ASME Sec VIII Div 2 - Part 5 Stress Classification Example Explanation 1

[Vicker85]

Hello Experts,

Table 5.6 in ASME Sec VIII Div 2 gives examples for Stress Classification. I am trying to understand when on a FLAT HEAD junction to shell would one consider the bending stress as Primary Bending or Secondary.
Of course I read through the Note (2) If the bending moment at the edge is required to maintain the bending stress in the center region within acceptable limits, the edge bending is classified as Pb ; otherwise, it is classified as Q.

What i am not clear is that how does one know if the edge moment is (or is not) assisting in maintaining the bending stress at the central region ? Do we need to perform any calculations for this OR some indicators from FEA results OR ... ?

 

[DriveMeNuts]

If you assume that your flathead is simply supported at its edge (i.e. not welded to any cylinder) and the calculated stress at the centre is less than 1.5xS, then all of the stresses of the 'actual' welded cylinder-head will be secondary because they don't contribute to keeping the stress at the centre of the head below the safe value of 1.5xS. The stresses at the cyl-head junction will therefore have a stress limit of 3xS.

Even though the stresses of the welded connection are secondary, the welded connection may provide some support to the head, reducing the stress to say 1.3xS, however the safe operation of the head doesn't depend on this stress reduction.

 

[Vicker85]

DriveMeNuts,

Thank you for that explanation! That helps a lot.
When you say - The stresses at the cyl-head junction will therefore have a stress limit of 3xS - I understand you meant only for the Bending Stress has a limit of 3S, right ?

So, in actual practice, one should evaluate stresses at flat head center (bending essentially) with simply supported condition as well in order to confirm if the the stress at flat head center < 1.5S - is that understanding correct ?

 

[TGS4]

abhi59 - I think that you understand the problem at hand fairly well now with the explanation that DriveMeNuts provided. However, there are a few things in that explanation that fall well-short of the mark.

First, you need to understand and appreciate the failure mode that you are talking about. In the case of the flat head and the S[sub]PL[/sub] stress limit (not necessarily 1.5S), you are dealing with the failure mode of plastic collapse. And you are comparing a membrane-plus-bending equivalent stress in the flat head your FEA to this limit. Similarly, in the cylinder-to-head connection, you also have to achieve a limit on the local membrane equivalent stress of S[sub]PL[/sub].

For the stresses that are categorized as secondary, you only evaluate them in the context of Protection Against Failure from Cyclic Loading: Ratcheting. There, the limit on the primary-plus-secondary membrane-plus-bending equivalent stress range is S[sub]PS[/sub], which is most certainly not 3S - you need to read the rules to better understand the limit.

 

[Vicker85]

TGS4 thank you for your explanation and elaborating beyond what i intended to ask! It is really helpful for my further investigations.

However, I should have mentioned in the question that my present analysis does NOT have cyclic loadings and therefore the Protection Against Failure from Cyclic Loading need not be checked in my present analysis. So, i am basically checking for Protection Against Plastic Collapse and Protection Against Local Failure.

So, in view of Protection Against Plastic Collapse - can you please confirm if my understanding is correct :
(1). At center of Flat Head : The stresses are essentially Bending Stresses (P[sub]b[/sub]) and this must be less than 1.5S - correct ?
(2). At the cylinder to head junction : Suppose the bending stress is Secondary(Q) then, Local Membrane stress(P[sub]L[/sub]) shall be less than 1.5S and primary-plus-secondary membrane-plus-bending (P + Q) shall be less than 3S (=S[sub]PS[/sub]) - correct ?
(3). From your explanation,

For the stresses that are categorized as secondary, you only evaluate them in the context of Protection Against Failure from Cyclic Loading: Ratcheting.

did you mean that Secondary stresses (like bending stress at head to shell junction) need NOT be checked at all for Protection Against Plastic Collapse if there are NO cyclic loadings ?

 

[TGS4]

Protection Against Failure From Cyclic Loading: Ratcheting MUST be checked regardless of whether the component is subject to cyclic loading or not.

But otherwise, no, secondary stresses need not be checked for Protection Against Plastic Collapse.

Your second understanding is incorrect. And for your first one, the limit is S[sub]PL[/sub], not 1.5S.

 

[DriveMeNuts]

(1) Agreed.

(2) This is a little confusing. The hand calculated longitudinal and circumferential membrane stresses need to be less than S (P[sub]m[/sub]) as per the basic thickness calculation equations, which protects against 'plastic collapse'. Strictly speaking, this doesn't apply the cyl-head junction as it is a structural discontinuity.
The Linearised Von Mises stress taken from the FEM is considered P[sub]L[/sub] and is allowed to be 1.5xS (S[sub]PL[/sub]). This limit is to protect against 'excessive distortion'.
The Membrane+Bending < SPS assessment is required to protect against Ratcheting. Even if there is no cyclic duty, it is required. I guess because pressurising and depressurising the equipment a few times over the life of the vessel could cause ratcheting to occur.

Treat the assessment of each failure mode as being completely separate. Calculate the stress from scratch for each assessment. They often turn out to have the same magnitude, however they need to be calculated separately because they contribute towards completely separate and unrelated mode of failure.