7 replies to this topic

[il747]

I am facing an issue while evaluating the PSV of an existing plant.

 

A heating medium (steam, tube side) vaporizes a hydrocarbon (n-pentane, shell side) in a vaporizer heat exchanger.

Vessel is constructed in accordance with ASME BPVC VIII Div1.

 

Design pressure of the system is 29 barg at the shell side.

Critical point of n-pentane is 32.7 barg and 196.55 °C.

 

For a blocked outlet scenario on the pentane side with heating medium continuing I am facing the following issue.

- Single pressure relief valve, 110% accumulated pressure = 110% x 29 = 31.9 barg which is very close to the critical pressure.

This gives excessive relief rates because of the relatively low heat of vaporisation (orifices size  ~ 66000 mm2 or 4x 8T10).

 

- Multiple pressure relief valves, 116% accumulated pressure = 116% x 29 = 33.6 barg which is above the critical pressure.

This gives much smaller relief loads because the heat transfer in the supercritical region is much smaller compared to the normal boiling, sub-critical, regime in combination with temperature pinching between steam and pentane side.

 

- Fire scenario, 121% accumulation, will also transition the system into the super-critical region.

 

Relief loads and maximum mass flux / orifice sizes have been derived using the rigorous method as per Ouderkirk and API STD 520 C2 respectively.

 

For 116% accumulation, blocked outlet, the required orifice area is around 3450 mm2, say a 4P6.

For 121% accumulation, fire case, the required orifice area is around 7400 mm2, say 2 times a 4P6

 

Would it be ok to install 2 x 4P6? (1 PSV for the process scenario and 2 PSV for the fire case)

One could argue that in this case there is only 1 PSV for the process (non-fire) scenario and a max. accumulation of only 10% is allowed.

Problem is that 1 x 4P6 at 110% would not cover the relief scenario because of subcritical, but 1 x 4P6 would be sufficient at 116%.

 

Or does the process scenario (blocked outlet) need to be covered by multiple relief devices in order to be able to use the 116% accumulation?

Then, I would end up with:

2 times 4L6 ( ~ 3682 mm2) to cover the process scenario (blocked outlet) and being able to use 116% accumulation.

And a 4P6 in addition to cover for the fire case as well.

 

Note: the system currently only has a 1x 4M6 PSV hence needs upgrading.

In case 3 PSVs are required it would become something like 4M6 + 3K4 (or 4L6) + 4P6.

 

I would appreciate your help in interpreting ASME BPVC VIII with respect to number of relief valves required or any other insights you might have.

 

[Bobby Strain]

I think you need to re-evaluate the relieving rate. I think you missed some important aspects. Your relief rate using only heat of vaporization near critical conditions seems in error. You should search for what you should do in this unusual case. I'm certain you will find this has been explored. Did you review the original relief calculation?

 

Bobby

[Pilesar]

Sometimes it works better (for thermodynamic reasons) to set relief pressure below system design pressure if the normal operating conditions allow.

[il747]

Thanks Bobby / Pilesar for your comments.

 

I have less of a concern with the calculated relief loads / orifice size. I have added more details in the below.

 

Would you be able to comment on my original question on which PSV configuration would allow the use of the 116% accumulation?

Quote:

 ==================================================================

2 x 4P6? (1 PSV for the process scenario and 2nd PSV for the fire case)

One could argue that in this case there is only 1 PSV for the process (non-fire) scenario and a max. accumulation of only 10% is allowed.

Problem is that 1 x 4P6 at 110% would not cover the relief scenario because of subcritical, but 1 x 4P6 would be sufficient at 116%.

 

Or does the process scenario (blocked outlet) need to be covered by multiple relief devices in order to be able to use the 116% accumulation?

Then, I would end up with:

2 times 4L6 ( ~ 3682 mm2) to cover the process scenario (blocked outlet) and being able to use 116% accumulation.

And a 4P6 in addition to cover for the fire case as well.

 

=====================================================

 

Thanks for your support.

 

 

For info. More information on relief loads / orifice sizes:

 

Blocked in scenario.

Originally, steam supply temperature was much lower. Pentane vapour pressure at that temperature was below PSV set pressure hence no relief for the blocked in scenario.

 

Currently steam supply temperature (from geothermal steam field) is higher.

Heat exchanger has been rated at 110% of PSV set pressure (normal boiling regime) and 116% (heating up supercritical fluid).

From this, duty and relief loads are derived.

For 110% the PSV needs to relief the resulting boil off rate

For 116% the PSV needs to relief the volumetric expansion of the supercritical fluid. (As per Ouderkirk: "Rigorously Size Relief Valves for Supercritical Fluids" CEP magazine, August 2002)

 

Fire case

Originally, equation from API std 521 5.15.2.2.2 was used: A = F' x A' / sqrt(P1).

This resulted in an orifice area of 6510 mm2 (but for some reason only an 'M' size 2323 mm2 was actually installed).

 

Using the rigorous method as per above, the required area is 7400 mm2

[Bobby Strain]

The steam pressure is more important than the temperature.

 

Bobby

[il747]

Hi Bobby,

 

Sorry for not making this clear: it is saturated steam.

 

Steam fields produce 2-phase fluid, steam plus high ammounts of water.

Water is knocked out before sending steam to the vaporiser. K.O. vessel is on pressure control.

We have taken the highest saturated steam temperature / pressure combination, i.e. the system reaching burst disc pressure on the steam side.

 

Cheers,

 

-IL

[Bobby Strain]

You don't need 121% of set pressure. Size 2 valves for 116% for the governing case. Or 110%.

 

Bobby

Edited by Bobby Strain, 29 March 2023 - 06:27 PM.

[sanderson51]

There are no rules that say that two PSVs have to be the same size in order to base sizing on 116% of set pressure.  A 3/4 X 1 in parallel with a 4 X 6 allows the capacity for both valves to be calculated at 116% set pressure.