ASME Section I, PG-32.1.2

[Raaden]

Hi everybody!

I have a short simple question for today.

The factor "t" is described as: actual thickness of the shell. So which is the definition for "actual thickness" of a seamless pipe? Should "t" be entered as the nominal wall thickness or should I use the minimal wall thickness. Because for me, "actual" sound as if I have to go out in the workshop and physically measure the pipe that I'm going to use for my design

Thanks

Pontus

 

[arto]

how about t = nominal th'k - mfg. tol [12.5%] - corr. allow. ?

 

[Raaden]

That was sort of the answer I was expecting, thanks arto

 

[PAN]

arto,
If the vessel is not in the corroded condition, should we minus corrosion allowance?

 

[GenB]

Actual means the ussable t of a pipe as the nominal less mfr's tol.- In some tube sizes and plate the actual is the
actual.
ER

 

[RNDguy]

Since ASME is a design code, for it "actual" would mean the actually used nominal thickness. While at the design stage, no pressure vessel or boiler really exists, so the term actual only means what the designer is finally adopting as the value of thickness. I am going by memory, but I believe somewhere in the code the issue of manufacturing tolerance is dealt with and it is stated that nominal thickness can be used without subtracting the manufacturing tolerance.

 

[GenB]

Yes you can use the nominal.
A lot of work has to be put into "the pipe" to meassure every inch and proove the AI that you are complying to Code,
it is certainly easy to deduct the Mfg tolererance allowed and forgrt about it.
your choice.
some other materials as tube you will forcefull deduct its tol.
ER

 

[aclark]

It seem that there is a lot of confusion on this subject. Here are the actual rules. When you are calculating the required thickness for a shell or nozzle made of pipe the 12.5% MUST be taken out the method that ASME uses is to multiply the nominal thickness by .875 you can see this in table X2.4 of the pipe spec.

In PG 32 and PG 33 the thickness of the nozzle is defined as the NOMINAL thickness. Section VIII is even more clear UG 37 also defines tn as the NOMILAL wall thickness. The reason for this is that you have already accounted for the tolerance when you designed the nozle for thickness.

If an AI doesn't understand this the you need to point this out and discuss. I have been an inspector for almost 30 years and an ASME team leader for over 10 and a lot of Code Cert holders don't understand this either.

 

[GenB]

aclark,
Then, can you please clarify the actual thickness original question
may be I still did not ge it!
"actual" thickness please

ER

 

[RNDguy]

Gentlemen,
Can I refer to PG-16.4 & PG-16.5, former is for plate undertolerance and the latter is for undertolerance on pipes and tubes. For plates, 0.01" undertolerance is allowed. For tubes and pipes, this is stated: "Pipe or tube material shall not be ordered thinner than that calculated from the applicable formula of this Section. The ordered material shall include provision for the allowed manufacturing undertolerance as given in Section II in the applicable pipe or tube specification." So what does this last sentence really mean? Does it not mean that nominal thickness can be used for design while undertolerance within limits as specified in Section II is maintained?
More than that, does not the ASME B&PV code state that factor of safety used in allowable stress includes fabrication and manufacturing tolerances? Applying 0.875 factor over and above inspite of all this seems like an overkill.
My 2-cents' worth. Thanks.

 

[prex]

I must respect aclark's experience, but I'm afraid that at least my interpretation of the code (as used in many years of design activity) is as follows.
Let's first of all recall what is said in UG-16:

(d)Pipe undertolerance. If pipe or tube is ordered by its nominal wall thickness, the manufacturing undertolerance on wall thickness shall be taken into account except for nozzle wall reinforcement area requirements in accordance with UG-40...

To me this is a clear statement and is equivalent to the statement of ASME I quoted by RNDguy(yes RNDguy, the meaning of that phrase is exactly the contrary of your interpretation!).
Its meaning is that the thickness of pipes as used in the design must be smaller or equal to the minimum thickness of pipe, that is the nominal specified thickness less any manufaturing tolerance (and less the corrosion allowance of course).
This is true everywhere in the code except only in UG-40, where the calculation of useful distances and lengths to calculate the area of reinforcement may be based onto the nominal wall thickness.
This also means to me, aclark, that the words nominal thickness of nozzle wall in UG-37 use the definition nominal = as used in the design and not the other one, encountered in UG-16 nominal = thickness per product specification. This is clear if you look at the definition of t for vessel wall in UG-37: the same nominal word is used, where the undertolerance is undoubtedly to be deducted (except the bonus of 6% or 0.01 in per UG-16).
So coming back to the original Raaden's question concerning ASME I, I think, paralleling the above way of reasoning with ASME VIII, that actual in PG-32.1.2 stands for the thickness by which the material is ordered (though I admit that this liaison is not so clear), and that nominal as used in PG-36.4.4 stands for the design thickness, that must be smaller than the minimum thickness.
Anyway I fully concur with generalbir: by using everywhere the minimum thickness (specified minus undertolerance minus CA) (as I always do) one gets, at a minimum absolutely negligible cost, an additional factor of safety for future evaluations, and spares the time of discussing the point with the AI or anyone else (including this thread).

prex

http://www.xcalcs.com

Online tools for structural design