7.2 Type II Leak Detection Methods & Test Pressure

Finding a refrigerant leak is a process of elimination: some methods only tell you refrigerant is missing, some pinpoint exactly where it's escaping, and using the wrong one — or the wrong pressurizing gas — can turn a simple leak check into a safety incident. This section walks through the detection methods a Type II technician is expected to know, plus the pressure-testing rules that keep those methods safe.

Electronic Leak Detectors

Electronic (halogen) leak detectors are the workhorse of modern leak detection. A sensor samples the air near a suspected leak point and reacts to halogenated refrigerant molecules, typically triggering an audible tone or visual indicator that intensifies as the probe nears the leak source. They're sensitive enough to catch leaks well below what a soap solution would ever reveal, portable enough to trace a line joint by joint, and standard equipment on virtually every commercial service truck. Their main limitation: they can be fooled by residual refrigerant clinging to a surface (a "false positive"), so technicians are trained to clear the area with air movement or wait before re-testing a suspect spot.

Ultrasonic Leak Detectors

Ultrasonic detectors don't sense refrigerant molecules at all — they listen for the high-frequency sound produced by gas turbulently escaping through a small orifice under pressure. Because the detection principle is acoustic rather than chemical, ultrasonic units work well in electrically noisy or contaminated environments where an electronic detector's sensor might struggle, and they can sometimes locate a leak from a short distance away without direct contact. They're especially useful on high-pressure systems and in industrial settings with heavy background odors or airflow that would otherwise dilute a chemical signal.

Soap Bubble Solutions

A soap-and-water (or dedicated commercial bubble) solution brushed onto a suspected joint is the simplest, cheapest leak-location method available. Escaping refrigerant produces visible bubbles right at the leak point, which makes bubble testing excellent for confirming and pinpointing a leak an electronic detector has already narrowed down to one fitting or flare connection. Its weakness is sensitivity: very small or slow leaks may never produce a visible bubble, and it can't be used on inaccessible, elevated, or hidden joints the way an electronic probe or dye trace can.

Fluorescent Dye

Fluorescent dye is injected into the refrigerant/oil circuit and circulates with the system's oil. Wherever refrigerant escapes, a small amount of dye escapes with it and leaves a visible residue at the leak site — one that glows brightly under an ultraviolet (UV) lamp, sometimes days or weeks after the dye was introduced. This makes dye testing especially valuable for slow, intermittent leaks that other spot-check methods might miss during a single service visit, since the dye keeps "marking" the leak location between visits. The tradeoff is that it's not instantaneous: the dye must be injected, the system run for a while, and a UV light used later to read the results, and the dye must be chemically compatible with the system's oil and refrigerant.

Halide Torches: Limited, Largely Legacy

The halide torch — a propane or acetylene torch drawing air across a copper reactor plate — changes the flame's color (typically green or blue-green) in the presence of a halogenated refrigerant. It was a mainstay for CFC and HCFC leak-checking generations ago, but it has fallen out of favor: it's far less sensitive than electronic detection, it doesn't react well (or at all) to some newer refrigerant families, and it introduces an open flame into a service environment, which is itself a safety concern. Expect it to show up on the exam mainly as a "legacy method with limited modern use" rather than a recommended primary tool.

Standing Pressure Tests with Dry Nitrogen

When you need to confirm whether a system holds pressure at all — rather than pinpoint an exact leak location — a standing (static) pressure test does the job. The system is pressurized with dry nitrogen through a pressure-regulating device, isolated, and monitored over time (often 24 hours or longer). A drop in pressure beyond what temperature changes alone would explain indicates a leak somewhere in the tested section; the test doesn't tell you where, only that one exists, so it's typically followed up with an electronic detector or bubble solution to localize the leak once a pressure drop is confirmed.

Never Use Oxygen or Compressed Air

This is one of the most heavily tested safety rules in leak detection: never pressurize a refrigerant system with oxygen or compressed air. Oxygen dramatically increases combustion risk — refrigerant oil that is completely harmless in normal air can support violent combustion in an oxygen-enriched atmosphere, and systems pressurized this way have caused fatal explosions in the field. Compressed air is also unacceptable: shop air carries moisture and contaminants, and when compressed and mixed with certain refrigerant/oil combinations under pressure, it can create combustible or unstable mixtures. Dry nitrogen is inert, moisture-free, and the only gas that belongs on the high-pressure side of a leak test — always introduced through a proper regulator, never directly from the cylinder.

"Low Charge" Signs Are Not Leak Inspections

A sight glass full of bubbles, a low receiver liquid level, or high superheat readings all tell you the same basic fact: the system is short on refrigerant. None of them, by themselves, satisfy the regulatory definition of a leak inspection, because they don't identify where the refrigerant is escaping. To count as an actual leak inspection, these indirect signs must be paired with a location method — electronic detector, ultrasonic detector, bubble solution, or dye trace — that can point to the specific leak site.

Test Pressure Limits and Isolating Components

Never exceed the lower of the system's designed low-side test pressure or the nameplate maximum working pressure of the weakest component being tested. High-side components are typically rated for substantially higher pressures than low-side components, so a pressure appropriate for testing a condenser can be dangerously excessive for a low-side evaporator coil or accumulator. Where possible, isolate individual components or circuits with service valves before testing — this narrows down which section of a multi-circuit system is actually leaking, reduces the volume of nitrogen (and refrigerant, if the system still has some in it) needed for the test, and prevents a known-good section from masking or being damaged by a bad one.

Leak Detection Method Comparison

The table below summarizes how each method is actually used in the field, along with its main strength and limitation — including the sight glass/receiver-level check, which is not a stand-alone leak inspection at all:

Scenario: Chasing an Intermittent Leak

A supermarket rack shows rising superheat and a bubbling sight glass — clear signs the charge is low, but not proof of where the leak is. The technician clears the area, runs an electronic detector along every accessible braze joint, and gets no clear hit. Suspecting a slow leak inside an inaccessible section of coil, they recover the remaining charge, pressurize the isolated section with dry nitrogen through a regulator set well below the coil's nameplate test pressure, and let it stand overnight. The next morning, pressure has dropped — confirming a leak exists in that isolated section, even though its exact point still needs to be found with a bubble test or fluorescent dye once the section is reopened.