Bolt Tensile Stress Area 1

[jtowne]

I want to specify the torque on a bolt with an axially drilled thru hole. I am subtracting the thru hole area from the tensile stress area(TSA) to get a modified TSA. Traditionally I use the equation T=kFd to determine the torque. What value of "d" do I use with the modified bolt? Is it the original major diameter or is it an equivalent diameter derived from the modified TSA?

 

[rb1957]

surely it predicts that a torque T will create a load F in the bolt ?

another day in paradise, or is paradise one day closer ?

 

[3DDave]

Not exactly - while the terms can be rearranged to give F in terms of T, that doesn't say the fastener can produce F. All it says is that F and T are proportional, which is an answer to a different question.

 

[rb1957]

i should've said a torque T will be created by a force F, or if you need (define) a force F in the bolt then you'll need a torque T to get it. and yes, it's the fastener's responsibility to carry this force (the equation doesn't make it so).

another day in paradise, or is paradise one day closer ?

 

[racookpe1978]

Only theoretically, under perfect assumed conditions.

But the perfect torque will only create a "assumed" pre-load at time_0, before the bolt and the assembled components stretch and relax.

But, a load will not create a torque (twisting motion of the bolt or nut) in real conditions. Theoretically, I assume an axial force on a bolt "could" force the nut to unscrew, but real world conditions of thread design will NEVER allow a torque strong enough to unscrew a nut or bolt.

 

[jtowne]

I want to thank you all for your insights on this topic. From Shigley "Contrary to first impressions, the subject is one of the most interesting in the entire field of mechanical design".

 

[jtowne]

I have determined the clamped material stiffness through FEA because it does not conform to the traditional conical frusta. The bolt's stiffness can be determined by k_bolt=(A*E)/L. Knowing the bolt and clamped material stiffness's I can determine the joint stiffness to be C=k_bolt/(k_bolt+k_material). Assume that I have tightened the joint to a predetermined value F which neither yields the bolt or the clamped material. Now I raise the temperature, dT. How do I determine the increase in both the bolt load and the members?

 

[rb1957]

do you have different materials, or is everything steel ?

another day in paradise, or is paradise one day closer ?

 

[3DDave]

Even if they are all the same - welcome to transient response. Usually not a problem, but sometimes ...

 

[rb1957]

at a simple level, if all the same material then not much load if you heat it. sure there'd be something if the heat is applied suddenly and excessively from the outer parts or from one side ... as you say transient effects.

another day in paradise, or is paradise one day closer ?

 

[jtowne]

Screw material is A2-70 stainless, and the clamped material is 6061-T6 Aluminum. Grip length is 3 mm. Screw preload is 2,000N.

 

[rb1957]

so the clamped material will expand more than the bolt, if I remember my thermal coefficients.

so this will increase the stretch in the bolt.

strain = c*deltaT, ... QED

another day in paradise, or is paradise one day closer ?

 

[Screwman1]

Watch out for contact bearing stress yielding when you increase the temp in that joint. Underhead yielding is often an issue with aluminum joints unless washers or enlarged heads are used.
For all of this, you would be well served to contact a fastener test lab to run some ultrasonic tension tests on your joint; that will give you actual results instead of theory.
Here is a possible source for testing:

http://pieng.com/