricky, but I'm sure we could all come up with a
simple enough solution.

Is this a reasonable approach?

Just wondering,
Mike

Google "offset rear triangle" for mentions of a bunch of currently or
recently made bikes with this. (Surly, Vision, Orbit, Specialized, some
more...)

Widening the Q sucks - it's already too wide on modern bikes.

For bikes that want/are 135 spacing but can do without the sealing of
mountainy hubs, it works pretty well to use 130 hubs respaced to 135
along with an offset rim.

Some more immediate questions in my mind than whether offset rear
triangles on upright bikes should be more common is just how good of an
idea it is to play with moving the left flange way closer to the
center, like a few manufacturers are doing currently, reducing the
tension disparity at the cost of lateral strength, and exactly how far
it can be pushed, and also whether offset rims are ever going to get as
big as they should be.

In article <[email hidden]>,

Mike Reed said:

Has anyone ever made a frame that has the dropouts off center?

Yes. See Tony Oliver's book _Touring Bicycles_.

Quoted message said:

If you moved them 9mm to the right, you could use the same spoke
lengths with an Ultegra rear hub (quick and dirty calculations). The
rim wouldn't be centered on the hub, so I guess that getting it
centered in the frame would be slightly tricky, but I'm sure we could
all come up with a simple enough solution.

Is this a reasonable approach?

Not really. Either the wheels end up out of plane with each other,
affecting the handling, or the chainline ends up being off. 7 speeds on
a 135 mm OLN hub is practically dishless and IMHO would have been the
best solution. While I have 8 and 9 speed bikes, I have always thought
that the benefits were minimal compared to the problems they cause with
the wheel structure.

In article <[email hidden]>,

Mike Jacoubowsky said:

Quoted message said:

Has anyone ever made a frame that has the dropouts off center? If
you moved them 9mm to the right, you could use the same spoke
lengths with an Ultegra rear hub (quick and dirty calculations).
The rim wouldn't be centered on the hub, so I guess that getting it
centered in the frame would be slightly tricky, but I'm sure we
could all come up with a simple enough solution.

Maybe I'm not understanding what you propose quite right, but the
right-side (derailleur-side) spoke position is determined by the
location of the largest cassette/freewheel cog. If you move that
further to the right, then you also have to move the chainring side
of the crankset further to the right as well, creating a
rather-massive offset on the right side, compared to what's on the
left. You could offset the seat tube to take care of that, but this
is heading into some very strange territory... a markedly
non-symmetrical bicycle. Since rear wheel dish has been successfully
addressed in a number of ways, I don't think it's a serious issue
that's in need of a frame redesign.

But maybe I'm just not understanding your proposal?

That's about it, I think. Tony Oliver's solution (see his book _Touring
Bicycles_) was to leave the chainline as normal and moved the dropouts
to the left, which then placed the rear wheel out of plane with the
front. But that creates other problems.

Mike Jacoubowsky said:

Quoted message said:

Has anyone ever made a frame that has the dropouts off center? If you
moved them 9mm to the right, you could use the same spoke lengths with
an Ultegra rear hub (quick and dirty calculations). The rim wouldn't be
centered on the hub, so I guess that getting it centered in the frame
would be slightly tricky, but I'm sure we could all come up with a
simple enough solution.

Maybe I'm not understanding what you propose quite right, but the right-side
(derailleur-side) spoke position is determined by the location of the
largest cassette/freewheel cog. If you move that further to the right, then
you also have to move the chainring side of the crankset further to the
right as well, creating a rather-massive offset on the right side, compared
to what's on the left. You could offset the seat tube to take care of that,
but this is heading into some very strange territory... a markedly
non-symmetrical bicycle. Since rear wheel dish has been successfully
addressed in a number of ways, I don't think it's a serious issue that's in
need of a frame redesign.

But maybe I'm just not understanding your proposal?

Nope, you've got it about right. I guess Q-factor vs. chainline would
be the big disadvantage here.

Quoted message said:

--Mike-- Chain Reaction Bicycles
www.ChainReactionBicycles.com

"Mike Reed" <[email hidden]> wrote in message
news:[email hidden]...

Quoted message said:

Has anyone ever made a frame that has the dropouts off center? If you
moved them 9mm to the right, you could use the same spoke lengths with
an Ultegra rear hub (quick and dirty calculations). The rim wouldn't be
centered on the hub, so I guess that getting it centered in the frame
would be slightly tricky, but I'm sure we could all come up with a
simple enough solution.

Is this a reasonable approach?

Just wondering,
Mike

Road Man said:

I think Mike is talking about building a symmetrical (not dished) rear
wheel, and offsetting the dropouts so that the chainline is
maintained.

However, I think the drive side dropout and flange could not be
relocated, since those are fixed by the cluster width and the
chainline. So to make an undished wheel the flange-flange spacing
would have to decrease, so that the vertical angle difference is
eliminated. That results in a new hub shell that is much narrower
than today's hub. Spoke tensions could be equalized, but what would
be the effect on wheel lateral stability?

Isn't this what American Classic hubs do? They claim a stronger wheel
and its pros and cons have been argued previously here.

Wouldn't lateral stability be no worse (actually somewhat better) than
it is on the drive side of a normal wheel? How important is lateral
stability outside of crashing, assuming it is adequate for the normal
stresses of bicycling? What circumstances would overstress lateral
stability of an AC wheel but not a "normal" wheel?

I think the dishing for 10s/130mm spacing requires heavier rims at
higher tension with less margin for error to get the tension split
correct between the left and right sides. My intuitive feeling is that
the AC approach is preferable to the dishing, but I have never used
their hubs, so I don't have any experience with them in real wheels.

John Stevenson said:

Didn't someone (Cinelli?) try and make a dishless rear wheel by moving
the sprockets outboard of the frame with a special hub and dropout
arrangement?

Cinelli made a dishless wheel that could be used on either the front or
the rear. The special freewheel attached to a spline on the side of the
wheel, which also lengthened the distance between the dropouts. But it
was kept centered between the dropouts by a spacer on the other side.

Pictures at
http://www.classicrendezvous.com/Market_highs/Cinelli_Bi-Valent_wheels.htm

This was, of course, a solution in search of a problem, so never caught on.

--

David L. Johnson

__o | Let's not escape into mathematics. Let's stay with reality. --
_`\(,_ | Michael Crichton
(_)/ (_) |

Nate Knutson said:

Google "offset rear triangle" for mentions of a bunch of currently or
recently made bikes with this. (Surly, Vision, Orbit, Specialized, some
more...)

Widening the Q sucks - it's already too wide on modern bikes.

As Nate said, the early Vision recumbents were built this way: dropouts
offset to the right, rear wheel built symmetrically between the
flanges. On a recumbent, the chain is usually so long that the crank
can have a normal "Q" and still work properly with the angled
chainline.

This has a couple downsides, relevant only to bicycle mechanics: the
rear wheel must be checked for dish (or lack thereof) on the frame, and
there's no easy way to realign the frame if it is bent. Symettrical
frames and "dished" wheels are easily verified independant of each
other, and thus it's easy to pick up on small alignment problems.

The later Vision recumbents had conventional symettrical stays and
dished rear wheels.

Jeff

Quoted message said:

Road Man said:

I think Mike is talking about building a symmetrical (not dished) rear
wheel, and offsetting the dropouts so that the chainline is
maintained.

However, I think the drive side dropout and flange could not be
relocated, since those are fixed by the cluster width and the
chainline. So to make an undished wheel the flange-flange spacing
would have to decrease, so that the vertical angle difference is
eliminated. That results in a new hub shell that is much narrower
than today's hub. Spoke tensions could be equalized, but what would
be the effect on wheel lateral stability?

Isn't this what American Classic hubs do? They claim a stronger wheel
and its pros and cons have been argued previously here.

Wouldn't lateral stability be no worse (actually somewhat better) than
it is on the drive side of a normal wheel? How important is lateral
stability outside of crashing, assuming it is adequate for the normal
stresses of bicycling? What circumstances would overstress lateral
stability of an AC wheel but not a "normal" wheel?

I think the dishing for 10s/130mm spacing requires heavier rims at
higher tension with less margin for error to get the tension split
correct between the left and right sides. My intuitive feeling is that
the AC approach is preferable to the dishing, but I have never used
their hubs, so I don't have any experience with them in real wheels.

I've got some AC hubs and my anecdotal impression is the shallow no-dish
works pretty well. The wheels are strong, even though they might make
some cringe when looking at them. As you mentioned, the most notable
improvement was how easy they were to build. Uniform tension all around.

-Collin

Reed met D'ohBoy

Reed meet D'ohBoy

Tim McNamara said:

In article <[email hidden]>,

Mike Reed said:

Has anyone ever made a frame that has the dropouts off center?

Yes. See Tony Oliver's book _Touring Bicycles_.

Quoted message said:

If you moved them 9mm to the right, you could use the same spoke
lengths with an Ultegra rear hub (quick and dirty calculations). The
rim wouldn't be centered on the hub, so I guess that getting it
centered in the frame would be slightly tricky, but I'm sure we could
all come up with a simple enough solution.

Is this a reasonable approach?

Not really. Either the wheels end up out of plane with each other,
affecting the handling, or the chainline ends up being off. 7 speeds on
a 135 mm OLN hub is practically dishless and IMHO would have been the
best solution. While I have 8 and 9 speed bikes, I have always thought
that the benefits were minimal compared to the problems they cause with
the wheel structure.

Yeah, I feel like I can deal with a little more cadence range than
Shimano gives me credit for. Since I switched to single-speed for
off-road, life has been much simpler... and faster.

I can't wait for all the Spinal Tap jokes when Shimano "goes up to
eleven!"

Quoted message said:

Reed met D'ohBoy

He likes to ponder stuff too?

outstanding humor sense

Collin O'Neill said:

Quoted message said:

Road Man said:

I think Mike is talking about building a symmetrical (not dished)
rear wheel, and offsetting the dropouts so that the chainline is
maintained.

However, I think the drive side dropout and flange could not be
relocated, since those are fixed by the cluster width and the
chainline. So to make an undished wheel the flange-flange spacing
would have to decrease, so that the vertical angle difference is
eliminated. That results in a new hub shell that is much narrower
than today's hub. Spoke tensions could be equalized, but what would
be the effect on wheel lateral stability?

Isn't this what American Classic hubs do? They claim a stronger wheel
and its pros and cons have been argued previously here.

Wouldn't lateral stability be no worse (actually somewhat better)
than it is on the drive side of a normal wheel? How important is
lateral stability outside of crashing, assuming it is adequate for
the normal stresses of bicycling? What circumstances would
overstress lateral stability of an AC wheel but not a "normal" wheel?

I think the dishing for 10s/130mm spacing requires heavier rims at
higher tension with less margin for error to get the tension split
correct between the left and right sides. My intuitive feeling is
that the AC approach is preferable to the dishing, but I have never
used their hubs, so I don't have any experience with them in real
wheels.

I've got some AC hubs and my anecdotal impression is the shallow
no-dish works pretty well. The wheels are strong, even though they
might make some cringe when looking at them. As you mentioned, the
most notable improvement was how easy they were to build. Uniform
tension all around.

The uniform high tension makes nipple self-unscrewing from heavier riders
much harder.

--
Phil

Quoted message said:

Mike Reed said:

Has anyone ever made a frame that has the dropouts off center?

Didn't Manitou's frame designer (I can't remember his name! Mr. Manitou
:-) do this back in the '80? I've always believed that it's an idea
worth the trouble. I suspect that the crank set chain line wouldn't
have offset quite as much the right rear offset. I.E. Say 4mm < the
rear movement to the right, would probably work ok. Maybe not shift
Shimano perfectly, but close enough. In fact the rear wheel wouldn't
have to be undished so much as to make it symmetrical. Every little bit
helps. Asymmetrical rim, right stays moved out a little, right crank
set moved a little right. Use an 'E' type front derail. Why not?

Regards, John