PSV for Tube Rupture - To relief Gas or Liquid Flow

[DoraeS]

Hi, I have a PSV sized for HEX tube rupture case, with tube side having high pressure gas, and shell side having low pressure liquid. When doing the PSV relief load calculation, one engineer told me that we need to calculate the high pressure gas flowrate through the broken tubes, and then take the calculated volumetric flowrate and specified it for liquid flow in PSV datasheet. My question is, is this correct? And why do we need to consider for liquid flow and not vapour flow? Thanks for all comments.

 

[Latexman]

If there isn't enough empty volume between the PSV and the HEX for disengagement, you may also need to consider two phase flow.

If there is a credible scenario that has liquid flow, it should be considered. Ditto, on vapor only and two phase flow. If more than one is credible, the scenario that has the largest area is selected.

Good luck,
Latexman

 

[CJKruger]

DoraemonS, there are numerous problems with putting a relief valve in this service.

My guess is: By the time you are done with all the endless discussions and studies to try and prove a relief valve will protect the equipment, it will be cheaper to simply replace the exchanger and follow the 10/13 rule.

 

[CMA010]

The ingress of the vapour pushes the liquid out? Therefore the volumetric vapour flow rate (at the low pressure side conditions) will be the (initial) liquid relief rate. You're probably going to end up with a huge relief valve, follow CJKruger's advice.

 

[sheiko]

CJKruger,

Do you mean a S&T heat exchanger is cheaper than a relief valve? Please explain.

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

 

[CJKruger]

sheiko, no a relief valve is probably cheaper than an exchanger. But below is the sequence of events on a project:

- Decide to use a PSV for tube rupture. Size it.
- Client question whether it will open quick enough.
- Consider doing transient study.
- Conclude it is too costly and time-consuming.
- Decide to install a rupture disk. Size it.
- Operations refuse to use rupture disk.
- Give up and replace exchanger.

 

[MikeClay]

CJKruger:

You forgot two the steps:
- Decide to install a buckling pin device.
- Operations determines that finding one certified
by ASME is difficult and refuses to insall one.

But the 10/13 rule is nice for the heat exchanger, but a leak case may still be valid and the connecting piping may not have the same pressure rating.

--Mike--

 

[CJKruger]

MikeClay, yes, and this brings up several interesting points.

1) My understanding is, that even if you design the exchanger for the 10/13 rule, you still have a tube rupture case. The only benefit of designing for the 10/13 rule, is that you can rely on downstream relief valves and do not need one on your exchanger.

2) This implies that the piping immediately around the exchanger must also be good for the 10/13 rule. As far as I know, many companies make the piping design pressure the same as the exchanger up to the first block valves.

But, I would be interested in hearing your experience/comments.