Testing pressure criteria: membrane + bending stress < 1.5*Rp/1.05

[Mihe]

Hi
I am verifying the strength of a thin-walled cylinder that is subjected to hydraulic pressure. The design is not strictly ruled by the pressure vessel code. However since there is no other applicable code Annex C “Design by analysis - Method based on stress categories” of EN 13445 is applied. I think the question is relevant also for ASME VIII division 2.
I have verified that the cylinder complies with the DBF formulas regarding cylinder wall thickness of EN 13445 and fatigue is not an issue for this component.

At the end of the cylinder there is a step that creates bending stress in the cylinder wall. The general primary membrane stress (Pm), local primary membrane stress (PL) and the Primary bending stress (Pb) has been calculated(linearized) by FEA and integration of the stress field according to the code. It is then to be verified that Pm<f and that PL+Pb<1.5f.

The design will be proof tested with 2 times the normal working pressure. For pressure test the stress limit is f=Rp/1.05. That means that in view of PL+Pb the stress limit 1.5*f will be considerably higher than the yield limit (1.5*f =1.43*Rp).
I get the point that the section will not be plastic through the full wall thickness until PL+Pb = 1.5*Rp but I want to be sure that I have not missed any rule or exception in annex C.

Question: -Is it correct that the pressure test load case would be ok in view of Annex C, if:
Pm<f (hence Pm < Rp/1.05) and
PL+Pb < 1.5*f (hence PL+Pm < 1.43*Rp)?
BR/Mihe

 

[DriveMeNuts]

Section 4.1.6.2 of ASME VIII Div 2 has similar equations.
I do find it intriguing that Pb can exceed Yield during the test. I suppose that a little plastic deformation is acceptable and can even be beneficial, and that plastic hinge is the failure method of concern.
Figure 3 of WRC 429 shows that plastic collapse only occurs when pure Pm + Pb exceeds 1.5 x Yield for a rectangular beam.

 

[r6155]

@ DriveMeNuts
What is the limit of a "little plastic deformation" for you?

ASME VIII Div 1 UG-99
This Division does not specify an
upper limit for hydrostatic test pressure. However, if
the hydrostatic test pressure is allowed to exceed, either
intentionally or accidentally, the value determined as
prescribed in (c) above to the degree that the vessel is
subjected to visible permanent distortion, the Inspector
shall reserve the right to reject the vessel.

Regards

 

[DriveMeNuts]

r6125,
There is a recent indepth thread about your query that will comprehensively clear up any lack of understanding that you may have.

 

[r6155]

I insist
"The Inspector shall reserve the right to reject the vessel if there are any visible signs of permanent distortion"

Regards

 

[TGS4]

r6155 - in this case, you are incorrect. A little bit of plasticity does not necessarily directly translate to plastic deformation that remains after the removal of the load. In fact, I would dare say that the limits indicated above retain an elastic core which precludes gross plastic deformation at all.

 

[r6155]

ASME say "visible permanent distortion", no plastic or another words.

Regards

 

[TGS4]

There only way that you get permanent deformation is plastic deformation. The two are synonymous.

 

[r6155]

TGS4
I know that, obviously.
As say DriveMeNuts "I suppose that a little plastic deformation is acceptable and can even be beneficial". I disagree, this is not acceptable.

Regards

 

[DriveMeNuts]

r6125,

May I remind you that the code of interest in this thread is EN 13445, not ASME VIII.

EN13345 sets the nozzle junction bending stresses to be Primary by default to ensure that the stresses at the junction remain within the elastic range. However, there is a note saying that these bending stresses can be classed as secondary by the user and that by doing so plastic deformation will result which may affect the aesthetic appearance. It says that these deformations don't affect the strength of the region. Therefore visible permanent distortion is acceptable.
ASME VIII Div 2 classes these bending stresses as secondary as default, so it will experience the same issue with visible distortion.

"Reserving the right to reject" does not equal "mandatory rejection" nor does it oblige the AI to reject the vessel as the default option. r6155, the AI will not spontaneously freak out at the sight of visible permanent distortion. The AI will investigate and seek advice as to whether the distortion is acceptable. The wording gives the AI authority and support so that the fabricator can't use undue influence or strong-arm the AI into allowing any potential dangerous stuff up onto the market.

Plastic deformation during a hydrotest can be beneficial for extending fatigue life.
Appendix 44 is entirely dependant on substantial plastic deformation.

It is not sacrilege for a vessel to have "visible permanent distortion".

 

[r6155]

@ DriveMeNuts
Thank you.
If I were the user I would not accept. I wouldn't give that manufacturer any more work.

Regards

 

[TGS4]

It is possible to have plasticity that does not equal permanent deformation. Local plasticity in bending happens in pretty much every vessel, often in multiple locations.

 

[GD2]

Mihe,

I am not familiar with EN 13445 but here is what ASME Sec VIII Div 2 tells about similar stress allowable for a hydrostatic Load Case for an elastic stress analysis.

I have noticed that your stated load combinations (P[sub]L[/sub]+P[sub]b[/sub]) is different than ASME VIII (P[sub]m[/sub]+P[sub]b[/sub]).

GDD
Canada

 

[r6155]

Visible permanent deformation is the lack of control of fabricator and AI. They do not made the job well.
I (as user) pay for a good work, I decide if acceptable or not.

"The AI will investigate and seek advice as to whether the deformation is acceptable". WRONG. AI cannot be judge and party.

Can anyone tell us what is the acceptable permanent deformation limit?: 5 mm, 10 mm ......... ?, how much then?

This situation is a commercial issue now ……… .or legal.

Regards

 

[TGS4]

r6155 - you are, unfortunately, conflating the Code-permissible limits on stresses during the pressure test and the note about permanent deformation in VIII-1.

VIII-1 has no limits on the stresses during the test condition, and so has this prohibition on visible permanent deformation.

EN-13445 and VIII-2 have specified limits on the stresses in the test condition that are set to preclude permanent deformation. I can confirm 100% that the limits in both EN-13445 and VIII-2 will, in fact, guarantee no permanent deformation.

 

[r6155]

@ TGS4. I insist with the question: "Can someone tell us what is the limit of acceptable permanent deformation?": 5 mm, 10 mm .........?, How much then? ". I am not interested in the value of the tensions reached, nobody can know after the test, it was an accident with the water pump.

Regards

 

[IdanPV]

@GD2,

ASME VIII-2 classified the membrane stress during the hydrostatic test as Pm (See Table 5.3) That is why you only find Pm in 4.1.6.2.
P.s
The document which you attached is very old, there were some changes in the Code.

 

[DriveMeNuts]

Actually, the AI would do nothing if they were concerned with visible deformation. Wouldn't give a big red REJECT stamp. Would just not sign anything and just do nothing.
The onus would be on the designer and fabricator to provide the AI with 3rd party audited document demonstrating that the deformation is safe and in compliance with the code. The AI would then be in a position to approve the vessel. The magnitude of deformation may not be the problem. Changes to the material properties or possible weld cracks at the location of concern may be the issue.

If the deformation is demonstrated to be safe and code compliant then it is a commercial issue. It is the purchaser's responsibility to specify no aesthetic deformation in the purchase documents. The designer would then have added material and the fabricator will charge you a healthy sum to accommodate that additional material. User pays.

It is often straight forward to work out if visible deformation is caused by the design or fabrication, but not always.

 

[TGS4]

r6155 - you can insist all you want on whatever question you have raised to deflect from the fact that your position regarding the magnitude of the stresses during the test condition is untenable. Until you admit that you were wrong about the test condition stresses, I will refuse to answer your questions in this thread.

 

[r6155]

I remember a problem at a reputable manufacturer: during the hydraulic test there was an incident with the water pump, the pressure was exceeded and there was a deformation (30 to 40 mm approx.) near the manhole. The Client was informed and the vessel was repaired over the weekend. This manufacturer (I worked at this shop in later years) did not admit his mistake. The investigation was not necessary. The business solution is mandatory and reputation prevailed.

Regards