Rupture Disk/Relief Valve Combinations 1

[swiss8018]

I'm currently evaluating the feasibility of implementing a rupture disk/relief valve combination in hydrogen cyanide service. We've been having polymerization issues that could negatively impact the performance of our current relief system. Although inhibitor addition is certainly an option, I do want to explore some other options as well.

Does anybody have experience regarding such a setup in HCN (or similar) service? How did you set it up? How do you deal with a leaking rupture disk (i.e. do you bleed off the leaking material, shutdown the equipment, etc)? Can a relief mechanism still function properly if the rupture disk is leaking by?

Any input is appreciated!

Swiss

 

[abcmex]

Swiss,
usually the rupture disk do not have leaks. That is why in some applications they are used ahead of the safety valves. They are open (ruptured) or closed (sealed).
Normally you will find a manometer between the disk and the safety valve to know when the disk is damaged.

perhaps will help:

http://www.bsbsystems.com/products/installation_instructions/pdf/77_4005Sept_2000.pdf

RGS

 

[cawse001]

The rupture disk allows you to specify a lower spec (generally with respect to corrosion resistance) material for the safety valve.

The main issue in materials selection is how long can you wait after a disk failure (not in an emergency relief situation) before being able to replace the disk/PSV.

When reviewing options be careful about the vacuum rating of the disk and its ability to handle thermal changes in the system.

 

[TrevorP]

Solidification, e.g. caused by polymerisation, can certainly have an effect on relief valve performance, and an upstream bursting disk is certainly the preferred option. However, even the performance of a bursting disk can also be affected by solidification on the disk. An inhibitor is therefore likely to be desirable or required, regardless of the relief system selected. Since solidification is a risk, I would certainly recommend use of a bursting disk. The disadvantage of bursting disks is that once they've gone, there is a path open to atmosphere, as it doesn't reseat like a relief valve. If this aspect is undesirable (And with HCN it will be!), then a bursting disk used upstream of a relief valve is certainly the preferred option.
As mentioned above, it is essential to monitor (and preferrably relieve - i.e. via an excess flow valve) the space between the relief valve and bursting disk. This is because the gases between the two devices will have a tendency to expand and contract with variations in ambient temperature. Since there is a fixed volume, they cannot do this, so the interspace pressure will increase. This increased pressure on the downstream side of the bursting disk will directly affect the bursting disk relief pressure, thereby affecting the integrity of the system it is trying to protect.

 

[EGT01]

Swiss8018,

I don't have much experience with polymer service but Fike makes a rupture disk (Poly-SD series) specifically for polymer service that is supposed to minimize the chance of polymer buildup across the pressure face of the disk. There may be other manufacturers as well.

Obviously, development of a polymer plug across the disk could alter its burst pressure but I would caution that plugging of the nozzle upstream of the disk could still be an issue. Seems I've heard of some situations where a purge fluid might be used in the piping of a relief system to help keep the relief path open.

Pressure buildup between disk and valve will affect the burst pressure of the disk. For lethal service applications, ensuring no pressure buildup becomes a little more problematic. For free vent installations, you will need to vent the space to some closed, low pressure system that won't affect your disk burst pressure.

In some lethal service cases, I've seen the space between disk and valve setup as a nonvented space. However, one of two conditions had to be met.
1. The sum of relief valve set pressure plus rupture disk burst pressure did not exceed the MAWP of the protected system.
2. A pressure transmitter was connected to the space between disk and valve and setup with a high pressure alarm to act as a "suitable telltale indicator".

In any event, you probably should have operating procedures that require operations to routinely monitor these installations for pressure buildup. Of course you probably still need a vent line to a closed, low pressure system to allow for depressure before line breaking. For the nonvented space installations, I've seen the vapor space between disk and relief valve hard-piped to the relief valve discharge which was routed to a closed, low pressure system. The interconnecting piping was fitted with block valves as needed.

 

[saxon]

swiss, If you use this type of setup, make sure that the RD is spec'd as a "non-fragmenting" disc. Fike makes them as well as others.

saxon