So this week my good friend “Jim” and I were talking about an “accepted practice” that has grown into an “industry practice” and how OSHA has definitely drawn the line on the “cord and plug” LOTO practice such that it does NOT include valves. This “practice” involves the concept of defining “exclusive control of an energy isolating device(s)” without the aid of a Lockout(s) or Tagout device(s). Last year, OSHA issued a letter (LOI) to a petrochemical company that asked the following two (2) questions: (I paraphrased the questions – see the full LOI at the end of this post for actual wording)
- Can I close a valve that goes to a pressure gauge and NOT lockout or tagout the isolation valve, since the valve is right next to the gauge that I will be changing?
- Can I close the inlet and outlet valve to a filter housing (or strainer), drain/vent the housing, all with valves right at the filter housing (e.g. arm’s reach) and change/clean the filter/strainer and not apply lockout or tagout devices to the isolation valves?
I will tell you upfront, I have lived this practice with BOTH of these exact scenarios and almost all of the companies I worked for allowed these tasks WITHOUT LOTO under some VERY STRICT and DEFINED circumstances. In fact, two of these facilities were VPP facilities and in one review this was looked at VERY HARD by the VPP assessors. We had Risk Assessments, written procedures, and VERY DEFINED TASKS where we allowed a SINGLE worker to do tasks (even LINE OPENING/EQUIPMENT OPENING task) without applying lockout devices to their isolation valve(s). We defined “exclusive control” as being
- a SINGLE worker who had BOTH
- arms-length AND
- 100% visual of the energy isolation device(s) protecting them
The task had a Risk Assessment done (e.g. a tad more than a basic JSA/JHA), a written procedure with annual training, PPE Hazard Assessment(s), Site Safe Work practice, and well-defined AND DOCUMENTED TASKS and circumstances as to when this practice was allowed. This “practice” was a spin-off from the “cord and plug” exception to LOTO (1910.147(a)(2)(iii)(A)).
However, as common as it is, OSHA does NOT permit this practice, as they made clear in their 2011 LOI. And although I have been a consultant since they “drew their line in the proverbial safety sand” and said NO, I will offer up an alternative argument as to how this “practice” became an “industry practice” which we now know violates 1910.147.
Here is what OSHA says… and the manner in which some of the companies I worked for viewed this practice and managed the risks…
(emphasis by me)
Changing a gage on a line. Scenario 1: A worker replaces the pressure gage device or other instrumentation on a process line or a piece of operating (running) equipment. The task involves three steps.
First, the device being replaced is isolated from its energy source by closing a valve.
Second, the pressure in the device is bled, usually by slightly loosening a connection (fitting) to bleed off any trapped fluid and pressure.
Third, the device is removed and replaced.
You stated that this work task can be generally completed in a short time period and is always WITHIN ARM’S REACH of the worker performing the task. You believe the worker has TOTAL AND EXCLUSIVE CONTROL OF THE VALVE AT ALL TIMES so as to prevent the UNEXPECTED RELEASE OF HAZARDOUS ENERGY. In the event that the worker leaves the work area, the valve is locked out and tagged out (LOTO) per your Energy Isolation Policy. Also, per your Energy Isolation Policy, other personnel are made aware when this task is being performed and personnel are trained and periodically audited to ensure that they never leave the job without proper LOTO. During your telephone communication with a representative of my staff, you compared this to the exclusive control provision for CORD- AND PLUG-CONNECTED ELECTRIC EQUIPMENT.
Question: Does the replacement of the pressure gage device or other instrumentation on a process line or a unit of equipment while that line or equipment is running REQUIRE lockout and tagging of the isolation valve?
Response: YES, with limited exceptions, the §1910.147 standard requires control of the release of stored energy via lockout or via tagout where the energy isolating device is not capable of being locked out. Unless the requirements of 1910.147, including written procedures and the application of locks or tags, or both; are followed, the PROXIMITY OF THE WORKER TO THE VALVE IS NOT A SUFFICIENT MEANS FOR THE CONTROL OF HAZARDOUS ENERGY.
You stated that the time period to complete the task is usually short and that you believe the worker, who is in arm’s reach, has total control of the valve at all times. In the scenario that you pose, it could happen that the worker is PULLED FROM OR LEAVES THE WORK AREA FOR AN EMERGENCY OR OTHER URGENT NEED, without locking out and tagging out the device, per your Energy Isolation Policy.
Regarding your belief that the operator has EXCLUSIVE CONTROL of the valve similar to that permitted by 29 CFR §1910.147(a)(2)(iii)(A), this PARTICULAR EXCEPTION IS LIMITED BY ITS OWN TERMS TO CORD- AND PLUG-CONNECTED ELECTRIC EQUIPMENT. OSHA also considered whether the other minor servicing and maintenance exceptions under §1910.147(a)(2) applies and determined that the EXCEPTIONS ARE NOT APPLICABLE TO THE SCENARIO you described.
Additionally, the standard requires that all sources of hazardous energy be controlled. If the pressure gage device (or other instrumentation) has another source of energy, such as an electrical connection, that energy source would also require hazardous energy control according to §1910.333(b), which also mandates written procedures as well as the application of locks and tags.
Cleaning/Changing a filter/strainer (NOTE: strainer is added by me to be sure we understand the full application!)
Scenario 2: A worker CLEANS OR REPLACES THE FILTER on a pumping system. Filters are generally set up in parallel (operating and standby) with isolation valves upstream and downstream of the filter and a drain valve on the filter unit. Alternatively, the piping to each filter can be set up (arranged) such that a duplex valve assembly (two ball valves in an integral housing with a common valve operating handle) can isolate an individual filter by the operation (turning) of a single handle.
To clean or replace the filter, the worker
CLOSES THE UPSTREAM AND DOWNSTREAM VALVES of the filter unit to be worked on or isolates the unit via the duplex valve assembly.
He or she then RELIEVES THE PRESSURE BY SLIGHTLY LOOSENING A CONNECTION OR SLIGHTLY OPENING THE DRAIN VALVE, then fully opens the valves, drains the filter housing, and
opens the housing to clean or replace the filter
Upon completion, the worker returns the filter to service.
Similar to Scenario 1 above, the task takes a few minutes to complete and the worker has exclusive control of both isolation valves, which are WITHIN ARM’S REACH at all times.
Question: Is it permissible to clean or replace a filter unit WITHOUT LOCKING OUT OR TAGGING OF THE LOCAL UPSTREAM AND DOWNSTREAM ISOLATION VALVES?
Response: NO. Please refer to the rationale given in the response to the first scenario.
So how was I trained on this concept of exclusive control, which is now known to NOT be an accepted practice by OSHA? It was very methodical…
This concept was allowed on non- and/or “LOW-hazardous material systems”. I NEVER saw this practice applied on a system containing a Highly Hazardous Chemical or Extremely Hazardous substance; however, we would permit this on “hazardous materials” that had:
- LOW Vapor Pressures (e.g. <85 mm Hg)
- LOW degree of harm (not an ACUTE inhalation hazard or contact hazard that would incapacitate the worker)
- SMALL quantity systems (e.g., should something go wrong, the LOPC event would be limited in scale based on the quantity of the HAZMAT available
- The STATE the HAZMAT was in (e.g., Vapor/Gas)
- The physical properties of the process (temps, the pressures involved, etc.)
The thought behind this was that if there was some type of uncontrolled LOPC event, the potential injury to the worker would not be one where they would become incapacitated, severely injured, AND the event would NOT extend beyond the local area where the work took place. We also designed the areas where these tasks were performed to have adequate MITIGATION devices such as TWO means of EGRESS, a Safety Shower/Eye Wash within 10-20′, well-lit areas, etc.
We considered other factors on which we could apply this practice, such as the NFPA 704 degree of hazard(s) and the physical state of the material was in. If the material carried a rating of 3 or 4 in either of the Flammable, Health or Reactivity, this practice was NOT even considered AND we performed line/equipment opening permit to these systems, EVEN though they may NOT be PSM/RMP covered process/systems due to the quantitates in the process and this, of course, utilized FULL LOTO protection.
NOTE: the NFPA degree of hazard 3/4 criteria originated from the IFC practice of segregating hazardous materials that required additional design/protection safeguards.
The task(s) we allowed this practice on was also ROUTINE, REPETITIVE, and INTEGRAL; however, the changing of a gage was NOT considered part of “normal operations”, but the “filter work” was considered part of “normal operations”. But as we know from other OSHA guidance, any “disassembly” involved with the task(s) would EXCLUDE these tasks from being considered “minor servicing” and both of the tasks OSHA is talking about in their 2011 LOI does involve “disassembly”.
Our view of “exclusive control” was defined as “in-view” AND “arms-reach” of the energy isolating device(s) AND the work was being done by a SINGLE worker with NO OTHER worker having access to these isolation devices. I can not stress these criteria enough and I MUST point out, again, that OSHA does NOT recognize this as “exclusive control”; however, OSHA does allow this concept on ELECTRICAL CORD-AND-PLUG equipment when the equipment goes to a ZES once the equipment is unplugged AND the worker doing the work has EXCLUSIVE CONTROL of the plug.
One of the reasons why we did NOT permit this practice on “hazardous materials” with higher vapor pressures and/or lower acute toxicity limits is that it has happened on more than one occasion where the worker doing the work actually, accidentally, OPENED one of their isolation valves. I have personally been involved incidents where the worker had to move a hose from around the filter housing and in doing so he did not roll up the hose, he just slid it over to the side. When he did this, he moved the hose so that it was against one of his 1/4 turn isolation valves. Later in the task, he “kicked” the hose further out of the way and when he did, the hose pulled on the 1/4 turn valve handle and CRACKED OPENED the valve. Luckily he was NOT over the filter housing at the time, as the caustic came out of the empty filter housing. So as we can see, “exclusive control” is intended to extend BEYOND just another person defeating our isolation devices; sometimes the worker can defeat the device(s) themselves by accident! A very important aspect of controlling hazardous energy, especially when the “hazardous energy” is one that is a “hazardous materials” with properties that can cause immediate and serious hazards in the event there is an LOPC incident.
I would like to add that the last company I worked for in 2001 ALLOWED this practice; but NOT when the isolation valves were 1/4 turn valves, as they too had experienced bad incidents with 1/4 turn valves releasing hazardous materials/energy. So any task where this practice was considered, had to have some type of valve OTHER than a 1/4 turn valve (e.g. engineering control!!!!). I would also like to note the company where I had my caustic incident also changed their Risk Assessment and now requires some other type of valve other than a 1/4 turn valve for this practice to be applied.
But here is more of our rationale for permitting this “exclusive control practice” in spite of OSHA’s newly stated position…
As I have written about before, in a chemical process, we have BOTH ROUTINE and NON-ROUTINE process openings – it is just a fact of life in the chemical industry. We view ALL process openings as “opening the process” and we RESPECT them ALL, but anyone who has worked in this environment recognizes the varying degrees of risks associated with these openings. For example, bulk transfers of hazardous materials (via a tank truck or railcar) at a facility is WITHOUT A DOUBT one of the most hazardous tasks that take place at the facility; however,
- when we remove the unloading hose cap from the connection to connect the transfer hose we ensure the unloading valves are CLOSED and there is NO pressure in the line – BUT I have never heard of anyone issuing a line break/process opening permit and/or requiring the unloading valves be LOTO’ed CLOSED before we remove the cap (mainly because this task would fall into “normal operations” within 1910.147); but none the less, we are OPENING the process and thus we have a risk of a release of hazardous energy yet we are not LOTOing any valves.
or
- when we disconnect the hose after the transfer (when the vast majority of exposures occur) we do not issue a line break/process opening permit and we are NOT locking out the unloading valves we closed before disconnecting the hose.
This is because we view these “openings” as a ROUTINE opening and we have taken our time and reviewed and analyzed the risks VERY CLOSELY and we have used our hierarchy of controls (engineering, administrative, PPE) to control these exposure risks.
Another example of a ROUTINE opening is “catching samples” from the process and again we have reviewed and analyzed the risks VERY CLOSELY and we have used our hierarchy of controls (engineering, administrative, PPE) to control these risks without issuing a permit for the task or including LOTO practices.
We logically moved this thought process and practice into other ROUTINE tasks using our same methodology; hence the idea of cleaning/changing filters or strainers (and as OSHA speaks to – changing instrumentation).
We did everything we would normally do as if this was a LOTO task, other than applying the lockout device(s) to the energy control device(s) as we “assumed exclusive control” when the isolation device(s) were within arm’s reach and line of sight.
I should also note that we treated water as a Hazardous Material when it was 140ºF or hotter and/or when it was over 50 psi in a pipe or hose over 1/2″ diameter.
2019 UPDATE: Here is a picture of what OSHA is talking about: NOTE: the red circle shows the valve that would be closed and the yellow square shows the gage that would be removed/replaced. CAUTION: the picture is of an NH3 refrigeration process and I would NEVER advocate/allow using this practice on an NH3 process, but I made these pics for a discussion the refrigeration industry is having on this matter; so they are for VISUAL AID purposes ONLY and IN NO WAY intended to show how this practice could be used on a system containing NH3!  Here is a picture showing the same setup, but how the strainer on the bottom of the valve would be opened and cleaned. The yellow circles show the valves which could be closed and the red circle shows where the process would be OPENED.  PLEASE NOTE that although these tasks are OUTSIDE OSHA’s definition of “minor servicing“, it was viewed as MINOR in our world of risks. This practice would have NEVER EVER been allowed for changing out a valve such as the one shown above in the pictures (regardless of size); FULL LOTO would have been applied! As can be seen in the pictures, a worker who is doing either task would have BOTH “visual” AND “arms-reach” to their isolation device(s) while doing this work. Placing a lockout device(s) in either scenario would be FULLY compliant with 1910.147; but what hazard is being created if locks/tags are not placed on the valve(s)? |
Some will argue an absurd situation, such as what if one of the valves leak by? My response…. how does a lock or tag on the closed valve help a leaking valve?
But a leaking ISOLATION DEVICE MUST be considered in the application of this practice!!! What if one of these valves leak by while I am doing the work; what are the hazards of the material in the pipe (chemical hazards, temperature, pressure, etc.)? But these deviations and hazards are why we WEAR FULL ENSEMBLE of PPE even when LOTO has been applied and we oftentimes require MORE than one (1) valve be used as the isolation means!
As you can see in the picture above, the valves in use do NOT even have handles on them, a common practice in ammonia refrigeration, so accidentally/inadvertently cracking one of them open (as in my hose and filter housing incident) would be near impossible. We would have also barricaded the work area (required via the procedure and a site written program on “barricading”) with RED and BLACK DANGER TAPE, which was a site SWP and thus NO ONE was allowed under the DANGER tape without first checking in with the owner of the tape (e.g. tape was tagged with a name, phone #, and hazard). So the probability of someone walking up with a valve wrench and opening one of these valves while I stand there doing the work…. even I am comfortable not putting my locks on the valves and most would consider me a “safety fanatic”.
So although OSHA has now clearly stated they do NOT accept the practice, it was done for decades and in many cases is still being done today. My hope is that the practice is VERY LIMITED in its application and is done with PROPER and ADEQUATE layers of protection should something go wrong, but the concern of someone walking up and opening a valve that is within my view AND arms-reach is something that is very low risk and I manage my safety via thorough risk analysis. I would NEVER suggest you or your facility violate OSHA compliance but this is a reality check – how much more time/effort does it take to apply LOTO devices? Sound logical? I’ll have to let you come to your own conclusions.
Here is the OSHA LOI: https://www.osha.gov/laws-regs/standardinterpretations/2011-08-30