Technicians would not bat an eyelid at applying lockout and tagout to an electrical main before opening a switchboard. Yet put the same technician in front of a manual ammonia valve and the accepted control may be a tag, a ribbon or someone saying, “Don’t touch that.”
That inconsistency has bothered me for years.
An ammonia refrigeration valve can control pressure, toxic inventory, cryogenic liquid and the movement of energy from a much larger connected system. Unintended operation can expose several people at once. Welders. Pipefitters. Operators. Anyone working on or near the line.
Why, then, is physical lockout still treated as unusual?
A tag communicates intent. A lock creates physical restraint. Verification proves the condition.
A lesson I have never forgotten
More than twenty years ago, I watched a highly skilled technician, my mentor in fact, open a valve that formed part of an isolation during a repair. The action released ammonia and gassed out the welders and those of us working nearby.
No malicious act. No reckless outsider. A competent and experienced person made one movement, believing it was the right one, and the isolation depended too heavily on memory, communication and valve position.
A physical lock would have introduced a deliberate barrier. To operate that valve, someone would first need to confront the lock, identify its owner and understand why it was there. Intent would be required to defeat it.
That is the value of LOTO. Not perfection. An additional layer of sufficiency between a human action and a serious release.
Why valve LOTO is often missing
The wrong mental model
Many people picture a large, expensive handwheel cover when valve lockout is mentioned. If that device does not fit, they conclude that the valve cannot be locked.
Custom has become the control
Ribbons are widely used in parts of the United States to indicate a leak or valve position. In Australia and New Zealand, a DO NOT OPERATE tag is common. Both communicate. Neither physically prevents operation.
The equipment is somewhere else
The lockout kit may be in a workshop, supervisor’s office or service vehicle. The procedure may be buried in a folder. Every extra step creates another opportunity for the control to be skipped.
Routine work feels harmless
Technicians may complete the same pump-down, oil drain or valve replacement many times without incident. Repetition can make exposure familiar. It does not remove the hazardous energy.
The adjustable cable changes the conversation
Many technicians are unaware that adjustable cable lockout systems can secure a significant proportion of the handwheel and lever-operated valves found in industrial refrigeration plants.
One device can often secure several valve types. The cable can pass through a handwheel, around adjacent pipework or through another stable feature so the valve cannot be operated without removing or defeating the device. It can also secure multiple related isolation points where the procedure and risk assessment allow that arrangement.
Is it perfect? No.
Is it suitable for every valve? No.
Does it significantly reduce the barrier to physical valve lockout? Absolutely.
The lockout arrangement still needs to be fit for purpose. The cable must not damage small-bore connections, instruments or insulation. It must not create a trip or entanglement hazard. The anchor point must be stable, and the device must genuinely prevent operation rather than merely make the valve look secured.
Ribbons and tags still have a role
Ribbons can indicate leakage, valve status or a special operating condition. DO NOT OPERATE tags communicate ownership and instruction. Both are useful visual controls.
But neither should be mistaken for physical restraint where lockout is reasonably practicable and necessary to control the risk.
Even rope has value as a conspicuous restraint. During my time at sea, we used it to tie several handwheels together in a visible lattice. Was it perfect? Again, no. Yet it made unintended operation more difficult and made the isolation impossible to overlook.
The point is not that rope should replace an engineered lockout device. It is that physical restraint changes behaviour and requires deliberate action to overcome. A purpose-designed, identifiable and lockable device does that job more reliably.
LOTO is only one part of ammonia isolation
This matters. Locking a valve does not establish a zero energy state.
An isolated ammonia line may still contain liquid, vapour, oil carrying dissolved refrigerant, trapped pressure or a blocked pocket that cannot be seen on the nearest gauge. A closed upstream valve may leak through. An automatic valve may reopen. A pressure gauge may be blocked, isolated or reading only one part of the system.
For line opening and intrusive work, LOTO must sit inside a complete isolation, evacuation and verification process.
Define the isolation boundary
Use the current P&ID, valve schedule and a physical line trace. Identify every source of pressure, liquid, vapour, automatic operation, backflow and trapped refrigerant that could affect the work.
Remove the inventory
Pump down, transfer or evacuate the section using a planned method. Closing a valve does not remove the ammonia, pressure or stored energy already inside the isolated section.
Isolate and secure
Operate the identified energy-isolating devices, apply personal or group locks and tags, and secure any control functions that could reopen a valve or re-energise the system.
Control trapped liquid
Identify locations where liquid refrigerant could be trapped between closed valves. Provide a safe transfer, relief or controlled pressure-management method before thermal expansion becomes another hazard.
Dissipate residual energy
Ventilate or purge the isolated section using an approved method and safe discharge location. Treat oil, frost, blocked connections and low points as possible stores of residual ammonia.
Prove the condition
Verify nil pressure using a suitable point and instrument, then verify the atmosphere with an appropriate gas detector. A gauge reading of zero alone does not prove that the system is safe to open.
Maintain control during the work
Keep locks under the control of the people exposed. Use a group lockbox where several workers or contractors are involved, and manage shift changes through a formal handover.
Return deliberately
Inspect the work, account for people and tools, replace safeguards, remove locks through the authorised process, and restore the system in a controlled sequence.
When a lock is not enough
The consequence of isolation failure determines the strength of the boundary required.
For higher-risk work, a single closed and locked valve may not provide sufficient separation from a live ammonia inventory. The task may require double block and bleed, a removed spool, a blind or blanking plate, disconnection from the energy source, or another form of positive isolation.
The decision should consider the possible release quantity, system pressure, valve integrity, work duration, line size, nearby people, escape conditions and whether hot work or another ignition source is involved.
Physical lockout remains essential, but it cannot compensate for a weak isolation design.
Put the control where the work happens
I am a massive believer in storing resources and administrative controls at the source of the hazard.
Call it Human and Organisational Performance. Call it Kaizen. Call it good work design. The label matters less than the outcome: make the safe action obvious, available and easier to complete correctly.
At a recurring isolation point, that can mean:
- a plant-specific isolation and pump-down procedure at the equipment;
- a clearly identified LOTO board with cable devices, locks, tags and lockout hasps;
- a current valve diagram or isolation map;
- defined verification and test points;
- the correct respiratory and other task-specific protective equipment nearby and maintained; and
- a simple way to report a valve that cannot be secured properly.
When the procedure and equipment are at the machine, distance and inconvenience stop being excuses. Better still, the arrangement becomes a visual cue. It reminds the worker that valve isolation is hazardous energy control, not merely an operating step.
Design for lockout
We should also stop expecting technicians to improvise around valves that were never selected or installed with isolation in mind.
New plant and modification projects should consider lockable handles, latch points, accessible double-isolation arrangements, safe bleed and verification points, clear valve identification and space to apply devices while wearing gloves.
Existing plants can be improved progressively. Survey the valves used for maintenance isolation. Trial suitable devices. Identify the exceptions. Engineer permanent lockout provisions where a cable cannot provide dependable restraint.
Small improvement. Significant reduction in reliance on memory.
People are part of the system
Workers do not ignore controls in a vacuum. They respond to the equipment, procedures, time pressure, local customs and tools available to them.
If the only valve lockout device costs hundreds of dollars, fits very few valves and is locked in an office, we should not be surprised when tags become the normal practice. If an adjustable cable device, personal locks and the correct procedure are mounted at the isolation point, the safe action becomes far more achievable.
That does not remove accountability. It gives competent people a system they can actually use.
The practical challenge
Lock it. Tag it. Prove it.
Walk your ammonia plant and look at the valves used for intrusive maintenance. Can they be physically secured? Are suitable devices available at the work location? Does the procedure remove inventory and verify the condition, or does it stop at “close the valve”?
Review your ammonia isolation controlsAdapted and expanded from Pádraic Durham’s original LinkedIn post. View the original on LinkedIn ↗