Here's my 2 shillings (farthings?) worth:
While torquing a fastener too close to the yield point may be detrimental,
fatigue life probably goes down if the fastener is undertorqued as well. You
do not want the fastener system be stetching and unstretching under load
cycles, as this cycling generates fatigue problems, it has to be clamped and
stetched enough to keep everything still. Think TR6 flywheels and Kas'
latest book.
I always thought the primary benefit of lubing the fastener was to get a
more accurate torque reading in the absence of measuring stretch, although
it make sense the target would have to be reduced a bit. Sticking friction
would have a big impact on torque reading in my mind. I would rather be
slightly overtorqued but know exactly where it is than be undertorqued and
thinking everything was great. Cheap torque wrenches are probably a bigger
source of variation than anything else I am guessing.
Mark
72 TR6
----- Original Message -----
From: "Michael D. Porter" <portermd@zianet.com>
To: "Jack W. Drews" <vinttr4@geneseo.net>
Cc: <fot@autox.team.net>
Sent: Tuesday, March 16, 2004 7:12 PM
Subject: Re: lube torqued fasteners?
> Jack W. Drews wrote:
>
>
> > result in a higher tension in the bolt. Furthermore, good practice on
> > setting a torque value is to stress the bolt to 80% of its yield.
>
> Most everyone has answered the question about lubed vs. unlubed torques,
but I think I need to clarify this business of
> torquing to 80% of yield. I'm sure there are exceptions, both in specific
instances and by manufacturer, but the GM
> engineering standard for fastener preload, as I recall, is 80% of _proof
load_. Proof load is defined by GM as 75% of
> yield. That means that GM sized and graded fasteners to nominally 60% of
yield (80% x 75%).
>
> Which means, simply, that once the maximum load is determined, the
fastener must be able to provide a clamping preload
> to exceed the load by a suitable safety factor (about 1.5) _and_ the
fastener itself had to be sized/graded to reach
> that figure at nominally 60% of yield strength.
>
> That may sound like overkill, but it resulted in very few fastener
failures, if good procedures were followed (I'm not
> saying that this engineering procedure protected the product from assembly
line errors).
>
> Structural and civil engineering standards are to preload to 2/3rds of
yield. I think the difference is that the
> accelerations of dynamic loads upon fixed structures is relatively smaller
than in automotive applications.
>
> Preload is preload, whether it's provided by a big fastener or a small
fastener--the difference is in the preload safety
> factor. Although I can't come up with a definitive source for this feeling
of mine, I think fasteners subject to
> cyclical loads are more likely to fail from fatigue the closer they're
preloaded to their yield point. That's the
> distinct advantage of ARP fasteners--their yield point is so high that
they can provide a superior clamp load at a
> relatively low percentage of yield strength. As well, because of Young's
modulus, they provide that higher clamping
> force with small elongation--an important consideration, I think, for
fatigue strength.
>
> All that said, I have no idea about what engineering standards were used
by Standard-Triumph to calculate the clamp
> loads and torques for the wet-sleeve engine. Without a good idea of the
grade of fastener being used, it's not easy to
> determine. Would be fun to take the standard torque/clamp load formula and
work backwards from the torques provided in
> the Triumph manual. For what it's worth, the factory manual doesn't
mention lubricating the stock cylinder head studs or
> nuts, or the rod bolts, before tightening.
>
> Cheers.
>
> --
> Michael D. Porter
> Roswell, NM
>
> Never let anyone drive you crazy when you know it's within walking
distance.
|