6 replies to this topic

[JoeWong]

Jaimin,

At choked flow condition, the flow passing through PSV will be more-or-less fix. Then the PSV downstream pressure will be the pressure drop (along the flare header) plus ATM.

HTH
JoeWong

[latexman]

The pressure at the outlet plane of the nozzle (not orifice) will be set by the pressure that gives sonic flow. Outside the nozzle there will be expanding jet(s) and shock wave(s). Across the shock there are step changes in pressure (loss) and entropy. The pressure downstream of the shock equals that stated by JoeWong, "the pressure drop (along the flare header) plus ATM." A shock wave is very, very thin and this irreversible pressure loss happens extremely fast. I recommend reading Shapiro's The Dynamics and Thermodynamics of Compressible Fluid Flow.

[pleckner]

I'm going to see if I can simplfy this a bit. You determined your worst case scenario, therefore you mave your maximum required relieving rate. Assumning we are discussing 100% gas/vapor flow, you apply the API equation to calculate the required PSV orifice you need. When you applied this equation, you are assuming a back pressure effect, even if it is zero. You then use a "1" as the back pressure correction factor in the API equation.

You choose a PSV. Now, you obtain the certified (stamped or rated) capacity of this PSV and use this flow in your calculation to determine PSV tail pipe and header size or pressure drop if the size if fixed. You can start at the flare and work backwards. The pressure at the flare (or whatever is your final destination) should be fixed and we'll say atmospheric. As the gas/vapor flows from the PSV outlet to the flare, pressure builds up in the piping, this is commonly called the built-up backpressure. The final pressure drop in the piping determined by fluid flow equations will give you a backpressure at the PSV outlet flange. You then check this against the charts to see if the backpressure is too high such that you either have too small of a tail pipe or you adjust the PSV size by using a backpressure correction factor of something less than one (the charts will give this to you) in the API equation. This could cause the required PSV size to go up in size.

PSV sizing can be trial-and-error if the tail pipe and header system is too small.

[pleckner]

There is a discontinuity between the PSV throat and the PSV outlet flange. You must break your analysis up into two parts, the first being what happens in the PSV and the second what happens in the downstream piping. The cirtical pressure ratio you keep referring to is that in the PSV, not in the piping.

Again, you calculate the pressure drop in the piping downstream from the PSV, usually backwards starting at the destination point where you know the final pressure, and determine the backpressure at the PSV flange. Again, this has nothing to do with the critical pressure ratio you are referring to. Once you have this value, you can set your piping design pressure. (Note that you can go forwards also but then this will have to be a trial-and-error calculation. Going backwards may be able to eliminate the trial-and-error.)

In general, if this is a balanced bellows PSV then the maximum backpressure should be something less than or equal to around 50% of the PSV set pressure. You can then add whatever safety factor you normally use (I use 10% or 25 psi, whichever is greater, above the maximum operating pressure) to set a design pressure.

If this is an open system, my design pressure will typically be the same all the way from start to end using the highest pressure I calculate.