Not logged inAmerican Welding Society Forum
Forum AWS Website Help Search Login
Up Topic Welding Industry / Technical Discussions / Mean Stress is larger than Alternating
- - By RickEng Date 02-03-2005 19:09
We use the Goodman diagram for castings and forgings.

The main point as I understand it is the Goodman approach reduces the
allowable stress amplitude Sa in proportion to the mean stress Sm, when Sm is tensile.

In other words it recognizes that in tension, the Sm does have an effect on Sa and Sr, and it should be reduced.

In the case of AWS D1.1 "A" plate for example, Sr is 24 ksi so Sa is 12 ksi.

If an Sm of 25.5 is applied, then the Sa must be reduced from 12 ksi to 7.62 ksi. if you apply a Goodamn diagram.

I guess this is where I confess I have not seen in the AWS where Sm is
addressed like this.

Basically the AWS S-N curves are for Sr = 2 Sa, which always has Sm < or = Sa built into the curves, yes?

So this would take care of any loading that went from 0 to peak and back down, much like in bridge loading, or any harmonic load where Sm = 0, yes?

So now I am wondering what do we do with loads where Sm > Sa ?, and Sm is tensile, in a design we are working on.

Is this the point where we start applying some double modified Goodman rule for weldments?

Is there any one out there applying Goodman diagrams like this to welded designs, when Sm>Sa?
Parent - - By dongp Date 02-24-2005 18:11
Fatigue design of welded joints is typically based on stress range since the residual stresses can be pretty high. With this argument, Goodman diagram approach is usally not applied in fatigue design of weldments. I suggest you check a new website www.battelle.org/verity for some of the most recent developments in the areas of fatigue design and life prediction of welded structures
Parent - By RickEng Date 02-24-2005 18:55
Thanks very much. I will look into it. I have also found good articles by the Gooderham institute along the same line. They say that the longitudinal residual stresses are routinely close to Sy and since all of the curves are emperical, that Sa + Sm = SY has already been accounted for, and no further reductions are necessary. But again, I will look into the www.battelle.org website.
Up Topic Welding Industry / Technical Discussions / Mean Stress is larger than Alternating

Powered by mwForum 2.29.2 © 1999-2013 Markus Wichitill