inteference fit 1

[mfinke]

Dear all,

i´m doing a analysis of a pressure resistant small tube under a bending moment.

In our first design the part looks like a hollow bolt. The head of this "bolt" was pressed in another part. When now applying a bending moment to the other end the hghest stress region is the radius between head and shank diameter.

Due to the notch we decided to change the design from "bolt like form" to a hollow cylinder. The boundary conditions are the same. One end of this tube is pressed into another part and at the opposite end there is a bending moment.

My question now is:
-due to the intefernece fit i introduce a pressure to the tube on the outer diameter. The stress at the loaction where the highest bending moment acts is a combination of stress from pressure and stress from bending.

So i´m not really sure about the state of stress and the most critical loaction

Can somebody help me?


regards

MAtthias

 

[corus]

It depends what you're looking for. Under fatigue you'd be looking at the change in stress. With the interference fit you have a constant stress whereas the bending moment, presumably, is the variable stress which you'd consider for fatigue purposes.
On the other hand if you're looking at failure through yield then you'd consider the combined nominal stress, away from any notches.

corus

 

[magicme]

hello

i cant say that i know the "most critical location" for your problem, but you should at least consider this.....

there is high, very local a longitudinal stress in the tube right at the circumference where it exits the interference fit (the end wall). Roark's third edition, Table XIII, case 9 addresses this. this stress would directly add to any bending stress in the tube due to an end load (or whatever else). my *guess* is that is the highest stressed region in your problem, but keep your eyes (and mind) open.

daveleo

 

[mfinke]

Hi daveleo,

i have found in the forth edition from 1965 in table XIII case 11 something similar to my problem, but this formulas are for thin vessels. The tube i´m analyzing is thick. D=6mm; d=4mm.

In this formula there is used lambda. What is this and how can i calculate it or find it somewhere?

thank you in advance

Matthias

 

[magicme]

Matthias

the dimensions you stated make the "tube" actually a thick-walled cylinder but the Roark case i stated earlier applied to thin walled tubes....i see no direct case for thick-walls, so you will have to play around with a few cases.

Roark (3rd ed) Table XIII, case 28 shows a thick walled cylinder under external pressure....you can calculate external pressure as a function of the interference fit. then calculate the radial and tangential stresses (formulas are there)......but that does not yet give you the longitudinal stress i talked about in my first message.....thats a trick..... at this point i guess i would take the calculated thick-walled pressure and apply it to the Roark's thin-walled tube case i talked about and calculate the longitudinal stress using the thick-walled external pressure applied to the thin-walled stress formula.

also... the thin-walled tube formula i am looking at (3rd edition) is the one with the short ring that is circumferentially compressing the middle region of the tube....there are no lambdas in that equation.....all the pararmeters in the equations are given at the very top of the table.

i know this is kind of a dance around the textbooks, but thats what makes engineering fun and interesting.

daveleo