determination of fillet weld length

I want to weld an 18ft x 5/8" inch plate to the web of a steel H-pile. If the plate will carry a load of 5 kips at the point furthest from the welded connection, how is the appropriate length of the weld determined?

Given: the total length of the pile is 18 ft. The flanges are 18" wide or 1/2 flange width = 9", which is the width of the plate. Fillet leg width is 1/4", each leg.

 

[MJCronin]

Is the load 5 kips at only one point along the 18 foot length of the plate or is the load 5kips per foot ?

AISC also specifies minimum weld size based on the thicknesses of the members joined.... This is another consideration.

Are the materials A36 or equivalent ? What type of electrode will be used E7018 ????

You know that there are a lot of hungry structural engineers out there with children dressed in rags and old calculators with dead batteries.....

MJC

 

AISC ASD doesn't specify weld length, weld size, yes, but only concerning the legs of the weld itself. As far as electrodes, materials, etc. it's not of concern as those can be represented by letters or symbols or whatever. I'm trying to determine if I should use a continuous weld vs. intermittant welds - if intermittant, how long would they be? The loading might as well be W in plf along the 18 ft length of plate that welded to the HP.


Here's an attempt at the drawing:

------- flange
|
| ! = loading in the -y direction
|______ plate
|
|
|
-------


one-half section view

 

[DaveVikingPE]

Manfred_x:

Check AISCS J2. There are provisions for intermittent welds, which may answer your question. The basic method is to first determine weld size assuming a <i>continuous</i> weld. When you've determined that, and it might be &quot;really small,&quot; you can beef up the weld size to something normal, like 1/2&quot;. Since the 1/2&quot; weld is a lot stronger than the smaller weld size, you can shorten the weld length appropriately, etc. Also check out Blodgett's and other's books. Have fun!

 

[ishvaaag]

I understand from your sketch load in -y direction is causing cantilever action 5 kip/lf*9 in=45 kip·in/lf moment to be fixed and taken by the web, no?
Take arm is 7/8 in, then F per linear foot at the weld 51.43 kip per linear foot. Take then into account safety factors if pertinent.
Check the fillet for this force perpendicularly to its axis.
Is dynamic the solicitation?
Thomas Magner has a Mathcad worksheet freely downloadable that * should* cover the case.
Also Pilkey's manual on paralell welds.
There are also web considerations; is a Von Mises check there satisfied?

 

[ishvaaag]

I have run the case as above described in one my Mathcad sheet on Pilkey's table 8-10 for the case, considering the loads as above. In more than the forces coming from moment those of shear also stand. The orthogonal fillet weld is subject to such forces, which are nonfactored, electrodes are E70, pile is for bridge application; then leg size comes to be 3/8 of an inch, continuous. Have not checked if meets min and max size, but it looks it will; nor is assumed weld to base in the leg be critical. No check of the web is possible with your data. No dynamic allowance nor special impact factor has been made.

 

[ishvaaag]

This is respect to the web Von Mises check. I have made a Mathcad sheet that checks what compressive stress the pile (and then web) will abe able to carry to meent the Von Mises check. If you have cantilevering plate only to one side of the the web , a 0.7 inch thick web can bear 46%Fy in compression, whilst a 0.6 inch thick only can bear 14%Fy compressive. Some similar check can be made for the plates cantilevering to both sides and hinging action summing at seams, for worse and lesser ability to carry axial loads.
So you see that the web thickness may very well limit the ability of the pile to carry the desired axial loads, for even being generally low for pile application, I think no less than 30%Fy at factored level will be put to use, and more even whilst driving the pile to overbear/ensure such later to be required capacity.