Flange Bending Moments 3

[butelja]

What is the proper way to convert a bending moment on a pipe into an "equivalent pressure" acting upon a flanged joint? I've heard that ASME Sec. III has equations for this, but since I don't work in the nuclear field, I don't have a copy of this.

I know that it can be shown that the maximum force/length due to an external moment is:

F`_max = 4*M / (pi*D^2)

But, should D be the gasket mean diameter, bolt circle diameter, or something else? Thanks in advance.

 

[KernOily]

I think there is something in 31.3 about this also, in one of the appendices. I'll check it out. Thanks!
Pete
P. J. (Pete) Chandler, PE
Principal Engineer
Mechanical, Piping, Thermal, Hydraulics
Processes Unlimited International, Inc.
Bakersfield, California USA
pjchandl@prou.com

 

[kstaylor]

The formula from ASME III for converting external forces & moments on a flanged joint to equivalent pressure is as follows:

Peq = (16*M)/(pi*G^3) + (4*F)/(pi*G^2)

Where:

Peq = equivalent pressure, psig
M = bending moment applied to joint due to weight & thermal expansion, in-lbf
F = axial force applied to joint due to weight & thermal expansion, lbf
G = diameter at location of effective gasket load reaction, in.

In the above equation, * represents multiplication, ^ represents exponentiation (i.e. raised to the power of), and pi is the ratio of circumference to diameter of a circle (3.14159...).

This is the same formula given in the "Kellogg book". Note that the value of Peq would then be added to the stated design pressure to determine the flange design pressure, PFD.

 

[butelja]

KSTAYLOR,

Thanks, that is EXACTLY what I was looking for.

 

[kstaylor]

Glad to help...

 

[KernOily]

Speaking of the Kellogg book, do any of you know where I can get a copy? When I went to Glynn Woods' 31.3 class he told us it is out-of-print.

Thanks!!! Thanks!
Pete
P. J. (Pete) Chandler, PE
Principal Engineer
Mechanical, Piping, Thermal, Hydraulics
Processes Unlimited International, Inc.
Bakersfield, California USA
pjchandl@prou.com

 

[TD2K]

What's the full title and author for this book Pete?

 

Hi Guys,

Regarding the Kellogg book - yes it has been out of print for some time.

Title - Design of Piping Systems
Copyright The M.W. Kellogg Company, 1941, 1956
Revised Second Edition (the last one) 1977
(by that time the Company was - Pullman Power Products)
Most of the original editing was done by Mr. David B. Rossheim.

Publisher John Wiley & Sons
A Wiley-Interscience Publication
ISBN 0-471-46795-2
Library of Congress Catalog Card Number 56-5573

This book was very important because it documented methodologies at a time when "piping stress analysis" was a "black art". The book is about 50 percent obsolete but the other 50 percent still applies (Pi is still Pi)

Another very good early documentation was the article written by John Brock for the Fifth Edition of the Piping Handbook (Sabin Crocker and Reno C. King) if you ever see this edition in a used book sale grab it.

Best regards to all, John.

 

[sprintcar]

You can occasionally find books like this at

http://www.alibris.com

(probably sell your old stuff there too!)

We found a few good deals in the past

Enjoy

 

[kstaylor]

I have occasionally seen the Kellogg book in used book stores. I found two copies in a used technical book store in Long Beach, CA about 12 or 13 years ago and bought one for $35 (should have bought both at the time, went back a little later to get the other but it was already gone).

I second what John Breen said about its continued usefulness. About half of the book covers material on Kellogg's Simplified Method and General Analytical Method of manual piping flexibility analysis, now badly outdated by computer codes. The other half is a wealth of information about material strength and theory of failure; design assumptions, stress evaluation and design limits; local components; exansion joints; pipe supports; vibration; and all kinds of charts and tables useful in checking computer results by simple manual methods -- all really valuable stuff, and still relevant in the computer age.

If you ever see a copy available, buy it pretty much regardless of price.

 

[KernOily]

Although the stuff in there is outdated since most analyses today are done by computer, there is still the time when one needs to do a manual analysis, for one reason or another. For example, in the layout of l-o-n-g steam lines for heavy oil steam injection, we usually do a quick layout and then use one of the hand stress-check methods to space the anchors, guides, and loops. This does several things: (1) gets the designers trained to do some easy stress work (2) frees up your CAESAR license to do more complex work (3) it is usually faster.

Glynn Woods has a brand-new book out on flexibility and the B31 codes. I got a flier on it last week but I went to his class so I already have most of his material. Woods is up there with Markl and Rodabaugh et al. and is rightly numbered amongst The Ascended Masters of Pipe Stress. He has a TON of info on SIFs that is totally interesting, at least to this nerd. For example, did you know that most of the SIFs calculated by the B31 methods are totally wrong when compared to empirical SIFs? That was a revelation to me. There is a bunch of current research ongoing to calcuate better SIFs.

There are a couple of other good books to get: "Piping Engineering", published by Tube Turns; my copy is third printing, 1975, and "Piping and Pipe Support Systems", Smith and Van Laan, McGraw-Hill, both of which are also most unfortunately out of print. I tell ya, it's jest a travesty and a crime to society, ain't it, when these great references go out-of-print... I have a recommended list of code/stress/flexibility references that I can fax you if you send me your fax number. Thanks!
Pete
P. J. (Pete) Chandler, PE
Principal Engineer
Mechanical, Piping, Thermal, Hydraulics
Processes Unlimited International, Inc.
Bakersfield, California USA
pjchandl@prou.com

 

Hello All,

Forgive me guys, I have this "sticking in my craw" and just have to let it fly:

The best historical papers on pipe stress analysis are by A.R.C. Markl and these were published in the 1950's by the American Society of Mechanical Engineers (ASME). The papers were gathered into a large book that was published by ASME. I understand that this old (out of publication) book has been reissued in paperback recently by ASME. The Markl papers were also included in the famous Tube Turns Book, "Piping Engineering" mentioned by Pete above. Another writer of important technical papers regarding pipe stress analysis is Everet C. Rodabaugh, he has added much original thought to the subject by doing research and writing the results in papers and other publications for the Welding Research Council. Take a look at the WRC web site and see all the piping papers that are published as WRC bulletins. Glynn Woods and Everet are now working together with the support of the B31 Committees to develop better SIF data for inclusion into future Codes.

While on the subject of WRC Bulletins, I would like to recommend to the group that you look into getting a copy of a recently issued WRC document:

Title: Guidelines for the Design and Installation of Pump Piping Systems.
V. A. Carucci and J. A. Payne
Welding Research Council Bulletin 449

The title is very descriptive of the contents - this is 48 pages of common sense guidelines to designing piping so that the pumps will survive. This is a must read and a great reference for your technical library.

On the subject of books, I believe that the best available contemporary piping design book is "Practical Guide to ASME B31.3 Process Piping", by G.E. Woods and R.B. Baguley (published by CASTI Publishing Company, 14820 29th Street Edmonton, Alberta, Canada, T5Y 2B1, Canada, Telephone (403) 478-1208, Fax (403) 473-3359 ). Another useful book (albeit, now out of print) is "Introduction to Pipe Stress Analysis " (2 volumes) by Sam Kannappan, 1992, (ISBN 0 89464 706 7), Krieger Publishing Company, Malabar, Florida, 32950. Mr. Kannappan is also a B31.3 Code Committee member. The Kannappan book shows pipe stress analysis approaches as they are done today with the contemporary computer programs. Another book, Piping and Pipe Support Systems, by P.R. Smith and T.J. Van Laan was published by McGraw Hill but I do not recommend it as I find it to be shallow, only about 1/3 of it is of value and these issues are covered better in other books. A good ("must have&quot but rather pricey reference book is the Piping Handbook, 7th edition, edited by M. L. Nayyar and published by McGraw Hill ,1992, IBSN 0 07 046881 8, (this book is an update of a classic reference which was originally written by R.C. King and S. Crocker; if you could find an old 5th edition of this handbook (on the internet as a used book) you would find excellent writing on pipe stress analysis by John Brock, including some wonderful historical information. One other standard reference (it does not address stress analysis but is a good reference for design) that is found in nearly all piping design firms is "The Piping Guide" by D.R. Sherwood and D.J. Whistance - ISBN 0-914-08219-1 (Syentek, PO Box 26588, San Francisco, California 94126). The "Piping Guide" is commonly used is community colleges for teaching layout and good practice. Piping Design for Process Plants, by H. F. Rase (Library of Congress Card Number 63-17483) is a good practical guide but not very deep in theory or technical background. It is still available through several book sellers (e.g., AMAZON.COM). The Seventh Edition of the Grinnell Book, Piping Design and Engineering, has recently been released and can be had by contacting your Grinnell representative ( you will pay either $100.00 or nothing for the book depending upon your relationship with your Grinnell Rep.). This book is useful but by no means as technical as the "Kellogg Book" Also, the methods shown for hanger sizing are not as accurate as a computer model. The "Kellogg Book" is, of course, out of print. The last issue was the "revised Second Edition" which was published after 1977 (ISBN 0 471 46795 2, Library of Congress Card Number 56-5573). About 50 percent of the book is obsolete and of only historical interest to the practical piping Engineer. The other 50 percent of the book is useful even today and provides good background information. I have seen the book listed on at least one used technical books site but I cannot remember where I saw it. It should be out there as a used technical book because a lot of them were sold. "Process Plant Layout and Piping Design", by Ed Bausbacher and Roger Hunt (IBSN 0 7913 0543 5 ), Auerbach Publishers, 1990, might be useful to you but, because it is pricey, I recommend finding a copy in a library and looking it over before you buy. I have been told that this book is now being published by a new publishing house but I lost track of who it is.

You may want to try Brown Book Store in Houston Texas for some of these books(713-652-3937,

http://www.brownbookshop.com,

orders@brownbookshop.com). Brown Book Store sometimes has "out of print" books. Other sites to visit who sometimes have book lists:

http://www.pipingdesign.com,

and

http://www.pipingnews.com.

There are special books covering special topics such as underground piping, plastic piping, large diameter piping and pentstocks, etc. Every year the Pressure Vessel and Piping (PVP)Division of ASME has a conference which produces about 200 good papers (about 40 percent piping subjects) and these papers are gathered into publications which are offered by ASME. You could write to ASME and get on their mailing list for publications.

Oh well, now I feel better – thanks for your patience.

Regards, John.

 

[mgp]

For anyone interested I have a copy of an interesting article on the subject. It is from an ASME conference in 1981 and co-written by L-C Peng from M.W. Kellogg.

The background for the "equivalent pressure method" (i.e. the Kellogg formula) is discussed, so it may help in the absence of the Kellogg Book.


The conclusion is as follows:

quote:
"The equivalent pressure approach have become a standard method in evaluating pipe load acting on flange connections. There are two acceptable ways of checking the pipe load using the equivalent pressure. They are rating table method and stresss calculation method. The rating table method is simpler but is much more conservative by ignoring the reserve strength. It is so conservative that it would probably disprove most of the installations which are operating satisfactorily. The stress calculation method is a more realistic approach which evaluate the actual reserve strength available in a flange.
When a standard flange is rated for a certain pressure, it normally posses sufficient strength to resist the rated pressure load plus sustantial reserve strength to resist the pipe load. This reserve strength varies from flange to flange and is not known until a stress analysis is performed. It is also highly dependent on the thickness of the connecting pipe when a bore of the flange is specified to be the same as the inside diameter of the pipe.
In designing a special flange, it is necessary to provide some allowance for the pipe load. The allowance can be either based on the actual expected pipe load or based on the load that will produce a bending stress equivalent to one half of the basic allowable stress at the connecting pipe. In any case care should be exercised in the final piping system design to keep the pipe load from exceeding the allowance provided."
End quote.

This is interesting because in the stress analysis you may get an equivalent pressure (via the Kellogg equation), but you need a max allowable pressure to compare with because stress programmes normally don't produce a proper flange stress calculation. By using the flange rating many flanges will fail the analysis. Consequently one can save many redesigns by doing an ASMEVIII calculation of the actual flange and connecting pipe.

For those interested send me an e-mail at mgp@kabelnettet.dk and I can mail a scanned copy (approx 2MB) of the article.

Regards
Mogens

 

[pstress]

I have a few points about the flange moment calcuations:


1. They seem potentially time consuming. I wonder if most piping stress engineers actually do this calculation for each line.

2. The methods seem generally conservative. They could mean expensive and unnecessary re-design of piping where stresses are otherwise fine.

3. They dont seem to address flange leakage.

...any comments on the above?

 

[mgp]

In my opinion, pstress is right in all 3 cases. The only problem is that there is not so many ways around it.

If the flanges should be properly calculated for leakage, then the ASME code is not the one to use. The new prEN1591 (preliminary european code) adresses this problem, but it is also time consuming. I found a paper from an ASME seminar

http://www.amtec-services.de/demos/GASKET_FACTORS.pdf

This includes some considerations about the various codes and leakage.

I think the normal approach is that if the ASME Code stress calculation is ok, then the leakage is also ok, when the bolts are tightened to withstand the test pressure.

How this is used in the stress calculations is another matter.

I have seen stress engineers ignore the flange check (depending on the criticality of the system), thus avoiding any non conformance in the output.

I have also seen many use an allowable pressure of 1.5 times the rated pressure for flange check in the stress analysis. I assume that the intention is to check if the equivelent pressure exceeds the test pressure, but where it comes from I was never able to find out. Perhaps someone can explain this?

I cannot see that this approach is correct. In my opinion you should use either the rated pressure or calculate the flange (ref quote in my previous post). This will work in most cases if only the system design pressure is set at a lower limit than the piping class pressure (= flange rating). The problem is that in many cases process engineers tend to pick the easy approach for selecting design pressure, rather than checking out the actual system pressure, also because this may make upgrades easier later on.

Another option which might save many redesigns would be to include a flange calculation for each size when the piping specs are established. This could produce a reference table for checking equivelant pressures. It will be time consuming, but only once.

For a spreadsheet for normal ASME calculations, check Thread378-19042
in the Piping & fluid mechanics engineering Forum

Regards
Mogens

 

[mgp]

I just fell over "CASTI guidebook to ASME B31.3" in our office. (It is also in John Breens list above)

It mentions an (interesting) alternative approach for leakage check which I had not seen before, presented by E.C.Rodabaugh in the
"ASME B31 Mechanical Design Document, appendix IV."

I have also come across a spreadsheet using the "Blick" method for calculating allowable bending moments - also relatively simple.

My questions:

Does anybody know more about the "ASME B31 Mechanical Design Document"?

Does anybody have experience with these methods and reason why they are not more widely adopted, apparently being less conservative, and (or) easier to calculate.


By the way a J.P Breen is mentioned in the acknowledgments section of the CASTI book. Might that be you John?


Regards
Mogens

 

Hello all,

The "Mechanical Design Committee Document" is the olde name for the "Model Document for B31 Chapter II" (Capter II of B31.1 and B31.3 is the "Design" Chapter). It is an "in house" working document that is maintained by the MDC in which the Committee develops and publishes (for review and comment within the Code) proposed new paragraphs for the B31 "Book Sections". The document is not available to the public.

Yes Mogens, I helped just a little with the CASTI book - very little. The book is all the work of Glynn and Roy and all the "good stuff" is their doing.

Regarding flanges, perhaps y'all know that the users of the B31 Codes are invited to ask questions in the form of official "inquiries". The appropriate Committee will discus the questions and they will usually issue official "interpretations" of the intent of the existing Code. An inquity resulted in the following "recent" Interpretation by the B31.1 Code Committee:

Subject: B31.1, Paras. 102.2.1 and 104.5.1(A), External Forces and Moments Acting on a Flange. Date Issues July 21, 1999, File: B31-97-042

Question (1): When selecting a flange on the basis of the pressure-temperature rating given in ASME B16.5, in accordance with B31.1, paras. 102.2.1 and 104.5.1(A), is it required to consider any external forces and moments acting on the flange?

Reply (1): Yes. Paragraphs 102.2.1 and 104.5.1, relate to pressure design only. In addition to pressure design, piping systems must be designed for external forces and moments (see para. 101.1)

Question (2): If a flange is designed in accordance with ASME Boiler and Pressure Vessel Code, Section VIII, Division 1, as allowed by B31.1, para. 104.5.1(A), is it required that the design pressure be increased to account for the external forces and moments acting on the flange?

Reply (2): No. B31.1 does not specify a method to be used to account for external forces and moments. The equivalent pressure method (see nonmandatory Appendix II, para. 4.2.3) is just one acceptable method used to account for external forces and moments.

Question (3): Is the method of analyzing a flange, in accordance with B31.1, Appendix II, para. 4.2.3, to prevent leakage of the flange,or is it to prevent failure of the flange itself?

Reply (3): The equivalent pressure method is intended to prevent overloading of the flange due to the combination of internal pressure and extermal moments. The failure mode due to overloading of the flange is not specified by the Code.

Question (4): When analyzing a flange, in accordance with B31.1, Appendix II, Para. 4.3.2, can the Committee provide a method of converting a moment about a centerline of the flange into an equivalent pressure?

Reply (4): The Committee does not recommend, endorse, or approve methods of analysis, except where stated basic design principles and formulas are supplemented as deemed necessary with specific requirements to assure uniform application of principles and to guide selection and application of piping elements.

Regards, John.

 

[mgp]

Thanks for the reply John

I clearly see that the "official" standpoint of the Code Committee is not to take any preference towards a specific method.

Still would it not be reasonable to assume that a method -
-from an un-published document within the Comittee,
-referenced in a published book,
-written by a Committee member (Mr. Glynn Woods, the author of the chapter, is a member of the Comittee)

is an appropriate method which will be accepted throughout the industry, and relevant authorities?

After all the "Kellog" method is also generally accepted but not officially approved by the Committee

What I mean is, engineers should be able to use this method feeling confident that by doing this, they made a safe and sound design (or should they not?)

Regards
Mogens

 

[vendora1]

Hi butelja,
I am going to design some vent silencer for steam, air and co2 gas.
So I try to download "A Reduced-Noise Gas Flow Design Guide" because I know that it is very helpfull reference, but I couldn't to do it.
I will be very appreciated if you could help me.
Best Regards,

 

A long thread, to which I will add.

The equivalent pressure is generally considered to be a rather conservative approach to evaluating the effect of external loads on flange performance.

Three reasons for this are:
1) In doing a Section VIII, Div 1 Appendix 2 type of calculation, increasing the equivalent pressure also has the effect of increasing the required bolt load to keep the gaskets tight (m times P) {this in actually not necssary} in addition to withstanding the external load.
2) The bolted joint is designed, again per Appendix 2, using allowable bolt stresses that are much lower than the normally desired bolt up stress (typically 50 ksi, see Div 1 Appendix S for a discussion, and ASME PCC-1 for a description of desired flange bolting practices). The margin between allowable stress (used for flange design) and bolt up stress provides a margin for accomodating external loads.
3) The treatment of moment as an equivalent pressure is rather conservative. The actual effect depends on the torsional ridigity of the flange, as demostrated by Dr. Koves (see "Analysis of Flange Joints Under External Loads," First International Symposium on Processs Industry Piping, Dec 14-17, 1993, MTI).

I believe Dr. Koves's treatment of external moment is included in the current draft of new flange rules under consideration for Section VIII, Div 1, and is also included in his flange analysis program K-Flange. Note also that the draft new rules for flange design also address leakage rates.

The equivalent pressure can be used, but it will be found to fail many succesfully working installations.