RE: cylinder studs: torque or length?

[Ben Barkow <dr.ben@xxxxxxxxx>]

> From: "Clive Liddell" <cliddell@xxxxxxxxxx>
> Subject: RE: cylinder studs: torque or length?
>
> Ben,
> Between us tinkerers and the BMW Engineers I know who's advice I would
> follow....
>
> Just as an aside:
> When angle torquing, in many appications, the material is taken to 
> close to
> yield point where the curve of increasing stress vs increasing strain
> becomes more asymptotic to the strain axis.  This gives a more constant
> "force" on the torqued item in service.  Now, using a torque wrench to 
> get
> to the same "force" is trickey because as the material approaches the 
> yield
> point there is less increase in torque proportionally as the nut/bolt 
> is
> tightened beyond that point - making the torque method (IMHO) a bit 
> dicey
> for important places where design engineers have decreed angle 
> tightning...
>
> Regards
> Clive Liddell (only an EE so what do I know about elastic limits, 
> limit of
> proportionality, yield points etc?)

Thanks for a helpful reasoned post.

I have no doubt that in The Factory the stretch method is best. And 
under ideal textbook conditions it may often make sense under the 
shadetree (or as we urban tinkerers say, "curbside").

But in practice, I ask you to count the number of joints between the 
stud nut and the engine block (Is it 5, give or take a washer?). Now 
consider the history of pulling and stretching and heating that EACH of 
your studs may differently have had - confident you know that history? 
If you switch nuts, are they all the same flatness because with the 
stretch method, every micro-millimeter counts?

Now, do you feel confident that when you tighten to a wispy specified 
torque of 15 ft-lbs, you are properly snugging-up all those joints with 
all kinds of unpredictable new and old stuff in those joints? The 
stretch method STARTS by torquing the heads to that wispy, smeary 
figure THEN it cranks in an unvarying stretch of half a thread pitch to 
a long stud of unknown prior stretching, heating, and whatever.

At the least, the torque method, used for centuries for all I know, (1) 
uses one method instead of two-in-series and (2) allowing for 
frictional issues, pretty much speaks to the heart of the issue: force 
per se.

Ben
After 40 seasons with BMW twins, still just a PhD Human Factors 
psychologist, so what would I know about real-world treatment of "soft 
data" or the cognitions of the humans at the end of the torque wrench

------------------------------