ASME Sec VII Div.2 part Structural discontinuity | Global plastic collapse | Elastic stress analysis

[shiraz883]

HI,

I have a flexible tubesheet forging for which I am evaluating the stresses on the small fillet radii region of the forging to satisfy the global plastic collapse as per ASME Sec VIII Div 2 part 5

I have made SCL's on the part as shown. The red marked portions of the forging will have linearized stresses - Pm and Pm+Pb evaluated against 'S' and '1.5S' respectively to satisfy global plastic collapse. But for the SCL's made on the small fillet radii portion - marked in yellow - see the attached picture, this region will have local membrane (Pl) and secondary membrane plus bending stress (Pl+Pb+Q) since this region is a discontinuity - a local discontinuity as per figure 5.1 also known as hopper diagram.

So, to satisfy global plastic collapse only Pl (local membrane) will be compared with the '1.5S' value and Pl+Pb+Q which is secondary membrane plus bending will be compared with '3S' to satisfy ratcheting and Pl+Pb+Q+F compared against '2Sa' to satisfy fatigue.

In effect there will not be a Primary membrane plus bending and hence the criteria as per global plastic collapse applies only to satisfy the limit of local membrane stress.

Is my understanding correct?

Thanks,
Shiraz



Shiraz
Sr. Engineer

 

[TGS4]

For all questions related to stress linearization and stress categorization, please refer to paragraph 5.2.1.2:

For components with a complex geometry and/or complex loading, the categorization of stresses requires significant knowledge and judgment. This is especially true for three-dimensional stress fields. Application of the limit-load or elastic–plastic analysis methods in 5.2.3 and 5.2.4, respectively, is recommended for cases where the categorization process may produce ambiguous results.

However, as a freebie - I will let you know that your limits are (mostly) incorrect, at least as far as the post-2007 Code is concerned. The limit on P[sub]m[/sub] is S (you got that correct). The limit on P[sub]L[/sub] and P[sub]L[/sub]+P[sub]b[/sub] is S[sub]PL[/sub], as defined in 5.2.2.4, Step 5. And with respect to the ratcheting limit, you ought to be looking at a stress range, and not a singular stress value. Likewise for fatigue.

Some guidance on categorization can be obtained from Table 5.6. With your component being a tubesheet (i.e. flat plate), there are additional considerations that come into play. Also, you can reference WRC 429 regarding placement of SCLs. Many of the ones that you drew may be considered invalid. Have you performed the SCL validations as recommended in Annex 5-A (5-A.3(c))?

 

[shiraz883]

HI TGS4,

I missed the limit on the Pl+Pb+Q as max of (3s or 2Sy) and you are right in stating that this is for ratcheting assessment wherein the stress range should be taken into acocunt and so also for the fatigue. And that is what I have done.

But my question was regarding the satisfaction of global plastic collapse in regions of discontinuity as the one shown in the figure that I posted.

I have read Annex 5-A (5-A.3(c)) and try to model the SCL's perpendicular to the midsurface of a section. However the other requirements that are to be fulfilled - stress monotonically increasing or decreasing and non-monotonic. I check that using the query picker in ansys and going from one edge or one vertex of SCL to the other and looking at the magnitude of eqv. stress except for concentration (in my case a fillet radii).

My main concern is that for the SCL in the yellow portion since there will be no primary membrane plus bending stress (owing to the region being a discontinuity) does comparing the Pl with '1.5S' satisfy the criteria for global plastic collapse?

Thanks,
Shiraz

Shiraz
Sr. Engineer

 

[TGS4]

Don't get me wrong - I completely understand your query.

I also have a long-standing policy of referring such questions to 5.2.1.2. Read the last sentence of that paragraph again.

 

[DriveMeNuts]

WRC 429 is also a good reference for learning the two separate forms of plastic collapse.
Which I don't think your geometry suffers from.