Can we use B31J to calculate SIF,SSI and flexibility factor for fittings where D/t ratio is 128?

Hi,

The ASME B31J "more applicable" formulas for SIFs are valid for D/t <= 100. This restriction is clearly stated by Note (1,b) of Table 1-1.

Therefore, for D/t = 128, I believe you should assess the SIFs by Finite Element Analysis. If you have B31J Essentials installed (as free software) together with Caesar II V. 9.00 or 10.00, then you can use FESIF module or FETEE (as applicable) to assess the SIFs and SSIs.

A problem might be the "olet" type branches, where the shell models are not quite accurate and I don't think you can use brick elements with FESIF "free" module.

If you have FEA Tools, then for "olet" type branches you may try Nozzle Pro to calculate the SIFS using brick elements.

I do not have installed FEA Tools now on my Office PC and I am not sure if Nozzle-Pro (latest version) can be used for brick element modelling. FE Pipe definitely can be used, but this means to have full PRG FEA Software Package available.

I hope these details will help you. Good luck!

Regards,

Thanks Dorin Daniel for your reply. In my case I have reinforced Tee connection for which I need SIF and SSi.
Can I use FEsif for reinforced Tee?
Can I get sif values for both header as well as branch?

YES, FESIF Module is entirely applicable.

Thanks again Daniel for quick response. But I think only B31J is available free.

Hi all,
Need your expert comment/opinion. I got SIF value from FESIF from other office for 48 x 24 reinforced Tee connection. Header and branch thickness is 9.525 mm .Pad thickness is also 9.525 mm. The D/T ratio for header is 128 hence we calculated SIF using FESIF software. The calculated SIF values are far less than B31.3 values. I feel that we are missing something or there is some input error. I have to rely on other office for FESIF software. Could you please identify where I am going wrong or is there anything which I am missing. Input and output reports are attached.

The new (B31J) SIFs can be quite different than the Appendix D values.
The run pipe values are much smaller (out-plane run SIF = 1.0!) and the branch values are similar in-plane. The B31J out-plane SIF for the branch is 12.0.
Appx. D SIFs for tees with reduced outlets are know to be under predicted.
When applying SIFs developed using FEA or taken from B31J do not use the reduced branch section modulus (Ze) DEFINED IN TEH cODES.

David,
Thanks for your reply and pointing out that reduced branch section modulus (Ze) should not be used in calculation as this is already taken care in FEA and B31J SIF calculation.
You compared SIF values with B31J but B31J validity has been demonstrated for D/T ≤ 100 only. In my case D/T is 128.
I think above SIF values (FESIF result) is for header and not for branch. I should check input options to get branch SIF values which will be higher than header SIF's.