From: jobst.brandt@stanfordalumni.org
Subject: Re: Rivnut success
Newsgroups: rec.bicycles.tech
Message-ID: <xL9ua.12045$JX2.733810@typhoon.sonic.net>
Date: Wed, 07 May 2003 15:38:05 GMT

Lindsay Rowlands writes:

> This may not be a big deal for the 'old man and the bicycle' types,
> but I've just had my first encounter with rivnuts and finally won. I
> needed to fit water bottle mounts on a frame I've set up as a fixed
> wheeler.

> I finally located some at the local aircraft maintenance hanger for
> $1 each.  For all intents and purposes aircraft bits and pieces are
> AF or non-metric.  I wasn't deterred and as it turns out 3/16 fine
> thread is so close to 5mm x .8 (most water bottle screws) that
> they're as good as interchangeable.

> After carefully measuring and drilling the frame, it was time to
> wrangle with the rivnuts.  I'd assumed it was just a matter of
> putting the rivnut in the hole and then tightening a water bottle
> bolt to expand the 'riv' part of the device.

Not a good idea!  I don't know many riders who believe that drilling a
hole in a frame tube is a reasonable concept.  I assume the rivnut was
securely tightened and painted to prevent motion, corrosion and water
intrusion.  Normally a water bottle thread is brazed or welded into
the tube, thereby becoming a structural reinforcement for the hole in
the tube.  A rivnut is not doing anything for strength and is not part
of the tube and I suspect it will loosen with time, having no preload.

Jobst Brandt
jobst.brandt@stanfordalumni.org
Palo Alto CA

From: jobst.brandt@stanfordalumni.org
Subject: Re: Rivnut success
Newsgroups: rec.bicycles.tech
Message-ID: <MDTva.13740$JX2.831196@typhoon.sonic.net>
Date: Mon, 12 May 2003 20:39:08 GMT

Mike DeMicco writes:

>>> After carefully measuring and drilling the frame, it was time to
>>> wrangle with the Rivnuts.  I'd assumed it was just a matter of
>>> putting the Rivnut in the hole and then tightening a water bottle
>>> bolt to expand the 'riv' part of the device.

>> Not a good idea!  I don't know many riders who believe that
>> drilling a hole in a frame tube is a reasonable concept.  I assume
>> the Rivnut was securely tightened and painted to prevent motion,
>> corrosion and water intrusion.  Normally a water bottle thread is
>> brazed or welded into the tube, thereby becoming a structural
>> reinforcement for the hole in the tube.  A Rivnut is not doing
>> anything for strength and is not part of the tube and I suspect it
>> will loosen with time, having no preload.

> I concur. Just recently, my Trek 1200 aluminum frame cracked around
> the seat tube at a crimped-in (Rivnut) threaded insert for the front
> derailleur mount. Had there been no insert (i.e., had the frame been
> built for a clamp-on derailleur) I doubt the frame would have failed.
> Clamp-on water bottle cage mounts are available and the safer
> choice.

This sound so much like the claims that cranks don't break and that
those that do were the result of any number of unrelated effects, like
for instance lubricating the spindle before installation.  The reason
that tale got into manufacturer's instruction sheets is because they
have a problem.  The results I see are ISIS and that Shimano is still
not satisfied with their solution as we see on the straight spline and
pinch bolts on their next generation.

http://www.lancearmstrong.com/lance/online2.nsf/Docs/5331302CD9C5EBBA86256D15006FD9B9

I am still waiting to see what they devise for crank failures at the
pedal eye, something far easier to solve than the spindle attachment.

Cranks break and holes in frame tubes, loaded in torsion and/or
bending, causes tube failures.  This is nothing new but as often
happens, R&D takes place on the customers shift.  If the stresses
around a 1/4" hole in a frame tube were visible, I'm sure few people
would even consider drilling such a hole.

Jobst Brandt
jobst.brandt@stanfordalumni.org
Palo Alto CA

From: jobst.brandt@stanfordalumni.org
Subject: Re: Rivnut success
Newsgroups: rec.bicycles.tech
Message-ID: <JJgwa.14081$JX2.846950@typhoon.sonic.net>
Date: Wed, 14 May 2003 01:12:09 GMT

Andy Dingley <dingbat@codesmiths.com> writes:

>> The Rivnut was designed for it, but the bicycle frame was not
>> designed for it. They can be installed in older steel frames.

> Anyone got any evidence of a frame that failed at the water bottle
> mounts, Rivnut or not ?

> It's like the way that thin swaged spokes don't cause wheel
> failures.  Yes, they're obviously less strong at that point. But
> that's not where the _structure_ fails.

> Never ask a metallurgist to do a structural engineer's job.

I think there is a lot of conjecture surrounding this subject and I
think it may arise from different frame tubes and different demands
put on the bicycle.  I know many riders who rode and some who still do
Cinelli frames with the Cinelli sloping fork crowns.  These are not
only heavy (solid steel) but are failure prone at the abrupt
transition from fork blade to the internal lug of the crown that
cannot be feathered to a gradual transition as an external lug can.

I broke two forks before I insisted to no longer have an internally
lugged fork crown.  That they don't break for many riders is fortunate
but the evidence is obvious when investigated.  The same is true for a
thin walled frame tube with a hole in its side.  Under torsion the
hole distorts elliptically at 45 degrees to the long axis of the tube,
the angle reversing with torque reversals somewhat like what one can
do with ones lips (upper left-lower right and reverse).

The best example would be to have a frame builder braze an "H" with
the drilled frame tube as the crossbar of the "H".  Then manually
torque the uprights of the "H" goal posts back and forth.  The hole
distortion is obvious and the cracks that result in time also.  No
doubt, many users might never achieve that condition but I certainly
did, consistent with how many of these things that "never happen" have
happened on my bicycle.  It may have a lot to do with where I ride.

Jobst Brandt
jobst.brandt@stanfordalumni.org
Palo Alto CA

From: jobst.brandt@stanfordalumni.org
Subject: Re: Rivnut success
Newsgroups: rec.bicycles.tech
Message-ID: <7ejwa.14119$JX2.849164@typhoon.sonic.net>
Date: Wed, 14 May 2003 04:03:15 GMT

Tim McNamara writes:

>> I think there is a lot of conjecture surrounding this subject and I
>> think it may arise from different frame tubes and different demands
>> put on the bicycle.  I know many riders who rode and some who still
>> do Cinelli frames with the Cinelli sloping fork crowns.  These are
>> not only heavy (solid steel) but are failure prone at the abrupt
>> transition from fork blade to the internal lug of the crown that
>> cannot be feathered to a gradual transition as an external lug can.

>> I broke two forks before I insisted to no longer have an internally
>> lugged fork crown.  That they don't break for many riders is
>> fortunate but the evidence is obvious when investigated.

> Hmmm.  My Ritchey has an internally lugged fork- not Cinelli but
> Ritchey's own unicrown-looking fork.  I'll have to keep an eye on
> that.

I don't know what shape his internal lugs have.  He is aware of such
things and may have designed his lugs with a better transition.
Cinelli ignored them, they being out of sight.  All those fancy
curlicues and shapes on the outside of most fork crowns are there in
lieu of a feathered transition.

>> The same is true for a thin walled frame tube with a hole in its
>> side.  Under torsion the hole distorts elliptically at 45 degrees
>> to the long axis of the tube, the angle reversing with torque
>> reversals somewhat like what one can do with ones lips (upper
>> left-lower right and reverse).

>> The best example would be to have a frame builder braze an "H" with
>> the drilled frame tube as the crossbar of the "H".  Then manually
>> torque the uprights of the "H" goal posts back and forth.  The hole
>> distortion is obvious and the cracks that result in time also.  No
>> doubt, many users might never achieve that condition but I
>> certainly did, consistent with how many of these things that "never
>> happen" have happened on my bicycle.  It may have a lot to do with
>> where I ride.

> Since you ride a lot and are a tall guy, on and off road and in
> mountainous terrain, you probably load the equipment more than most
> riders.  And you (as far as I can tell from the photographic
> evidence) don't buy the latest and greatest equipment every year.
> I've noticed from the photos that your bike doesn't appear to have
> water bottle bosses (nor water bottles for that matter).  Is this
> why?

No.  I used to have a bottle on there on former bicycles (brazed in
blind threads) but found that plastic water didn't taste nearly as
good as chilled drinks in grocery stores and restaurants.  Not being a
racer, I allow myself to stop for refreshments.  It seems my body
conserves water fairly well.  On hot days I stick a bottle in my
jersey pocket if it looks like to far to the next fuel stop.

> I've seen, in person and photos, cracks in bike frames that appear to
> propagate from a notch or corner.  I haven't seen one start at a round
> hole, and the lore of bike frame building seems to suggest that round
> shapes and holes are OK.  I've always wondered about that, but I've
> never seen a crack develop at, for example, a brazing vent hole or a
> hole for internal cable routing.  Sounds like you're saying, however,
> that none are good and that all can result in tubing failure (at least
> in tubes loaded in torque).

Vent holes are far smaller than a 1/4" hole for a Rivnut and they are
generally locate in a no torque zone like at the fork end or in the
"rear triangle" that is made of thin tubes because there is no
significant torque there.  The "rear triangle" is a tetrahedron that
is stressed in tension and compression.

Pedersen made a bicycle in which all tubes were loaded only in tension
and compression.  He had a fetish against torque and bending... dirty
stresses.  Most people who see his design do not recognize it for that
fetish that he pursued without compromise.  That is why the bicycle
looks so odd and of course is ungainly enough to remain a collectors
oddity.

http://www.dursley-pedersen.net/

Jobst Brandt
jobst.brandt@stanfordalumni.org
Palo Alto CA

From: jobst.brandt@stanfordalumni.org
Subject: Re: Rivnut success
Newsgroups: rec.bicycles.tech
Message-ID: <9MDwa.14329$JX2.864133@typhoon.sonic.net>
Date: Thu, 15 May 2003 03:24:53 GMT

Mark McMaster writes:

>> I think there is a lot of conjecture surrounding this subject and I
>> think it may arise from different frame tubes and different demands
>> put on the bicycle.  I know many riders who rode and some who still
>> do Cinelli frames with the Cinelli sloping fork crowns.  These are
>> not only heavy (solid steel) but are failure prone at the abrupt
>> transition from fork blade to the internal lug of the crown that
>> cannot be feathered to a gradual transition as an external lug can.

> I can attest to the potential for internal crown forks breaking, as
> I just had one break.

> The fork has a semi-sloping internal crown, had somewhere between
> 20,000 and 25,000 miles on it, and came with my Nobilette SLX frame
> bought new in 1991.  Mid-ride a few weeks ago, the bike suddenly
> developed an odd creaking, and when braking to a stop to check it
> out, I noticed that the fork seemed to bend backward much more than
> it should under brake force (and more than I thought possible for an
> intact fork).  Upon closer inspection I found that one fork blade
> had a crack running almost all the way around just below the crown,
> and was hanging on only by a sliver of metal.  The crack was just at
> the lower edge of the crown/leg brazing bond.

Not to worry, the non-scientific community of riders will tell you it
doesn't happen and when it does, it's your fault.  I'm sure with your
grasp of engineering that you can understand why it occurs and that it
is likely.  As I said, I have had to such failures, both with the
crack at the place you describe, the remaining "hinge" being at the
front on both my occurrences.  I found this interesting, because it
explains why the fork has about a 15 degree rake.  The idea being to
put the fork mainly in compression for the road roughness impact
rather than for braking loads as is often suggested.

Jobst Brandt
jobst.brandt@stanfordalumni.org
Palo Alto CA

From: Mark McMaster <MMcMstr@ix.netcom.com>
Newsgroups: rec.bicycles.tech
Subject: Re: Rivnut success
Date: Fri, 16 May 2003 07:39:28 -0400
Message-ID: <3EC4CDF0.2090202@ix.netcom.com>

jobst.brandt@stanfordalumni.org wrote:
> Mark McMaster writes:
>
>
>>>I think there is a lot of conjecture surrounding this subject and I
>>>think it may arise from different frame tubes and different demands
>>>put on the bicycle.  I know many riders who rode and some who still
>>>do Cinelli frames with the Cinelli sloping fork crowns.  These are
>>>not only heavy (solid steel) but are failure prone at the abrupt
>>>transition from fork blade to the internal lug of the crown that
>>>cannot be feathered to a gradual transition as an external lug can.
>>
>
>>I can attest to the potential for internal crown forks breaking, as
>>I just had one break.
>
>
>>The fork has a semi-sloping internal crown, had somewhere between
>>20,000 and 25,000 miles on it, and came with my Nobilette SLX frame
>>bought new in 1991.  Mid-ride a few weeks ago, the bike suddenly
>>developed an odd creaking, and when braking to a stop to check it
>>out, I noticed that the fork seemed to bend backward much more than
>>it should under brake force (and more than I thought possible for an
>>intact fork).  Upon closer inspection I found that one fork blade
>>had a crack running almost all the way around just below the crown,
>>and was hanging on only by a sliver of metal.  The crack was just at
>>the lower edge of the crown/leg brazing bond.
>
>
> Not to worry, the non-scientific community of riders will tell you it
> doesn't happen and when it does, it's your fault.  I'm sure with your
> grasp of engineering that you can understand why it occurs and that it
> is likely.  As I said, I have had to such failures, both with the
> crack at the place you describe, the remaining "hinge" being at the
> front on both my occurrences.  I found this interesting, because it
> explains why the fork has about a 15 degree rake.  The idea being to
> put the fork mainly in compression for the road roughness impact
> rather than for braking loads as is often suggested.

In my case, the blade was cracked both front and back, with
the remaining "hinge" being in the middle.  The internal
crown on this fork was a bit unusual in that it had an
additional external reinforcing point (much like the point
on a lug) on the side of the blade closest to the tire.
This reinforcing point extended to about 1/2" below the
internal portion of the lug.  When I discovered the crack,
it extended around nearly entire circumference of the blade
except for the a small portion at the reinforcing point.
The blade remained attached by only a small sliver of the
reinforcing point.

 From later examination, it appears the last portion of the
blade to go before I discovered it was a small portion
exactly opposite the reinforcing point, at about the center
of the face directly away from the tire.  In other words, it
appears that there were two cracks, front and back, that
grew until they met at one side of the fork (they were
prevented from meeting at the other side by the reinforcing
point).

Mark McMaster
MMcMstr@ix.netcom.com

From: jobst.brandt@stanfordalumni.org
Subject: Re: Rivnut success
Newsgroups: rec.bicycles.tech
Message-ID: <geaxa.14730$JX2.892086@typhoon.sonic.net>
Date: Fri, 16 May 2003 18:38:04 GMT

Mark McMaster writes:

> From later examination, it appears the last portion of the blade to
> go before I discovered it was a small portion exactly opposite the
> reinforcing point, at about the center of the face directly away
> from the tire.  In other words, it appears that there were two
> cracks, front and back, that grew until they met at one side of the
> fork (they were prevented from meeting at the other side by the
> reinforcing point).

It's good to hear that the fork blade was stressed equally by braking
and road shock, that is, it was stressed in tension forward and back
effectively the same and failed equilaterally.  My forks also had the
long tang on the inside.  Now that you mention it, it seems the
ancients knew something they didn't talk about.  No engineers analyzed
these failure back then either.

Jobst Brandt
jobst.brandt@stanfordalumni.org
Palo Alto CA

From: jobst.brandt@stanfordalumni.org
Subject: Re: Rivnut success
Newsgroups: rec.bicycles.tech
Message-ID: <gJZwa.14581$JX2.881623@typhoon.sonic.net>
Date: Fri, 16 May 2003 04:23:40 GMT

anonymous writes:

> Would it be fare to say that if your frame isn't butted, then a
> couple of holes near the middle couldn't hurt much?  Being that the
> tube there is thicker than it would need to be.

The thicker ends on tubes are for bending.  Torsion is the main enemy
of holes.  Of course the thicker the tube the safer, but then some
thick tubed frames are made of low grade tubing.  If it's a clunker, I
wouldn't worry.  It's not as though it was going to break in half all
at once.  Cracks grow slowly.

Jobst Brandt
jobst.brandt@stanfordalumni.org
Palo Alto CA