Balanced Bellows for Thermal PSV?

[processpeeps]

I'm specifying a PSV for thermal expansion case. Tailpipe will discharge downstream of a globe valve back into a header. Header pressure is 30 psig. Set pressure is 85 psig. Is a balanced bellows PSV required due to backpressure (> 10% of set pressure), or is a conventional PSV acceptable for a liquid thermal PSV (very low flow rate)?

Has anyone ever seen/specified a balanced bellows 3/4" x 1" or 3/4" x 3/4" PSV? I have only seen conventional in the past.

 

[Ashereng]

Maybe take a look at the Farris 2700 series?

http://www.cwfc.com/farris/catalog.html

This is the link to the spec.

"Do not worry about your problems with mathematics, I assure you mine are far greater."
Albert Einstein
Have you read FAQ731-376 to make the best use of Eng-Tips Forums?

 

[pleckner]

The general rule is that the variable backpressure should not exceed 10% of the set pressure for conventional valves. Note I write "variable". When you send the specification to the vendor, you report the backpressure and whether it is constant or variable. If the backpressure is relatively constant, no problem for the conventional relief valve.

 

[RJB32482]

So let me get this straight, if I have a PSV set at 100 psig with a CONSTANT backpressure of 20 psig, I'm OK with using a conventional relief valve? THough that the 10% rule covered both constant and variable backpressure?

Thanks.

 

[EGT01]

Just looking at Farris and Consolidated, I believe 1x2 is the smallest you can get a balanced bellows design. But as Ashereng indicates, both Farris (model 2700) and Consolidated (model 19096) have smaller valves that come in a balanced piston design. The balanced piston design is similar in action to the balanced bellows but doesn't quite have the backpressure tolerance that the bellows offers.

Using a balanced valve design has its advantages but also disadvantages too. For either balanced bellows or piston design, the bonnet of the valve needs to be properly vented so that pressure won't buildup in the bonnet in case of a leak developing in the bellows or piston seals. This becomes more of a problem if you are dealing with hazardous materials but even if not, any leak is an additional maintenance problem.

As Pleckner points out, constant backpressure is not necessarily a problem for a conventional valve. Relief valves do have limits for backpressure that don't necessarily follow the pressure rating of the outlet flange. The backpressure limits are generally included in the manufacturers catalog.

The catch to using a conventional valve with constant backpressure is having to compensate the spring setting for the constant backpressure so you do not exceed the maximum pressure allowed for the equipment being protected. For a conventional valve, the set pressure is the sum of the spring differential set pressure and the superimposed (constant) backpressure. So if you want a relief valve to open at 85 psig (assuming same as protected equipment MAWP) and it has a constant backpressure of 30 psig, then you need to specify a spring differential set pressure equivalent to 55 psig.

For further information regarding the definition of pressure related terms for relief valves (for example Cold Differential Test Pressure) and other considerations of backpressure on conventional valves, see API Recommended Practive 520, Part I.

My preference would be to first try and use a conventional valve, then look at balanced valve design as second choice.

 

[Ashereng]

From the OP, I assumed that the header pressure of 30 psig is constant. I know he didn't say that, but that seems like a normal header pressure going to the KO.

In any case, I took the request for suggestion on a 3/4" x 1" valve, non conventional, and the 2700 by Farris is a good one.

"Do not worry about your problems with mathematics, I assure you mine are far greater."
Albert Einstein
Have you read FAQ731-376 to make the best use of Eng-Tips Forums?

 

[pleckner]

The 10% rule refers to variable backpressure. Let me quote from the Farris 2600 series cataloge, page 84, "The conventional valve may be used when the variation in backpressure does not exceed 10% of the set pressure, provided the corresponding variation in set pressure is acceptable." Remember, the conventional valve is a differential device, much like a rupture disk.

You don't really have to worry about what the differential spring setting will be unless you are manufacturing the valve yourself. All you need to do is tell the vendor you want a conventional valve, tell them the set pressure and the expected backpressure and tell them if it is constant or variable (by some %). The vendor will come back and tell you if you can't use a conventional valve. They will design the valve with the proper spring setting.

As far as the limitations in constant backpressure, this is true but it is a mechanical restriction only and is usually quite large.

 

[Latexman]

The test bench set pressure (open to atmosphere) + constant backpressure = opening pressure.

Good luck,
Latexman