Pressure drop across a relief valve 4

[roker]

Hello,

Could someone clarify what is the process occuring across the RV orifice: some times hundreds of bars are dropping across the RV, to what kind of energy this process is transformed? is it friction (heat)?

regards,
roker

 

[dcasto]

The process is the same as a control valve. There is no change in enthalpy, no work added or removed. The entropy will change, increasing to become more random. The temperature will normally decrease (there are points where a gases like methane will get warmer with a small pressure drop somewhere aroung 6000 psig level). Liquids may enter a two phase, they will also cool slightly. So the change in entropy is reflected in a change in temperature.

 

[Latexman]

It is my experience that RVs (relief valves) are modelled, NOT as an orifice, but, as an isentropic nozzle. A pressure safety valve (PSV) is usually located on a nozzle on the top head of a pressure vessel. The top head of the vessel *acts* like an ideal converging nozzle from the large diameter of the vessel to the inlet diameter of the PSV. The pressure (usually the MAWP) at the top of the vessel straight side is treated as the stagnation pressure and temperature (velocity ~ 0). It is accelerated to the PSV nozzle size at decreasing pressure at the nozzle exit until the nozzle outlet pressure reaches the back pressure on the PSV or the mass velocity reaches a maximum (sonic flow).

Being isentropic, it is usually treated as frictionless.

Good luck,
Latexman

 

[Latexman]

Btw, control valves are usually modelled as isenthalpic, not isentropic.

Good luck,
Latexman

 

[roker]

Hello Latexman,

Thank you, as I understand the DP across the PSV is converted to kinetic energy at the PSV outlet and than to friction loss on the outlet piping to the flare.

What disturbs me is that a DP of sometims hundreds of bars is acomplished on a "disk" of a few milimiters.

regards,
roker

 

[CJKruger]

Latexman,

To find the conditions in the PSV "nozzle throat", you do a isentropic flash from the inlet.

To find the PSV outlet (tail pipe) conditions, you do a isenthalpic flash from the inlet.

 

[Latexman]

CJKruger,

Thanks. I should have said, RVs are modelled, NOT as an orifice, but, as a flow nozzle.

Good luck,
Latexman

 

[Latexman]

roker,

I'm wondering what you mean by "a DP of sometims hundreds of bars is acomplished on a 'disk' of a few milimiters."

The disk of a PSV sits on top of the full nozzle and has the spring force pushing it down so it opens at the "set pressure". Most disks have a diameter larger than the nozzle diameter and, possibly a lip at the outside diameter making what is called a "huddling chamber". Once open, the pressure and flow of the fluid act on this larger area to fully open the PSV. This is what "pops" the valve open and keeps it open until the pressure is somewhat less than set pressure.

But, I don't believe the thickness or diameter of the disk comes into play with determining the pressure drop. Yes, it has to be thick enough to have the mechanical strength it needs. The majority of the pressure drop comes from the fluid no longer being confined and expanding.

Good luck,
Latexman

 

[roker]

CJKruger,

Correct me if I'm wrong but by flash did you mean expansion?

Latexman,

It will be helpful to understand at the end of the relief process what is the product of the energy release.

regards,
roker

 

[Latexman]

It's not clear to me what you are asking, but the end of the tailpipe does experience thrust which has to be designed for.

Good luck,
Latexman

 

[Latexman]

roker,

I have written a FAQ on the issue of what is the process occurring across the PSV. It is a derivation of the numerical integration technique I use for sizing PSVs. It is faq1203-1293 I have found that if you understand the derivation of a solution method, you most likely understand the problem and it's solution. Take a look.

Good luck,
Latexman