B31.3 Reduction in MDMT without Impacts

[JackWayne]

I am having a difficult time in figuring out the Stress Ratio in Fig. 323.2.2B of B31.3 2004 Ed. What exactly is the Nominal Pressure Stress? I've figured the minimum wall thickness necessary for the pipe; going with the ID formula I come up with a minimum thickness of .599" + C.A. of .0625" = .6615". I am looking at using 6" XXH (.864")SA-106C material. The pipe MAWP is 5000# @ 350°F. I am trying to determine the MDMT (from what I can tell using 323.2.2A the material is only good for 25°F on it's own. How do I come up with the Stress Ratio to determine if I can go lower (it needs to go to -20°F). This may be basic but as I am used to ASME Sec. VIII (the thickness is under 1" P1 material so it isn't an issue for Vessels) I am a bit confused!

 

[rneill]

Not sure why, but I'm getting different results. My software is based on the OD formula in B31.3-2008 (Eqn 3a) and indicates that for 6" A106C pipe at 5000 psig, with a 0.0625" corrosion allowance, I need 0.826" wall thickness ...

This includes a manufacturing tolerance of -12.5% on wall thickness and is based on the reduced allowable stress value at 350 F (interpolated between the 300 and 400 F values) of 23.1 KSI.

There really shouldn't be a reason why the ID formula (3b) and the OD formula (3a) would provide such different results so I'm not sure what's up although the way the calculations work, you essentially have to gross up the corrosion allowance by the tolerance as well and that is often done incorrectly in hand calcs.

Based on the calculated wall thickness of 0.826", I would select the Sch XXS pipe which I then interpolate to have a MDMT of 24 F without considering the stress ratio (basically the same number you had).

Since I calculate the Sch XXS pipe to have an actual MAWP of 5278 psi, and the maximum is only 5000 then my stress ratio for pressure is 94.7%. Note that per the notes to Fig 323.2.2B, I do need to consider separately components with established component ratings (e.g., 5000/component rating), and I do need to consider combined longitudinal stresses as well which would typically require something like Ceasar II (or a lot of hand calcs).

However, Based on Pressure Alone, I would be allowed to take about a 5 F reduction in the MDMT since I am at 94.7% of the allowable pressure (or stress). This doesn't get you anywhere close to the required MDMT of -20 F.

At this point, you're going to be looking at an impact tested material and it is probably best to select a grade that includes impact testing as part of the mandatory requirements as opposed to doing your own impact testing after the fact.

 

[JackWayne]

Thanks rneill, as soon as I get back to the office I will do some more playing, I certainly could have screwed up the hand calcs.

 

[rneill]

I had a couple of other thoughts to pass along ...

1. Impact Test Exemption in ASME VIII Div 1:

I had a situation about 6 months ago where we actually impacted tested some 3" Sch XXS (0.60" wall), A106 Grade B pipe that a client had mistakenly installed. This was for 5000 psi service and we needed an MDMT of -9 F but the client had installed the A106 assuming that B31.3 accepted this to -20 F (their spec had called for A333 Grade 6 pipe but ...). Despite only being impact tested at -9 F, the pipe failed miserably. Consequently, despite the exemption in ASME VIII, I would not assume that the material would actually have acceptable impact test values. Myself, I do not allow the use of the UG-20(f) exemption above 1/2" in any specs I write.

2. Connections:

Just wondering what you are using for connections (e.g., flanges) - I think you'd need API 10,000 to get a rating of 5000 psi at 350 F since neither ASME Class 2500 nor API 5000 flanges would typically have this rating. Perhaps you're using hub and clamp connectors?

 

[bernoullies123]

The stress ratio is defined in the General notes of Fig. 323.2.2B (2008 ed.)

This thickness calc is the same by using either the OD or ID calculation per Interpretation 21-13

http://cstools.asme.org/csconnect/pdf/CommitteeFiles/26668.pdf