Topic Options
#35728 - 05/27/10 07:09 PM Sustained stress index (0.75 SIF or 1.0SIF)
Nald Offline
Member

Registered: 02/20/09
Posts: 119
Loc: Malaysia
Dear engineers,
My concern is about Sustained stress index of 0.75*SIF & 1.0*SIF. For the articles, books and even codes I've read, I did not found any concrete basis when to use these two options. I know that these two options was stated as per (#1-34 (2/23/81)) and (#6-03 (12/14/87)) but this seems confusing.Based on mechanical engineering news-jan 01. It was stated "So for the time being I would recommend that this factor of 0.75SIF>= 1.0=SSI be used in analysis" and "0.75SIF as a multiplier is probably not 100% right but it is more approriate than 1.0".

My question is, What specific condition to use 0.75 or 1.0 as Sustained SIF Multiplier?

Please share your Idea..!

Any responce subject to this matter is highly appreciated!


Regards,
Nald



Edited by Nald (05/27/10 07:57 PM)
_________________________
Regards,
Nald

Top
#35731 - 05/27/10 10:11 PM Re: Sustained stress index (0.75 SIF or 1.0SIF) [Re: Nald]
Borzki Offline
Member

Registered: 09/16/04
Posts: 759
Loc: Traz
You can search for B31.3 Code Interpretation 6-03 dated Dec. 14, 1987, which permits the users to ignore SIF (SIF=1, in that case) for sustained and occasional loads but there's a cautionary note that states:"However, see ANSI/ASME B31.3, para 300(c)(5)".In my own opinion, correct me if I'm wrong fellow stressers, I would go to the conservative side which is not to ignore SIF and add some restraints on my system as needed.I think the 0.75SIF is to avoid collapse..Or if your system is fine with the actual SIF used then there's no need to worry about that.





Top
#35750 - 05/28/10 06:17 AM Re: Sustained stress index (0.75 SIF or 1.0SIF) [Re: Borzki]
Edward Klein Offline
Member

Registered: 10/24/00
Posts: 334
Loc: Houston, Texas, USA
Part of the issue is the nature of how the SIF was developed. It is based on experimental data explicitly on cyclic loading and was not directly intended for purely sustained loads. The reluctance to discount the SIF altogether is due in part to recognizing that, while we consider pressure and weight as "sustained" loads, we know that systems in operation do vary - pressures do actually cycle.

So, the confidence to ignore the SIF altogether depends upon the confidence that the system will run in an essentially steady state. As stress engineers, we typically get a minimal window in how a system ends up operating. You are lucky if you get to talk with a process engineer or an operator of a unit to get some insight to how the process actually runs.

Where I am at, we start with the full SIF as the default basis. Usually, that works as there are typically other factors (particularly equipment load limits) that tend to limit system stresses well below allowables. In cases where the sustained stress does run over the limit, we will routinely check with the 0.75 B31.1 multiplier to see if the system will pass. If that doesn't work, we will then makes changes to the support arrangement.

There are cases where we have chosen to remove the SIF multiplier. But, those are carefully examined and considered for special cases where a very limit number of cycles are expected and changing the support arrangement poses significant technical or economic challenges.
_________________________
Edward L. Klein
Pipe Stress Engineer

All the world is a Spring

Top
#35757 - 05/28/10 08:01 AM Re: Sustained stress index (0.75 SIF or 1.0SIF) [Re: Edward Klein]
John Breen Offline
Member

Registered: 03/09/00
Posts: 482
Loc: Pittsburgh, PA (& Texas)
Nald,

Stress Indices are NOT used in B31 Codes, SIF's are used in B31 Codes - these are not the same as Stress Indices.

The B31 Code for Pressure Piping, Stress Intensification Factors were developed based upon cyclic testing of NPS 4 piping components. The testing showed that local bending stresses are developed in these components that are greater than the bending stresses that are developed in lengths of straight piping under the same loading.

Regardless of the loading (primary or secondary) the local bending stresses in the (other than straight) components are greater. As a precaution to address potential collapse, the Committee decided to include a portion (or all) of the SIF (0.75 times the SIF from Appendix "D", but with a resulting SIF that is never less than 1.0 as it was NOT intended to lower the SIF) in the equation for calculation of stresses due to sustained pressure and weight and in the equation for calculation of stresses due to sustained loads and occasional loads.

In the case of sustained loads or occasional loads ALWAYS multiply the SIF (as calculated by Appendix "D") by 0.75. If the product of this multiplication is greater than 1.0 use the product as the SIF to intensify the bending moment. If the product of this multiplication is LESS than 1.0, use 1.0 times the SIF (the entire SIF as calculated by equations of Appendix "D") to intensify the bending moment.

Additional reading:
http://mydocs.epri.com/docs/public/000000000001012078.pdf
(Interpretation 1-27 on page 17 of:
http://cstools.asme.org/csconnect/pdf/CommitteeFiles/23712.PDF

Regards, John


Edited by John Breen (05/28/10 08:13 AM)
_________________________
John Breen

Top
#35759 - 05/28/10 08:30 AM Re: Sustained stress index (0.75 SIF or 1.0SIF) [Re: Nald]
Dave Diehl Offline
Member

Registered: 12/14/99
Posts: 2382
Loc: Houston, TX, USA
B31.3 now has Code Case 178 to provide an equation for longitudinal stress due to sustained loads. It has a moment multiplier called the stress index. This index is the greater of 0.75i and 1.0.
We should see this Code Case moved to the base Code in the 2010 Edition.
That doesn't tell you the basis of the term (0.75i) but at least it is "standardized".
I have "heard" that an index of 0.75i on the sustained moment will have the calculated stress, when compared to Sh, approach the collapse limit of the component.
_________________________
Dave Diehl

Top
#35779 - 05/30/10 05:21 PM Re: Sustained stress index (0.75 SIF or 1.0SIF) [Re: Dave Diehl]
Nald Offline
Member

Registered: 02/20/09
Posts: 119
Loc: Malaysia
Borzki, Edward Klein, John Breen & Dave Diehl;
Thank you so much for your suggestions, explanations and references that you provided me.
It helps me understand regarding this matter.


Regards,
Nald


Edited by Nald (05/30/10 05:28 PM)
_________________________
Regards,
Nald

Top
#35783 - 05/31/10 02:36 AM Re: Sustained stress index (0.75 SIF or 1.0SIF) [Re: Nald]
ammangz Offline
Member

Registered: 01/17/10
Posts: 5
Loc: Philippines
Mr. John Breen,

Quoting your previous reply to Mr. Nald:

"In the case of sustained loads or occasional loads ALWAYS multiply the SIF (as calculated by Appendix "D") by 0.75. If the product of this multiplication is greater than 1.0 use the product as the SIF to intensify the bending moment. If the product of this multiplication is LESS than 1.0, use 1.0 times the SIF (the entire SIF as calculated by equations of Appendix "D") to intensify the bending moment."

I just would like to ask if this could be done by setting the B31.1 SIF Multiplier to 0.75 in the Configuration Setup Dialog? If in case some nodes have values less than 1.0, then their SIF values will be overridden by 1.0 times the SIF value. Please comment if what i have in mind is a correct way to do this.

More power!

_________________________
Best regards,

ammangz

Top
#35828 - 06/01/10 08:08 AM Re: Sustained stress index (0.75 SIF or 1.0SIF) [Re: ammangz]
Dave Diehl Offline
Member

Registered: 12/14/99
Posts: 2382
Loc: Houston, TX, USA
Ammangz,
Two issue combine here - Code requirements and CAESAR II implementation.
The (implied) Code requirement for calculated stress due to sustained load is that the moments be increased by (0.75)SIF with the added requirement that this product (0.75)SIF not be less than 1.0. So, if, say, SIF=1.1, this product will give a multiplier = 1.0 (rather than 0.825). That's the Code.
Like you say, the CAESAR II Configuration setting for SIF Multiplier will allow you to change that 0.75 to whatever you want. But CAESAR II will not let the product of (multiplier)SIF be less than 1.0.
If the product (multiplier)SIF is less than 1.0, we do not update (multiplier)SIF with (1.0)SIF as you state above. We replace the product with 1.0.
_________________________
Dave Diehl

Top
#42258 - 04/07/11 02:23 AM Re: Sustained stress index (0.75 SIF or 1.0SIF) [Re: Dave Diehl]
bluna Offline
Member

Registered: 02/05/08
Posts: 4
Loc: Spain
Good Mornig,

Is correct, according to B31.3 code, use a factor 0.0001 as sustained SIF multiplier?
There is any exception or special condition to use that factor in a calculation?

Thanks a lot for your help.

Regards,
bluna

Top
#42276 - 04/07/11 07:11 AM Re: Sustained stress index (0.75 SIF or 1.0SIF) [Re: bluna]
mariog Offline
Member

Registered: 09/29/07
Posts: 798
Loc: Romania
I quote a sharp answer from
http://65.57.255.42/ubbthreads/ubbthreads.php?ubb=showflat&Number=17866&page=2

Don't confuse CAESAR II switches with the Code. The Code says (or implies) stress=(index)*Moment. That index can be "i" or "0.75*i" but never less than 1.0.

CAESAR II has this: index=(multiplier)*i with the stipulation that the index minimum is 1.0.
So, in CAESAR II if:
multiplier=1, index=i but not less than 1.0
multiplier=0.75, index=0.75*i but not less than 1.0
multiplier=0.0001, index=0.0001*i but not less than 1.0, so index=1.0


Best regards.

Top
#42282 - 04/07/11 08:14 AM Re: Sustained stress index (0.75 SIF or 1.0SIF) [Re: mariog]
bluna Offline
Member

Registered: 02/05/08
Posts: 4
Loc: Spain
Thank you for the answer.

Regards,
bluna

Top
#42360 - 04/12/11 02:32 PM Re: Sustained stress index (0.75 SIF or 1.0SIF) [Re: mariog]
PKU Offline
Member

Registered: 08/18/06
Posts: 78
Loc: Aberdeen
This discussion is quite useful and users need to address this by invoking Code case 178 to satisfy B31.3 2010 editions. Have anyone verified the sustained code stress generted in extended report?
_________________________
PKU

Top



Moderator:  Denny_Thomas, uribejl 
Who's Online
0 registered (), 124 Guests and 0 Spiders online.
Key: Admin, Global Mod, Mod
May
Su M Tu W Th F Sa
1 2 3 4
5 6 7 8 9 10 11
12 13 14 15 16 17 18
19 20 21 22 23 24 25
26 27 28 29 30 31
Forum Stats
12065 Members
14 Forums
16973 Topics
75151 Posts

Max Online: 303 @ 01/28/20 11:58 PM
Top Posters (30 Days)