Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations cowski on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

Hydrostatic Relief Valves with Anhydrous Ammonia 1

Status
Not open for further replies.

Rooney

Mechanical
Feb 6, 2002
2
My question pertains to the placement of HRVs on extremely short sections of pipe between a manual shut off valve and a blind flange. The fluid being used is anhydrous ammonia and we have the ability to lock the manual valves open if necessary. All of the resources I have found require an HRV on a section of pipe between 2 or more shut off valves, but none of them make mention of the situation between 1 valve and a blind flange. Any help would be greatly appreciated.
 
Replies continue below

Recommended for you

It would seem that ANY blocked in section of pipe would behave similarly, whether it is between 2 valves or a valve and a blind.

Another option may be an expansion chamber with a rupture disc isolating the chamber from the blocked in section. The disk is normally instrumented to alarm when blown. This system ensures total containment. These are used in liquid chlorine systems, but I'm not sure about ammonia.
 
Butelja is right on the money. I've seen some specs which didn't require PRVs on sections of piping (less than certain lengths) that could be blocked in with valves. The reasoning is that valves leak 'slightly' and with most situations like this, slight leakage is all you need to prevent an overpressure.

If one of the valves is a blind flange, it's fair to say that leakage is going to be zip from that direction so overpressure becomes even potentially more of an issue.
 
Leakage isn't really the problem. It's a very cold fluid warming to ambient temperature while trapped. Very low pressure today can evolve into a suprisingly high level by tomorrow morning. It's the exact reverse of steam, which collapses into a vacuum.
 
But TD2K point was that since you are dealing with a small volume the leak from the valve will make sure that the pressure does not increase that much.

Best Regards

Morten
 
Thank you all for your replys. The one thing that no one addressed was locking the valve in the open position during normal operations. In this scenario, the short piping section is always in constant communication with the rest of the piping run, so there can be no pressure buildup. The only time the valve would be closed would be for maintenance involving that short piping section. Does anyone know if locking the valves open is allowed instead of installing the HRV?
 
At the refinery I'm currently working at we have all sorts of piping runs which can be potentially blocked in because of the need to change line-ups. Theis results in lots of cross connected lines with valves at each end. For the in-plant piping, I have not seen any relief valves even for steam traced piping. The refinery relies on its standard operating procedure(s) that essentially state 'don't block in both ends of a pipe at the same time'. Now, on the longer runs of line down to the wharf, those long runs of line do have some relief valves.

I also previously worked in an ethylene plant years ago which has low temperature liquids. We did not provide relief valves on piping that 'could' be blocked in just because there were block valves. On the other hand, you don't have the cross connection issue to the same degree as in a refinery.

On another job I did on the North Slope for an NGL injection skid, the client did want relief valves on all sections of piping that could be blocked in.

I think at the end of the day you'll need to look at when these valves are going to be used. If they are routinely used as part of normal operations, it would be much easier in a Hazop to say that misoperation is likely. If they are only used during specific times, you might be willing to rely on operating procedure depending what the plant philosophy and history is (if you routinely blow out flanges, that's a sign procedures likely aren't enough <g>).

The type of valves is another issue. On my previous post, I should have added that on some of the specs where relief valves are discussed for blocked in piping, they do state if the valves are truly bubbletight, then relief valves may be necessary even on short runs.
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor