AC Cools Fine Until Humidity Spikes, Then Floods the Pan: Decision Tree

Why this matters

A system that runs dry for weeks and then floods the secondary pan the first muggy afternoon is a latent-load problem masquerading as a drain problem. The condensate trap and line are usually fine at low humidity because the coil produces little water; once outdoor dew point climbs into the 70s, condensate generation can triple and any marginal slope, trap, or P-trap break shows up as overflow. The tech who only snakes the line returns the next humid day. The trap, the slope, the air seal, and the actual moisture-removal rate all interact, and the failure surfaces only at high latent load. This tree separates a true drainage restriction from a trap-priming fault from a coil that is producing more water than the drain was ever sized to carry.

Symptom presentation

The clean version: float switch trips or secondary pan shows water only on days the outdoor dew point exceeds roughly 68 to 72 F. Indoor temperature is at setpoint, no error code, supply air is cold. Primary pan and the first foot of drain line look clean. The customer reports "it only does it when it is humid out," which is the diagnostic key, not noise.

Variants that change the tree:

• Water gurgles or burps at the trap audibly: trap is breaking under negative pressure, not clogged.
• Secondary pan floods but primary stays dry: blocked primary drain port or pan slope reversed.
• Drips from the front of a cased coil onto the blower deck: coil sweating or condensate blow-off, not the drain at all.
• Float switch trips with a dry pan: switch fault, covered in a separate tree.

Quick checks before tools

Read the outdoor dew point on a real hygrometer or weather data, not the air temperature. Then measure the coil's water production rate the simple way: pull the trap clean-out and watch flow during a stable cooling cycle at high humidity. A 3-ton coil at 78 F / 65 percent RH return can shed 8 to 14 pints per hour. If the line carries a trickle while the pan rises, the restriction is downstream.

Confirm trap geometry. A draw-through coil (blower downstream of the coil, the common residential arrangement) sits under negative static, so a trap with too little water seal gets sucked dry and the pan backs up because air, not water, occupies the line. Measure trap depth against the air handler's external static: trap water column must exceed total negative static at the drain pan, typically a minimum of 2 inches plus an inch per 0.5 in. w.c. of static.

Check pan slope toward the drain port with a torpedo level. A pan that reads level or back-pitched holds water at the float regardless of line condition.

Isolation tree

Branch 1: drain line restriction.

• Pour a measured gallon down the clean-out with the unit off. Slow drain or backup confirms a partial clog. Blow the line with nitrogen or a wet/dry vac at the termination, then verify free flow.
• Biofilm slime is the usual culprit; a clear line that fouls every humid season points to no biocide tab and a flat horizontal run that holds standing water.
• Verify the termination is not submerged or trapped at a second low point downstream.

Branch 2: trap priming and air seal.

• With the blower on cooling speed, hold a manometer at the pan drain. If negative static exceeds the trap seal depth, the trap pulls dry and the pan overflows. Deepen the trap or install a properly sized trap kit for that static.
• A trap with no vent on a high-static draw-through system will glug and break. Add a vent tee downstream of the trap, sized per the equipment instructions.
• Blow-through coil (blower upstream) puts the pan under positive static and needs the opposite trap logic; confirm which arrangement you have before deepening anything.

Branch 3: coil producing more water than the drain carries.

• A 3/4 inch PVC primary at minimum slope (1/8 inch per foot) carries roughly 12 to 15 pints per hour comfortably. A high-latent coil at design dew point can exceed that. Measure both production rate and line capacity.
• Reduced airflow raises latent removal and water rate: a loaded filter or fouled coil drives more condensate, not less. Check total external static.
• Oversized equipment short-cycles and never reaches steady latent removal at low humidity, then dumps a backlog the moment it runs long on a humid day.

Branch 4: condensate blow-off and coil sweat.

• Face velocity over roughly 450 to 500 FPM on a wet coil carries droplets past the pan onto the blower deck. Verify blower CFM against coil rating.
• Uninsulated drain pan or line below dew point sweats and drips outside the pan; this is not overflow.

Confirming the diagnosis

Re-test on the next high-dew-point day or simulate by running the system to a low indoor setpoint with the space humidified. Pour-test the line clear, confirm trap seal exceeds static, and verify the float stays dry through a full hour of high-latent runtime. Log outdoor dew point, return dry bulb and wet bulb, total external static, trap depth, and measured pints-per-hour at the clean-out. That record proves the fix held across the actual load that produced the failure.

Remediation summary

References

• ASHRAE Handbook HVAC Systems and Equipment, Chapter on Air Handling Units - condensate drain and trap sizing for draw-through and blow-through coils.
• AHRI 210/240-2023 - residential coil latent capacity rating conditions.
• 2021 IMC Section 307 (Condensate Disposal) - primary and secondary drain, slope, and auxiliary pan requirements.
• 2021 IECC Section R403.7 and OEM air handler installation instructions - trap depth versus blower external static.