ASME Sec VIII Div 2 - Part 5 : 5.3 - PROTECTION AGAINST LOCAL FAILURE - HYDRO-TEST 3

[Vicker85]

Hello Experts,

Do we need to assess PROTECTION AGAINST LOCAL FAILURE for Hydro-test condition as well ? I have checked for PROTECTION AGAINST PLASTIC COLLAPSE by Limit Load Method with applicable load factors.
As i read through 5.3.2 Elastic Analysis — Triaxial Stress Limit, it guides on the limitation on algebraic sum of the three linearized primary principal stresses from Design Load Combination (1) of Table 5.3.

Does that mean that we need to check ONLY for load combination (1) [P + P[sub]S[/sub] + D] ? Note that, in the same table, we also have Load Combination (9) [P[sub]T[/sub] + P[sub]S[/sub] + D + 0.6W [sub]pt[/sub] ].

Further, again for Hydro-test condition, do we need to assess for Ratcheting ?

 

[DriveMeNuts]

A Hydrotest ratchetting assessment isn't required. During the Hydro-test, the vessel may experience plastic deformation which is similar to a single ratchet cycle. This is expected. It is also good for increasing fatigue resilience.

I'm not sure about local failure.

 

[Vicker85]

DriveMeNuts,
Thank you for clarifying on the Ratcheting assessment for Hydro-test.

Let us see if some other experts could share their views if assessment for PROTECTION AGAINST LOCAL FAILURE for Hydro-test condition is required ?

 

[IdanPV]

abhi59,

1. It seems that you are using an old edition of the Code.
The Pressure Test Load Combination in table 5.3 of the 2017 Edition it's in clause #9 but in the new, 2019 Edition it's appears in clause #18.

2. You wrote that you were using the Limit Load Method, so why you are referring to Table 5.3 which is applicable for the Elastic Analysis Method?

3. The Loads Cominations in section #18 of Table 5.3 or in the Test Condition of Table 5.4 (1/ßT[PT + PS + D + 0.6W pt]) are applicable for clauses 5.2.2.5 (Elastic Stress Analysis) and 5.2.3.6 (Limit-Load Method).

4. As you wrote, clause 5.3.2 is asking for the principal stress from the Design Load Combination of Table 5.3.

5.Note that you can't use the Limit Load Method in order to demonstarte protection against local failure

https://www.eng-tips.com/viewthread.cfm?qid=473062

 

[Vicker85]

IdanPV,

Thank you for your reply. Please note the following for your replies:

1). My bad! Indeed, I was referring the Old Code Edition (2017). I had both the Editions open and I quoted the older one while I posted my query!. Thanks for pointing that out.

2). What I meant was - I was trying to justify the clause 5.3.1.2 (which says When protection against plastic collapse is satisfied by the method in 5.2.3 (which happens to be Limit Load Method), either method listed below is acceptable. ) before proceeding to using assessment of Protection Against Local Failure by 5.3.2 ELASTIC ANALYSIS — TRIAXIAL STRESS LIMIT. And the triaxial stress limit refers to Table 5.3, which is applicable for assessment of local failure by Elastic Stress Analysis.

3). Agreed. That is what I have followed to demonstrate protection against plastic collapse by Limit Load method under Test condition (using Table 5.4 (1/ßT[PT + PS + D + 0.6W pt]).

4). So, that basically means that Local Failure under Hydrotest conditions cannot happen even if the stresses are very high for once for a, relatively, short period of time!! I wonder what could be the local failure mechanism covered ? Could it be that local failure would also happen as an incremental failure locally by developing cracks at the high stressed points if allowed to happen over a long period of time (Operating or Design Loading) ?
. In my opinion, cracks could develop instantaneously as well under Test Conditions! I have read TGS4's post in other discussions where he has highlighted about how triaxiality mechanism serves for local failure.

5. Agreed. For local failure, its either Triaxial limit by Elastic Analysis or Local Strain Limit by E-P Analysis. And thanks for that link - I have read that earlier - Very interesting and important discussions in that post!

 

[TGS4]

The check on local failure, with the specified design margin, does not need to be achieved for the test condition. That it achieves the required design margin at the design conditions is sufficient.

 

[Vicker85]

TGS4,
Thank you for the confirmation on that point! Very Helpful.

However, I will again ask my question above on why is this criteria of local failure not required to be achieved for Test Condition -
That basically means that Local Failure under Hydrotest conditions cannot happen even if the stresses are very high for once for a, relatively, short period of time!! I wonder what could be the local failure mechanism covered ? Could it be that local failure would also happen as an incremental failure locally by developing cracks at the high stressed points if allowed to happen over a long period of time (Operating or Design Loading) ?

 

[DriveMeNuts]

I may be wrong but I think of local failure as being where the three principal stresses increase simultaneously so that the Von Mises Stress never extends beyond the yield surface and therefore the material doesn't yield. Therefore the material exhibits something similar to brittle behaviour and spontaneously ruptures if the sum of the three principal stresses gets too high. I don't associate brittle failure with crack growth or incremental failure.

I understand that the Elastic method of making sure the sum of the three Principal Stresses remains less than 4xS is a rule of thumb type guestimate which is probably conservative enough to also cover the Hydrotest.

I haven't used the Elastic Plastic Method and don't know how accurate it is compared to reality. I "assume" it is also conservative enough to cover the Hydrotest.

 

[TGS4]

It's not that Local Failure won't happen - it's that the design margin required for the design condition does not need to be achieved for the test condition, and that the margin at the design condition adequately covers you for the test condition.

Take a look at the formulation for the von Mises stress. What is the von Mises stress when the 3 principal stresses are equal? Zero. That means in a state of hydrostatic (tensile) stress, the von Mises approach to plasticity falls apart. Effectively, the material switches behaviour from ductile to brittle. The inverse of this is that under hydrostatic compressive stresses, materials will behave in a a much more ductile manner than the uniaxial testing would imply. This is what permits wire drawings and similar forming procedures.

 

[Vicker85]

Thank you all for your valuable insights!