Challenge with CLASSES in ASME viii-2 Edition 2017 1

[Mm.Kaiser]

Hi,
As you know, in edition 2017 of asme viii-2, it has introduced the classes for vessels which made some challenges for me.
According to edition 2015 part5, I always perform FEA with S.F=2.4 and in E-P analysis, load factor 2.4. But in new edition, it seems that i have to consider my cases in CLASS1 which S.F=3 and load factor in E-P analysis=3.
These new criteria will make the thickness thicker although my special vessel parts and equipment according to previous version have been worked for more than 10 years without problem.
Please give me an advice.
thanks.

 

[TGS4]

Use Class 2. Ω=2.4 in that case.

 

[Mm.Kaiser]

Wonderful. Thanks TGS4.
Please explain a little more for the method to choose classes. I have to satisfy client.
Note that i am designing some special parts which don't have clear rule in part4.
Also Do you mean beta for Ω sign?

 

[TGS4]

Choosing the Class comes in Part 2. It's usually a decision that the User makes in the UDS. In your situation, it's likely the appropriate choice.

Yea, I meant β=2.4. Mistype with the special characters...

 

[Mm.Kaiser]

One more question:
What about PE certification? During these years we never engaged Professional Engineer to certify the UDS or MDR because it's not mandatory in my country. Note that i live in Asia. Also User's UDS is not update with new rules and classes.
Isn't there any conflict to choose Class 2 without PE certification?

 

[TGS4]

Division 2, until the 2017 Edition that included Class 1 that allowed no PE certification, has always required PE certification. I have no answer for why you were able to skirt the rules before. Nevertheless, the rules are the rules.

User's UDS should be updated appropriately.

 

[DriveMeNuts]

Mm.Kaiser, if you are interested, in Europe a design margin of 2.4 is also used on the European code and it is mandatory to have the calculations, operations manula and hazard assessment reviewed by a notifying body.
I understand that the difference between the US and the EU is that the in the US, the PE becomes partially responsible for the safe operation if the equipment and is therefore liable if the equipment fails during service. In the EU, the notifying body is just a checking service and is not liable at all.

The only way you could have skirted the rules is to fabricate the vessel "in general compliance" with Div 2 and not actually have it U2 stamped. I'm sure your country will have required that the calculations at least be checked by an independent engineer. Well, I hope so???

 

[Mm.Kaiser]

DriveMeNuts, i appreciate your kind attention.
Like Eu we have Native standard but not for Finite element analysis.
Note that We have to approve MDR by independent consultant company. َFurthermore, we consider all ASME criteria, examinations and NDT except PE certification. Also we have our vessels U Stamped not U2.
Thanks again.

 

[DriveMeNuts]

I've read your original post again. So in your original post you are saying that you have been designing Div 1 vessel and using β=2.4 for any FEM analysis?
To get a U-Stamp, for your existing vessels you should have been using a β of 3.5. This is the standard design margin for Div 1. The new Appendix 46 states what should have been happening for the last couple of decades.
Just because you used Div 2 design by analysis method to design a component on a Div 1 vessel didn't mean you could use the Div 2 design margins. You should have continued using the div 1 margin of 3.5.
This is a common (or convenient) error that slips through with div 1 vessels because the ASME AI is a specialist and fabrication inspection. They don't review the design report. The AI relies on the independent consultant company knowing what they are doing. They often don't.

The auditing of the Div 1 and Div 2 Class 2 calcs is completely different. I expect that the PE for an Div 2 Class 2 vessel is ASME accredited and signs their life away when they review a vessel. For Div 1, a random agency with unknown competence acts a checking service.

 

[Mm.Kaiser]

You are right.
General vessel designs according to div1 with S.F=3.5. Ustamp is for this part.
But some parts of vessel doesn't have exact rule in vessel in div1 (such as special quick openings) and should be calculated according to div2-part5 with SF=2.4.

 

[XL83NL]

Like Eu we have Native standard but not for Finite element analysis.

Where are you located?

 

[marty007]

Mm.Kaiser,

But some parts of vessel doesn't have exact rule in vessel in div1 (such as special quick openings) and should be calculated according to div2-part5 with SF=2.4.

This is wrong.

ASME VIII-1 paragraph U-2(g) up until 2019 has stated: "This Division of Section VIII does not contain rules to cover all details of design and construction. Where complete details are not given, it is intended that the Manufacturer, subject to the acceptance of the Inspector, shall provide details of design and construction which will be as safe as those provided by the rules of this Division."

The design margin for ASME VIII-1 is 3.5 on tensile. To design equipment that will be as safe as those provided in VIII-1 you should have been using a design margin of 3.5 for all VIII-1 vessels.

How can you say that you have been designing equipment to be as safe as those of VIII-1 if you are using a lower design margin?

Cheers,
Marty

 

[DriveMeNuts]

Mm.Kaiser,

Please read the following from the ASME publication from 2013, "ASME PTB-4 - Section VIII Division 1 Example Problem Manual"
It clearly states that you have been producing vessels which are not compliant with Div 1.

5.2 Paragraph U-2(g) – Design-By-Analysis Provision without Procedures
Paragraph U-2(g) states that VIII-1 does not contain rules to cover all details of design and construction. When complete details are not given in VIII-1, it is intended that the Manufacturer, subject to the acceptance of the Inspector, shall provide details of design and construction which will be as safe as those provided by the rules of VIII-1. Paragraph U-2(g) essentially permits the engineer to design components in the absence of rules in VIII-1 with the criterion that the design margin in VIII-1 be maintained. As discussed below, this requirement may be satisfied by using VIII-2, Part 4 or Part 5, as applicable, in conjunction with the allowable stress and weld joint efficiency from VIII-1.

 

[TGS4]

It is my personal opinion that the design of any component for a Section VIII, Division 1 vessel should use ß=3.5 for demonstrating Protection Against Plastic Collapse when performing a Design By Analysis per Part 5 of Section VIII, Division 2.

 

[Mm.Kaiser]

I thank all of you for spending time.
TGS4, i trust you and believe in your knowledge. But what about my real experiences in construction?
I have been design and construct lots of equipments like quick opening closures via FEM and S.F=2.4 and they are working without any problem. In my cases, Elastic results (stress linearizations) with s.f=2.4 are as same as elastic-plastics. (i mean if in elastic analysis, membrane and bending stresses don't pass the allowables, in E-P analysis the solution will not converge too and vice versa.
Is it logical for me changing the s.f to 3.5 and increasing the thickness subsequently?

 

[TGS4]

I will grant you that quick opening closures are a bit of a different beast. Provided that you are following the (non-mandatory) guidance in Appendix FF, especially FF-3, what you are saying might fall under "engineering judgement" in the first paragraph of FF-3. Of course, this also assumes that you are following the mandatory rules in UG-35.3. I suppose where this gets dicey would be in the interpretation of UG-35.3(b)(4), and how your AI views your calculations.

Nevertheless, I do not design quick opening closures, so I do not have additional insight on that specific topic.

 

[Mm.Kaiser]

TGS4,
If you decide to design and construct a 'quick opening closure' or something else like 'Split Tee' with finite elemnt analysis, what will be your CLASS and related ß factor? Think that there is no USD and you design it for yourself!!
Although the Question is ridiculous but your answer is detrminative for me.

 

[DriveMeNuts]

Many Pressure vessel codes in Europe use a design margin of 2.4 with a similar level of NDT and design verification to ASME VIII Div 1.

This would suggest that Design margin of 3.5 used by Div 1 results in robustly built pressure vessels (built like a brick out house).

This perhaps means that your closures are "safe". However, just because VIII-1 results in robust components, doesn't mean that you can just take it upon yourself to reduce design margins because you have experience. This is an area where the code does not allow the use of judgement and experience.

U-2(g) does not specify that the sub component be just "safe". It says that it needs to "be as safe as" the VIII-1 rules. Further ASME publications specify "as safe as" to mean using a margin of 3.5.

Thankfully, U-2(g) has been re-written and Appendix 46 added to prevent this convenient error from occurring to future fabrications.

The selection of VIII-1, VIII-1 C1 and VIII-2 C2 is a decision based on balancing the cost of the additional engineering, design verification and NDT for VIII-2 C2 against additional fabrication costs of VIII-1. If the use of 'design by analysis' is required, then VIII-2 C1 is pointless in most cases.

The decision may also be influenced by a customer requirement or there not being many VIII-2 C2 certified fabricators out there.

 

[Mm.Kaiser]

DriveMeNuts,

If i do a FEA on a part with design factor 3.5 (AS you addressed in ASME PTB-4), so...
when do you use CLASS 2 with S.F=2.4 in your analysis?

I have checked lots of products from famous European manufacturers with FEA and most of them will fail if we use S.F=3.5. For example most of Piping Tees will fail with Div1 margin and they are using S.F=2.4.

Note that when i use pipe material API-5L-X65 and forge material A694-F65 with Sy=450Mpa and Su=530Mpa, the allowable stress with S.F=2.4 is equal to 220 Mpa and with S.F=3.5 is 151 Mpa. You can see the significant margin.
unfortunately it is Impossible for me to use S.F=3.5 because my products will not remain competitive with more material.

 

[TGS4]

Mm.Kaiser - to answer your question:

- Quick Opening Closure - I simply don't know.
- Split tee enclosure - it is likely that I would match the design margin against burst from the Code of Construction for the pipe/vessel that it is repairing.

As I said before, Quick opening closures are a unique beast in this industry. Nevertheless, I will discuss the topic of design margins with my colleagues at next week's Code Committee Meetings. (Note, the meetings are open to the public - all are welcome to attend -

https://event.asme.org/Boiler-Pressure-Vessel-Code-Week)