Relief valves and Deficiencies

[SrChemE]

Would like to know if anybody had to develop plan to address PRD deficiencies?

I am currently working on a list of PSVs with the following deficiencies: incorrect setpoint, Inlet/Outlet Piping issues, orifice/valve sizing issues, oversized, undersized.
The way I thought about prioritizing them was to have the ones in hazardous chemicals on the top of the list, together with the ones that are Safety Critical. The next level should probably be based on cost.
Is there a better approach? Am I in correct the way I am prioritizing them?
Can I have some ideas on how to prioritize the remaining ones?

 

[The Obturator]

Can you advise on what you are trying to achieve here ? Are you for example, trying to establish a list of "reliable" SRV's ? I am asking since some of the criteria you already stated may not always be the obvious answer. Has the vendor selected properly against correct provided information ? Has the valve been properly tested after shipment and prior to installation ? etc.

Furthermore, are you interested in process applications only. spring operated or pilot operated SRV ?



Per ISO, only the term Safety Valve is used regardless of application or design.

 

[SrChemE]

The defects originated from an audit. The PSVs undergone engineering evaluations that were performed by several process engineers. Now the deficiencies from the audit need prioritized and a "plan of attack" put in place for determining a path forward. This is part of our PSM improvement goals.
I was just wondering if anyone had had to prioritize relief valve deficiencies?

 

[EmmanuelTop]

The easiest way is to prioritize deficient PSV's based on the consequences of their deficiency. Sounds very much straightforward, regardless of the type of deficiency.

It would be great if you can also factor the likelihood of failure (something like the chance that a relief might be required from a PSV during "normal" operation - some systems are simply closer to this point than others, e.g. live process vs. storage). But even without frequency, the consequence ranking should do.


Dejan IVANOVIC
Process Engineer, MSChE

 

[SrChemE]

Dejan, thanks for your suggestions. I will rank the PRVs based on your recommended likelihood of failure and consequence of failure.

 

[vladrath]

At the plant I'm at the people who came up with a plan of attack prioritized on the most hazardous applications, safety critical and then 'everything else'. Essentially what you did.

Once you have those buckets filled I would go with.

1. Incorrect orifice (I'm assuming this takes care of undersized/oversized).
2. Incorrect setpoint.
3. Incorrect inlet/outlet lines.

 

[gocougs1]

From my 5 years of experience with refinery PSVs, I see three levels of priority for PSV audit deficiencies:

High/Emergency
[ul]
[li]Resulting overpressure > 3x MAWP (if you ever find one of these, triple check your calculations, and make sure it is a single jeopardy scenario. I have yet to find a PSV this grossly undersized)[/li]
[/ul]

Medium
[ul]
[li]Resulting overpressure is 1.5-3x MAWP[/li]
[li]Inlet or outlet lines are smaller than PSV nominal body size (i.e., a 2J3 PSV has 1.5" inlet piping)[/li]
[li]Inlet line losses > 7% of set pressure, for a non-pilot PSV, at rated capacity[/li]
[/ul]

Low
[ul]
[li]Resulting overpressure < 1.5x MAWP[/li]
[li]Minor set pressure adjustments[/li]
[li]Inlet/outlet line losses are slightly above allowable per code/plant standards.[/li]
[li]Thermal relief valve issues (i.e., a 3/4"x1" PSV)[/li]
[/ul]

Most PSV defiencies are going to be in the "Low" priority and fixing them within 5 years or next turnaround is appropriate. "Medium" priority findings should be fixed within a year ideally. For a "High/Emergency" priority finding, a senior engineer once told me they would push to shut the refinery down and fix it immediately.

 

[georgeverghese]

Another parameter you may want to include is the likelihood of a PSV lifting when there is an overpressure scenario. For this, the following may influence :

a) Whether there is a lower level high pressure trip loop to isolate the source of the overpressure, and how dependable is this trip loop?

b) Whether you have sufficient process safety time (PST) to prevent the PSV lifting - what is the additional pressure rise that would occur upon activation of the instrumented trip loop, given the flow through the SDV as it tends to close.

In recent times, it has become good engineering practice (but not mandatory) to ensure that all penultimate trip loops are engineered such that the PST is adequate to keep the max accumulated pressure within the limits of activation of the higher level PSV.