SIF for Weldolet on a Header

hi

in our current Project in one system the Header size is 16" with sch 10s Titanium pipe and the branches are 4" with sch 10s same titanium mtl
the Branch connection is a WELDOLET
if i am using this it gives a SIF of 4.05
due to this high SIF the branches are failing in the blast case
we are asked to put (0.75*Actual SIF) for blast case
Eventhough there is a Vertical Rest and Holddown very near the Branch (500 mm) it is failing in the vertical blast case
can any one suggest how to solve this

Sometimes, trying to make things too rigid worsen the situation (I know that look strange). Try some gaps in supports or increase flexibility.

Regards,

Check c2 help instruction. You have an option to neglect SIF in SUS and OCC but it must not be less than 1.

but in this case i can't neglect the SIF value

since it is a Big bore pipe (16") with less wall thickness and if we are taking a branch of 3" using weldolet the SIF value is so high
eventhough i did't model the branch pipe simply i break the node and use weldolet connection SIF
in that case it self it is failing

for heavy thick pipes there is no problem only for the thin wall pipes are failing

in such cases how to solve this issues

What is the relevance of SIF for a blast scenario?


Regards,

But the real problem for you is that you "are asked to put (0.75*Actual SIF) for blast case" which mean that you have to obtain the approval to do this.

Regards,

If you are so concerned brace the branch from main header. Please read through the balst requirement either in Fabig technical note or DNV codes. Are you able to select the right zone, right overpressure, right drag load? Are you taking into account grating opening i.e free space etc.? You can go upto yield of material. Have you provided holddown supports for header etc.? Blast doesn't cause problem if propely supported.

"eventhough i did't model the branch pipe simply i break the node and use weldolet connection SIF in that case it self it is failing"

Of course, it is the header that determine weldolet SIF, not the branch. (See Appendix D of B31.3)

In any case at 0.3 bar drag pressure, you can expect a lot of problems.

One solution is to increase the thickness of the line, but I think nobody want this.

The other is the one that PKU gives you i.e. to ignore the SIF for this particular case.

Regards,

Code SIFs for these cases are way off from reality.

http://65.57.255.42/ubbthreads/ubbthreads.php?ubb=showflat&Main=7883&Number=36067#Post36067

Say your header is anchored and guided after that and the header is straight, you will not see any failure. As soon as your header is changing direction, it will show failure. The answer is self explanatory. This is the reason I said SIF is not required as the bending is not due to branch.