PSV for Heat Exchagner Tubes Failure

[jolivers]

Hi everyone,

Is there a direct formula to calculate the required orifice area of a safety valve only with the following parameters?

- Cross-sectional area of the heat exchangers tubes
- Maximum allowable low side pressure (in that case, the shell)
- Maximum allowable high side pressure (tubes)

If not, please, tell me the best way to calculate it.

Thanks in advance.

 

[MortenA]

You should consider a double tube rupture. I would recommend a solution with a bursting disc AND a small PRV in parallel with a slightly low sp. The PRV would handle the shell shut-in case and slow pressure increase , where a ruptured disc would be too much trouble, and the disc would be your safety element that would handel prv failure AND the burted tube.

wrt. to the flow rate: Consider the area*2 (two end and choked flow across the opening.

Best regards

Morten

 

[don1980]

This question is answered in API 521 5.19.3. When a tube breaks, there will be flow through both ends of the break. Conservatively, you can regard this as two orifices.

 

[don1980]

BTW, Morten.......rupture disks are fickle devices. They're usually very reliable at opening when needed, but they're also prone to open when you don't want them to. Thus, they can be a significant liability and headache from a plant reliability perspective. The only time I'd put a disk on an exchanger is when there's a risk of a dynamic failure - a potentially explosive release of energy. This can occur when the shell side is full of liquid and the tube side has a high pressure liquified gas. A disk is the only thing that can work in that case. Otherwise, I'd avoid using a disk.

 

[MortenA]

As i said, install a PRV with a lower set-p. This valve would reduce the risk of an unwanted release. IMO the increased value in operational safety exceeds the (potential) operational problems from the BD.

 

[don1980]

I'm not saying you're wrong - just saying we'll agree to disagree on this point. Having a PRV with a set P lower than that of the disk won't do much to reduce the risks of premature/nuisance disk failure. That's exponentially true if the system is liquid-filled. Unfortunately, we have to use a disk for those dynamic tube rupture cases, but those cases are not common in a chemical plant or refinery.

 

[Jim111]

One question popped up, about heat exchanger with two tube( one for steam and one for ammonia liquid), between the two tubes is methonal. The RV of shell side is 6890 lb/hr( Methonal), the tubeside of ammonia is in liquid condition(-32 C@ 900 Kpa), flow rate is 399kg/hr). I just do not know how to check if the RV is good for pressure releaving when have tube rupture in ammonia tube.

Thank you for your time.

Jun