Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations GregLocock on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

PSV Reaction Force Calculations for Pipe Stress Analysis - General Clarifications

Status
Not open for further replies.

Vulture860

Mechanical
Oct 25, 2012
7
I'm calculating the safety valve reaction forces using the equations out of B31.1 Nonmandatory Appendix II (for a steam PSV) and out of API 520 Part 2 (for a flare gas PSV). I have a few questions regarding these calcluations:

1. My understanding is that B31.1 Nonmandatory Appendix II is for steam applications only while API 520 Part 2 can be used for both steam and for all other vapors. Is this a correct statement?

2. When looking at open discharge systems, what does the P component in the API 520 reaction force equation physically represent? The definition is "the static pressure within the outlet at the point of discharge". I've always understood that the pressure component of these equations was your delta P at the discharge (conservatively using set pressure minus atmospheric), which results in some pretty substantial pressure thrusts at higher set pressures. However from just thinking about this, I feel that the end of a discharge piping likely won't see the set pressure at the outlet considering the gas is expanding through the PSV orifice into the larger discharge header. It seems B31.1 outlines an approach to calculate P1 (assumed at very edge of discharge piping before it encounters atmoshperic pressure). However, I don't have a good understanding of what they're physically calculating and I assume this approach only works for steam applications.

3. I've seen arguments for not having to account for reaction forces in closed discharge systems. Is there any truth to this, I simply don't have a good understanding of how the relief vapor would act physically on a closed discharge system.

4. I typically apply the B31.1 dynamic load factor value (max = 2.0) to all PSV applications, including reaction forces caluclated out of API 520. Does anyone see an issue with this approach (too conservative, not conservative enough, etc)?

Any advice or guidance to any these items would be greatly appreciated. I am attempting to develop a standard approach to analyzing PSV's from a pipe stress analysis standpoint and would like to hear how others approach these applications.

Thanks.

 
Replies continue below

Recommended for you

The equation doesn't care what the differential, or outlet pressure actually is, because it takes the conservative approach and uses an outside, or outlet pressure of 0 thereby calculating the maximum possible differential pressure in all possible cases no matter what the outlet pressure boundary conditions might be.

You always have to account for unbalanced pressure forces. The differential pressure of in-line valves is usually easily resisted by axial forces in the pipe wall. As often is true, pipe walls have a rather substantial capability to carry these axial stresses and soon dissipate them into the supporting frames, or into the surrounding soil via friction forces in the case of underground pipelines. You can never ignore axial forces from any source, only make a determination that they are, or are not significant to your problem at hand.

PSV thrust loads are applied at the point where they act, the centroid of the valve, with the appropriate directional vectors, which will normally put significant bending loads, in addition to the axial loads, on the pipe, as it is only the pipe which is often the sole support for the valve itself.

OMG%20something%20else.png
 
You may also want to think about what the maximum unbalanced pressure force might be on a closed valve located at the end of a pipe, turned various directions from the pipe's longitudinal axis, and how the resulting load on the pipe might be resisted.

OMG%20something%20else.png
 
Thank you for the reply, much appreciated to get some insight on this.

To clarify the 'P' value portion of your response, let me propose at quick hypothetical.... say i have a PSV set at 200 psig set pressure and a 6" sch 40 outlet. Per the rules of API 520, the approach to determine the pressure component P*A would mean I'd multiply 200 psig x 5.761 in^2 = 1152.2 lbf? (not accounting for overpressure/accumulation/etc for this simple example).

Also, do you agree that B31.1 Nonmandatory Appendix II is meant for steam applications only?
 
P X A is independent of fluid.

If the applied force was calculated from thrust, then I believe you would see Force = mass x acceleration of the fluid, in which case you would need to distinguish between fluids.

OMG%20something%20else.png
 
MpisPipeStress,
The force in the 6inch pipe will not be the set pressure times the cross sectional area of the pipe. Look at API 520. The required P for the PxA portion of the thrust is the stagnation pressure at the point of concern - not the static pressure and certainly not the set pressure.

It's amazing how many pipe stress engineers get the application of API 520 wrong. At the last company I was at the "guru" applied the thrust equation to relief valves which were passing the gas at sub-critical flow conditions even though API 520 states that the equation is for "choked" flow only. The first thing to determine is whether you have critical flow or not.
 
Biginch,
Think you need to take a look at the API 520 Code and particularly the equation in Section 4.4.1.1 of the API 520 Part II 2011 Edition. Low and behold in that equation there are two terms contributing to the Reaction Force - a momentum term and a pressure term with the pressure term being P*A which is what I was alluding to. Appologies willbe accepted graciously.
 
I didn't insult you, did I? Where? If I did, I am sorry.
I simply said P * A wasn't the API 520 method, it's just a method.

OMG%20something%20else.png
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor