As an Amazon Associate, Akermin earns from qualifying purchases. Product links on this page are affiliate links.
The gas changes one thing about detection and several things about everything else. A 2L classification means the refrigerant can reach a flammable concentration in a small room, so the work changes shape around that fact: charge size against room volume, ventilation before and during the job, ignition sources treated as a planning item, a nitrogen pressure test instead of a trace charge, recovery and vacuum equipment specified for a flammable refrigerant, and fixed monitoring mounted low because the vapour is denser than air. The sniffer in your bag is the least of it, and swapping it for a newer one changes almost none of the above.
A flammability class is a set of site rules, not a sensor setting
It is easy to read a classification as a property of a bottle. It is more useful to read it as a set of constraints on a building. The reason the industry cares whether a refrigerant is 1, 2L, 2 or 3 is that the answer feeds into how much of it may be charged into a system serving a given room, what the room must have in the way of ventilation or mechanical extract, what the equipment itself must do when it detects a leak, and what training and procedures the people working on it need. None of that is a setting on an instrument.
This is why replacing a leak detector does not make a workshop ready for the work. A detector reports a concentration at a point in space at a moment in time. It cannot reduce a charge, it cannot ventilate a room and it cannot keep a flame away from a joint. When the first mildly flammable system arrives on a maintenance contract, the honest audit covers the procedure rather than the toolbox, and the toolbox question comes last.
| Step in the job | How it was done with a non flammable charge | What a mildly flammable charge requires |
|---|---|---|
| Pressure test | Nitrogen, or nitrogen with a trace of refrigerant | Nitrogen, or nitrogen with an inert tracer, and no flammable trace charge |
| Leak location | Halogen sniffer along the joints | Same instrument, plus a check that it is suitable to use in the atmosphere present |
| Brazing and hot work | Recover, then braze | Recover, purge, verify the atmosphere, then braze |
| Recovery and evacuation | Any serviceable machine and pump | Equipment specified by the manufacturer for a flammable refrigerant |
| Room protection | Rarely considered | Charge limit against room volume, ventilation, and monitoring where required |
| Monitor placement | Not applicable | Low, because the vapour is denser than air |
Density decides where a sensor goes
Vapour density is the quietest and most consequential difference in this whole subject. The common A2L refrigerants are substantially heavier than air as vapour, so a leak does not fill a room evenly. It sinks, spreads across the floor and accumulates in the lowest available volume, which may be a pit, a duct, a stairwell or the space under a skid. A detector mounted at eye level because that was a convenient place for the cable can read nothing while the floor layer is already significant.
The same physics governs an entirely different industry, and the hardware there makes the logic visible. The Leakbee hardwired RV propane detector is specified for 12 V direct current with three mounting brackets for surface and flush fitting and a five year life, and it exists to be fitted near the floor of a vehicle because propane vapour pools there. It is a propane device and nothing in its bullets states a response to refrigerants, so refrigerant coverage is not stated in the listing bullets and should not be assumed. What transfers is the placement reasoning, not the instrument. Service work on fuel gases has been dealing with dense vapour for decades, as the guide to gas leak detectors for propane heater service sets out.
Pressure testing without putting refrigerant in the air
The habit worth dropping first is pressurising a system with a trace of its own refrigerant to make a leak findable. It vents a greenhouse gas on purpose, it is increasingly restricted, and with a mildly flammable charge it creates precisely the atmosphere you would rather not stand in while holding a heated probe. Dry nitrogen does the pressurising job better, because you can take the system above its operating pressure, because it is inert, and because it works on a system whose charge has already escaped.
Adding a small fraction of hydrogen to that nitrogen turns the pressure test into a leak test. The Robinair LD9-TGKIT tracer gas service kit is built around that method, stating a hydrogen gas leak detector designed for a mixture of five percent hydrogen in ninety five percent nitrogen, a full colour display with message and error screens, a sweep mode, automatic calibration and reset to ambient air, a true mechanical pump, an audio mute and three ultraviolet lights in the 395 to 415 nm band for dye inspection. The point is not the feature list, it is that the test gas is inert in use and the system under test contains no refrigerant at all while you work. Where a sniffer still fits into that sequence is covered in the roundup of gas sniffers.
Fixed monitoring and the clock inside the sensor
Where a standard calls for a fixed monitor, that device becomes a maintenance item with a lifespan, not a fit and forget fixture. Every sensing element ages. Electrochemical cells deplete whether the instrument is used or not. Catalytic beads lose activity after poisoning or a heavy exposure. Metal oxide films drift and need a settled baseline in clean air. Heated diode elements are consumed and are often sold as replaceable for exactly that reason. The five year figure stated for the propane detector above is typical of the order of magnitude involved, and it is a limit rather than a target.
Two habits follow. Record the installation date of every fixed sensor and the replacement interval the manufacturer states, and treat both as hard dates rather than suggestions. And check response rather than only calibration, because a sensor that has been poisoned can hold its zero perfectly and still fail to respond to gas. The mechanisms behind that decline are set out in the piece on why gas detectors do not last forever.
One more thing belongs here, because it is a common misunderstanding about what is already in the van. The Inficon 718-202-G1 is sold as a combustible gas detector, and an instrument of that kind responds to a mildly flammable refrigerant only once the concentration approaches the region where the mixture itself could burn. That is far above the concentration at which you want to know a joint is weeping. A combustible gas instrument is a tool for judging an atmosphere, not for finding a small refrigerant leak, and owning one does not close the gap.
Rebuilding the service routine around the class
Rewrite the job sheet in the order the work actually happens. Before travelling, confirm the refrigerant and its class, the system charge and the volume of the room it serves. On arrival, establish ventilation and identify ignition sources, including the ones that are not flames, such as switches, portable heaters and vehicles. Recover fully before any hot work and verify the atmosphere afterwards rather than assuming the recovery was complete. Pressure test with nitrogen, with a tracer if you want the leak found at the same time. Use the halogen sniffer for pinpointing once the system is charged again, and keep a separate instrument for judging whether the air in the room is safe.
Finally, give the fixed equipment the same attention as the portable equipment. Monitors mounted at the wrong height, sensors past their replacement date and alarm thresholds nobody can explain are all common, and all of them are invisible until the day they are needed. Where several hazards share a plant room, the instrument class widens again, which is the territory covered by the overview of four gas detectors.
Frequently Asked Questions
Why does vapour density matter so much for sensor placement?
Because a refrigerant vapour that is heavier than air does not mix evenly into a room, it sinks and spreads along the floor, filling the lowest volume first. A sensor at head height can read nothing while a layer near the floor is already at a concentration that matters. Placement follows the physics of the gas rather than the convenience of the cable run, and for a dense vapour that means low.
Can I still use a trace charge of refrigerant to find a leak?
Putting a flammable refrigerant into a system specifically so that it can escape into a room is a poor choice when an inert alternative exists. Nitrogen, or nitrogen carrying a small fraction of hydrogen as a tracer, does the same job without venting a greenhouse gas and without creating a flammable atmosphere around the joint you are about to work on. It also lets you test a system that has already lost its charge.
Does every tool in the room need to be rated for a flammable atmosphere?
The question is not about every tool in general, it is about what could be present at the moment the tool is used. Planning the work so that a flammable concentration never forms, through charge limits, ventilation and recovery, is the primary control. Equipment suitability is the backstop for when that control fails, and the manufacturer statement for each device is the only authority on whether it qualifies.
How long does a gas sensing element last?
It depends entirely on the element. Electrochemical cells have a published service life measured in years and age whether or not they are used. Catalytic beads lose activity through poisoning and through repeated exposure to high concentrations. Metal oxide films drift. Heated diode elements are consumed and are often designed to be replaced in the field. The common thread is that every one of them has a finite life and none of them announces the end of it.
Is a fixed monitor needed in every plant room?
That is decided by the installation standard that applies, the charge size and the room volume, not by preference. The reason monitoring appears at all is that a mildly flammable refrigerant released into a confined space can reach a concentration that matters before anyone notices, and a monitor is the only thing that will be watching at three in the morning. Where one is required, its placement and its sensor life become part of the maintenance schedule.