WRC 198 or WRC 448?

[StevenHPerry]

Which is more applicable to the scenario below: WRC 198 or WRC 448? Or is there another standard that might fit better? I have only access to the descriptions available at the forengineers.org at this point and would like to recommend the firm buy one or the other but not both.

My problem:
I need a method of evaluating the stresses in the pipe wall for a W-type I-beam (preliminarily, W12x40 if that matters) welded to 54"OD 3/16" thick pipe (acting as lugs for the supporting spring cans).

I have access to WRC107/297 but these methods do not cover such a cross section.

I have reviewed Blodgett's Design of Welded Structures but wish to have a more "official" code, bulletin, or practice.

Other than WRC 198/448 are there any other tricks out there? Can one give a W12 an effective rectangular dimension to be used with WRC107?

Thanks,

Steve Perry

 

[JohnBreen]

Steve

Please understand that Welding Research Council Bulletins ARE NEITHER CODES NOR STANDARDS. The ONLY "official" document you have to comply with is the applicable Code mandated by your local jurisdiction. The good news is that these Codes allow you the latitude of utilizing more rigorous methodologies to assure compliance.

What Code does your local jurisdiction require you to comply with?

If you are trying to support a pipe with a rolled "W Shape", you may have to go to a FEA analysis to satisfy the requirements of the prevailing Code or Standard.

The piping Codes are based upon beam theory and beam theory will not predict the highest local bending stresses in a pipe wall at the location of welded appurtenances.

Judging from the section modulus of the W shape that you are welding to the pipe, you have a significant loading. You are likely going to need a rather thick pipe wall to weld this W shat to (to accommodate the weldment) so you will have to look at the sustained local bending stresses at the welds and also evaluate the local thermal stresses caused by the restraint of the (colder) W shape.

Regards, John.

 

[BillBirch]

Steve,
If you are interested in the local stresses in the pipe, I would suggest using a more regular shaped (elliptical?) reinforcement pad between the I-beam and the pipe wall.

 

[BigInch]

I presume there is a very good reason you must weld the W to the O and there is no way that you can put the pipe on top of the W and the W on top of the spring cans. Right?

http://virtualpipeline.spaces.msn.com

"We can't solve problems by using the same kind of thinking we used when we created them." -Albert Einstein

 

[LSThill]

StevenHPerry (Mechanical)

Regards to FLUOR DESIGN GUIDE PIPING STRESS ANALYST has the WRC 198 PROGRAM IN TI-59.

Reference to Dennis Moss Secound Edution WRC 198.

Also,

http://www.Paulin.com

FE-PIPE Software dose this type of Analysis.

L S THILL

 

[LSThill]

WRC BULLETIN 448 Evaluation of Welded Attachments REPLACE WRC 198

Introduction

The purpose of this report is to summarize the background of Code Cases N-122-2 and N-318-5. Case N-122-2 is included in this report as am appendix. Case N-318-5 is Appendix E of report 2. Frequent reference will be made in the following text to these Cases. The definitions of symbols used in this report are those of the Cases. Other symbols are defined where used in the text and/or table.

N-122 and N-318 were first published around 1980; some four years after the publication of Welding Research Council Bulletin 198. The motivation if the development of N-122 is described by the following extract WRC198:

"A piping system design requires the specification of a corresponding support system whose function is to transmit force to the piping system or to limit its deformation. Structural attachments which connect the pipe system with its supporting elements are classified into two general categories, those integral to the piping system and those that are not . Lugs welded to the pipe are examples of integral attachments. whereas roller supports and various types of clamps are examples of nonintegral attachments. Article NB-3645 of Section III of the ASME Boiler and Pressure Vessel Code requires that the effects of attachments in producing thermal stresses, stress concentrations, and restraints on pressure-retaining members shall be taken into account in checking for compliance with design criteria."

The preceding quote is directly relevant to N-122 (Class 1 piping); WRC 198 covered only CLASS 1 piping. However,having a relatively simple method of calculating stresses at lugs, the analogous N-318, of the Class 2 and 3 piping, was developed.

During the years between 1980 and 1994, N-122-1 and N-318-1,-2,-3,-4, were published. These mostly involved clarifications of how to use the Cases. in 1994. N-122-2 and N-318-5 were published. These editions included a major change in that the coefficients of the equations of CT, CL and CN were divided by two. This report deals, in large part, with the bases for making this change.

The review of this report leads to several suggestions for revision N-122-2 and/or N-318-5. These are discussed in Sections 4.3, 4.4 and 4.5 of this report and are summarized in Section 8.


Regards to FE/Pipe "bend with Stanchion" page 1.8.4 compare to WRC Bulletin 448 January 2000 Evaluation of Welded Attachments on Pipe and Elbows

Report 1: Background and Suggested Revisions:
ASME Code Cases N-122-2 and N-318-5 "Lugs on Straight Pipe" , E.C. Rodabaugh & E.A. Wais
Report 2: Evaluation of Welded attachments on Pipe and Elbows
E.C. Rodabaugh
E.A. Wais
G.B. Rawis



L S THILL

 

[4Pipes]

I'd forget lugs of any shape until I confirmed that 54" x 3/16" did not require a full 360 degree ring. I would suspect that a ring would be required for handling if nothing else. Moss, Bednar, Roark would all be good starting points.

 

[LSThill]

WRC Bulletin #198
Abstract
Analysis of the stresses in a cylindrical shell cause by external loading through a welded attachment. The source equations are from WRC-198 by Dodge, Rodabaugh and Moor. The program will analyze a rectangular or a circular
attachment for bearing loads, bending moment, shear forces and tensional moments. The output consists of two sets of stresses, Dodge’s original analysis, and the modified analysis, applied to the Primary Stress Indices
equation for ASME Section III.

The first output data set is based on Dodge’s original paper which calculated the stress base on the section modulus of the pipe. The program uses all of the equations 1 and 2, and equations 10, 11 and 12 for this data
set. The Second data set is based on the contributions of Rodabaugh and Moore, which adjust the stresses so that they are based on the section modulus of the welded lug. The second set was intended to be added to the Primary Stress
Indices that are outlined n ASME SECTION III, Paragraph NB 3652. The data set uses all of equations 3 and 4 to adjust the stresses for data set 1 to get the stresses found in Equation 21. Equations 10. 11 and 12 are also part of the
second data set.

LIMITATINS: The analyst must e careful which data set is to be used, Dataset #1, nominal stress in the pipe, is intended to be stand-alone stress value for the pipe at the point of attachment, not regarding the internal pressure
stresses. Dataset #2, nominal stress at the base of the attachment, is intended to e added to all other stresses calculated for the point of attachment by independent means, and then compared with the allowable, as
presented in Paragraph NB3652 of ASME SECTION III.

The author of WRC-198 suggests certain size limitations for the welded attachments. The are:

Cod/OD < 0.3 for circular attachments
(B1/0.3)² + (B1/1.2)² < 1.0 For rectangular attachments

The analysis found I WRC-198 is not the same as that found in WRC-107, and therefore should not be expected the yield the same results. WRC-107 is based on Bijlaard’s equations and is a very involved analytical approach to attachment or nozzle stresses. WRC-198, on the other hand, is an empirical
approach which develops stress indices for the welded attachments based o the shape and dimensions of the attachment, and the diameter-to-thickness ration
of the cylinder.

On page of WRC-198, the author summarizes their comparisons with WRC-107 as
follows:

“In summary, for thrust and moment loadings, comparisons
among the WRC-107 method, Dodge’s direct calculations of
Bijlaard’s analysis, and the stress-index method, developed
in Ref.2 (egs 4a, 4b and 4c herein) show that the stress-index method tends always to be conservative. For square loaded areas, the WRC-107 method gives about the same results as the stress-index method. For loaded areas that are not square (i.e., B1 = B2), WRC Bulletin 107 uses an approximate method (based on Ref. 7) which Appears to be adequate for thrust loading over the parameter range
Considered and fair for moment loadings within the parameter rang ¼ <= B1/B2 < = 4; within this range, the magnitude in conservative, but the location and direction may be in error.On the other hand, the WRC-107 method may be misleading for Moment loadings on integral lugs which are long and narrow (i.e. for cases where B1/B2 < = ¼ or B1/B2 > = 4.”

(Ref. 7 is a paper by Bijlaard entitled “Stress for Radial Load and External Moments in Cylindrical Pressure Vessels”.)
The same equations are used by WRC-107 and WRC-198 for shear loading.


JohnBreen (Mechanical) please add any additional comments for WRC 448 comments and team




L S THILL

 

[JohnBreen]

Steve Perry,

I guess I am getting back to "what do you want to do" at this point.

If you are looking to satisfy the ASME B31.3 Code, you can do that by using FEA (suggest FE/Pipe) and the analytical methods of the ASME B&PV Code, Section VIII, Division 2. With 56 inch OD and 3/16 inch wall bearing on that "W" shape, you are certainly going to need a reinforcing plate (aka, "doubler plate", etc.) or a saddle to distribute the loading over more contact area. In fact, you may need a stiffening ring at the support point to keep the 56 inch OD pipe "round" in cross section. That "point loading" will likely include a LOT of punching shear. I do not think that looking for shortcuts ("tricks") will get you where you want to go. You are going to have to address ALL the local membrane stresses and compare them to Section VIII, Div. 2 allowable stresses.

As an alternative to FEA, you can use a beam theory model (Caesar II) of the piping system to get an approximation of the loadings at the support points but the calculated beam bending stresses will be meaningless. The design will then become similar to designing supports for horizontal vessels. This type of design usually takes the form of designing saddles and sometimes it includes the stiffening rings that I have mentioned. This is treated quite well in Chapter 11 of the classic book, "Process Equipment Design" by Brownell and Young, John Wiley and Sons, ISBN 0 471 11319 0.

Regards, John.