Another way to Calculate SIF.... 2

[cantiqa78]

Hello Everyone,

I have this Dummyleg on a bend in my piping system that had failed under Sustained load when i'm doing the analysis with CAESAR II, so i check the SIF and yes it gives me a big number of SIF. The Code is ASME B31.3, and there is no SIF calculation for this case in ASME B31.3 APP.D, so i have to do some FEA, but the problem the company don't have this program and i do not have much time to run an analysis with FEA program right now. So is there another way to calculate the right SIF?
And another question that i have does ASME B31.3 Use SIF for calculation of Sustained and Occasional Load like ASME B31.1 ?

 

[DSB123]

cantiqa78,
You say the SIF number in the Sustained case from Caesar is big. How big? Normally the user has to input the SIF's for attachments such as dummy legs. You could take the loads on the trunnion (i.e. dummy leg) from the Caesar analysis and perform a manual calculation in accordance with a "local" load assessment (for example using the Kellog method) to assess the pipe stresses.

You also ask whether the SIF is used in the calculation of the Sustained plus Occasional stress in a ASME B31.3 analysis but you are performing an ASME B31.3 analysis so you should be aware of what the requirements are shouldn't you ? If you are performing a stress analysis to a particular Code then you really should be aware of what the actual Code requirements are and not "hope" that the software program being used takes everything into account for you.

 

[cantiqa78]

Thanks DSB123,

Yes i want to do a manual calculation with Kellog Method but i don't have the reference, i only have "Design of piping System" By MW KEllogg and the "Design Manual of Pipe attachment subject 3810" by Kellogg, if this one is what you meant, i don't found about SIF Calculation there, can you refer to me which page is this on Pipe attachment(Trunion) Design Manual of Kellogg, or maybe one that i got is not complete in the page.
My another question about SIF calc.at ASME B31.3, yes i am aware about this, but i only one to know the reason behind that, all i know that SIFs are use for cyclic load under bending moment, and Sustained and Occasional are not include in it so why ASME B31.1 put this 0.75i at Sustained and Occasional Calculation and what about at ASME B31.3?
Once again thanks...........

 

[JohnBreen]

Do you understand WHAT a Stress Intensification Factor is? Do you know the differences between SIF's, Stress Indices, Stress Intensifiers, etc.?

The ASME B31 Codes for Pressure Piping have very exact methods for determining SIF's and if you are going to use an SIF with a B31 Code analysis you cannot justify any other methodologies. You can use FEA to determine the SIF's or you can do fatigue testing of the component. You will find that FEA is less time consuming.

In Code Case 178, B31.3 provides an equation for calculation longitudinal stress for SUSTAINED loads. The equation combines stresses (Tresca method) and it includes SIF's in calculating the bending (moment) stresses.


By the way, B31.3 will soon have a major modification of SIF's and Appendix "D". Stay tuned.

Regards, John.

 

[DSB123]

Hi JohnBreen,
Any idea when the revised Appendix D will be issued? Is there an intention to ratify the SIF's across the various ASME Codes (Seems the way forward)?

 

[nickelkid]

visit the library @

http://www.paulin.com/

enjoy

 

[JohnBreen]

DSB123 asks:

Any idea when the revised Appendix D will be issued? Is there an intention to ratify the SIF's across the various ASME Codes (Seems the way forward)?

Last week at the B31.3 meetings in Puerto Rico there was a discussion within the Sub-group on Design (SG-B) of an agenda item for reviewing SIF work done by Ed Wais and Ev Rodabaugh that will result in changes to Appendices "D". The same issue is on the agenda for the B31 main Mechanical Design Technical Committee (MDC) for next week in Vancouver. B31.3 will wait for some input from MDC. I would guess it would take three more meetings before any B31.3 Appendix "D" proposal is ready for balloting (18 months). Also, at this time ASME ST LLC (

http://stllc.asme.org/)

has a request for proposals out for addressing this issue and work resulting from that effort will also impact any balloting date.

There is a general project underway to unify similar design rules across the ASME B31 Code for Pressure Piping. After Markl's work (it is important to understand how limited that work was - carbon steel, NPS 4, schedule 40, etc.), there has been a significant amount of "Markl-like" testing done and there is good data available. Also, B31J (Standard Method to Determine Stress Intensification and Flexibility Factors for Piping Components) will be published in 2008 and it applies equally to all B31 Code Sections. Also, there is now a significant amount of high cycle data available that will lead in changes to the basic "Markl" S-N curves. All these issues will apply equally across the range of ASME B31 Section Code books and so there will be an effort to bring a level of commonality to these Codes as these SIF rules change.

Regards, John.

 

[DSB123]

JohnBreen,
Thanks for the information and clarity in your response. However it seems to take an "age" to consider/reconsider and the publish any changes within the ASME Codes.

 

[LSThill]

JohnBreen (Mechanical)

John, is this a part of PVRC & Paulin Research work?

Regards

Leonard

 

[LSThill]

John Breen,and Team Members:

visit the library

Fatigue
Markl, SIFs, and ASME VIII-2 Fatigue Design (January 2008)
This brief article addresses the question of "What are SIFs, where did they come from, and how do they relate to ASME Section VIII-2 fatigue designs?" Read more...

Markl, SIFs, and ASME VIII-2 Fatigue Design
Did you know that PRG has the only automated FEA Stress Intensification Factor (SIF) calculator in the world?

PRG software has been automatically calculating SIFs for varieties of piping components for more than fifteen years.

This brief article addresses the question of "What are SIFs, where did they come from, and how do they relate to ASME Section VIII-2 fatigue designs?"

Since B31.3 2007 (and earlier) references finite element methods and ASME Section VIII-2 fatigue methods in Appendices 4 and 5, this is information that should be understood by every senior piping designer.

Markl, FEA, and Div 2. Appendix 5
One benefit of PRG software is the automated calculation of piping stress intensification factors (SIFs). SIFs are automatically calculated in FE-SIF, Nozzle/PRO, and FE/Pipe. In fact, there is no other FEA software in the world that provides automatic SIF calculations.

Also:

What if the SIF is Not Accurate? (January 2008)
Article discusses effects of using inaccurate stress intensification factors (SIFs) in pipe stress analysis. Read more...

What is a Stress Intensification Factor (SIF)? (January 2008)
Article explains the use of Stress Intensification Factors (SIFs) in pipe stress analysis and how they are derived using experimental and finite element analysis methods. Read more...

Use SIFs and Flexibilities in CAESAR II® (January 2008)
Methods are presented for using accurate stress intensification factors (SIFs) in the CAESAR II pipe stress analysis program. Read more...

FE-SIF is specially designed for these calculations and intended to be an “everyday product” for piping engineers to utilize in concert with their usual piping analysis software.

This topic will provide a brief introduction to how SIFs are calculated, how they relate to ASME Section VIII Division 2 fatigue design, and where did SIFs come from.

A Stress Intensification Factor (SIF) is defined as the ratio between the peak stress and average stress in a given component:

SIF = Actual Peak Stress / Nominal Stress in Part
A. R. C. Markl and his team (1950’s) developed the original SIFs still used in ASME piping Codes today.

In his study, Markl determined that girth butt-welds typically resulted in stresses approximately 1.7 to 2.0 times the stress in non-welded piping. As a result, all of the piping codes have been “base lined” to include the factor of 2.0 for girth welds:

B31.3 SIF = Actual (Peak Stress) due to Moment M
Stress in Girth Butt Weld due to Moment M

OR
B31.3 SIF = Actual (Peak Stress) due to Moment M
2 * (Moment M) / (Section Modulus Z)

In terms of ASME Section 8, Division 2, Appendix 5 and finite element analysis (FEA) work, we could use the following equation interchangeably with the previous equations:

SIF = Range of Peak Stress due to M
2 * (Moment M) / (Section Modulus Z)
= 2 * (Pl+Pb+Q+F)
2 * (M / Z)

OR
SIF = Alternating Peak Stress due to M
(Moment M) / (Section Modulus Z)
= (Pl+Pb+Q+F)
(M / Z)

The peak alternating stress (PL+PB+Q+F) is usually determined from finite element analysis. Normally, the peak stress is the product of the secondary stress and a fatigue strength reduction factor (FSRF). For instance:

PL+PB+Q+F = (PL+Pb+Q)*FSRF / 2.0
FSRFs are determined from testing or taken from references such as WRC 432.

As discussed in NUREG/CR-3243, the mean curve fitted to Markl’s fatigue test data gives a relationship between the stress range in a butt weld pipe and the number of cycles to cause a thru-wall fatigue failure:

i * M / Z = Sf = 490000 * (N)-0.20 (Equation 1)
where

i = stress intensification factor
M = bending moment
Z = section modulus
N = expected number of cycles
Sf = allowable cycling stress


The mean curve described by Equation 1 is shown in the figure below.



Equation 1 has been normalized based on the peak stress range in a girth butt-weld (i.e. i = 1.0). As a result, there is an inherent factor on the peak stress range “S” of 0.50. All peak stresses given by Markl SIFs are half of their actual values due to Markl’s use of girth butt-welds as a baseline.

The factor of two makes the alternating peak stresses used in Division 2 Appendix 4 & 5 very easy to implement in terms of the Markl failure criteria (Equation 1). One could conclude that by using a factor of 0.50 on peak stress, Markl has essentially reduced the stress range to an alternating stress component. Using this conclusion, we can use Division 2 Appendix 4 & 5 peak stresses with the following equation:

Pl+Pb+Q+F = 490000 * (N)-0.20 (Equation 2)


N = (Pl+Pb+Q+F / 490000)-0.20 (Equation 3)
Equation 2 gives the ASME Section 8, Division 2 alternating peak stress (Pl+Pb+Q+F) that would cause a through-wall fatigue failure with a 50% probability of failure.

Equation 3 gives the number of cycles to failure for a given ASME Section 8, Division 2 alternating peak stress (Pl+Pb+Q+F).

FE-SIF automates these calculations and provides accurate SIFs for all piping components, regardless of their geometry or design.


L S THILL

 

[4Pipes]

Cantiqa,

Are you saying that you have done a Budlong calc as implemented in Caesar and applied these sifs to the stress calc?

What scf have you used for the trunnion from elbow calc? What happens if you reduce it by a small amount.

Are you using a sif reduction for the sustained case?

Folk look down their noses at the Budlong calc and use Kellogg instead. Budlong does produce high sifs especially on thin wall. Is this such a surprise given the structurally poor shape of a trunnion from an elow? Budlong is as least based on a coherent argument. Kellogg, simply compares a trunnion from an elbow to that on straight pipe. Kellogg gets away with it because for most, but not all geometries, it is conservative. Budlong will pass some geometries and loadings that are failed by Kellogg.

 

[JohnBreen]

Hello all,

I would point out that David Diehl wrote an interesting article regarding bend/elbow (with and without trunnions) in one of the olde COADE Caesar II newsletters (October, 2002) that you might want to download and keep for future reference.

Go here for the pdf file:

http://www.coade.com/newsletters/oct02.pdf

Regards, John.

 

[JohnBreen]

...........and if you can read Spanish.......

http://www.pucp.edu.pe/congreso/cibim8/pdf/20/20-07.pdf

John