Welding plates to existing steel members

[David Scrivener]

I am considering repairs to a historic steel bridge. The bridge has corrosion to the beam flanges and stiffeners. How practical is it to weld new plate onto existing plate? Also what methods can be used to replace corroded rivets that are accessible from only one side?

 

[CWEng]

The first step would be to determine if the old steel is even weldable. It wasn't considered in the original design since it was riveted.

 

[David Scrivener]

Thank you for the response. The existing steel bridge dates from 1910. I assume weldability could be determined from samples, but if it was not considered in the design it is unlikely to give suitable results.

 

[BridgeSmith]

Replacing rivets can be a difficult process, even with access to both ends of the rivets. They were generally driven in hot so that they would deform to fit the holes tightly, even if the plates weren't completely aligned. Consequently, they often times cannot be driven out, and instead must be drilled oversize to remove all of the rivet.

Even if the steel is weldable, if the the plan is to keep the bridge is service, the fracture toughness of the original steel must be considered for fatigue resistance, and it wasn't typically very good. Bolted cover plates could work.

Unless the corrosion is fairly isolated, it will likely be more cost-effective to replicate the bridge using modern steel, and put the original on display somewhere.

Rod Smith, P.E., The artist formerly known as HotRod10

 

You first need to characterize the existing steel, starting with chemical analysis. (n.b. PMI/XRF is NOT chemical analysis.)
If possible remove enough sample to do a small tensile test and Charpy impact testing at the lowest anticipated service temperature (NBC will tell you that).
Any of the above findings could rule out welding.

Bigger picture though, the corrosion issue makes me nervous. That also needs to be characterized, but only with the help of a specialist corrosion engineer.

"Everyone is entitled to their own opinions, but they are not entitled to their own facts."

 

[weldstan]

I evaluated a similarly dated bridge now over the San Antonio River along the River Walk, but this is a foot bridge. Chemical analysis was done to confirm weldability. Only the bottom corroded portions were replaced and strengthened with new A-36 plates. Some of the rivets were drilled out and replaced with bolting with the bolt heads resembling rivet heads.

 

[BridgeSmith]

...with the bolt heads resembling rivet heads.

The typical twist-off high-strength bolts have a smooth domed head that resembles a rivet.

Rod Smith, P.E., The artist formerly known as HotRod10

 

[bridgebuster]

David, clarification please


On the "inaccessible" side, is there room for an ironworker to hold the head of the bolt? If not you may have to go to a blind bolt; only problem they don't have nearly the capacity of an A325 bolt. In general, rivet removal should go smoothly if you don't have too many plies; after 3 plies it can get difficult. If they can't be punched out try drilling them out but you'll go through a lot of drill bits - Hougen bits are expensive. If you can trust an ironworker with a torch it's a lot quicker than drilling.

Regarding welding, as the others said, you should take a sample. A 1910 bridge could be A7 steel, which is often weldable but you need to preheat it. Check AWS D1.1 and/or D1.5

 

[Tmoose]

" A 1910 bridge could be A7 steel..."

I'm thinking that is ASTM A7.

https://www.astm.org/DATABASE.CART/WITHDRAWN/A7.htm

https://www.hudsontoolsteel.com/technical-data/steelA7

 

[CWEng]

You should probably check out this AWS specification. It has guidance on old steels if I remember correctly:

D1.7/D1.7M:2010 GUIDE FOR STRENGTHENING AND REPAIRING EXISTING STRUCTURES