Index Home About Blog
From: jbrandt@hpl.hp.com (Jobst Brandt)
Newsgroups: rec.bicycles.tech
Subject: Re: Frame Alignment Question?
Date: 20 Jan 1999 16:19:46 GMT

Aka Tom N writes:

> My friend has a bike that pulls to the left. By pulling, I mean when
> I ride it with no hands on a straight stretch of level road it veers
> to the left.  Both his bike and my bike are identical, meaning size,
> geometry, fork, drive train, tires, seat, and color, but my bike
> rides as straight as an arrow.  The only thing we haven't checked is
> the frame alignment.  Could a tweaked frame cause this problem?  How
> do you check the frame alignment?  I have access to a surface plate
> and dial indicators...can someone explain in detail.

You didn't say whether the bicycle was new and had never fallen,
however, it is definitely frame alignment.  I assume your wheels are
in straight but even that has hardly any effect.  What is unclear, is
whether it is the frame or the fork.  An easy check for the rear of
the frame is to ride behind and see whether the rear wheel is directly
centered behind the seat tube and whether the front and rear wheels are
in plane, leaving a single track on the road after riding through
water.

For front to rear alignment, stretch a string from one rear dropout
around the head tube to the other rear dropout and measure the
distance between the string the seat tube on both sides.  This only
helps identify where the problem might be.  You should probably let a
skilled frame repair do the work.  It requires some slight but careful
bending.

Most of these "pulling to one side" effects come from a bent fork and
that is best done by removing the fork and straightening it, something
that is best done by someone with the means of measuring it.  However,
a common fix for this is to bend the fork blades sideways toward the
side to which the bicycle prefers to go.  Even this must be done
carefully, one blade at a time and in 5mm increments.  Inexpensive
bicycles are easily "fixed" this way.  The bicycle can be made to ride
straight this way but that does not mean that this was what was wrong
in the first place.  It merely means that the fork effect counteracts
whatever caused the asymmetry in the first place.

Jobst Brandt      <jbrandt@hpl.hp.com>


From: jbrandt@hpl.hp.com (Jobst Brandt)
Newsgroups: rec.bicycles.tech
Subject: Re: Hot Resetting Rear Triangle
Date: 26 Aug 1999 00:54:44 GMT

Rick Knowlan writes:

> In discussing my idea to upgrade my old, beloved lugged Reynolds
> 531d-b framed bike, a friend suggested today that I consider
> spreading the rear triangle to accept 9 speed.  I measured it, and
> it looks like it is currently 116 mm. Going to 130 would be a "real
> stretch".

> Since I plan to strip and refinish the frame, does it make sense to
> have it "hot reset" by a competent frame builder?  Has anyone tried
> this?

Don't!  If you knew how much "cold setting" a frame gets anyway when
built and how little it affects the steel for the size of these
deformations, you wouldn't even consider it.  Heating the tube to
visible red is far worse than putting a slight bend in the rear
triangle and then aligning the dropouts.  This is something done
regularly with the tools also found in bike shops.


--

Jobst Brandt      <jbrandt@hpl.hp.com>


From: jbrandt@hpl.hp.com (Jobst Brandt)
Newsgroups: rec.bicycles.tech
Subject: Re: Hot Resetting Rear Triangle
Date: 26 Aug 1999 15:19:18 GMT

Rick Knowlan writes:

> My apologies for a misleading post that did not accurately state my
> question.  Sorry I wasted your time.

> When I said "hot resetting", I did not mean heating the tubes and
> bending them.  I meant (and should have said) heating the lugs to
> loosen the brazing (or is it silver soldering?) and resetting the
> lugged joints a fraction.  That would, I imagine, require "longer"
> bridges, too. Has anyone tried resetting previously fastened lugs?
> Any success?  Would the heat required change the properties of the
> tubes?

Oops.  Even worse.  That means the position of the lugs needs to be
redone after the frame is widened (without heat).  All this torching
around the dropouts will leave you with lugs that look like an amateur
built the frame, taking forever to get the lugs right with multiple
re-heating and re-brazing.

Who started all this anyway.  I get the feeling that "cold setting =
BAD" is another one of those rumors bike shops started to keep riders
from doing simple frame adjustments themselves.  Such admonitions have
a way of running away on their own, never to be stopped.  It's like
steel frames getting soft, aging tires before use, the curse of
rotating mass, and a slew of other remotely true but factually wrong
concepts.

Jobst Brandt      <jbrandt@hpl.hp.com>


From: jbrandt@hpl.hp.com (Jobst Brandt)
Newsgroups: rec.bicycles.tech
Subject: Re: Hot Resetting Rear Triangle
Date: 30 Aug 1999 17:38:23 GMT

Troy Courtney writes:

> In my posting I clearly stated that stays should always be done one
> at a time with the bottom bracket locked down and support placed at
> the bridge junctures to prevent the bridges from being separated
> from the stays.

> If I used the methods other than what I stated and what you espouse,
> then I should just sell all my tools and quit repairing frames and
> quit building customs frames...because I surely would not be
> qualified to be doing either.

It's not that complicated and it can be done accurately without a lot
of tools other than a caliper able to reach 130mm and dropout aligning
levers, typically ones like the Campagnolo set the clamps into the
dropouts.

Placing ones foot neat the root of one chainstay, frame laying on its
side, manually bend the upper rear triangle until half the distance is
accurately achieved.  Turn the frame over and repeat so that the full
distance is achieved.  Use the dropout aligners to make the dropouts
parallel.

This assumes the frame was initially on center.  It it was not, then
this can be determined by the head tube to rear dropout string
measurement and sighting over the head and seat tube to see if they
are parallel.  A crocked frame that has several misalignments is ha
headache because one error invades the other so they are not easily
individually corrected.  But then that was not the intent of this
discussion, only hoe to spread the rear end of the frame.

Jobst Brandt      <jbrandt@hpl.hp.com>


From: jbrandt@hpl.hp.com (Jobst Brandt)
Newsgroups: rec.bicycles.tech
Subject: Re: Hot Resetting Rear Triangle
Date: 30 Aug 1999 23:30:31 GMT

Rick Knowlan writes:

>> It's not that complicated and it can be done accurately without a lot
>> of tools other than a caliper able to reach 130mm and dropout aligning
>> levers, typically ones like the Campagnolo set the clamps into the
>> dropouts.

>> Placing ones foot neat the root of one chainstay, frame laying on its
>> side, manually bend the upper rear triangle until half the distance is
>> accurately achieved.  Turn the frame over and repeat so that the full
>> distance is achieved.  Use the dropout aligners to make the dropouts
>> parallel.

> Is it possible that this procedure may result in a small radius bend with
> accompanying high stresses?

No more so that other methods one might use.  The angles involved here
are so small that you can only see the deviation if you sight along
the tube.  The way you describe it, it sounds like a huge bend.

> My LBS is willing to try cold setting my frame but has cautioned
> that frames sometimes fail during this process (a risk I'm willing
> to take provided the LBS exercises due diligence), and several have
> developed cracks that rendered them useless within a few months.  I
> don't know whether this is evidence of shortcomings in their skill,
> in the whole enterprise of cold resetting, or in their willingness
> to help me avoid buying a new frame.

They're partially right about that.  Some frames have cold joints with
no brass in them, on the verge of breaking.  I think you'll be better
off knowing about this than continuing to ride around with a crack
that's happening.  Some of these have only an external meniscus of
brass and nothing in the joint, something a nice paint job will cover.

> I have watched people bend steel tubing in fabricating shops, and
> they usually use dies that prevent the tubes from crimping and
> spread the bend over a longer radius.  Has anyone tried this?  Am I
> being overly cautious?

That's true, but they aren't bending the tube 1/2 degree or less as is
required for this adjustment.  Again, this is a subject that is blown
out of proportion by two orders of magnitude (100x), just like the
rotating mass BS that assumes enormously exaggerated accelerations.
These effects exist but do not apply here.

Jobst Brandt      <jbrandt@hpl.hp.com>

From: jbrandt@hpl.hp.com (Jobst Brandt)
Newsgroups: rec.bicycles.tech
Subject: Re: Frame Damaged?
Date: 3 Jan 2001 22:58:37 GMT

Nick who? writes:

> Can anyone tell me how I can tell if the chain stays have been bent?

Run a string from one rear dropout over the head tube and back to the
other dropout, attaching the string similarly on both ends.  The
distance to the seat tube should be equal.  If not, the dropouts must
be pulled/pushed in the direction that makes this equal doing it one
side at a time while recovering the correct dropout spacing with
the second move.

However, that said, since the rear triangle is seldom exactly
symmetrically stiff, you should already notice that the dropout
spacing is too wide or narrow if the rear end was bent sideways.  What
bothers me about this story is that the bike shops you asked didn't
demonstrate the distortion they claimed to detect or not.  They sound
like a bunch of liars.  In what Gomorrah do you live?

Jobst Brandt      <jbrandt@hpl.hp.com>


From: jbrandt@hpl.hp.com (Jobst Brandt)
Newsgroups: rec.bicycles.tech
Subject: Re: Frame Damaged?
Date: 5 Jan 2001 02:33:10 GMT

Mike Jacoubowsky writes:

> Jobst: What was left out of the original post was that this was for
> an insurance quote.  That puts the shop in a touchy area.  On the
> one hand, the frame may appear to be just fine, and the string test
> etc come out OK.  On the other, if the accident was someone else's
> fault, and there is *any* indication of something amiss (impact
> evidence on the frame etc.), they may feel safer protecting the
> customer and assuming the worst for the insurance claim.

Insurance money comes out of the same till and truth is not altered by
whose fault it is.  I hope I am not seeing a double standard here for
real and insurance damage.

> What I'll typically do is check out the frame, note the lack of
> obvious damage or alignment on the tag, but also note that there's
> no way to be certain that it didn't suffer unseen damage from the
> accident.  Is it likely?  Heck no.  But every once in awhile you do
> see something strange that can be traced back to an event that
> seemed, at the time, insignificant.

That's why we have skilled bicycle mechanics who know the difference
and give a straight answer to a straight question.  I have been
involved in enough bicycle claims against the business to know that
all the ones I came across were liars and fabricators.  They all
failed to win their plausible case that was obvious from the beginning
to any one with some experience.

> By the way, we get quite a few people wanting us to write up phony
> insurance claims for them...adding in stuff that wasn't on the bike,
> having us write that it was a more expensive model than it really
> was, or wanting us to note that the bike is unsafe to ride.  Pretty
> amazing what some people want us to "partner in crime" with them.
> Sorry, but no.  But we will make sure we don't close off any legit
> avenues of compensation.

They may do that because bicycle shops are not all on the level and
are complicit in these claims, realizing that they can be the one who
makes the repairs and sale of new equipment.  As I said, I have seen
plenty of it and the stories are often complex... too complex.

Jobst Brandt      <jbrandt@hpl.hp.com>


From: jbrandt@hpl.hp.com (Jobst Brandt)
Newsgroups: rec.bicycles.tech
Subject: Re: Frame Damaged? - The Result
Date: 5 Jan 2001 02:45:24 GMT

Nick who? writes:

> Well I tried the string method twice and came up with 2mm and 3mm
> difference, respectively.

You mean the dimension was 2 and 3mm or you measured the difference
twice and got different answers?  1mm difference is a straight frame.
2-3mm difference is OK when you realize that it's a differential where
2mm means the position is 1mm from being exactly centered.  Like a
dishing tool.

> Even allowing for error and inaccuracy of the measurement method, I'm
> inclined to think that the stays don't need any realignment.

I agree.

Jobst Brandt      <jbrandt@hpl.hp.com>


Index Home About Blog