Minimum straight distance between a regulator and a safety valve

[sheiko]

Hello,

I was wondering if you know any criteria about the required minimum distance (straight line) between a regulator (actually a pressure-reducing valve) and a pressure safety valve?

I'm wondering because we are experiencing PSV chattering in a system where a PSV (set at 80 mbarg) is located 5D (D = line size, 2" in my case) after a pressure regulator, reducing pressure from 10000 mbarg (10 barg) to 50 mbarg. Seems that the cause of misoperation is the turbulence created by the regulator...

Any experiences? References for min. straighr line criteria?

Thanks.

"We don't believe things because they are true, things are true because we believe them."

 

[iainuts]

If this is a pop acting RV, it has nothing to do with turbulence. Generally, snap acting RV's such as used on pressure vessels work great where the volume to be relieved is large and the inlet losses on the pipe to the valve are small (ie: less than ~5%). On piping however, there is no large volume to vent because you have a pipe and the entire pipe isn't going to vent down anyway. The only portion of the pipe that will vent down is that portion which, when venting, has a pressure loss up to the valve of less than roughly 5%. Dynamic pressure losses for your typical pipe when a RV lifts are very high, so the pipe volume being vented is very small. The result is a blow down of this small section of line, followed by the valve closing. I did an analytical study on one such line and found the line got blown down in milliseconds. The inertia of the RV poppet was the only thing that slowed the chatter of this valve to the speed of a machine gun.

Note also that regardless of how well you size the RV/regulator pair, the problem won't go away. First, you can't get a RV that's EXACTLY the capacity of the failed open regulator. You want a bit of extra margin anyway. Even if you could, the reg doesn't necessarily fail wide open and you don't necessarily have the actual flow rate even if it does (Cv values are aproximate).

If you have any doubt about this being the cause, calculate the line losses when the RV is open, the flow rate for your failed open reg, and determine the volume you are blowing down and how fast it blows down. I think you'll be impressed by how short the blowdown you calculate is.

The solution is to get a relief valve that doesn't pop open like a typical ASME coded snap acting valve for pressure vessel service. There are a number of good manufacturers out there that make these, but it depends on your size and flow rates. Note that these are generally not ASME coded valves, but they don't have to be for ASME piping.

 

[sheiko]

Thank you!

BTW, the PSV is pilot-operated type. Fluid is gas nitrogen. Relieving flowrate is 335 Nm3/h. T = 15 deg C.
Instrument engineer said it is impossible to find conventional PSV for this capacity and set pressure (80 mbarg) combination.

And any idea/references concerning the mimi. straight distance required between PCV and PSV?


"We don't believe things because they are true, things are true because we believe them."

 

[mbt22]

Sheiko,

I suspect the 80 mbarg may be the problem. That seems awfully low for any relief valve. Is a liquid filled U-tube an option?

Matt

 

[sheiko]

Thanks Matt,

In fact, we made some tests and it appeared that the problem is coming from the regulators upstream the PSV. Indeed, they start "pumping" before the PSV start chattering. Apparently, the regulators would be oversized , which causes the flow to become erratic and the PSV to open and close. Normal flow for regulator is 15 Nm3/h and max capacity is 335 Nm3/h, which means the regulator is operating at 5% of max. capacity.


"We don't believe things because they are true, things are true because we believe them."