UPDATED on 2/6/19, based on feedback from my friend and Refrigeration PSM Extrodinare Brian Chapin over at RCE (A former Partner in Safety). I have shared his analysis and my rebuttal/agreement…
By a long shot, these are the top three (3) issues found with Ammonia Refrigeration RMPs:
- The NH3 is “Liquified by Pressure” – not by refrigeration!
- Process Purgers are NOT the same as “Purge System” as listed in the Process Controls section of the RMP
- If you’re not on the Coast – your process was most likely not built to withstand as Cat 1-5 Hurricane! The States of NE, MO, and IA have not seen a Hurricane since before the ice age. LOL
Here is a brief explanation of how these three (3) should be listed in your RMP:
Gas Liquidfied by Pressure
Although a mechanical “refrigeration” process is “refrigerating”, the ammonia within in it is NOT refrigerated; it is merely the refrigerant. We see “Gas liquified by refrigeration” at bulk fertilizer facilities where NH3 is stored in large atmospheric storage tanks at temps around -32F. There is actually a refrigeration process that is pulling the NH3 vapors from the top of these tanks and compressing them and then condensing them back into a liquid, which is then sent back to the storage tank. The NH3 is kept as a liquid because of the refrigeration. In a mechanical refrigeration system like we see in cold storage, food processing plants, and even petro-chemical the ammonia is stored as a “Gas liquified by pressure”.
We have to understand that this is much more than a reporting error, as even as early as last year we came across RMPs that were going on 20 years old that still assumed a “refrigerated liquid” in their Off-Site Consequence Analyzes!!! So rather than having a WCS of 3-4 miles, they claimed less than 1 mile and a huge difference in population impacts. So understanding how our HHC/EHS is stored and handled is ABSOLUTELY CRITICAL in understanding our process hazards.
Brian’s Challenge: When a refrigerated vessel (such as an LTR) is the worst case how is it not liquid by refrigeration? It’s actually in a vacuum.
My Response: Guessing one could make the argument that it is indeed a “refrigerated liquid”, if the Worst-Case Scenario (WCS) was from this vessel AND the WCS involved a passive mitigation device in the way of a dike or some type of engineered secondary containment would allow the liquid NH3 to form a pool deeper than 1 cm during the WCS. Then the WCS could be assessed as a “refrigerated liquid”. How many of these scenarios have you come across? 95% of the ones I come across, the WCS is the receiver (liquified by pressure), but I am guessing I may need to clarify if you tell me you have seen a WCS from refrigeration process that had a secondary containment system around the LTR (or engine room).
We agreed that neither of us has come across a mechanical refrigeration system which utilizes NH3 as its refrigerant AND is equipped with a secondary containment system around the process vessels (or even the mechanical/engine room). So here is the bottom line… we may indeed have a vessel with “refrigerated liquid ammonia” because the vessel operates under a vacuum; however, in order for us to claim our WCS involves a “Gas liquified by refrigeration” we have to also meet the 2nd part of the criteria…
For a liquid or a gas liquified by refrigeration where the released refrigerated liquid forms a pool deeper than 1 cm, the release duration should be the time required for a pool formed by the released substance to completely vaporize.
This 1 cm depth is vitally important in how the NH3 will “vaporize”. A “pool” of liquid ammonia (at least 1 cm deep or deeper) will actually remove the heat from the surface containing it and thus cause the liquid to vaporize at a SLOWER rate thus lessening the WCS considerably vs. assuming all the liquid will flash off in 10-minutes. Without some type of engineered secondary containment that has been put in place, for this reason, a facility should NEVER claim “a gas liquified by refrigeration” from a mechanical refrigeration process using NH3 as its refrigerant.
Of course, this is just my professional opinion – what do you say? Let me know by e-mail or don’t be bashful challenging me in our discussion groups on FB, LI, Twitter, etc.
What is a “Purge System” as defined by EPA’s RMP?
EPA defines it as:
A system that replaces the atmosphere in a container with an inert substance to prevent the formations of an explosive mixture.
These systems are commonly found in flammable processes and they typically involve the use of Nitrogen (i.e. inert substance). These safety systems are NOTHING like the “auto purgers” found on many mechanical refrigeration processes using NH3.
Hazards Identified
This one is certainly not specific to ammonia refrigeration, but it seems almost all of the RMPs we have worked on or come across list Earthquakes, Floods, Tornadoes, and Hurricanes as major hazards identified in their RMP. When should these be listed? The RMP E-Submit User’s Manual (Febuary 2018) states the following:
(emphasis by me)
12. Earthquake – Report earthquakes as a major hazard ONLY IF THEY OCCUR or are likely to occur at your site such that YOU PLAN AND DESIGN FOR THEM.
13. Floods (Flood Plain) – Report floods as a major hazard ONLY IF THEY OCCUR or are likely to occur at your site such that YOU PLAN AND DESIGN FOR THEM.
14. Tornadoes – Report tornadoes as a major hazard ONLY IF THEY OCCUR or are likely to occur at your site such that YOU PLAN AND DESIGN FOR THEM.
15. Hurricanes – Report hurricanes as a major hazard ONLY IF THEY OCCUR or are likely to occur at your site such that YOU PLAN AND DESIGN FOR THEM.
Some facilities may actually be in a seismic zone and have to incorporate this hazard into their design – thus we would then list “earthquakes” as a hazard in our RMP, BUT ONLY THEN!
Some facilities may actually be in a hurricane zone and may have incorporated this hazard into their design – thus we would then list “hurricanes” as a hazard in our RMP, BUT ONLY THEN!
I hope no one has a plant in a “flood plain”; however, we have a number of clients who built a plant decades ago and at the time the property was not listed in a flood plain; however, FEMA recently updated their “flood plain” areas and now the plant(s) sit in a recognized flood plain. A GOOD PHA will identify this in the “facility siting” as a flood can pose SERIOUS hazards to storage vessels, even pressure vessels!!! So making sure your vessels are high enough to withstand some level of flooding is absolutely critical.
Lastly, anyone working with their RMP, even when you have someone like me to prepare the RMP, should download a copy of EPA’s RMP E-Submit User’s Manual (Febuary 2018) and READ IT (cover-to-cover).