Periodic Review of Unchanged Relief Systems

[Latexman]

Most of our processes get changed over time. More capacity is needed, new chemicals come along, process condition changes are recommended by R&D, etc.

In parallel, relief calculation methods improve and standards change.

The wheels of progress keep turning on all fronts as time goes on. Some turn fast and others turn slow.

When we have a process change, we review existing reliefs and design new reliefs, if needed, using the best relief sizing technology we have at that time.

I would guestimate that 80% of the reliefs in our plants get a review because of a change every 5-10 years. Our processes are batch with hundreds of products. Many new products get introduced every year. Things may be different in a commodity chemical plant.

Is there a requirement, either regulatory or your company policy, to periodically review unchanged relief systems? How frequently? Must relief sizing methodology be updated at that time, or is the old methodology grandfathered?

Good luck,
Latexman

To a ChE, the glass is always full - 1/2 air and 1/2 water.

 

[shvet2008]

post-USSR
Being done first time calculation is never done again. Many of PSV are "as is" and have no calculation record.

 

[don1980]

Some companies do have a policy of reviewing and revalidating PSVs every 5 yrs, but my understanding of OSHA PSM is that that is not a specific requirement. OSHA does mandate that we review and revalidate the Process Safety Analysis, but I don't see anything in the OSHA PSM regulation indicating that that means documenting a revalidation of each PSV every 5 yrs.

There are a lot of engineering consulting companies that make a lot of their money performing PSV revalidations. Naturally, they'll say that PSA revalidation means revalidating each PSV. That's an interpretation that they're free to make, but that's not the only interpretation of the reg. Many companies have a practice similar to yours. PSV revalidation occurs naturally due to plant changes and re-designs that automatically cause a re-analysis of the relief devices within the scope of those projects.

 

[jari001]

My site does the PSA review every 5 years and what the senior engineers tell me is they will do a detailed review of the sizing calcs and standards at that time and determine if significant enough changes have occured in any guidances (determined via intuition/experience rather than a checksheet) to warrant looking at any PSVs individually. My site is in the aseptic vaccines processing game, so processes and products don't change to the extent that regular reviews of PSVs are a byproduct of doing business.

And since I'm in pharma, any deep dive into PSVs would definitely go to a contractor . The couple times a deep dive has been done, the site's engineers did spell out which PSVs in partiular needed evaluation so they could head off the "review 'em all" speech from the contractor.

 

[gocougs1]

Every 15 years is the general consensus for a full unit-wide PSV re-evaluation for the refinery I work for. There is no real basis for this other than what people think is appropriate. In a 15 year period there will typically be significant changes to the code and larger plant-wide changes may occur. In the end I think it really depends on how thoroughly the existing PSVs were evaluated. Also, for PSVs that are well documented with clear assumptions and calculation results, it will be much less effort to re-validate them than PSVs that have zero documentation/basis.

 

[EmmanuelTop]

Latex, see attached document from HSE UK (Health & Safety Executive), particularly Appendix 2 - Process Plant Operation within Safe Limits, and Appendix 4 - Relief, Blowdown, Vent and Flare Systems:

http://www.hse.gov.uk/offshore/ed-loss-of-containment.pdf

Reviews of safe operating limits are recommended every 2-3 years, even if there are no changes to the system - reviews not specifically related to relief systems only but for all elements that are a part of the Loss of Containment risk management system, or technical integrity management system. I don't know if PRV/PSV reviews and the corresponding sizing methodologies are defined as a separate PSM requirement, but this is something that an efficient PSM system would definitely capture - the case(s) when the changes in calculation/sizing principles are significant and may lead to the finding that some of the overpressure protection devices are ineffective. Your question is actually a great litmus that can show how effective and thorough is PSM system in any operating entity.



Dejan IVANOVIC
Process Engineer, MSChE

 

[David272]

In a batch chemical plant environment, the emergency pressure relief system may be reviewed when the materials made on the plant are changed. I did this for one plant, but took the approach of demonstrating that the solvents used in the new process had lower molar latent heats than the previous solvents and that the vessels were not over-filled. The system looked right and I knew that the people who had designed the original system were very highly respected. So effectively I was assuming that the original assessment that was done to an earlier code was still effectively valid for that plant. I have also come across a similar approach at a design house, which keeps copies of the older codes so that it can check that the designs on old plant were correctly implemented at the time that the plant was originally designed.