One phrase that drives me crazy while working with my friends in the Ammonia Refrigeration industry is,
“Come on, Bryan; we’re not a refinery or chemical plant; that is unnecessary here.”
For example, recently, we were asked to aid in an investigation into a reportable release from a PSM/RMP-covered process, and during our review of what happened, the facility failed to mention that their process was built with an Emergency Pressure Control Systems (EPCS), a term used by IIAR and many of the other codes and standards that pertain to mechanical refrigeration processes. These systems are now a requirement in most states and are intended to reduce releases of NH3 (as well as other refrigerants) to the atmosphere. Obviously, in this case, something did not work as planned. After a quick visit to the EPCS location, we found the required maintenance isolation valves on BOTH SIDES of the EPCS closed. Further investigation indicated that they had most likely been closed for many years, as the last time any work was done on the EPCS was eight years earlier (when we assumed the valves were closed and not re-opened).
A root cause failure that led to the release?
We can argue that on another rainy day, but I think most of us can agree that when we have a “safety system” installed, and it has been made INOPERABLE, and this played a role in the process safety event – this is a serious concern that needs to be investigated and understood.
NOTE: the state where this process is located REQUIRES an EPCS in the state Fire Code:
605.10 Emergency pressure control system.
Permanently installed refrigeration systems containing more than 6.6 pounds (3 kg) of flammable, toxic or highly toxic refrigerant or ammonia shall be provided with an emergency pressure control system in accordance with Sections 605.10.1 and 605.10.2.
NOTE: the 2015 IMC also requires a EPCS:
1105.9 Emergency pressure control system. Permanently installed refrigeration systems containing more than 6.6 pounds (3 kg) of flammable, toxic or highly toxic refrigerant or ammonia shall be provided with an emergency pressure control system in accordance with Section 606.10 of the International Fire Code.
(PLEASE TAKE NOTE that I used IFC 2018 in writing this article and the reference above to Section 606 of the IFC is outdated. The requirements are now found in Section 605 of the IFC)
In all fairness and full disclosure, ANSI/IIAR 2 – Standard for Safe Design of Closed-Circuit Ammonia Refrigeration Systems does NOT make an EPCS a “requirement”, as all of the information about these types of safety systems is located in Appendix I – Emergency Pressure Control Systems which is an “Informative” appendix to the standard. But I can tell you as soon as one of these systems are installed, whether for compliance with a state code or company standard, DESIGNING it, INSTALLING it, OPERATING it, and MAINTAINING it must be done per a RAGAGEP, of which ANSI/IIAR 2 – Standard for Safe Design of Closed-Circuit Ammonia Refrigeration Systems, 2014, Appendix I – Emergency Pressure Control Systems is a really good resource to use.
IIAR-2, 2014 Appendix I contains the following requirements (which are NOT mentioned in the IFC 605.10 by the way)
I.4.4 Isolation Valves
I.4.4.1 Each crossover valve should be provided with a stop valve on either side to allow isolation of the crossover valve for maintenance.
I.4.4.2 Isolation valves should be locked in the open position during normal operations.
It is these isolation valves, a vital necessity for maintenance, that MUST BE CAR-SEALED in the OPEN position in order to fully comply with our design, install, operational and maintenance RAGAGEP (IIAR-2). I am not going to try and connect this safety system to ASME Section VIII Appendix M-5 as if this safety system is a Pressure Relief System, but I will say this safety system is a “PRESSURE CONTROL SYSTEM” and it would not be a big step for OSHA/EPA to take. This EPCS is just one layer of protection before an RV(s) lift and its entire purpose is to act as an INDEPENDENT LAYER of PROTECTION to prevent a Loss of Primary Containment event.
These two (2) isolation valves on either side of the EPCS will cause a failure of a safety system, with a high probability that it will lead to an LOPC event (e.g., a consequence). Therefore, BOTH of these valves meet the criteria of being a “safety critical valve,” of which any deviation will increase process hazards. Thus we need to CONTROL these two (2) valves by some “official” means… Car Seal Program, which is laid out nicely in ASME VIII, Appendix M-5.
Had these two (2) valves been OPEN when the process had its pressure excursion, the EPCS would have prevented the LOPC event (i.e., the Reportable Release) and saved the facility a lot of headaches and the risk of an EPA inspection. Should the EPA decide to do an inspection and identify that the process had a designed safety system to aid in managing pressure excursions and that the facility had this safety system out of service without a MOC or any type of interim controls, things could turn ugly quickly.
The good news is that the facility is in the process of fully implementing a Car Seal Program. Interestingly enough, we found several other valves in their process that they felt (with some minor coaching) should be included in their program; all in all, they have 16 safety critical valves in their program. Having this car seal program in place and functioning when/if EPA/OSHA were to perform an inspection will go a long way toward showing “good faith”; in fact, next month I am at the Corporate Offices training engineers on Car Seal and how it should be part of all process safety management programs.
PLEASE SEE my other posts related to Car Seal Programs. As shown below, there is a FREE RESOURCE that anyone can obtain on how they should design, implement, and manage their Car Seal Program.