Fitting Alterations 1

[garnet1]

By adding an integrally reinforced fitting ie: WOL, TOL etc, to an elbow, extruded tee or pipe cap, would you consider this as altering the registered fittng? I have an A.I. that says I need to support this with detailed calculations. I disagree. I am using B31.3 304.3.2 to support my claim. Example: 1/2" TOL welded to the bottom of an extruded tee for a drain.

 

[aybee]

wouldnt think there would be a problem with doing what you suggest provided the weld strength is good for whatever loads are acting

 

[deanc]

I agree with the AI. You have now created a "nonstandard"
piece and you need to justify the design. Should not be that painful.

 

[Scipio]

Actually I always interpreted 304.3.2 as strictly referring to branches off of piping runs, not fittings.

 

[jte]

garnet1-

How does your AI justify the existence of the Bonney Forge Elbolet?

http://www.bonneyforge.com/specs/branch/elbolet.cfm

This is a fitting explicitly designed to be attached to another "standard" fitting.

deanc- The engineering effort to justify these designs can be very painful indeed. For example, what is the standard thickness in various parts of a tee? Each fabricator designs their own, and they won't release their proprietary data.

jt

 

[garnet1]

I asked the same question about elbolets. I got a very indirect ansewer. Inexperience in the field is most of the problem I think. It makes it difficult to carry out good engineering practice when you stalled having to deal with such issues.

 

[Scipio]

jte,

Actually each manufacturer may design their own, but they have to work within established tolerances. In your example of a tee, for instance, in the case of a butt-welded tee ASME B16.9 has a wall thickness tolerance of not less than 87.5% of nominal thickness, which should be adequate for reinforcement calculations, so you know if you've got a 8" Schedule 80 tee you've got a minimum wall thickness at any given position of 0.4375".

Have to admit I never really thought of it much since I've only worked on branch connections on pipe runs, but it got me looking, here's what I've found, using a butt-weld B16.9 elbow on B31.3 piping as an example. First off, article 1.3 of B16.9 states "Fabricated laterals and other fittings employing circumferential or intersection welds are considered pipe fabrication, and are not within the scope of this Standard". That says to me, once we add one of those elbolets to our butt-weld elbow, if I consider that an intersection weld, that elbow is no longer under B16.9 design.

If that belief is correct, then our butt-weld elbow/elbolet fitting is no longer a listed component as defined by B31.3 paragraph 303, and a design calculation is required per section 304. That's where I lose track of things, as strictly speaking nothing in 304 really seems to apply to this particular example.

 

[deanc]

Well stated Scipio. It must be remembered each manufactuer,assembler,or repair firm is responsible for the design of thier work. If the fitting is accepted standard and addressed by ANSI-fine. If not or it is a proprietary design,such as an "elbowlet" the design needs to be addressed.

The AI has the right and the authority to ask for any justifcation he wishes. "We have always done it that way" has no meaning.

If you are not prepared to justify your design,you have no busniess holding a stamp.

 

[garnet1]

I put this on the same level as a U-stamped vessel for example, where nozzles under NPS 2" need no further reinforcement calcs. (Obviously not the same specifics but same thought) Would this mean that no more drip leg attachments using integrally reinforced wol's on the end caps without formal cals? Or thermowell / corrosion coupons installed on 90's? What calcs would a person use anyway? Proof of fitting integrity after the hole has been made, or just required weld amount or both? Good discussion. Something that isn't thought about too much.

 

[jte]

Hi Scipio (and deanc!)-

B16.9 may establish minimum acceptable thicknesses - in this case the corresponding min. pipe wall thickness after tolerance - for fittings, but that doesn't establish the minimum thickness necessary for the fitting to be safe. Take an 8" x 8" x 8" Sch. 80 tee. If it were built 0.4375" thick it would most likely deform under pressure due to the huge stresses at the size-on-size intersection. The tee must be significantly thicker to have adequate reinforcement. How much thicker and where that extra steel is distributed is the proprietary part.

I'm actually more familiar with B16.5 fittings, so I'll jump to B16.5 Annex D and note that paragraph D1.2 equation (1) is essentially 1.5*(P*r/S) with S defined as 7000 psi. So an 8" Sch 80 Cl-150 flanged tee should have a min wall (not flange) thickness of 1/8" bumped up to 0.31" in Table 10 ("Actual values in the dimension tables... are approximately 0.1 in. to 0.2 in. heavier...&quot.

Paragraph 6.1 of the main body of text states in part, "Additional metal thickness needed to withstand assembly stresses, shapes other than circular, and stress concentrations must be determined by the manufacturer, since these factors vary widely. In particular, 45 deg. laterals, true Y's, and crosses may require additional reinforcement to compensate for inherent weaknesses in these shapes."

Again, this code specifies a minimum to meet the code - but not the minimum necessary to safely retain pressure without excessive deformation. That part is explicitly "determined by the manufacturer."

I appreciate your insight - let me know what you think!

jt

 

[deanc]

Gentlemen:

This is what makes the Code fun. Esp.with a hard headed inspector and engineer.

Suggestion: Justify your fitting for pressure(ANSI or by cals.)Then justify your weld size,by default in the Code or calcs.

You could qualify as a non-standard design and proof test.

What I am getting at is to provide the AI with something he can buy off on. If the Code has the answers-fine. If not use
an engineering justifcation.

Life could be better for all,ask your AI what he wishes to see. Been doing this for alot of years and the "dance" is important.

jte: I agree with you on your last entry,sound engineering must always be used beyond the Code requirments.

Regards:deanc

 

[garnet1]

Yes Dean is fun...especially when there is a hard headed inspector on the other end LOL. I would agree with your last statement, in that proper welding, examination, testing and documentation is required (and made available to the AI). I ask though, if you have pressure rating of the initial fitting ie sch 80 BW elbow A234 WPB and 300 psi design pressure, use a class 3000 WOL in accordance with B16.11, why the calcs if you install the WOL in accordance with manufacturers specs. Using good practice etc in staying away from any other stress zones, welds etc.

 

[deanc]

OK,How about this. Show the AI the ANSI sections that justify the pressure for the fittings or refer to them in a signed letter. Then justify your weld size to the Code section to which you are building. If it is by default list the paragraph. If by calc.show them. You and the AI need something to hang your hat on,not the fitting manuf. recommendations.

Does this make sense?

 

[garnet1]

I agree.....

 

[jte]

Sheesh... another accusation of being a hard headed engineer! Oh well, guess I'll have to go home and drink some beer to soften it up. deanc- if you'll be at the PVP conference let me know, I'll buy you a beer too!

jt

 

[twjag]

garnet1,

You can address this question to Bonney Forge.
Years ago, BonneyForge provided tables with full dimensions for WOL, TOL, etc., that gave you all the dimensions you needed for calculating reinforcement per the various ASME PV & piping codes. They no longer publish these dimensions.
Bonney Forge should be able to give you the dimensions needed for the calculation, or provide a supporting calc to you.

 

[bvi]

Technically, the AI is right, although almost no one does detailed calculations in these circumstances. Modification of the elbow makes in unlisted. However, in my opinion, one should go the the manufacturer of the branch connection fitting, as suggested by twjag. The fitting manufacturer is required to have proven the design, which should include proving the design for use with elbows if that is the intended application. Assuming the manufacturer has proven per 304.7.2 (e.g. by burst testing or extensive succesful service experience), documentation showing this is required to be available for the owner's approval. If the manufaccturer is unwilling or unable to do so, you are faced with qualifying the component yourself, as an unlisted component, or with convincing the AI that this is unecessary.

 

[LSThill]

HI Team members:

If Piping Standard Specfication Branch Table use in the above piping installation no additional calculation is needed. The Piping Standatd Engineer check all posibility before the Piping Specification was approved and issued.

The tools use in the above is FE-Pipe for the final check by the Pping Standard Specification Engineer.

Leeonard@thill.biz

http://www.thill.biz

 

[LSThill]

HI Team members:

If Piping Standard Specfication Branch Table use in the above piping installation no additional calculation is needed. The Piping Standatd Engineer check all posibility before the Piping Specification was approved and issued.

The tools use in the above is FE-Pipe for the final check by the Piping Standard Specification Engineer.

Leeonard@thill.biz

http://www.thill.biz