Ansys does not comply with ASME Section VIII Div. 2 part 5

[shiraz883]

thread569-380066

Hi,

This is in reference to the thread in which TGS4 says that "ansys does not comply with ASME VIII div.2 part 5". Can someone elaborate on that? Because I went through the ANSYS help topics and searched "stress linearization" and saw the same formulae as used in the code. I do not see a reason for non-conformation of ansys linearized stress with that dictated by the code.

Thanks,
Shiraz

 

[TGS4]

Have you reviewed 5-A.4.1.2?

 

[shiraz883]

Yes I did.

Please find the attached comparison between ASME and ANSYS linearized stresses.

 

[shiraz883]

The earlier post allowed only one attachment. Here is the ansys linearization snapshot from ansys help topics

 

[TGS4]

See Step 2 (b). That's the critical parry.

 

[shiraz883]

Thanks TGS4,

I see that a macro is already there wherein this critical parry is taken care of. Saw it few weeks ago and there was a manual evaluation (in one of the axis-symmetric models) by taking the values of Sx,Sy,Sz,Sxy and Seqv using PDEF command in ansys and doing manual calculation by applying simpson's rule of integration over these data points.

Thanks
Shiraz

 

[TGS4]

You're completely missing the point.

Step 2. Calculate the bending stress tensor.
(a) Bending stresses are calculated only for the local hoop and meridional (normal) component stresses, and not for the local component stress parallel to the SCL or in-plane shear stress.

The point is that when assembling the bending stress tensor, it is composed of the membrane-plus-bending stresses ONLY from the hoop and meridional directions. In the other orthogonal direction (through-thickness), as well as for all of the shears, only the membrane stress is to be used when forming the tensor. ANSYS generates that tensor using membrane-plus-bending for all 6 component stresses, without an option to revert any components to membrane only.

 

[rickfischer51]

Sounds like a need for an enhancement request. Or maybe a job for APDL.

Rick Fischer
Principal Engineer
Argonne National Laboratory

 

[TGS4]

Have been requesting it since 2002. And have been asking other users to request it, too. Even mentioned that ABAQUS has had this capability. We'll see...

I have a work-around, in any case.

 

[rickfischer51]

What's the workaround? Macro?

Rick Fischer
Principal Engineer
Argonne National Laboratory

 

[TGS4]

Some APDL and some MathCAD worksheets.

 

[rickfischer51]

I contacted our Ansys tech support guys. They are putting in an enhancement request, and they are sending me a macro. I'll keep you posted.

Rick Fischer
Principal Engineer
Argonne National Laboratory

 

[rickfischer51]

I just got an email from my Ansys support guys that R19.1 is now BPVC compliant. Yippie Ki Yay!!

Rick Fischer
Principal Engineer
Argonne National Laboratory

 

[TGS4]

I saw that, too. Now we just need to ensure that engineers know and understand what needs to be done to ensure compliance.

 

[DriveMeNuts]

Remember that ANSYS only sets the normal Bending stress parallel to the X Axis to zero.
So you need to Orientate the local Co-ordinate system so that its X axis is parallel with the SCL.

Does anyone know how ASME came up with this rule to set the through thickness bending to zero? Is it introduced to remove the effects of a compressive peak stress on the inside surface of a pressure vessel under internal pressure?

Does it work for structures such as the internals of a Pressure Vessel as well?

PTB-3 doesn't provide an explanation.

 

[TGS4]

The explanation is in WRC 429. There is no physical "bending" that can occur in the through-thickness direction - even though the through-thickness stress in a thin shell should change from the negative of the internal pressure to zero at the outside surface. And shear "bending" makes no physical sense, either - the shear stress distribution for a valid SCL should be small and parabolic, and zero at the surfaces.