I have been asked to explain the difference between actions an operator may take during a process leak and the actions that the same operator may take as an “emergency responder” to the leak.
First, take a look at a 2003 OSHA LOI in regards to OSHA’s position on Emergency Response. The letter is specific to an Anhydrous Ammonia process, but the rationale can be applied to any PSM covered process.
Bottom line is that there is a DISTINCT difference between the actions as “operator” takes as an operator and the actions he/she may take as an “emergency responder” and this would INCLUDE any DEFENSIVE ACTIONS he/she may take in their attempt to control the emergency. These actions are defined by two different OSHA standards as well, and I will use the different standards to explain the requirements and where the lines are drawn in the sand between an “operator” and when that operator becomes an “emergency responder”.
The specific question that was asked of me is “Can Ammonia operators use process controls located in the engine room to close valves, change set points of equipment, in order to manipulate the process to slow or stop a leak in a separate area? The answer is YES, but the actions the operator takes falls under 1910.119(f)(1)(i)(E) Emergency Operations (I am assuming that the process is covered under PSM and/or RMP). There are not a lot of refrigeration businesses that have “emergency operations” procedures, as most claim their ESD buttons are their “emergency shutdown procedures” and everyone evacuates the process when the system has a major leak. If the business is really going to have the operator close valves, change set points of instrumentation, and manipulate the process using process controls or manual valves, THESE ACTIONS HAVE TO BE in an OPERATING PROCEDURE that is annually certified and trained on at least every three years.
Let’s go through the process of an ammonia leak on an evaporator in a freezer. An employee sounds an alarm of the ammonia leak and the NH3 operator thinks he can stop the leak in the freezer by closing certain valves and manipulating set points on other equipment so that the flow of NH3 to the freezer is shut down. This is all well and good, as long as there is a defined operating procedure for the actions the operator will take. The procedure should define exactly what valves he/she is too close (e.g. using valve tag #’s is the most accurate way to list the valves) and what set points they are to revise and what the new set points are to be. MAKE SURE your units of measure MATCH exactly what the process controls readout (e.g. PSIG vs. PSI) so that personnel do not make data entry errors. This emergency operating procedure MUST STILL contain “consequences of deviation” and “steps to correct or avoid the deviation”. This is so if the matter gets worse as there may be something else causing the leak or their actions actually worsen the leak, they are instructed to “SHUTDOWN” the process using the emergency shutdown procedure or the engineered system shutdown. If their actions are successful in stopping the release, do not forget you just had a PSM/RMP incident that will need to be investigated per the standard(s) requirements.
Let’s for the sake of this discussion say that his/her actions did not stop the release and that entry into the freezer is needed. Not sure of a situation where this would occur, but for the sake of argument let’s assume the facility wants to send two operators into the freezer. At this point and time, the operator is no longer an operator but is now an “emergency responder” that falls under 1910.120(q) and all the requirements that section requires for a response to an uncontrolled release. As OSHA made clear in their LOI that I linked to above, there is a DISTINCT difference between an Emergency Action Plan (EAP) and an Emergency Response Plan (ERP). In this case, the ERP is a must have (see 1910.120(q)(2) in place to manage this response.
To summarize…the Operators emergency actions MUST BE documented in an SOP that is annually certified and at the point where they leave the controls and don specialized PPE to enter a hazard area in an attempt to control the release they are then guided by their Emergency Response Plan. In the PSM/RMP world, ALL ACTIONS that are taken by operators to control the process, especially those actions that are in response to a serious deviation such as a leak, MUST BE documented to prevent “freelancing”. These “emergency procedures” need to be SPECIFIC in nature and also include “consequences of deviations”, as well as “steps to correct or avoid “the deviation. In other words, how will the operator know if their actions are helping or worsening the leak situation? They need to know if their actions are not working to control the leak so that they can initiate EMERGENCY SHUTDOWN per the ESD procedure(s).
Note: the person who asked me to comment on this, also asked if a MOC would be needed to change the set points of process controls in this situation. The short answer is NO, AS LONG AS these changes were specifically spelled out in the “Emergency Procedure” and this procedure included the consequences and steps to correct or avoid the consequences of their actions taken under the operating procedure. A MOC WOULD BE NEEDED if the “emergency procedure” was not specific in what the operator was to change the set points too! These “emergency procedure” activities cannot be a “crap shoot”, but rather they MUST BE a clearly defined activity that has been closely analyzed so as to be sure it does not worsen the situation or lend itself to “freelancing” by personnel. The time during a leak is not to be “experimenting” by closing certain valves to see if it helps. We have to KNOW ahead of time and “emergency procedures” need to be reviewed with a BROAD AUDIENCE to ensure that actions being taken are SAFE and COMPLIANT, and most certainly they will get us the desired results of STOPPING the leak.