We are performing an analysis of a B31.3 piping system at low operating temperature (-82 degF) using the unlisted material shown below.

A519-4130 75K Material
95 ksi min. tensile strength @ 200 degF
75 ksi min. yield strength @ 200 degF

Per Table 323.2.2 [Note 3] "Impact testing is not required if the design minimum temperature is below -20 degF but at or above -155 degF and the stress ratio defined in Fig. 323.2.2B does not exceed 0.3.”

Per Table 323.2.2B [Note a.3] the stress ratio is defined as the “combined longitudinal stress due to pressure, dead weight, and displacement strain (stress intensification factors are not included in this calculation) divided by S at the design minimum temperature. In calculating longitudinal stress, the force and moments in the piping system shall be calculated using nominal dimensions and the stresses shall be calculated using section properties based on the nominal dimensions less corrosion, erosion, and mechanical allowances”.

Per B31.3-302.3.2d1/B31.3-302.3.2d2 the basic allowable stress (S) is 31,667 psi.

Our approach is to take longitudinal stress (the sum of the Caesar calculated axial stress and bending stress) at each node and divide this value by the basic allowable stress (S) determined from 302.3.2d1/B31.3-302.3 to yield the Stress Ratio. If the calculated stress ratio is less than 0.3 the pipe qualifies. Does this approach appear valid?

Furthermore, since Table 323.2.2B [Note a.3] states that stress intensification factors are not included in the longitudinal stress calculation. Does Caesar apply SIFs to the Axial and Bending stresses shown in the Stress Extended Report or are the SIFs applied only to the Code Stress?

Thank you to everyone who helps with this inquiry.

Values will be with SIFs.

Thank you vadher for your response.

In light of your answer is there a way to obtain the longitudinal stress where the SIFs are not included in the calculation?

As in Caesar only one case you make SIF as 0 in calculation.
So consider sustain case with SIF zero.

This case won't give you the displacement strain.
So add T1 case and then check stresses of system.

This is the maximum we can do with Caesar to fulfill the code requirement.

If need any further assistance mail me sjvadher@gmail.com

Sapelous,

"Yes", in the Configuration (on the SIF & Stress pane), set the "Sustained_Case_Multiplier" to 0.00001.

Note, this "Code Interpretation" may now violate the rules of B31.3 2012. Verify what you're planning to do with the client and associated inspectors.

And build your load case with all components you wish to include (e.g., W+T1+P1+D1) and set the stress type to SUStained.
Be sure to check your work before accepting these results.

We investigated using the OPErating case (W+T1+P1+D1) to determine the code required longitudinal stress (sum of the Axial Stress and Bending Stress in the Stresses Extended Report). Furthermore, to comply with B31.3 Fig. 323.2.2B(a3) "stress intensification factors are not included in this calculation" we set the program to Allow User SIF @ Bends, allowing us to override all the SIFs with a user-defined value of 1.0 including the bends. Finally, the longitudinal stress was used to determine the stress ratio.

What is the reasoning behind using the SUStained case for this analysis? Can we achieve the same results using the OPErating case as suggested above?

Additionally, Richard, you stated that "this 'Code Interpretation' may now violate the rules of B31.3 2012". Can you elaborate?

Thank you all for your input on this inquiry; all help is sincerely appreciated.

The method you chose overrode the SIFs for your elbows but what about the SIFs for tees?
In CAESAR II we have a special B31.3 setting that can be used to adjust the SIF for teh sustained case. That's the 0.00001 multiplier Rich suggested. (0.00001 times any SIF will give an SIF=1.)
In CAESAR II, this switch does not work for load cases defined as "operating".
But don't use this load case to evaluate B31.3 sustained stress because B31.3 - 2012 does not allow you to set the sustained stress index to 1.0.

Coincidentally we did not have any tees in our model. However, I did check the output report to ensure that all the SIFs=1 for the OPErating case using the approach of overriding all the SIFs with a user-defined value.

Hypothetically, in a future scenario if we did have a tee couldn't we simply specify the SIF(i)=SIF(o)=1 at the tee?

Thanks.

Yes, but then you would have to run two models - one for B31.3 stress compliance and a second for this temperature check. The first is standard modeling with tees names and the second with tees removed and elbows set to SIF=1.

A little thought and work-around and you can probably use the same model for both.

Dave
It is interesting to note "A little thought and work-around and you can probably use the same model for both", I had similar situation and I created two files. Can you explain how the same file can be used managing SIF's?

I was hoping that changing stress type would allow me to manipulate the application of SIFs in CAESAR II.
I no longer have that hope.
I cannot offer a way to run, in a single model, BOTH regular stress checks AND this operating stress with SIFs=1 throughout.