7.1 Leak Detection Methods

Why Leak Detection Drives the Whole Rule

Every leak-repair obligation in EPA Section 608 starts with one question: is this appliance actually leaking, and where? Until a technician can answer both halves, the 30-day repair clock, the verification tests, and the recordkeeping rules cannot begin. That is why the Type II (medium- through very-high-pressure) and Type III (low-pressure) portions of the exam test leak detection so heavily — it is the gateway skill for the entire refrigerant-management program.

The Clean Air Act's prohibition on venting (40 CFR 82.154) means a slow leak is not just an efficiency problem; it is an unlawful release of an ozone-depleting substance (ODS) or a high-global-warming-potential (GWP) hydrofluorocarbon (HFC).

The exam never asks you to operate a specific brand of detector. Instead, it asks you to choose the correct category of method for a scenario and to know each method's headline limitation. Memorize the trade-off, not the device.

The Methods, Side by Side

Electronic Detectors: The Workhorse

The electronic detector is the method the exam treats as the default for finding the general area of a small leak. Three sub-types appear in study material:

• Heated-diode — most sensitive; cracks refrigerant molecules on a hot ceramic element and reads the ion current.
• Infrared (IR) — highly selective and stable; reads how the gas absorbs IR light, so it shrugs off contamination.
• Corona-discharge — older and least sensitive; reads current change as gas passes a high-voltage field.

Good technique matters more than the device:

1. Calibrate before every job and let the probe warm up.
2. Move the probe slowly — roughly one inch per second.
3. Scan low points first (vapor sinks), then valve stems, flare/brazed joints, and Schrader cores.
4. Shield the area from wind, which blows the leak plume away from the sensor.
5. Confirm any hit with soap bubbles before you cut into the system.

Standing Pressure and Vacuum-Decay Tests

These confirm a leak's existence and feed the EPA verification requirement, but they will not point to the joint:

• Pressure decay: charge with dry nitrogen (plus a trace of refrigerant), log starting pressure and ambient temperature, wait the agreed interval, then look for a drop. Always temperature-correct — pressure falls as the room cools, which can mimic a leak.
• Vacuum decay: pull a deep vacuum, valve off the pump, and watch the micron gauge. A steady rise means a leak; a rise that plateaus suggests moisture boiling off, not a leak.

Nitrogen Is the Standard Test Gas

Dry nitrogen is inert, cheap, and moisture-free — ideal for pressurizing a system to hunt leaks. The hard rules:

• Always use a pressure regulator (a cylinder holds well over 2,000 psig).
• Never exceed the appliance's nameplate working pressure.
• Never substitute oxygen or shop compressed air; mixed with compressor oil they can detonate.
• A trace of refrigerant added to the nitrogen lets an electronic detector find the leak; that trace must be recovered, not vented.

Example: A grocery walk-in cooler loses charge slowly but a careful electronic-detector sweep finds nothing because the freezer fans keep blowing the plume away. The technician pressurizes the idle system to nameplate pressure with dry nitrogen and a trace of R-448A, shuts the fans off, and re-scans. The detector now alarms hard at a brazed suction-header joint; a few drops of soap solution form a steady bubble, pinpointing the exact leak for repair.

Where Leaks Hide

Leaks concentrate at mechanical stress and joints: flare and brazed connections, valve stems and Schrader cores, the evaporator U-bends (vibration and corrosion from condensate), rubbing/chafing line sets, and relief-valve seats. Knowing the usual suspects lets you sweep efficiently instead of waving a probe randomly over the whole unit.