Combined Stresses In Weld Joint Of Flange And Web

[sanjaychetan]

IN PLATE GIRDER DESIGN OF WELD JOINT BETWEEN WEB AND FLANGE,VARIOUS STANDARDS AND BOOKS LIKE GAYLORD,MUKHONOV ETC REQUIRE THE JOINT TO BE CHECKED FOR ONLY SHEAR FLOW AND VERTICAL SHEAR DUE TO ANY CONCENTRATED LOAD FROM FLANGE TO WEB(IN CASE WEB IS NOT MILLED ).MY DOUBT IS THAT THE WELD WILL ALSO BEND IN THE SAME RADIUS OF CURVATURE AS THE WEB TOP OR FLANGE'S WEB SIDE SURFACE. SO BENDING STRESSES SHOULD BE INTRODUCED IN WELD AS WELL.SO WE SHOULD CHECK THE WELD FOR COMBINED BENDING AND SHEAR STRESSES.BUT THIS IS NOT REQUIRED AS PER STANDARDS. I GUESS I HAVE GOT IT WRONG SOMEWHERE BUT CAN'T FIGURE IT OUT. CAN ANYONE PLEASE ENLIGHTEN ME ON IT AND SUGGEST SOME REFERENCE?

 

[flyby911]

it's not necessary that the weld suffers bending in this case..just like the RFT bars in an RC beam..
the beam bends but the RFT bars suffers a tension force only and not bending..
the flange plate bends while attached to the web plate so they bend together...what prevent slippage between them is the friction force between them that is transfered through the weld as a shear force parallel to the flange plane (that's called the shear flow)

 

[shin25]

The shear flow in the beam happens due to bending moment. The weld should be of sufficient capacity to handle the resulting tension and compression forces from bending moment.

 

[Lion06]

The welds will only resist the shear flow. You design the section to resist the bending moments. The weld resists the shear flow to ensure that the section acts compositely. There is no need to include the bending moments in the weld check - check shear flow only.

 

[miecz]

sanjaychetan-

As you say, the whole section takes on a curve due to the bending action, and the weld takes on the same curve, causing some bending in the weld. The bending stress, however, is very small compared to the shear stress and the codes ignore it.

 

[shin25]

Bending moment in a beam can be caculated from the shear forces. In turn, the shear flow in a beam can be calcuted from the bending moment.

The theory behind designing the weld should be the same as finding out the number of studs needed to make a composite connection for a composite member. This theory calculates the capacity needed from the bending moment.

 

[miecz]

shin25-

While it's true that the shear flow can be calculated from the bending moment, it's not because the weld is resisting bending moment. The weld is resisting the shear flow at the section, which is a function of the shear at the section.

 

[flyby911]

in that particular case

the shear flow = Q.S/2I.b

Q=max shear on that beam
S=first moment of area of the flange plate about the beam's horizontal axis
I=moment of inertia of the C.S
b=web plate's thickness