Backing Bar Removeal

[ninjasomke]

I have a situation where fabrication steel backing bar needs to be “partially” removed, so that field backing bars can be installed for field erecting. All welds to be CJP and “back gouge” is not possible. Are there any code rules that would prevent the “partial” removal of the shop backing bars?? AWS section 5 is not very clear on this.

 

[gtaw]

Backing shall be continuous for the full length of the joint. Splices, where required are to be complete joint penetration groove welds.

If there is an issue regarding the intent of the structural welding code, the Engineer, retained by the Owner, is the individual with the authority to add to, modify, or delete the requirements of the code.

Best regards - Al

 

[ninjasomke]

Thank you for the input. As an engineer, what is your opinion? After the Northridge earthquake I have removed them from beam flange to columen connections.

 

[Duwe6]

Having been intimately involved in in-the-field column splices and moment connections, backing bars are a lot of trouble and make the item hard to fit up. What works best IMHO is a groove-weld prep with a 'meaty' land. This land is to be fit tight to the mating member - easy for the ironworkers, as they just swing in the piece and tack it on the land. Then when the groove has a root and hot-pass, they put a fat fillet on the back side. Member stays straight. The finished weld is a partial-penn groove weld with fillet reinforcement on BOTH sides. NOTE: All decent welders will add a reinforcing fillet to groove welds in tee-joints, like moment connections between beams and columns. That reentrant corner that a weld-cap forms at the column is very bad for stress-concentration, and it naturally 'calls to the welder' to be faired in to eliminate it.

If you are worried about fatigue problems due to the partial-penn, this detail can be changed to full-penn by backgouging the land-side of groove in the field. In this case, detail the bevels to be on the INSIDE of the beam. This allows the iron-heads to easily skim-gouge all the way across the beam, and not have to 'fight' the web. Pet peeve of mine - moment connections are usually detailed to allow the easy cuts to be made in the shop, then only a small 'rat hole' weld access cope is shop-made. Takes a good hour of work in the field just to get the rat-hole recut to allow the backing bar to be 'sort of' fit. Then when that full-penn moment made that traditional way is UT'd, the tech sees massive indications at the root, all due to a less-than-perfect fit of the backing bar in the web area. To add further insult to injury, the precut bars that the iron-heads get given are the exact length as the width of the beam's flanges. This leaves the poor welder with no run-out tab so he can make the full-penn weld full thickness at the ends of the flange.

All of the above means that even if full-penn, vs. partial-penn with fillets both sides, is needed, it is faster [thus cheaper] to backgouge. And if you cannot bring yourself to detail a backgouge, at least make the SHOP cope the web so that that d*amnable bar can be tacked in place without further work while hanging up in the air trying to get the beam welded.