(?1+?2+?3)?4S criteria in new ASME Div. 2 3

[BPVFEA]

I am confused with this clause. Please help me out.
In Cl. 5.3.2 of new ASME SEC. VIII Div 2 (2007), for 'Protection against local failure' it is stated that
"Following Elastic analysis criterion shall be satisfied for each POINT in the component. The sum of the local primary membrane plus bending principal
stresses shall be used for checking this criterion. (?1+?2+?3)?4S"

My query is

1) The stress needs to be evaluated at particular point (node) or needs to be Linearised across thickness?
2) If it is at a particular point, we can find total ?1+?2+?3 (Primary+seconday mem+bending) at each node easily.
But how can we take out secondary stresses separately. e.g at flange to hub fillet some secondary stress would be present due to local discontinuity.
3) If stress to be evaluated through thickness, then my understanding is we have to place one end of SCL through the end of fillet and not through fillet (see fig 5.A.11).
In ANSYS I can find Linearised membrane+bending at Inner, center outer part of SCL. I will add up M+B for ?1+?2+?3. But here also question remains how to take out secondary stress for this.

Thanks in advance

 

[prex]

Not sure to understand your question.
To analyse the component, you need to separate local membrane and primary bending, right? So why not use those values, that you already have.
Note also that that limitation is very rarely of significance. It may be limiting only where a very high state of compression in all directions exists.
And of course membrane and bending stresses are concepts that exist only after a linearization has been done.

prex

http://www.xcalcs.com

: Online engineering calculations

http://www.megamag.it

: Magnetic brakes and launchers for fun rides

http://www.levitans.com

: Air bearing pads

 

[TGS4]

The evaluation is to be done at every point.

Note that this is not a new requirement, it is the same requirement that was in the 2006 and earlier editions of the Code. You were always performing that required check, right? How did you do it then?

The intent of the linear-elastic local failure check is to ensure that a state of tri-axiality does not exist anywhere in your component. If S1=S2=S3=some positive number, then your calculated Seqv would be equal to zero. However, there is a known failure mode whereby components can fail with very low Seqv, but high tri-axiality, and this check (or more specifically, the elastic-plastic local failure check) is to catch those cases.

 

[BPVFEA]

Thank you Prex & TGS4,
I have been recently moved to analysis. I have to learn a lot understand and apply these concepts.

I am still not clear about how to seperate any secondary stresses. because the code mentions to consider PRIMARY stresses only.

 

[prex]

The separation of secondary stresses is a big issue, on which you'll find many discussions on these forums: two recent examples thread794-249156 and thread794-248964.
Generally speaking you normally obtain general membrane and bending by formula, local membrane by formula or FEA (using only the loads producing primary stresses), the balance is almost always secondary.
The main characteristic of a secondary stress, that may help in identifying them, is that it is self limiting: the deformation caused by the loading tends to reduce the stress.
The main characteristic of a primary stress, that may help in identifying them, is that they must satisfy the laws of equilibrium, in other words they must equilibrate the external loads acting on the component (Note: an external load is not necessarily mechanical, like pressure, a thermal expansion originated in a structure separate from the component under examination is also external).

prex

http://www.xcalcs.com

: Online engineering calculations

http://www.megamag.it

: Magnetic brakes and launchers for fun rides

http://www.levitans.com

: Air bearing pads