What you'll see

A ductless head (or the outdoor unit) is flashing a light pattern or showing an error code, and it's either not running or running poorly. Mini-splits are inverter-driven and far more electronically self-aware than a basic split system — they'll tell you what they think is wrong. The catch: every manufacturer uses its own code list, so the first step is always to read the specific code for that brand. After that, almost every fault falls into one of a few families, and the common real-world causes within each family are similar across brands.

Don't troubleshoot a mini-split like a conventional system. Forcing contactors and back-feeding 24V doesn't apply — these run on inverter boards, communication signals, and thermistors.

Walk it in order

  1. Read the exact code for the brand. Count the blink pattern on the indoor head, read the display, or check the outdoor board's LEDs/display. Get the manufacturer's meaning for that code — it names the subsystem. This is non-negotiable; a generic guess wastes the trip. (Reference the brand's code list by the unit's model; don't assume another brand's numbering.)
  1. Group the code into a family. Most codes fall into:
  • Communication error (indoor↔outdoor): the two units can't talk.
  • Sensor/thermistor fault: a temperature sensor is open/shorted/out of range.
  • Electrical/inverter fault: voltage, current, IPM/compressor drive, or overload.
  • Refrigerant/pressure/temperature fault: low charge, overheat, high pressure, discharge temp.
  • Fan/motor fault: the DC fan motor or its feedback.
  1. Communication error → check power and the interconnect wiring. This is one of the most common mini-split faults. Verify proper supply voltage at the outdoor unit, then check the interconnecting communication conductor and grounds. Causes: a loose/wrong terminal landing (S1/S2/S3 type wiring landed out of order), a nicked or undersized interconnect cable, moisture in a connection, reversed line/neutral on some designs, or a failed board. Confirm the wiring landing matches the diagram terminal-for-terminal before condemning a board.
  1. Sensor/thermistor fault → measure the thermistor. Mini-splits lean heavily on thermistors (room, coil, discharge, ambient). A flagged sensor is usually open, shorted, or has drifted. Unplug it and measure resistance against the temperature/resistance spec; compare to the actual temperature. A cheap thermistor or a connector that vibrated loose fixes a lot of these — check before you blame the board.
  1. Electrical/inverter fault → check voltage and current. Codes for voltage, IPM (inverter power module), or overcurrent point at supply problems or the drive. Confirm steady supply voltage under load (low/sagging voltage trips inverters), check connections, and look for an overamp condition from a mechanical problem. IPM/compressor-drive faults can mean a failing inverter board or a compressor problem — verify supply and load before deciding.
  1. Refrigerant/temperature fault → check charge and airflow. High-discharge-temp, low-pressure, or overheat codes often trace to low charge (frequently a flare leak on the line set), a restriction, or blocked airflow at either coil. Mini-splits are unforgiving about flare quality and proper evacuation — a slow flare leak is a top real-world cause. Gauge it (where the unit has ports) and inspect the flares.
  1. Fan/motor fault → check the DC fan and feedback. A DC fan motor that's obstructed, failing, or has a bad feedback signal throws a motor code. Spin it (power off), check for obstruction, and verify the connector.
  1. Only then suspect the board. Boards do fail, but they're the last suspect after wiring, thermistors, charge/flares, supply voltage, and airflow check out. Confirm the obvious external causes first — mini-split boards get replaced needlessly all the time.

What "normal" looks like

  • Supply voltage: steady at the rated voltage under load; inverters are sensitive to sag and dips.
  • Interconnect wiring: landed terminal-for-terminal per the diagram, correct gauge, tight, dry.
  • Thermistors: resistance matches the temperature/resistance curve for the actual temperature; connectors seated.
  • Flares: properly cut, deburred, and torqued to spec; line set evacuated to a deep vacuum at install (the 500-micron-class standard).
  • Charge: factory charge plus the line-set adjustment for length per the data plate.

Common faults & what they mean

  • Indoor/outdoor communication error: miswired or damaged interconnect, loose terminal, moisture, or a board. Verify the landing first.
  • Thermistor open/shorted/drifted: flagged sensor code; measure resistance vs temperature. Common and cheap.
  • Low voltage / inverter (IPM) fault: sagging supply, loose connection, or drive/compressor problem. Confirm voltage under load.
  • Low charge / flare leak / restriction: discharge-temp or pressure codes, weak performance. Inspect flares — the usual culprit on ductless.
  • Blocked airflow (either coil): overheat/pressure codes. Dirty indoor filter/blower wheel or dirty/blocked outdoor coil.
  • DC fan-motor fault: obstruction or failing motor/feedback. Check the fan and connector.
  • Failed board: real but last — only after externals are cleared.

Tech tips & gotchas

  • Read the brand's code first — always. There's no universal code list. A blink pattern that means "communication" on one brand means something else on another. Look it up for the model.
  • Communication faults are usually wiring, not boards. Confirm the interconnect is landed terminal-for-terminal and the cable is sound before you sell a board. A landing swapped at install is extremely common.
  • Flare leaks are the ductless Achilles' heel. Low-charge and discharge-temp codes very often trace to a sloppy flare. Inspect and re-flare to spec; don't just add refrigerant.
  • Thermistors are cheap and frequently the answer. Measure resistance vs temperature before condemning the board — a drifted or unplugged sensor mimics big failures.
  • Inverters hate low/sagging voltage. A marginal supply or a long, undersized circuit trips inverter faults that look like board failures. Measure voltage under load.
  • Don't apply conventional-system reflexes. No jumpering contactors, no back-feeding 24V. These are communication-and-inverter machines — work the code, the wiring, the sensors, and the charge.

Safety / code notes

  • De-energize and verify before opening the unit; inverter capacitors can hold a dangerous charge well after power-off — wait and verify discharge before touching the drive.
  • Recover and handle refrigerant per EPA 608 (40 CFR Part 82, Subpart F); many mini-splits use R-410A and newer ones use A2L refrigerants (R-32, R-454B) — follow the A2L handling and leak-check practices and any room-size/charge limits in the listing.
  • Interconnect wiring and the disconnect must meet NEC requirements; size the circuit per the unit's MCA/MOCP nameplate.