Elbows and Tee-Sections in B36.19 S/S pipe Sch40 4

[jackdaniels]

HELP PLEASE !!
I am reviewing some unfamiliar plant equipment and am using the P&IDs to help me find my way around that part of the plant. These drawings show ASME/ANSI B36.19 schedule 40 pipework is specified, albeit at at varying diameters. Whilst not being the main focus of my review, I am checking for 'ageing' of the pipework (wall-thickness) by ultrasonically examining them. Having read B36.19 - and searching ASME standards, I can't seem to find anything that derives wall thickness of elbows and tee sections. My question is simple: does the calculated design wall thickness (that required sch 40 pipework) simply 'read-across' to elbows and tee-sections?

 

[SNORGY]

I think yes, provided the elbows and tees are manufactured and registered to a recognized standard. Then the computation of minimum wall thickness is the same as for straight pipe.

Regards,

SNORGY.

 

[jackdaniels]

Cheers, SNORGY

I have no original design calcs to check thru, hence my quandry.

Couldn't help but think that 'loading' might be increased by any effects of the fluid 'going round the bend' as opposed to a straight piece of pipe section. But aside from erosion/corrosion in some tee-sections, we seem to be there or therabouts on sizes. The replacement tee-sections we have in stock seem somehow thicker, more beefier than the measurements suggest that the in-plant ones are.

 

[XL83NL]

Have you checked ASME B16.9 table 13, 'Tolerances'?
FOr US Customary units, go to table I-12.

 

[jackdaniels]

Great response guys, thanks.

Section 2.1 of the ASME B16.9 actually states 'The allowable pressure ratings for fittings designed in accordance with this Standard may be calculated as for straight seamless pipe of equivalent material...'.

So I am progressing - without original design cals available - with the elbows + tee-sections min wall req'ts same as piping reqt's. It is as I thought; but at least I now have some reasoning, and an ASME statement, behind my decision... rather than just my own theory.

And I had a weekend to mull it over, always good to have that option!

Jackdaniels

 

[4Pipes]

This is a really good question. I thought that these fittings are based on having a burst test equal to that of straight pipe of the same schedule. Hence the wording of Section 2.1 above. The actual thickness of the fitting at any location could be anything necessary to achieve this same burst test as straight pipe. "In order to meet design or manufacturing requirements, it is expected that some portion of formed fittings may have to be
thicker than the pipe wall with which the fitting is
intended to be used". Unless you know the "as-burst tested" thickness profile used by the particular manufacturer, how do you know what thinning to allow at each location. The only place where you know the thickness is next to the weld prep.

 

[XL83NL]

jackdaniels,

please be aware that, in e.g. B31.3, for design calculations of wall thicknesses, one determines the 'rating' of fittings by taking the nominal wall thickness of seamless pipe, and reducing that wall thickness with 12,5% (on top of the 12,5% mill tolerance).

allthough this isnt a straight answer to your question (which I found hard to understand - probably cause Im dutch), its something you might want to consider when determining wall thicknesses per code requirements!

 

[jackdaniels]

@4Pipes - all the tolerances seem to position a welding joint from the elbow centre-line etc, or have data for wall thickness only where the welding joint is... similar with the tee-sections.
I think you are correct on the burst tests, but i have even less information on this!!!

@XL83NL: So if I understand this correctly, there should be a 12.5% margin on the mill size of the pipe being made, due to manufacturing variation;
then a 12.5% tolerance on the pipe size rating due to the nominal size being low;
and then a margin on the design size, due to choosing the Schedule 'up' from a design wall thickness calculated.

Thanks to all for your postings.
Whilst still unclear on actual 'rating sizes' for fittings, I now feel theat I have enough information to mandate the replacement of certain sections of our piping. When we order new, we will pay for the knowledge of the suppliers with regard to wall-sections of fittings.


Many thanks, and kind regards,

Jackdaniels

Jackdaniels.

When theory and practical don't agree, it's the theory that's wrong!

 

[XL83NL]

@ jackdaniels;

No, that's one times a 12,5% tolerance too much. There should only be 2 reductions of 12,5% on wall thicknesses; one for mill/manufacturer's tolerances, one for fittings. Please note this is for ASME B31.3 LISTED fittings (like e.g. B16.9 tees, reducers, elbows, etc.)
However, if you e.g. only use pipe + flanges, you have no fittings, and thus only have to reduce the wall thcikness once (for the mill tolerance only).
Also, please refer to para 302.2.2 of B31.3:2008, and keep in mind that all the above is for calculating the required wall thickness given a known design pressure.

 

[jackdaniels]

Thanks XL83NL,

No problem, so if the fittings have a wall section = B36.19 sch40 design min wall thickness PLUS 25%, they meet design?



Jackdaniels.

When theory and practical don't agree, it's the theory that's wrong!

 

[augern]

@jackdaniels

That is not quite right. A new Sch 40 fitting will have a design pressure rating equivalent to that sch pipe of the same material. The fitting will have a greater minimum wall than sch 40 pipe. 4Pipes is correct. Read section 2.2 B16.9.

 

[jackdaniels]

Cheers augern,

Having read section 2.2 of B16.9 I now know that to understand what the fitting's wall thicknesses should be I have to establish the value from the designer's calcs; simply reading across from straight pipe is risky. Another option would be to get them recalculated, as we might not have the calcs to be made available to me.




Jackdaniels.

When theory and practical don't agree, it's the theory that's wrong!

 

[sjrfc2]

If you're looking to get some thought on minimum thickness for the elbows, you might want to look at ASME B31.1 para 102.4.5. This gives you a min for the intrados and extrados. My guess is that it's origin is rooted in the torus formulas.

 

[XL83NL]

sjrfc2, I believe those intrados/extrados formula's are for the minimum required wall thicknesses after bending a straight piece of pipe into a bend (at least, in B31.3, there's a similar section, so based on that, and assuming the analogy for this part on the .1 and .3 editions are similar ... )
Im not 100% sure though, you have to check the reference I made to B31.3 with B31.1.

 

[jackdaniels]

Thanks guys, I thought I was confused, now I am just not sure of the level of confusion I am at.

I can't help but feel that it might be clearer if a body (ASME/ANSI?) were to write a standard stating what each fitting would meet, eg sch 40 fitting will suit same limits of a sch 40 pipe (could list the required wall thickness).

But would this bring in room for error when an instance of a higher rating fitting is needed? (Although I can't think when this might occur). Such as a sch 40 pipe needs one so-called 'sch 80 fitting' but the remainder can be sch 40?

It also raises a question of whether fittings are made specific for a customer, or, for want of a better description, to a 'schedule number'?

Jackdaniels.

When theory and practical don't agree, it's the theory that's wrong!

 

[sjrfc2]

Sorry I didn't mean to indicate that the intrados/extrados applied to the ASME B16.9 fittings. I think the mfgs will need to define wall based on their mfg process so that they meet the burst test requirement. I agree that Sch 40 fittings will be matched IRT service with the Sch 40 pipe (due to the burst test requirement).

I agree the intrados/extrados formulas are for bent piping where you're going to get backwall thinning and thickening in the crotch. It seemed like original question centered around UTs of existing piping and how to assess thickness of sections in place (which may have been subject to erosion/corrosion). If there were some inconsistencies in wall thickness it seemed as though a check against the intrados/extrados formulas was a good sanity check IRT B31.1 requirements.