Fatigue performance of crane girder connections 1

[valleyboy]

Hi all

I have a question regarding fatigue assessment for a steel structure.

I am looking at the design of crane girder to column connections. The connections are subject to fatigue check to BS 7608. I am approaching assessment using a stress spectrum defined by the building owners, using the Miners Summation method. At present, I have classified the weld details (all full penetration butt welds) and carried out a fatigue assessment. The sum of Miners summation marginally exceeds 1.0.

In view of this I was considering the use of post weld treatment (weld toe grinding) as a solution.

British Standards permit an increase of 30% of the fatigue strength for this treatment.

My question is how to go about calculating a revised number of cycles considering this increase in strength? Is it permissible to simply increase the number of cycles previously calculated by 30%, or do I need to calculate a modified S-n curve and recalculate a number of cycles. If the latter, is anyone aware of any guidance on how to assemble an improved S-n curve?

VB

 

[rb1957]

"an increase of 30% of the fatigue strength" ... that can be interpreted two ways ...
1) increase the cycle range by 30% for a given endurance (ie raise the curve, load is usually on the vertical axis) or
2) increase the endurance for a given load (ie shift the curve to the right).

i think 2) (what you suggested) is the more conservative.

i'm surprised you're not designing for infinite life.

Quando Omni Flunkus Moritati