The Heat-or-Cool Test to Force a Fault

Why this matters

The hardest call to win is the one where the part works fine while you are standing there and quits the moment you leave. Intermittent faults waste truck rolls, burn customer trust, and tempt you into the worst habit in the trade: replacing parts on a hunch. A heat-or-cool test forces the fault to show itself on demand, so you diagnose what is actually breaking instead of guessing. Almost every trade has a version of it.

The principle

Most intermittent failures are temperature-sensitive. Metal expands when it warms and contracts when it cools. A hairline solder crack closes when cold and opens when hot. A weak electrical connection passes current at room temperature and drops out once it heats under load. A swelling capacitor or a marginal semiconductor behaves at one temperature and misbehaves at another.

If a fault changes with temperature, you can drive it. Apply controlled heat or controlled cold to one suspect at a time and watch the symptom appear or disappear. When it flips on cue, you have found your part. This is the single most reliable way to corner an intermittent.

Tools you need

  • A source of localized heat. A heat gun on low, or a hair dryer for gentle work. Keep the airflow narrow and aimed.
  • A source of localized cold. Canned freeze spray (the inverted-can refrigerant aerosol sold for electronics) is ideal because it cools one spot fast. A bag of ice in a pinch.
  • Your normal meter and the symptom you are chasing. You must be able to see the fault: a reading on a meter, a relay that clicks, a motor that stalls, a display that blanks.
  • Eye protection, and a way to run the equipment safely while you apply heat or cold.

Step by step

  1. Reproduce or stage the symptom first. Get the equipment into the state where the fault matters. If it only fails warm, let it run until it is warm. If it only fails on a cold morning, you are going to chase it cold. You cannot force a fault you cannot observe.

  2. Pick your direction. If the unit fails after it has been running a while, it is almost certainly a heat-related fault, so your tool is cold: cool things down and the symptom should clear. If the unit fails on first start when everything is cold and clears once warm, your tool is heat: warm the suspect and the symptom should clear. Match the test to the failure pattern.

  3. Isolate one suspect. Narrow to a single component, connection, or board section before you spray. If you flood the whole assembly with cold, everything changes at once and you have learned nothing. Aim at one capacitor, one connector, one relay, one solder joint.

  4. Apply in short bursts and watch. Hit the suspect with a two or three second burst of cold (or a steady stream of warm air) and keep your eyes on the symptom. A temperature-sensitive fault flips quickly. A relay re-energizes, a reading snaps back into range, a display wakes up, a motor catches.

  5. Confirm by reversing. When the symptom changes, let the part return to ambient and watch it drift back to the fault, then trigger it again. A real find repeats. If you cannot make it flip twice on purpose, you have not confirmed anything yet.

  6. Walk it outward. Once you have the rough zone, narrow the spray pattern. If cooling a whole connector clears the fault, cool one pin at a time. The point that controls the symptom is your bad joint, bad pin, or bad part.

Reading the result

  • Cold clears the fault, heat brings it back: classic thermal-expansion break. A cracked solder joint, a connection that opens when hot, or a semiconductor failing as it warms. Cold shrinks the metal and re-closes the gap.
  • Heat clears the fault, cold brings it back: a marginal part that needs warmth to conduct, or moisture and contraction opening a joint when cold. Common in equipment that only acts up on the first cold start of the day.
  • Nothing changes with either: the fault is probably not thermal. Move on to vibration (tap-test the suspect with an insulated tool while watching the symptom), to load, or to voltage.

What not to do

Do not bake a whole board or freeze an entire assembly hoping something reveals itself. Broad and uncontrolled tells you nothing about which part. Keep heat off plastics, sensors, and anything rated for low temperatures unless you mean to test it. Never aim heat at fuel, refrigerant lines under pressure, or sealed components that can rupture. And never let the test itself become the repair: once you find the part, fix the part, do not leave a connection that only works because you warmed it.

References

  • Trade-standard intermittent-fault diagnosis (thermal-cycling and component-isolation method)
  • Manufacturer service documentation on thermal-sensitivity testing of electronic controls
  • See related: Confirming a Part Is Actually Bad; The Scene Tells the Story Before You Touch