Flare Groove Welds

Date 10-16-2002 13:26

Recently, we came across some cracking at the ends of flare bevel groove welds on brackets welded to rectangular tubing. The bracket was welded on three of four sides. The fourth side which is at one end of the flare groove weld was not welded. After seven years of use a crack was found in the tubing at the end of flare groove weld. We found similar cracking in the same area on other pieces of the same equipment. The equipment had been in use for only 3 to 7 years. To avoid the problem the joint is going to be changed and all sides can be welded.

I would like to know if there are any guidelines (or requirements) for wrapping flare groove welds? Here in Canada, our welding code has requirements for wrapping fillet welds; however, I haven’t found any requirements for flare groove welds. Are there any recommendations regarding wrapping the ends on flare groove welds in AWS D1.1 (or any other welding code for that matter)? Does anyone have a rule of thumb for this situation?

Has anyone out there experienced similar problems with flare groove welds?

By jwright650 (*****)

Date 10-16-2002 19:07

I've seen similar cracking but it was the weld not the base matal and it happend an hour after welding.(well, it was noticed about an hour after welding)

I work for a structural steel fabricator and occasionally we have lintels over windows and doors that are made up of a TS8x4x1/2 x 30'-0 long rectangular steel tube with a 3/8" x 1'-0 wide steel plate across the bottom(normally it's offset to catch the brick or block over the windows or doors). Drawing calls for cont. welds both sides. We had cracking of the welds when the opposite side was welded. The cracks were exactly in the center of the weld bead. We used FCAW (1/16" dia. E71T-1 w/ 100% C/O2) and I'm not for sure the exact cause of the cracking. I was only guessing, but I thought maybe the welder was welding too much too close together. I had instructed him to weld 6" at a time alterating sides and ends until he finally completed a cont. weld at each side. That seemed to help, no more cracking occured after he started staggering the heat around.

By Niekie3 (***)

Date 10-16-2002 19:37

Are these weldments in fatigue loading?

Regards
Niekie

By Sean

Date 10-17-2002 11:37

The brackets are used to support a lifting cylinder and so the stresses would vary from a minimum to a maxium around the brackets. So yes, the weldment was in fatigue loading.

By Niekie3 (***)

Date 10-17-2002 18:08

All weldments in fatigue service WILL eventually fail. Unlike unwelded components that have a fatigue limit, a welded component has no fatigue limit. You therefore have to decide on what is a reasonable fatigue life for the component and ensure that your design gives this life. After this life, accept that it will have to be replaced.

It is very difficult to give a lot of guidance on the design of your component to increase the life, without having a better idea of the service conditions and present design. There are some fatigue design codes for structures that give excellent guidance on how to design for fatigue loading.

Just to give you an idea of the potential complexity of this problem, I can point out that often an improvement in fatigue life can be obtained by reducing the strength of the component. This is so because under certain conditions this results in a redistribution of the stresses which in turn results in moving the maximum stresses from a welded area to an unwelded area. The unwelded area will typically have a better fatigue capability. Another obvious suggestion is to reduce the stress concentrators as far as possible. Usually this means using continuous welds rather than intermittent welds. In your case, it would probably mean that you need to weld the bracket all round rather than just on three sides. If your cracking is taking place on the ends of the welds, (Where they terminate) then this is almost certainly the case.

Hope this helps

Regards
Niekie Jooste

By Sean

Date 10-18-2002 14:02

Thank you for your reply Niekie. This bracket is connects the lifting cylinder on an aerial device. What welding related design codes would you recommend for this type of fatigue loading?

In the welding design codes that I have available (AWS D1.1 and CSA W59) I have not found any specific requirements for weld termination on flare bevel groove welds or for that matter requirements for continous welds on fatigue loaded structures. If there is a greater chance of fatigue cracking on intermittent partial joint penetration groove welds why isn't there a code requirement specifying how they should be terminated or precluding their use in fatigue loading conditions? Or have I missed something in the codes?

By H.Dibben (*)

Date 10-18-2002 19:39

Partial penetration welds and intermittant welds are both prohibited under fatigue loading. (See section 12.4.14 of W59, and I am sure AWS D1.1 has similar provisions.)

The only problem is that sometimes partial penetration welds are unavoidable. For example, how can you make a flare bevel weld that isn't partial penetration. I am not sure how you can comply with the code in a situation like this.

By Niekie3 (***)

Date 10-20-2002 18:52

The design code for structures that I have been exposed to, that has great guidelines for handling fatigue loading, is BS 8118. This is however aimed at Aluminium. The equivalent for steel, I believe, is BS 7608.

What is great about this code is that you can take your weld detail, and by comparing it to some sketches, you can calculate a final fatigue life. Therefore, by changing your weld detail you can see how it affects your fatigue life. This is a slightly different approach to that taken by codes such as D1.1 (I am no expert on this code) which merely allows or disallows certain weld details.

Hope this helps.

Regards
Niekie Jooste