On 7/12/2018 10:59 AM, Frank Krygowski wrote:
On 7/12/2018 11:39 AM, Jeff Liebermann wrote:
On Thu, 12 Jul 2018 00:22:57 -0700, John B. Slocomb

wrote:

On Wed, 11 Jul 2018 20:16:33 -0700, Jeff Liebermann

wrote:
Nope. I'm not trying to measure if the tubing is strong
enough.
Assuming identical lengths of tubing, I wanted to see if
the presence
of a Rivnut significantly changed the tension required
to bend or
break the tubing when compared to the identical tubing
that did not
have a Rivnut inserted. If I'm able to pull hard
enough, I should be
able to eventually break both tubes. If they break at
the same
tension, then I'll declare the Rivnuts are safe to use.
If there's a
substantial difference in tension, then I'll declare the
Rivnuts
weaken the frame.

I'd think it obvious that any hole drilled laterally into
a tube would
have an effect on the bending strength of the tube. The
question
wouldn't be whether the rivnut changed the strength of
the tube but
whether the tube was strong enough with the rivnut
installed in the
tube.

Again, I expect some reinforcing effect from the Rivnut's
clamping action. For an analogy: Did steel frames fail in
significant numbers when water bottle bosses were brazed on
and tapped? I doubt it. While a crimped-on Rivnut wouldn't
add as much strength (as a guess) I think it may add enough
to get the strength back up to that of the un-drilled tube.

Also, I'm pretty sure a bike down tube sees little if any
bending stress, except perhaps in a crash. The stresses of
concern are torsional.


Agreed.

Reading between the lines, what others are apparently
suggesting is
that even with a hole drilled into the frame, the tubing
is still
sufficiently strong to consider the bicycle rideable. In
other words,
if the drilling a hole and installing a Rivnut decreased
the bending
strength by 10%, I would agree that the hole and Rivnut
don't pose a
risk. However, if it decreased the strength by 50%, I
would consider
it a hazardous modification. I'm not too sure what to do
about
numbers in between or even if the 10% is realistic.

To complicate matters, there's the problem of the notch
required by
Rivnuts to prevent rotation. That's a stress riser by
anyone's
definition and will probably be the start of any break
during testing.
I'll try to position it where it will do the least damage.

The notch is small enough to be enveloped in the crimped
portion of the Rivnut. I doubt that it's effective as a
stress riser. Really, I doubt that it feels any significant
stress, other than compression from the clamping or crimping
action.


In practice, crashed frame tubes don't deform at the bottle
bosses. The general truism is that a brazed joint is as
strong or stronger than the steel tube so a brazed insert
poses no risk. That may not be exactly correct but we work
with it.

Some builders at the cusp of change between 'no brazed bits'
and 'braze every possible thing' fashions (like Galmozzi)
brazed bolts on the tube and so nuts secured the bottle
cage. Odd looking but worked as well as anything.

--
Andrew Muzi
www.yellowjersey.org/
Open every day since 1 April, 1971