I love the Line Break and Equipment Opening (LEO) SWP because it is one of the most hazardous tasks a worker will perform within a covered process AND the fact that there is no OSHA standard dictating how a business manages these risks. This lack of “OSHA Control” spoon-feeding us safety requirements causes most management groups to be lost in the risk and allows us (safety engineers/professionals) to use traditional risk controls without having to show “what does OSHA require“. Throughout my career, my facilities utilize some of the most toxic HAZMATs and with each company, I learned more and more about the hazards of breaking open the smallest of “tubing” associated with the highly toxic and flammable HHC/EHSs.
One controversial item that we always come across in LEO discussions/training is how we isolate a pipe run or piece of equipment before we “open” it. From my first day on the job as a safety engineer, using single valve isolation on pipe/equipment that had contained an EHS/HHC was NEVER allowed when the process was “live” on the other side of the valve. And trust me, when PSM came online and we began having to have a formal safety plan for each “break” we ran into this routinely!
In this article, I will explain the risks of using a single valve for LEO isolation, which oftentimes means hundreds of thousands of pounds of the HHC/EHS is on the other side of the valve and the process is LIVE (pressure, temps, etc.). BUCKLE UP buttercup as this is going to be a wild ride in real safety engineering!
Without question, LOTO plays a major role in LEO activities; achieving a Zero Energy State (ZES) within the pipe/equipment is a HUGE STEP in achieving a safe and successful opening. But when our process was built, the design engineers most likely did not build it with LEO in mind, so we are left with a process in which our ability to properly isolated and verify ZES is often in question. I wanted to state this up-front as this single fact is one of the reasons why LEO is such a hazardous task.
But do traditional LOTO practices suffice? Can we simply close and LO a single isolation valve and consider the system isolated at a ZES? Well, the answer is… “it depends”. Just like we have discussed before with LEO, wherein the process the opening is being made, the chemical(s) involved, the pressures and temps involved, and even the location of the worker(s) doing the LEO all come into play when considering the risk of the task. Of course, we hope the chemicals have been fully evacuated, the pressure is 0 psig, and the temp is less than 120F; but those are hopes and dreams and we do NOT perform LEO under hopes and dreams!
Imagine this scenario,
- a pump receives the HHC/EHS from a bulk tank and pumps the HHC/EHS to the next stage of the process.
- the pump needs maintenance, but between the pump’s inlet and the bulk tank is a single (1) 2″ isolation valve.
- the HHC/EHS in this scenario is Anhydrous Ammonia
- the bulk tank is at 85% capacity which is 306,000 pounds of NH3 stored as a liquified pressurized gas; there are two (2) identical tanks that are manifolded to two (2) pumps for a total of 715,000 pounds of NH3
- The pressure in each tank is @ 165 psig due to ambient conditions
- the back up pump, next to the one needing maintenance, has been placed into service and the process flows continue during your maintenance work
- If the pump(s) are aligned to the discharge manifold, know that we have flow/pressure on the pump outlet isolation valve as well (it is not deadheaded, but it could become deadheaded)
- The two (2) pumps can be aligned to either tank
Would you be comfortable working behind a single isolation valve under this setup?
What steps would you take to compensate for this scenario (i.e. single valve isolation with 700,000 pounds of NH3 stored as an LPG @ 165 psi)?
We have been dealt this hand and it is up to us to work through it! Here is my take on what I required under this situation…
1st, LESSEN the quantity available should something go wrong. If there is any space in the sister tank, move as much as we can over to it. Then, through proper planning, we use up as much of the NH3 in the one tank as we possibly can and then we valve it out. But do not get too comfortable, as there may be only one (1) valve in place to isolate the two tanks! I have seen this more than once in my career. When I was a plant Safety/PSM manager I had a unit manager that even went as far as bleeding the tank to a “scrubber” so as to ensure we got as much NH3 as we could out and the pressure in the tank was well below 10 psig. (NOT a ZES, but a hell of a lot better than liquid @ 165 psig!)
2nd, we will do this work behind a single valve THIS ONE TIME; so maintenance and the nested contractor are fabbing up the new inlet piping arrangement that will include a Double Block and Bleed arrangement on the new piping. And the same will be done on the outlet piping. This will allow for proper evacuation and positive isolation for future pump work.
3rd, once the single valves (inlet and outlet to the pump) are LO’ed we now much deal with the NH3 that was trapped between these closed valves. This task is also a SERIOUS RISK as we tend to have a lot of failure opportunities in this “homemade” evacuation system we are forced to devise. I have written about this several times, but I want to make clear – I LOVE DEADMAN VALVES and will ALWAYS require their use when “bleeding/venting” off an HHC/EHS. I am so adamant about the uses of these valves that I will oftentimes require the isolation plan to be arranged such that there is an engineered bleed/vent valve BETWEEN the two isolation valves. Yes, I am well aware that this may entail the portion of the process needing to be evacuated to grow substantially (i.e. we can’t use the pump’s discharge valve as there is no engineered bleed/vent valve and we will have to go downstream and find a bleed/vent and then past that use the next isolation valve!). I have had an engineering manager REQUIRE this to a point where the environmental manager got angry over the amount of “waste” that was going to be generated from this “simple LEO job”. Evacuating the HHC/EHS in a CONTROLLED fashion such that we lose very little of the HHC/EHS to atm is a big deal – CONTROL = LOW EXPOSURE!
4th, the PPE for the ENTIRE WORK SCOPE will be LEVEL B ensemble (e.g. Full chem suit with SCBA, chem boots and gloves – we may even go with a fully encapsulated B suit depending on the work location)
OR
the unit personnel will have the option of installing “BLANKS” on the inlet and outlet flanges at the pump. This decision will of course be dependent on any additional work they may find during their original scope of work. If things go as planned, they will most likely do their work in LEVEL B; however, if they find issues that will make this job last longer than ______ (fill in the blank based on your risk tolerance) then they may choose to use the blanks (a positive means of isolation) which will allow them to work with lesser PPE demands than that of LEVEL B. BUT PLEASE KNOW THIS – after the work is done, the REMOVAL of the blanks (both of them) will be managed as a LEO and will be done under the LEVEL B PPE.
Is any of this required by OSHA? Not one bit is in any standard; however, I challenge anyone to argue these measures are not necessary for a LEO under this process design. Could OSHA cite anyone not doing this – heck if I know, but I stopped doing safety to make OSHA happy some time back in the early 1990’s. This is how we bring value to the business… taking a risk that OSHA has not written a standard on and use our safety tools to get the risks down to a tolerable level for those involved.