Relieving Pressure Too Close To Critical Point 2

[jrthomas]

Here's what I've got: Tank MAWP = 430 psig; Existing PSV set pressure = 420 psig; Critical Pressure of contents (F-404a) = 541.2 psia; Sizing Scenario = Fire Case (F-404a is not flammable, but client wants to see the calcs).

OK, so here's my question: How is the relieving pressure calculated when the set pressure < MAWP? The reason this matters is because it affects the value for the heat of vaporization of F-404a that we "plug" into the sizing equations. A lower relieving pressure means we get away from the critical pressure, get a larger value for heat of vaporization, and this allows us to use a smaller valve.

I'm pretty sure the relieving pressure in this case is either 535 psia (based on MAWP) or 522.9 psia (based on set pressure), there's not really that much of a difference. However, the suggestion was made to lower set pressure to 350 psig in order to lower relieving pressure to 438.2 psia. Now there's quite a bit of difference.

In my mind, relieving pressure should be based on set pressure, but API 520, 8th ed., Tables 2 thru 6 destroyed that idea. In their examples, relieving pressure is exactly the same, no matter if set pressure is equal to MAWP or set pressure is less than MAWP.

 

[jrthomas]

Sorry for posting again so quickly, I wasn't sure how to edit my original post.

I just wanted to say that preliminary calcs show the existing valve to be way undersized if we use a 535 psia relieving pressure (and the associated very small heat of vaporization). It would be great if we could utilize the existing valve by simply lowering the set pressure (and therefore the relieving pressure). But I'm not sure if it's that simple, so that's why I'm asking.

 

[don1980]

jrthomas - the relieving pressure is a design value that you choose. In other words, you're free to size the PSV for any relieving pressure you choose, so long as that value doesn't exceed the vessel's allowable accumulation (1.21xMAWP in this case).

Protecting vessels containing supercritical fluids from fire exposure is a real problem. Unless the relieving pressure is sub-critical, a PSV won't provide any real protection at all. You can sometimes design around that by lowering the set pressure enough and making the valve big enough to keep the pressure below the critical point (keep it in the region where there's an Hvap). If there's no Hvap, you still need to install a PSV for code complaince, but it's pointless to size it for fire exposure because that valve won't do any good, regardless of it's size.

 

[moltenmetal]

Vessels containing gases or supercritical fluids or high boilers need thermal relief, non-reclosing relief devices or automatic depressurization systems if there's a credible risk of prolonged exposure to fire. But is there a credible risk of prolonged exposure to fire in this case?

 

[jrthomas]

don1980,

Thank you for your post. This issue of set pressure being different than MAWP has made me evaluate the way I look at PSV sizing. I want to have a high level of confidence that what I am doing is right, whether it is evaluating credible scenarios and their flow rates, or plugging in numbers to a PSV sizing equation during the final step.

So, focusing on this particular example of a storage tank containing F-404a: We have already completed the first step of the PSV sizing process by determining credible scenarios, and are currently focused on determining the required relieving rate for fire exposure (which depends on the relieving pressure). Using good engineering judgment, the set pressure for the PSV installed on this F-404a tank should be at least 220 psig (approx. 200 psig max operating pressure + 10%). The maximum set pressure is limited to 430 psig (MAWP). IF a set pressure of 350 psig is chosen, are you saying that the relieving pressure can be between 350 psig (0 psig overpressure) and 520.3 psig (1.21*MAWP)? Surely relieving pressure can't be "any relieving pressure I choose" (like less than set pressure, for example). It must have both upper and lower bounds. I hope that I am understanding this correctly.

Best Regards,

Jacob

 

[don1980]

Jacob - The relieving pressure can be any value between the set pressure and the maximum allowable accumulation (350 - 520 psig in the case you describe). Choose a value that: (1) is below the critical pressure, and (2) results in a valve size that's not excessively large.

Note that the Hvap decreases as the pressure increases, until you get to the critical pressure at which point Hvap no longer exists. It's helpful to have a mental picture of this in your head - go to the internet and call up an enthalpy-pressure diagram. Most any diagram will do. You need the valve to be large enough so that the peak pressure (relieving pressure) doesn't reach the critical point.

As long as you do the calculations correctly, you can be confident the actual pressure during a fire won't exceed the relieving pressure that you chose. That's because sizing calculations are inherently conservative.

 

[TD2K]

ASME section 8 Div 1 states the maximum allowable overpressures most of us are used to: 10% for a single safety valve, 16% for multiple relief valves, 21% for fire. However, that's the maximum pressure. There's nothing wrong (or nothing that prohibits you) from designing your relief valve for a lower accumulated pressure.

Typically, I apply the 10%, 16% or 21% to the set pressure but if you have to in a marginal application, the real question is do you keep the vessel pressure from exceeding the maximum allowable under code? That can mean your PSV has a higher than usual accumulation at (for example) 110% of vessel MAWP.

In your case because of the critical point, setting the PSV lower than the vessel MAWP gets you a smaller safety valve because of the increase in latent heat.