UB-16(b) Interpretation 1

[TJOrlowski]

Colleagues:

A hypothetical question:

Construction Code ASME Section VIII, Division I, latest edition/addenda. A TEMA type BEM shell & tube, with single course construction of the shell. Shell size 6" IPS, out of SB-619 UNS N10276 Alloy C276 pipe. Design MAWP 75 PSIG + FV @ 325F, with a .0625" inside corrosion allowance. No RT.

What is the minimum required wall thickness (in decimals, not wall schedule) of the shell? How did you arrive at your answer? I have an engineer that I believe is mixing and matching code sections in a design he's submitted for quotation.

-TJ Orlowski

 

[gr2vessels]

TJ, you didn't mention if the MAWP is equal to the design pressure. If you use the MAWP to determine the wall thickness, the value could be higher than using the design pressure.
I guess your problem comes from ASME II, Table D, note G14 (division with 0.85 for long stress calcs). The rest is a formula, paper and pen, nothing fancy.
PS. Give me the design pressure and I'll run the calc for you.

 

[TJOrlowski]

Gr2vessels,

Sorry, my question was not clear. Design pressure is 75 PSIG + FV @ 325F. Relief devices will be set at < 75 PSIG, so MAWP is moot. I agree that this calculation should be pretty straight forward.

-TJ Orlowski

 

[TJOrlowski]

Nobody huh? I was expecting responses from all the usual suspects on this. . .

-TJ Orlowski

 

[Duwe6]

Pressure - only thickness will be about 0.015". On small Hx's made from pipe, mimimum thickness gets very small, even with bending stress and corrosion allowance figured in. The pipe-mill hydro on 6NPS pipe is on the order of 1000 psig.

 

[SnTMan]

...then, applying UG-16(b), I'm getting 1/16 + 1/16 CA for 1/8 total.

Applying UG-16(c), I get Duwe6's 0.015 + 1/16" CA + pipe wall tolerance, 12 1/2% of 0.0775, total 0.087

UG-16(b) governs.

Regards,

Mike

 

[TJOrlowski]

SnTMan,
I'm in total agreement. I have an engineer (and also a programmer at TWO well-known pressure vessel software outfits that we license) that I believe is (are) combining UG-16(b) and UG-16(c): asserting that pipe wall tolerance needs to be added to he UG-16(b) minimum of .125".

I had to go back to make sure I hadn't missed something in the new addenda. Our AI and his supervisor are also both in agreement that the UG-16(b) minimum is .125". I just wanted to see what some other fabricators thought.

-TJ Orlowski

 

[jamesl]

I see the HX is designed for FV, but no length of shell was given. It seems to me the designed is controlled by external pressure regardless of the length. The minimum required thickness (exluding UG-16(b)) will be more than 0.015+0.0625CA in order to have FV.

Of course, UG-16(b) will govern eventually.

 

[TJOrlowski]

120" shell length, but yes, we'll always be going back to UG-16(b).

-TJ Orlowski

 

[jamesl]

If the shell is 120" long, required thickness for FV is 0.0584"+0.0625"=0.1209", still slightly less than UG-16(b) requirement.

 

[SnTMan]

I suppose it could be considered a bit of a gray area.

UG-16(b) says "regardless of product form", leading you to disregard product form tolerances, UG-16(c) says "after the minimum wall thickness is determined" with no reference to how it is determined, i.e. possibly per UG-16(b).

Regarding software, it seems to me that the program I use most often gets more buggy with each new release. What can you do?

Regards,

Mike

 

[Duwe6]

My interpretation is that with pipe you can either use "actuals" or the nominal less the 12½%. And unless the actual slice of pipe in available at the time of the vessel calc's, you are back to the 87½% of SA-53/106 nominal.

I cannot interpret "regardless of product form" with as-delivered wall thickness. 'Form' means Pipe, Tube, Plate, Formed Head, etc.

 

[SnTMan]

Duwe6, I interpret that "regardless of product form" means that mill tolerance can be disregarded, as when that language formerly appeared in UG-37 for example.

So for the UG-16(b) example above, nominal wall selected would be 1/8" or greater, tolerance disregarded.

(Would normally say Regards here, but not this time

Mike

 

[TJOrlowski]

I'd interpret UG-16(d)'s (reading from the 2011a addenda) "after the minimum wall thickness is determined" as: using the applicable code section or appendix. Therefore the pipe undertolerance allowance stays in that calculation. UG-16(b) makes no reference to allowing for undertolerance.

I understand it to be: take the GREATER OF your applicable calculation, or 1/16 + CAi + CAo. Our AI, AIA, and it appears that those in this thread agree.

As to software - my view is that they save time on "cookie-cutter" designs. But as soon as you get a project where your specifications are a little wide of your scope (like a small diameter C276 HX built to TEMA B with a 1/16" CA, where 9 out of 10 like this would have no CA), they actually end up costing time.

I spent less than 60 seconds determining that a Sch.10 shell would be OK for this unit originally. Then when one of our engineers said that we'd have to use Sch.40 - it has cost the both of us hours upon hours.

-TJ Orlowski

 

[SnTMan]

I'd as soon the software stick to just doing calculations, and let me figure out the other Code stuff. The software I use calculates MDMT's (often WRONG), extreme fibre elongation for seamless pipe (WRONG), MAWP's, but NOT for a flange pair with different design pressures as across a tubesheet.

Just had it flag a CS exchanger at 650 F as being in the creep range, and use 3S allowables. WRONG. Couldn't override it.

More that half the output is stuff for which I have no interest and no trust.

<End of rant>

Regards,

Mike

 

[jamesl]

Sch 10 pipe is not good enough for this case. For pipes, you cannot use nominal thickness or actual thickness. You have to use nominal minus specified undertolerance.

 

[TJOrlowski]

jamesl,
I understand what you're saying, but I don't interpret the language in the Code that way. The words "regardless of product form" in UG-16(b) tell me that I should DISREGARD whether I'm using pipe or rolled and welded plate, and use a minimum of 1/16" + CA.

Applying your interpretation: my customer would be better suited, and the Code satisfied, if I bought a piece of 10 GA C276 plate, and made a rolled and welded shell (I wouldn't even have to perform RT on the shell, my required thickness would be < .125" even with a .60 efficiency on the long joint, so UG-16(b) would still govern). I don't have a roll with both top roll small enough and enough length to make the cylinder in a single course, so there's going to be at least one circ seam in the shell (still satisfies Code though). Call me naive, but I don't believe that to be the intent of the Code.

I suppose I could just have the C276 stockist perform an eddy current on the the piece of pipe. If it's >= .125" throughout, we're good to go.

-TJ Orlowski

 

[jamesl]

If you use a sch 10 pipe, nominal thickness is 0.134". From the Code's view, its minimum thickness is 0.134" X 0.875=0.117". It is below the required 0.125" per UG-16(b). It is also less than required by external pressure as indicated in my previous post.

If you roll a 0.125" sheet, you will be OK from the Code's view. Also full RT or no RT has no impact on external pressure rating.

If the MTR shows a sch 10 pipe, even your thickness measurement is 0.125" throughout, it does not count per Code. There was a code interpretation very clear about this issue.

Of course, if the HX is not stamped, it will be another story.

 

[TJOrlowski]

jamesl,
Again I have to respectfully disagree. Paper documents do not take precedence over actual values.

If you're telling me that empirical data has no value in the construction and certification of Section VIII, Div.1 vessels and heat exchangers, the other side of that coin is that I would be able to use an MTR showing passing physical properties as a defense for putting inferior material (material that didn't pass my verification of the MTR) into a Code unit.

Paper documents support the actual product being built, not the other way around.

-TJ Orlowski

 

[jamesl]

TJ,

Here is Code interpretation VIII-1-86-11:

Question (1): Does UG-16(d) allow the selection of nominal pipe size material which has an actual minimum thickness which is equal to or greater than the calculated minimum wall thickness, regardless of the fact that the undertolerances provide material which is less than that required?

Reply (1): No.

Question (2): Does UG-16(d) require that the nominal thickness less the manufacturing undertolerance be equal to or greater than the minimum thickness required?

Reply (2): Yes.