As an Amazon Associate, Akermin earns from qualifying purchases. Product links on this page are affiliate links.
Infrared and heated diode sniffers both find halogenated refrigerant, but they get there differently. An infrared sensor measures absorption of light at a wavelength the refrigerant molecule absorbs, so it responds to concentration and is comparatively tolerant of oils, moisture and other gases. A heated diode passes air over a heated ceramic element; halogenated molecules break down at the surface and change the element current, which gives very high sensitivity to tiny leaks but also makes the sensor respond to any halogen, including the chlorine in some cleaning solvents. For locating a micro leak in a clean space the diode’s sensitivity is an advantage, and the heated diode refrigerant leak detector illustrates the type with a stated 0.05 oz/yr threshold. For repeatable readings in a greasy plant room or where you want a concentration trend rather than a yes or no, infrared is the steadier instrument, as with the infrared refrigerant leak detector rated to find leaks below 0.03 oz/yr. Neither is an alarm, neither measures ppm as a calibrated gas monitor does, and neither replaces a pressure test on an empty system.
What the two sensors physically measure, and why that decides the job
A refrigerant sniffer does not weigh refrigerant. It responds to molecules that reach the sensing element, which is why the same leak reads differently depending on how close the probe is, how fast air moves past it, and how long you dwell. The two technologies differ in what happens at that moment.
A heated diode sensor is a small ceramic bead with a heater and an electrode. Air is drawn or diffused onto the hot surface. Halogenated molecules, meaning compounds carrying fluorine, chlorine or bromine, break apart at that temperature and release halogen ions that change the current flowing through the bead. The change is amplified and shown as a bar, a number or a tone. The mechanism is why the diode is so sensitive: a handful of molecules is enough to shift the surface chemistry. It is also why the diode is not selective. Any halogen-bearing vapour can trigger it, including the chlorine in some degreasers and the fluorine compounds in certain aerosol propellants.
An infrared sensor shines light through a small chamber and measures how much is absorbed at a wavelength the refrigerant absorbs. More molecules in the chamber means more absorption, so the output tracks concentration in a roughly linear way within the calibrated range. Because the measurement is optical, oils, moisture and most non-halogenated gases pass through without a response. That is the core trade: the diode is a superb leak finder and a poor gas identifier, while infrared is a better identifier and a slightly less frantic leak finder.
Neither sensor is a calibrated analyser in the sense a plant gas monitor is. The listing for the RALBIL heated diode unit states a maximum sensitivity of 0.05 oz/yr in the stationary state, and the listing for the Fieldpiece DR82 infrared unit states it finds leaks below 0.03 oz/yr. Those figures are leak-rate sensitivities under stated conditions, not a measurement range in parts per million. If you need a ppm reading against a calibration certificate, you are looking at a different class of instrument.
Sensitivity figures: what 0.05 oz/yr and 0.03 oz/yr actually mean
Refrigerant leak rates are usually quoted in ounces per year or grams per year. The arithmetic is worth doing once so the numbers stop being abstract. One ounce is about 28.35 grams, so 0.05 oz/yr is roughly 1.4 grams per year, and 0.03 oz/yr is roughly 0.85 grams per year. Spread over a year, that is a leak you would never see as an oil stain and would struggle to find with bubbles, because the gas escapes too slowly to form a visible bubble at any useful rate.
That is the honest case for a sniffer over a soap solution. A bubble test needs a leak fast enough to inflate a film, so it finds the leaks that matter for a service call today. A sniffer with a stated threshold of a fraction of an ounce per year can find the slow leak that has been topping up a system for months. The catch is that the threshold is a best-case figure. It assumes a clean background, a still probe held at the right distance, and a sensor in good condition. In a real plant room the practical limit is set by whatever else is in the air.
Be careful comparing a stated leak-rate sensitivity with a stated detection limit in ppm. They are not the same quantity and cannot be converted without knowing the geometry of the leak and the airflow. If one listing gives oz/yr and another gives ppm, the honest move is to treat them as different units describing different tests and not to rank them against each other.
Selectivity, cross sensitivity and the false alarm that wastes an afternoon
The most common complaint about refrigerant sniffers is that they alarm in a clean room. With a heated diode the explanation is usually chemistry. The sensor responds to halogen, and halogen turns up in places you do not expect: chlorinated brake cleaner, some flux residues, certain dry cleaning vapours, and the propellant in some sprays. A diode that screams near a bench where someone stripped a joint with solvent is doing exactly what its physics dictate.
Infrared is more selective because it looks for absorption at a wavelength tied to the refrigerant’s molecular bonds. That does not make it immune. Some refrigerants absorb at overlapping wavelengths, and a blend can present a mixed spectrum, which is why infrared instruments are calibrated for a family of refrigerants rather than a single gas. But the everyday nuisance sources, oils, grease, water vapour and most solvents, are largely invisible to it. In a busy mechanical room that difference alone can decide which instrument you reach for.
The practical rule: if you work in clean domestic and automotive spaces and want the smallest possible leak found, a diode is a reasonable default. If you work where solvents, oils and other gases are present, or where you need to explain to a client why the reading went up, infrared gives you a cleaner story.
Sensor life, contamination and the maintenance nobody budgets for
A heated diode is a consumable. The hot bead is slowly attacked by the same halogen chemistry that makes it work, and it is poisoned by exposure to high concentrations, oil mist and moisture. Sensitivity falls before the sensor fails outright, which is the dangerous part: an instrument that still beeps but is half as sensitive will miss the leak you were called to find. The listing for the RALBIL detector notes that the probe can be replaced, which is the feature to look for, because a sealed sensor means a disposable instrument.
Infrared sensors are also consumable but on a different schedule. The optical cell can fog with oil or dust, and the source has a finite life. The listing for the Fieldpiece DR82 states a 10-year sensor life and an IP54 rating against moisture and oil, which is the manufacturer’s claim rather than a measured figure, but it tells you the design intent: a sensor you do not expect to replace often. Treat any stated sensor life as an upper bound under good conditions, not a guarantee.
On both types, the maintenance nobody mentions is the filter and the probe. A blocked inlet filter slows response and makes the instrument look insensitive. A probe left in a dirty van picks up oil that later outgasses into the sensor. Replacing the filter and wiping the probe is the cheapest sensitivity restoration available, and it is skipped far more often than it should be.
Where each technology is the wrong tool, and what to use instead
Neither sensor tells you the system charge, the pressure integrity, or whether a leak is on the high side or the low side. A sniffer tells you refrigerant is present at the probe tip. That is all. If the system is empty, there is nothing to sniff, and a pressure test with dry nitrogen is the correct instrument. If you need to prove a repair holds, a standing pressure or vacuum decay test is the correct instrument. A sniffer cannot do either job and should not be sold as if it can.
Neither is a safety device. A refrigerant sniffer is not a combustible gas detector, not an oxygen meter and not a carbon monoxide alarm. In a plant room where a refrigerant leak could displace oxygen, the instrument you need is an oxygen meter, and in a space where a flammable A2L refrigerant could accumulate, the instrument you need is a flammable gas detector rated for that gas. A refrigerant sniffer does not satisfy confined space entry pre-checks and does not substitute for a four gas monitor. This is the single most important limitation to state plainly, because the shape of the tool invites the wrong assumption.
There is also a category boundary worth naming. If your real question is temperature, not gas, an infrared thermometer is the right instrument and shares nothing with a refrigerant sniffer but the word infrared. The two are often confused by buyers searching for the physics rather than the job.
Reading a listing: what a specification can and cannot tell you
Listings vary wildly in what they publish, and the gaps matter as much as the numbers. A useful comparison table for the two physics looks like this, with cells filled only where the supplied listings state something.
| Attribute | Heated diode | Infrared |
|---|---|---|
| Sensing mechanism | Heated ceramic element, halogen response | Optical absorption of refrigerant |
| Stated leak sensitivity | 0.05 oz/yr (1.5 g/a) stationary | Below 0.03 oz/yr |
| Refrigerants covered | CFC, HCFC, HFC, HFO, blends, halon | CFC, HCFC, HFC, HFO, blends |
| Selectivity to non-halogen gases | Low; responds to any halogen | Higher; optical absorption |
| Stated sensor life | Not stated in the listing bullets | 10 years (manufacturer claim) |
| Ingress protection | Not stated in the listing bullets | IP54 |
| Display of leak level | 9-stage LED display, 3 sensitivity levels | Large backlit LCD |
| Replaceable probe or sensor | Probe replaceable | Not stated in the listing bullets |
The table shows the shape of the decision. The diode publishes a leak threshold and a sensitivity adjustment; the infrared publishes a threshold, an ingress rating and a sensor life. Neither publishes a ppm calibration, a response time in seconds, or a cross sensitivity list, so those questions cannot be answered from the listing and should be asked of the manufacturer before purchase.
Two other instruments in this space are worth naming because they show how the same job can be presented differently. The Robinair LD7 listing describes a colour display with bar graph and sweep modes, which is a presentation choice rather than a change of physics; it helps you trace a leak back to its source. The Bacharach H-10 Pro listing is sparse, giving only brand, product number and category, so no sensing technology or sensitivity figure can be attributed to it from the supplied text. Where a listing is that thin, the honest position is that the specification is unavailable, not that the instrument is inferior.
For a fuller treatment of the two families side by side, see the dedicated comparison of infrared and heated diode leak detectors, and for the buying side of each, the guides to heated diode detectors and infrared detectors.
The short version for a buyer standing in a supplier’s doorway
If you chase small leaks in clean domestic, automotive and light commercial spaces and you accept that the sensor is a consumable, a heated diode gives you the most sensitivity for the least outlay. If you work in dirty plant rooms, need a reading you can defend, or want a sensor you replace rarely, infrared is the calmer instrument. If you need a ppm figure, an alarm, or a pressure integrity test, neither is the right purchase, and the correct instrument is a calibrated gas monitor or a pressure test rig. Buy for the job, not for the word infrared on the box.
Frequently Asked Questions
Is infrared always more accurate than a heated diode?
No. Accuracy and sensitivity are different things. A heated diode is often more sensitive to very small leaks, which is why listings quote thresholds like 0.05 oz/yr. Infrared is more selective and gives a reading that tracks concentration more predictably, but that is not the same as being more accurate at finding a leak. The right choice depends on whether you need to find the smallest possible leak or to avoid false positives in a contaminated space.
Can a refrigerant sniffer detect a flammable A2L refrigerant leak safely?
A refrigerant sniffer can indicate that refrigerant is present, but it is not a flammable gas detector and does not satisfy the safety requirements for an A2L space. If the risk you are managing is ignition, you need a flammable gas detector rated for the specific refrigerant. Do not treat a sniffer reading as a clearance to work.
Why does my heated diode alarm when no refrigerant is present?
The sensor responds to halogen, and halogen appears in solvents, some aerosols and flux residues. A diode alarming near a bench where chlorinated cleaner was used is behaving as designed. Ventilate the area, let the sensor settle, and re-test. If the alarm persists in clean air, the sensor may be contaminated and due for replacement.
Do I need both an infrared and a heated diode detector?
Most technicians do not. A single good instrument plus a bubble test and a pressure test covers the majority of work. Two sniffers make sense only if you routinely work in both very clean spaces where maximum sensitivity matters and dirty spaces where selectivity matters, and even then the second instrument is a convenience rather than a necessity.
What does a sensor life of 10 years actually mean?
It is the manufacturer's expected life under good conditions, not a guarantee. Exposure to high concentrations, oil mist and moisture shortens it. The listing for the Fieldpiece DR82 states 10 years and an IP54 rating, which suggests a design intended for plant room use, but you should still expect sensitivity to fall gradually and plan for replacement.
Can a sniffer tell me how much refrigerant has leaked?
No. It tells you refrigerant is present at the probe and roughly how strong the response is. It does not measure the system charge, the leak rate in the system, or the total loss. For that you need a pressure test, a weighing scale on the recovery cylinder, or the system's own charge records. A sniffer is a locator, not a meter.