Liquid PSV Inlet/Outlet Pipe Sizing 4

[ChEMatt]

I have a PSV that needs to be upsized from an E to an F orifice as the capacity needs to increase. The body is changing from a 1x1 to a 1.5x2. The orifice area is going from 0.212 in2 to 0.337 in2 as a result.

I'm hoping to avoid having to re-weld the inlet and outlet TOL's for the new PSVs and keep most of the existing piping.

The PSVs are located fairly close to the piping and I'll confirm we don't exceed dP limits on the inlet and outlet.

(sch 80)
1" pipe has an inside area of 0.7190 in2
1.5" pipe has an inside area of 1.767 in2
2" pipe has an inside area of 2.952 in2

Given that inside area of the 1" pipe is larger than the orifice size of the PSV, why can't I reduce down on the outlet of the PSV from a 2" outlet to the existing 1" pipe? This is brine water service so I'm not concerned with vaporization.

Thank you for your help!

-M

 

[Latexman]

why can't I reduce down on the outlet of the PSV from a 2" outlet to the existing 1" pipe?

If it don't exceed dP limits on the outlet, you can.

Is this a pop or modulating PSV?

Why did you mention vaporization? Is this cold brine or hot brine?

Good luck,
Latexman

To a ChE, the glass is always full - 1/2 air and 1/2 water.

 

[ChEMatt]

I mentioned vaporization to preclude the possible question as to whether or not I accounted for two-phase flow.

It's a pop PSV.

It's produced water, so, probably kind of warm, maybe 80F, 100F worst case. I will, of course, double check. But it was at atmospheric pressure before entering the pump. Sounds weird to say that about water...

Thank you for your reply!

-M

 

[Latexman]

Remember, being a pop PSV, the dP's have to be evaluated at the capacity of the PSV, not the sizing flow rate. The larger orifice PSV will increase that flow quite a bit, and dP [∼] (flow)[sup]2[/sup].

Good luck,
Latexman

To a ChE, the glass is always full - 1/2 air and 1/2 water.

 

[ChEMatt]

I can only give you one star per thread. Oh well. Thank you for that reminder! But, I did evaluate at rated flow.

-M

PS Seems like, looking at an H orifice at 0.865 in2, if the dP in the piping downstream doesn't create significant backpressure, you could get away with using smaller piping than the orifice. That can't be right tho, no?

 

[Latexman]

Many years back Code required the outlet pipe be the same size as the PSV outlet. Now, the dP limits rule the day.

Good luck,
Latexman

To a ChE, the glass is always full - 1/2 air and 1/2 water.

 

[met11]

If it helps, I've almost never had a case where I've been able to reduce the outlet pipe size to below the flange size and still keep the DP at less than 10% of set pressure. Those rated flows can get you. So if someone showed me a case where the outlet pipe was smaller than the outlet flange, it would either have to be a super short pipe run, or I'd assume they did the calculation wrong.

 

[don1980]

Why do we inherently have difficulty meeting the pressure drop limits when using some particular PSVs and not others? This is an unintended consequence which stemmed from the original creation of API 526 (API orifice sizes, and the PSV bodies into which these orifices are installed). As the orifice sizes increase, the bodies sizes (and their connection sizes) don't increase proportionately. The result is that some orifices (particularly the larger ones) are too big for the PSV body/connections. That is, some orifices allow a flowrate that inherently causes too much pressure drop for the PSV's inlet and outlet size.

See the table that I've attached. Specifically look at the ratios for inlet-area/orifice-area and outlet-area/orifice-area. This explains why you almost never have pressure drop problems with 1D2 PSVs, and why you almost always have pressure drop problems with larger PSVs like 4P6 and 8T10. When up-sizing the inlet and/or outlet piping isn't enough to solve a pressure drop problem, the most cost-effective solution is to restrict the lift of the PSV. Effectively, this allows you to reduce the rated capacity (the flowrate used to calculated the pressure losses) for the PSV. Since pressure drop increases proportionately to the square of the flowrate, it doesn't take much of a flow reduction to make a big impact on the pressure drop. Obviously, the reduced flowrate must be high enough to satisfy the required flow. Contact you PSV supplier when implementing this option. Based on your required flowrate, they will determine the necessary lift reduction, and they will supply the necessary part which is installed by a professional valve shop. The vendor will also provide the new rated capacity for this restricted-lift PSV.