Lamalar Tearing of steel plates loaded transverse

[Dinosaur]

Hey folks. I am working on a project where the designer wants to weld a plate to a second plate to form a "T" shape and then he wants to load the "stick" of the "T" to the full capacity of the plate (e.g. this is part of an elaborate flange splice to a cross girder). I am concerned the "bar" of the "T" will tear even though the detail calls for a CJP weld. Does anyone have any references to precautions against loading a steel plate across its thickness as described here? Thanks.

 

[kelowna]

It depends a lot on the steel grade and steel thickness you are using. It also depends on the weld detail

Traditionally laminar tear was a problem on that loading situation and the detail was discouraged. Steel plate lamination processes accumulated deffects in the middle of the steel plate and created points where internal tears could create if plates were loaded transversally to the plate thickness.

These days, steel quality is much better and the issue is not so prevalent.

Myself, I do not like the detail but it is very common on cable stayed bridges, for example, where the anchor plate on the deck for the cable stays is welded to the top flange of the longitudinal deck girders. The plate is aligned with the web underneath. Clear example of tension trasfer through the plate of the top flange
(Hope my explanation makes sense).

You can order a special material quality for the plate (in Europe I think it has the letter Z) which has been checked for internal deffects. Many people, myself included, would request UT inspection of the plate after the weld is completed to ensure that no internal tears are present.

Also, talk to a welding engineer. In a previous job I was involved, the welding engineer specified stringent pre-heat conditions for that particular detail. He also mentioned that other welding engineers favour 'mail slot' details or require gouging of the material and then weld passes to feel the hole on the plate prior to welding the other plate, apparently with the idea of heat treating the area.

A lot of 'vudu' if you aks me. Ideally, avoid the detail and stop worrying

 

[Dinosaur]

Kelowna,

Thanks for your input. You have given me some interesting thoughts to bring up. I do not have the authority to prevent the use of this detail, but I have other resources to influence the decision. We have had some plates for regular plate girders that opened up after the flanges were stripped from the original 60inch-ish wide mill plate. The internal delams were very tight and were only discovered when the torch cut the edge through a delaminated area. However, my fear is thst the "bar" of the "T" will not be delaminated but only have a weak plane that will delam sometime after it gets substantial stress. I believe the "T" is formed of two 1-1/2 inch thick pieces. Not the thickest, but still thick enough to concern me. I believe the "T" is designed to be loaded in compression since it is on the bottom and this may be much-ado about nothing, but it smells bad. The "Mail Slot" approach sounds a bit extreme, but it sure eliminates the load path I am concerned about. Currently the "T" is detailed to be made from a single bevel groove weld. Lots of heat in that baby.

 

[kelowna]

Interesting about the lamination problems you describe. Never heard of something similar, although now I will keep my eye open for it

The material you will be using is not that thick. In a previous job (cable stayed bridge) the designer specified a 70 or 80 mm anchor plate (over 3")welded to a 50 mm (2") girder top flange(not sure about the actual dimensions, but around those figures. Mind you, it was Z grade steel, the welding qualification was extremely careful, and those babies got tested to death!

I did a quick google search. Maybe Z grade material is the easiest way to go if possible. It might be expensive but you do not need a lot. In this bridge, the plate section was butt welded to the rest of the flange (normal grade steel) either side.

http://www.bluescopesteel.com.au/index.cfm?objectID=A61C29C7-1513-11D4-89F600C04FCF6B8F

Good luck

 

[graybeach]

Dinosaur,
I do not like this detail either. When it is unavoidable, I have required that the plates be tested in accordance with ASTM A770 - Standard Specification for Through-Thickness Tension Testing of Steel Plates for Special Applications.

 

[Dinosaur]

Thanks folks. I think there is some useful advice here and my own thoughts have been ratified.