Playbook: RTU Compressor & Refrigerant Faults

Adapted from HVAC FDD Reference v1.0, Remediation Playbooks (pp. 168–169).

Step 1 — Verify the fault

1. Supply-fan prerequisite (RTU-0010): before interpreting capacity, temperature split, or refrigerant signatures, compare the final supply-fan command with independent proof for that fan. A fail-to-start contests the airflow premise of RTU-0002 through RTU-0006. A commanded post-heat fan run must keep the final command true; purge, smoke control, and local hand modes omitted from the command are host NO_EVAL, not timer exceptions. Follow the proof-of-operation playbook for the mismatch itself.
2. SAT tracking (RTU-0011): confirm sat_sp is the final active mode-specific target, then verify stable supply-fan proof and mechanical heating/cooling status. Exclude startup, setpoint/mode steps, defrost, post-heat, demand response, normal DX off-cycles, and OEM limiting. Use yTooWarm/yTooCold only as direction evidence, not a root-cause verdict.
3. Compressor short-cycling (RTU-0001): pull the compressor run status trend and count starts per hour — more than 6 starts/hr indicates short-cycling. Check minimum on-time per cycle: less than 5 minutes is abnormal. (Albayati et al. 2023 achieved 95.7% accuracy on RTU fault classification with semi-supervised learning; the trend check remains the ground truth.)
4. Evaporator fouling (RTU-0002): calculate the temperature split RAT − SAT during steady-state cooling and compare to the baseline split for the current compressor stage. A 25% or greater reduction indicates fouling. Typical baselines: 8 °C (14 °F) at stage 1, 12 °C (22 °F) at stage 2.
5. Condenser fouling (RTU-0007): measure condenser leaving air temperature minus OAT and compare to baseline for the current stage and OAT. A 30% or greater increase indicates fouling.
6. Refrigerant charge (RTU-0008/0009): with the compressor settled at a steady stage, measure suction superheat and liquid subcooling at the service ports and compare to the unit’s charging chart for the current conditions. High superheat with low subcooling indicates undercharge; subcooling well above the chart with normal-to-low superheat indicates overcharge. Rule out condenser airflow restriction (RTU-0007) first — it moves the same readings; low-ambient head-pressure control can mimic overcharge on a correctly charged unit.

Step 2 — Remote triage

1. Confirm command, proof, temperature, and stage timestamps are fresh and aligned, and that each command/status pair has the same equipment scope.
2. Review local/remote state, smoke and freeze safeties, purge and post-heat states, compressor lockouts, and recent overrides or service activity.
3. Correct only verified BAS binding or sequence defects. Never bypass smoke, freeze, condensate, high/low-pressure, electrical, or OEM safeties, and do not repeatedly reset compressor or fan lockouts.
4. For RTU-0011, compare the tracking direction with economizer command, compressor/heating stage, fan proof, and OEM limit history before changing setpoints or tuning. A bad active-target binding is not a capacity fault.

Step 3 — On-site service

Only qualified HVAC/refrigeration personnel may open panels, enter OEM service mode, or work on a refrigerant circuit. Follow the manufacturer’s procedure, lockout/tagout requirements, and applicable refrigerant-recovery rules before approaching belts, capacitors, contactors, fans, or compressors.

1. Short-cycling — check in order of likelihood:
   1. Thermostat differential: increase from 1 °F to 2–3 °F to prevent rapid cycling.
   2. Refrigerant charge: low charge causes low suction pressure, tripping the low-pressure safety. Check subcooling and superheat against manufacturer specs.
   3. Run capacitor: a weak capacitor makes the compressor struggle to start. Test with a capacitance meter — replace below 90% of rated value ($50–$150).
   4. Oversized equipment: if the unit is significantly oversized for the load, cycling is inherent — staging controls or a compressor VFD may help.
2. Evaporator fouling:
   1. Replace the air filter — the most common cause and the cheapest fix ($50–$200/bank).
   2. Inspect the evaporator coil; clean with coil cleaner and low-pressure water ($200–$500).
   3. Check the evaporator fan motor — degraded motors reduce airflow across the coil, mimicking fouling.
   4. Check for ice on the coil — icing indicates low refrigerant charge or a failed defrost cycle.
3. Condenser fouling (RTU-0007):
   1. Inspect the condenser coil from outside the unit — cottonwood seeds, leaves, and debris are the most common culprits.
   2. Clean the coil from the inside out with a garden hose or pressure washer ($0–$200).
   3. Check the condenser fan motor and blade condition.
   4. Check that adjacent RTUs are not discharging hot air into this unit’s condenser intake — rearranging discharge hoods can fix this.

Step 4 — Confirm resolution

1. Short-cycling: monitor starts per hour over 48 hours. Target: fewer than 6 starts/hr with minimum 5-minute on-time.
2. Evaporator fouling: recalculate the temperature split — it should return to within 15% of baseline. (Note: this resolution target is tighter than RTU-0002’s 25% alarm threshold — the fault clears well before the coil is fully recovered, so confirm against the 15% target, not against the alarm clearing.)
3. Condenser fouling: recalculate the condenser split — it should return to within 20% of baseline.
4. Supply fan: through a normal controller-owned cycle, verify the final command and independent proof agree after commissioned pickup/dropout times. Do not force operation or bypass an interlock.
5. SAT tracking: after the causal repair, trend final active setpoint, SAT, fan proof, and conditioning proof across normal heating/cooling cycles. The error should remain inside the commissioned band once settled.
6. Schedule preventive maintenance: quarterly filter changes, annual coil cleaning. For multi-RTU sites, service all units in the same visit.