Splicing Universal Beams

[AussieAnton]

Hi All,

Has anyone got any experience with splicing univeral beams together (end to end). I've found some dodgy sections in a structure and was planning on getting the suspect sections cut out and replaced.

My initial thoughts were simply a straight cut with full penetration butt welds all round but someone has asked whether we should be offsetting the cuts (and welds) on the top and bottom flanges.

Also, any thoughts on weld specs? Is low hydrogen required? MIG or stick better?

Thanks
Anton

 

[Tmoose]

It could depend a lot on the loading.
If the joint falls at a point of zero moment, only shear exists, and splice plates on the web are most efficient.
If the loading is unknown, duplicating original strength and stiffness with low stress concentration takes some care.

Welding straight across flanges is generally to be avoided.

Dan T

 

[paddingtongreen]

Loading condition is paramount. I have no problem with full penetration weld going across the flange, especially if you UT or RT them. I would weld the web first with a single sided weld with a backup strip, then the same with downhand welds to the flanges. You will have to figure on a rathole at the top and bottom of the web.

Michael.
Timing has a lot to do with the outcome of a rain dance.

 

[hokie66]

Michael is correct. Full penetration welding will restore the capacity of the beam. That said, I prefer splices to be away from the maximum moment location.

 

[AussieAnton]

Apparently the requirements for spicing of channel sections used for truck chassis is that they either cut diagonally (when viewed in elevation) or step the cut so that the weld on the top flange and the weld on the bottom flange are offset from one another.

Anyone ever come across this? I've tried to think what the benefit is of doing this, but for simple bending, can't think of one.

Thanks

PS: What are UT and RT short for?

 

[metengr]

UT - ultrasonic testing
RT - radiographic testing (aka x-ray).

 

[AussieAnton]

Ahh, thanks.

 

[Tmoose]

Is your application a truck frame, or a stationary structure subjected only to static loads, or a stationary structure subjected to dynamic loads, or something else altogether?

When loads are significant and dynamic, as in on and off-road machinery, welds across flanges in tension are rare. I think this is true of reinforcement or patch plates (often diamond shaped or handsomely sculptured) as well as spliced beams.

http://img.directindustry.com/images_di/photo-g/wheel-excavator-292936.jpg

http://backhoe-excavator.com/wp-content/uploads/2010/08/Cat-Excavator.jpg

I suspect it is in part due to issues of stress concentration and material non-uniformities in the weld and HAZ.

This tech note points out the performance trade offs between "strength" and elongation of welds whether loaded longitudinally or perpendicular, even in statically loaded applications.

http://www.jflf.org/pdfs/papers/design_file2.pdf