Nominal Wall Thickness for 3'' Sch 80 Elbow and Tee 2

[nuk4]

Your advise is needed.
ANSI pipe thickness for a schedule 80 3'' pipe (carbon steel) is 7.62mm. Please can somebody tell me the corresponding wall thickness for a 3'' 90 degrees elbow, and a 3'' x 3'' Tee? What formula was used? Any internet resource of interest will be appreciated.

 

[jcfoley]

nuk4,
Most of the fittings are based upon a specific code as per the schedule of the pipe. ANSI has a lot of standard for these types of fittings. Butt Weld Elbows & Tees are covered under ASME/ANSI 16.9. If you specify this in your purchasing documents, you can then rest assured that your specified fittings are equal to or better than your pipe in strength. Hope this helps as you did not state why you need this wall thickness. I assumed that you were doing stress calcs.

Chris Foley
Midland, TX

 

[nuk4]

jcfoley, thanks for your response. ASME B31.3 and ANSI B31.4/8 is the relevant code for my stress calculations involving defect assessement for corroded piping.

 

[jcfoley]

nuk4,
In this case, I would suggest that you buy these codes. I can't ever remember seeing a wall thickness requirement for a fitting of this type (that does not mean that they do not exsist). I have seen this for fittings "created" in the field as found in the ASME BPVC. I might suggest that you get with your supplier and see if they can help you out. I am sure that a supplier like Weldbend will get you the information that you need.

Chris

 

[pipingdesigner]

Am I missing something here ?

Doesn't a Sch 80 3" pipe have the same wall thickness as a Sch 80 3" 90 Elbow and a Sch 80 3"x3"x3" tee ?



pipingdesigner

http://www.pipingdesigners.com

 

[jcfoley]

That is how I have handled the analysis in the past. I have never looked at the ANSI codes for these fittings for a specific wall thickenss. However, the fittings that I have examined, seem to be thicker than the like schedule pipe (this could be a supplier or tolerance issue). Does anyone have 16.9 so this can be verified?

Chris

 

[jte]

Pretty much by definition, an ell or (especially) tee must have a thicker wall than the corresponding pipe. If you are making a connection by stubbing one pipe into another, you'd do some area reinforcement calc's and probably add a repad (or more likely, use an integrally reinforced branch fitting such as a weldolet). So why would you expect a tee to have the same nominal thickness as the piping with no reinforcement. Ells are similar in that if you model their thickness as uniform, their stress distribution varies depending on location. So the fabricators have some incentive to optimize their designs to save a few ounces of steel per fitting which adds up to big bucks.

As to codes/standards, B16.9 does not specify thickness. from paragraph 2.2, Design of Fittings, "The design of fittings shall be established by mathematical analysis... contained in nationally recognized pressure vessel or piping codes, or at the manufacturer's option by proof testing in accordance with paragraph 9 of this Standard. In order to meet design or manuacturing 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..."

As to the original issue of a fitness for service analysis, if you're really concerned, then a FEA with at least shell elements (not beam elements like Caesar II) would be called for. I'd use Section VIII-2 acceptance criteria, unless you're in Section III territory.

jt

 

[jcfoley]

Great Information jte...a star for your efforts...

However, am not sure that we have given nuk4 the answer he wanted. Not only do you not have a standard thickness, the thickness is variable over the fitting.

nuk4, for what it is worth, I would try jte's suggestion of using FEA to analyze the fittings.

Chris

 

[JStephen]

First off, Weldbend's catalog shows the wall thickness of their ells and tees. It is the same as pipe.

I would assume that they mean this as the "minimum" thickness and that any additional thickness which may be present would be to allow them to maintain the minimum thickness after forming. They show the thickness tolerance as +0.12", -0.0" for 18" and smaller, +.019", -0.0" for 20" and 24".

Weldbend does show pressure ratings for fittings "based on code for pressure piping." "Ratings apply in accordance with Sections 1 and 5 of the code for pressure pipe."

 

[NozzleTwister]

nuk4,

Generally, the wall thickness of your tees and elbows will match the pipe. As mentioned above, in reality these fittings will be a little thicker than the pipe. The forged tees have integral reiforcement built in and the blank for forming elbows may be a little thicker to compensate for thinning during forging, but you will specify them by the wall thickness that you require based on straight pipe wall thickness calculations.

Nobody has mentioned materials yet, the statement above assumes that your fittings are equivalent (same allowable stresses) to your pipe.

In services where the Pipeline Codes, B31.4 & B31.8, are used, it's not unusual for your pipe and fittings to have different allowable stresses. Sometimes high strength pipe is used with weaker fittings that are a thicker wall. Also, with pipe bends that are commonly used with these codes, the thinning during manufacture does need to be addressed when purchasing a bend.

NozzleTwister
Houston, Texas

 

[nuk4]

Many kudos for all your response! In reality, tees and elbows are a little thicker than the same welded/seamless carbon steel[austenitic steel] pipe of the same pipe schedule {this can be verified when measured using the DM4 DL [krautkramer]thickness Gage}. For example,3''Sch 160 has the following wall thickness data:
a)pipe = 11.13mm
b)Elbow = 12.52mm
c)Tee = 12.52mm
For b] and c] the assumption is that the mill tolerance is 12.5% .Therefore the wall thickness for the elbow and tee is obtained by finding 1/8th of 11.13mm and then adding to 11.13mm to obtain the corresponding wall thickness.I find this difficult since am unable to see this in any of the ASME/ANSI applicable codes, or has anybody a clue to this? Am worried since the integrity of individual components when combined in a piping system needs to be checked to ensure that not only the individual components are fit to purpose but also the piping system as a whole.

 

[NozzleTwister]

nuk4,

You're a little off-base on you're conclusions from you're assumptions. You are right about this not being covered in the ASME/ANSI applicable codes (now it's ASME as the ANSI has been dropped).

The mill tollerances are covered in each ASTM material specification, ASTM A53, ASTM 106, ASTM A234, ASTM 312, ASTM 403, etc.

NozzleTwister
Houston, Texas

 

[MarlaMike]

Hi Guys

Sorry to interupt but its worth noting that ASME B16.9 only governs the wall thickness of the fittings at the extreme ends, basically therefore the thickness at the middle could be anything. It is common practice to use say sch100 pipe to make the fittings and just remachine the ends. Dont know if it helps

Mike

 

[jte]

nuk4-

You mentioned that you are assessing corroded components. In this case, you can refer to API-579 sections 4 )and maybe 5) and appendix A.5.5 (for elbows). Formula A.193 provides a minimum thickness of an elbow as a function of the pipe diameter, bend radius, and circumferential location on the elbow. Section 4 comes into play to evaluate whether your particular thickness profile will work. Not as easy to read and plug and chug as say B31.3 but it can be done!

jt

 

[JohnBreen]

Yes, B16.9 really only standardizes the end-to-end and center-to-end dimensions (and some "squareness" tolerences). As stated previously, the wall thickness at the ends (matching the pipe) is to be the same as the schedule of the matching pipe so that there will be no "high-low" at the weld fit-up. The geometry is not standardized beyond that. The Standard states that the fitting must be able to pass a "proof test" - e.g., in a pressure test "to destruction" the matching pipe will burst before the fitting fails.

Most manufacturers have their unique "shapes". Some are very "cylinder-to-cylinder" in shape, some are barrel shaped and some have pronounced spherical "cheeks". The variation in thickness through the fitting is simply the manufacturer's choice for the best use of material.

The crotch radii are not standardized on B16.9 Tees and that can really make a big difference in the stress intensification factor for stress analyses. The testing that was the source of the SIF's was done in the early 1950's and used the fittings of only ONE manufacturer. The Code says if you have better SIF data than that which is provided in Code Appendix D you can use it but not many of us have such "better data". Consequently, most stress analyses are approximate.

Good luck, John.