At piping branch point(Tee)when user defined SIF(i)&SIF(o) are used in the analysis using CAESAR-II,

1.It is mandatory to define branch type at one of the element (out of three intersecting elements), in case design Code uses EFFECTIVE SECTION MODULUS for branch. For example ASME B31.3

2.If design code uses EXACT SECTION MODULUS for branch in the calculation, then user should not enter branch type at any element (three intersecting elements). Only user need to enter values of SIF(i)& SIF(o)in all three intersecting elements.

Intergraph members are requested to confirm above understading. Kindly correct it in case above information does not hold true.

Immediate response is highly appreciated !

Dave / Richard please response to understand what CAESAR-II follows for the calcualtion in case of user defined SIF.
Thanks.

I believe you have it right.
Why not build a model to confirm? A header with 4 reducing tees (tees that qualify for the Z adjustment found in B31.3). Supply SIFs for all; call out B31.3 for two and B31.1 for the others. Call out a tee for one in each Code. Compare the results.

Thanks Dave for your prompt response.

The method you suggest to check the model, is not suitable as both the code B31.3 and B31.1 uses an effective section modulus for the branch.

But i tried with a simple model,
First model- with all 3 elements with user deifned SIF and no branch type defination.
Second model with all three element with user defined SIF and one element with type of branch. Note that SIF values in both the model are same.

As a results, first model produces higher branch stresses compared to second model. I understand first model uses exact section modulues of branch and second model uses effective section modules for branch.

If you don't specify a tee type, CAESAR II doesn't really know in-plane versus out-of-plane. If CAESAR II assumes that in-plane is "local x axis" cross global "Y". So "yes" you can obtain different results (depending on the orientation of the tee)) based on whether or not you specify a "tee".

Thanks Rechard and Dave.

I conclude that out of three intersecting elements, at one of the element, user should specify a "Tee type" in order to pickup correct in-plane vs Out-of-plane directions by CAESAR-II.

Dear Richard & Dave

Please clarify my doubt..

Header 26” – 660.40 mm STD(9.53 mm) & Branch 10” – 273.10 mm STD(9.27 mm)

D/T =660.40/9.53 = 69.29 && d/D =273.10/660.40 =0.41

These ratios are comes under the validity of ASME B31.3.

But still in my case problem with high SIF values in stresses. So I opt to use FE SIF to calculate the SIF values.

My doubt is :

Based on ASME B31.3 -- In-plane SIF = 7.357 ; Out-plane SIF= 9.476.

FESIF results (SIF thru HEADER based) -- In-plane SIF = 1.58 ; Out-plane SIF= 0.49 (So it is equal to 1)

FESIF results (SIF thru Branch based) -- In-plane SIF = 4 ; Out-plane SIF= 13.45.


If it is user defined SIF, As per above Discussion I have to enter SIF @ 3 points and I have to define Type.

For Example:

In HEADER: (Node 20 – TEE point – unreinforced)

Element -10 -20 ; @ 20 – FE SIF values (SIF thru HEADER based) entered

Element - 20-30 ; @20 – again FE SIF values (SIF thru HEADER based) entered

IN BRANCH:

Element - 20-40 ; @20 which SIF I have to Enter in branch whether “SIF thru Branch based” or “SIF thru HEADER based”. ??

Technically which one is correct to enter SIF values in branch ??

Any comments ...

Any help ...

Durga,
Enter the branch SIFs for the branch pipe.
Do not identify the type of branch in CAESAR II (e.g., UFT) as that will have the program calculate an effective section modulus (Ze) in B31.3. Your FEA SIFs reflect the intent of the Ze so Ze should not be used.
Confirm the proper use of in- and out-of-plane SIFs.

Dear Dave,

Thanks for your response..

So for the branch I have to input “SIF thru Branch based”. Clear…

But from the above discussion between you, Richard and pradeepd -- concludes that “out of three intersecting elements, at one of the element, user should specify a "Tee type" in order to pickup correct in-plane vs Out-of-plane directions by CAESAR-II.”

1. From your statements what I understand if the user defined SIFs are from FE then no need to enter the TEE type. Then for branch also CAESAR-II takes Zexact for calculations there by stresses are increased. Why Ze is not to be used when user calculates SIF from FE ? Can you pl explain..

2. But User defined SIFs can be from various sources like FESIF, French COADE method or simply with some factor like 1.414*SIF.. then what about the remaining cases whether we have to define or not..

Please Correct If I am wrong.

1. The effective section modulus (Ze) used in B31.3 is that Code's approach to adjusting stress at reduced outlet connections. If you look at B31.1 you will see that that book has a different SIF for the reduced outlet and NO adjustment to Z. To call a CAESAR II intersection a "tee type" using B31.3, will invoke Ze and that is not appropriate if the adjustment is made in the SIFs.
2. To quote B31.3 300(c)(3): "Engineering requirements of this Code, while
considered necessary and adequate for safe design, generally employ a simplified approach to the subject. A designer capable of applying a more rigorous analysis shall have the latitude to do so; however, the approach
must be documented in the engineering design and its validity accepted by the owner."
So, when using more rigorous analysis, you are responsible for your assumptions and the assumptions of your analysis tools. I think it would be wise for you to confirm CAESAR II operation under these more rigorous analyses.

Dear Dave,

Thanks for your explanation.

Yes I agree with you. If we are using rigorous analyses like FE-SIF again no need to adjust Z value.

But Richard Said “If you don't specify a tee type, CAESAR II doesn't really know in-plane versus out-of-plane. If CAESAR II assumes that in-plane is "local x axis" cross global "Y". So "yes" you can obtain different results (depending on the orientation of the tee) based on whether or not you specify a "tee”.

Now, I don’t want to adjust Z value for Branch connection so I didn’t define TEE type. In this situation how CAESAR-II knows in-plane versus out-of-plane?

Can you please explain what is the relation between “Defining TEE type & for taking In-plane and Out-plane SIF by CAESAR-II”.

Please Correct If I am wrong.

Any comments..

The software will assume an orientation and default in-plane axis as I described above. This may or may not align with your orientation of the tee. If you don't specify a tee type with the SIF values, CAESAR II has to rely on its default orientation.

It is better to define the tee type.

Dear Richard & Dave

Thanks you for your explanation.

But I am in confusion.. Because

1As Richard said “It is better to define the tee type” because If user don't specify a tee type with the SIF values, CAESAR II has to rely on its default orientation. YES I AGREE..

As Dave said “If we are using rigorous analyses like FE-SIF again no need to adjust Z value. Not required to define TEE type”. YES I AGREE WITH THIS STATEMENTS ALSO..

Please tell me Technically which one I have to follow Now.. whether I have to define TEE type or NOT.

It appears we are not going to be able to explain this to you in this Forum. Open a Service Request on SmartSupport and send in your job. Point out the specific tee you have questions about and we'll answer you.

Dear Durga,

Addition to your question, just share with you.

In one of project, we are using structuarl members with Pads at elbow as trunion supports not Piping, hence calculated SIF by FEM, however, to model in CAESAR, we are selecting the " TEE TYPE", Reinforced tee, RPad thickness, SIF(I), SIF (O) from FEM at all 3 location points of bend. Equivalent Pipe size also calculated.

Client considered Trunion is "TEE" < even there is no fluid inside trunion> welded connection with Elbow.

May be useful to you.

Dear experts, correct me, if my analysis technique is not correct.

It is highly conservative, but no option, as instructed by Client.

Regards
SND

Dear Richard & Dave.

Thanks for your Reply,

This query is not related to any particular job. It’s a general query for a CAESAR-II user.

If we are inputting “user Defined SIFs” then whether we have to DEFINE TEE type or not??

As per the above discussion I didn’t find any conclusion on this. Kindly clarify this doubt.

The answer is "Yes".

Thank you very much Richard for conclusion on this topic..

If we are inputting “user Defined SIFs” then user has to DEFINE TEE type.

Dear Dave Do you have any comment on above conclusion..

I think it was stated several times that calling out a tee in CAESAR II will allow the program to apply an effective section modulus if such a calculation is required (by Code) for the tee.