As I have stated already this week, I am a HAZOP kind of guy. I have used just about every methodology under the sun over the past 25+ years, and I have come to a personal conclusion that in almost every occasion, a HAZOP would have been the best methodology to utilize for a correct Process Hazard Analysis. My next favorite tool is to use a Failure-Modes-Effect-Analysis (FMEA) on specific pieces of equipment within a process, like say a compressor. Heck, I have even suggested to a couple of compressor manufacturers that a “competitive advantage” would be to have an FMEA done on your compressor and provide this analysis to customers who would be using the compressor in a process handling a highly hazardous chemical/extremely hazardous substance. Imagine the time savings across all customers and the marketing gold mine by being able to say our “compressor comes with a certified FMEA” (I know – I hate to use the word certified, but you know the marketing people will!!!)
But even when we choose a What-if or What-if Checklist as our methodology to conduct our PHA, there are some boundaries in which these tools NEVER INTEND to cover! One such “what-if” scenario that drives me CRAZY is to ask:
What if the Relief Valve has an incorrect setpoint?
An absolutely legitimate question to be asking in terms of process safety, but it is NOT a question I would ask in a PHA? This error would be covered in an array of other PHA activities…
Let me state this up front – this is NOT a slam against anyone’s checklist as I understand why the question exists in the checklist. I am just saying that it is not a typical PHA question.
Using the HIGH-PRESSURE scenario from my earlier postings, we got all the way down to our final layer of protection… the RV. However, as I stated in earlier postings, we can NOT count the RV (or any safeguard listed) until we have CONFIRMATION that the PSI is accurate and that the safeguard is included in an inspection/testing program.
As for the PSI being accurate, the RV documentation must include the design basis and sizing documentation. The design basis will cover the type and set point for that specific RV. As I noted in our scenario, the vessel has two (2) RVs: the first RV is set at 85% of the MAWP and that RV discharged to a containment/destruction system. The other RV was set at the MAWP and discharges to the atm at a safe location. BOTH of these RVs would have DIFFERENT design basis and maybe even sizing.
But in our HIGH-PRESSURE scenario, we ask what could cause this scenario to begin and then was ask ourselves what we have in place to PREVENT-PROTECT-MITIGATE. The RV lifting is actually our POINT OF RELEASE, so it is NOT intended to PREVENT the HIGH-PRESSURE event, nor is it INTENDED to protect assets once it has lifted. (NOTE: it is designed to protect the vessel from catastrophic failure, so it is more of a “mitigating” layer – rather than a protection device in the realm of our PHA scenario). So when we get to the point of listing our RV(s) as a safeguard, we have to ask the following questions:
- Are they documented in the PSI?
- Are they included in an inspection/testing program and does that program require ANNUAL visual inspection and does that annual visual include a written procedure and training?
- Are their procedures for changing the RVs and is there a “quality assurance” aspect to these procedures to ensure human errors are controlled?
Finally, our true last line of defense against having RV’s with the wrong set points in service is our Pre-Start Up Safety Review (PSSR). But we must be able to show that the PSSR requires someone with the proper skills/training, is needed to verify the RV nameplate with the Vessel nameplate AND the PSI Design basis during the actual PSSR.
With all of these controls in place, asking the What-If question about the RV having an incorrect setpoint is not a PHA scenario. It is part of the PSI evaluation and PSSR and controlled through proper MI procedures and Quality Assurance practices.