concepts · updated 2026-08-01
Water over air in radiant cooling
confidence: high volatility: cold verified: 2026-08-01fresh
The archive's consistent preference for water over air as a thermal medium is not stated in one place — it is the cumulative conclusion of the Cool Cell brochure, the building thermal storage essay, t
The archive’s consistent preference for water over air as a thermal medium is not stated in one place — it is the cumulative conclusion of the Cool Cell brochure, the building thermal storage essay, the pool-heater cooling documents, and the night-sky experiments. This page now cites two primary Baer essays directly: “Cooling with Night Air” (1984), where Baer measured the limits of air-based night cooling in his own house, and the standalone “Storage” essay (2002), which makes the water-and-radiant-ceiling case with numbers. A third filed source (response_to_baer.pdf) is a short handwritten note from Baer to Bruce W. Davis, dated July 8, 2000, commenting on the reception of the Corrales Comment irrigation column — not a technical document about radiant cooling; it belongs more properly with the Corrales Comment page.
The archive’s water-over-air argument
Water over air is one of the archive’s most consistent themes:
| Medium | Storage capacity | Transport | Night-sky coupling |
|---|---|---|---|
| Air | Very low (1 BTU/ft³·°F) | Ducts, fans, electricity | Poor — requires large surfaces |
| Water | ~62× air by volume | Pipes, gravity, thermosiphon | Excellent — roof radiators cool water directly |
The Cool Cell system is built on this preference: water in ceiling reservoirs absorbs interior heat during the day, then circulates to roof-mounted radiators at night where it dumps heat to the sky. No compressor, no refrigerant — just water, gravity, and the temperature difference between a building and outer space.
Baer tested the air side first — and measured its limits
The preference was earned empirically. In “Cooling with Night Air” (SUNPAPER, January 1984), Baer analyzed the traditional air-based strategy — open windows at night, close them by day — and measured his own house at 5:30 AM on August 23, 1983. Both drum-mass rooms were underventilated on still nights: the guest room, with only ½ sq ft of top vent, was “miserably underventilated,” and even the living room’s 10 sq ft of vent left the ventilation coefficient far short. His conclusion: natural convection of night air alone is insufficient for most rooms, and effective air flushing demands large vent areas (14.4 sq ft inlet and exit in his worked example). The essay also marks his first explicit engagement with Harold Hay’s movable insulation as the paradigm for two-mode buildings.
The water side of the ledger is stated in Baer’s standalone “Storage” essay (May 2002): the heat generated by people, lights, and appliances — about 250 BTU (75 watt-hours) per square foot of floor per day — “must be absorbed by the ceiling or walls during the day then disposed of at night to the sky”; 3 gallons of water can absorb that much in rising 10°F. “The most comfortable form of heating and cooling is radiant; the ceiling is the ideal surface to accept heat and is also good at emitting it.” He judges the sensible heat of plain water “a perfectly satisfactory thermal storage,” with phase-change slurries as an optional upgrade that would turn one day’s storage into two.
The Corrales Comment note
The filed source (response_to_baer.pdf) is a personal note Baer wrote to Davis the day the irrigation column appeared: “Here’s my advice about engineering for slobs (like me — like us). Not a peep from anyone in response. I suppose most regard it as offensive.” The postscript offers “more on H₂O if it interests you. In fact a nightmare about getting it straight out of our air” — a hint toward the desiccation satire that became Sunny days ahead.
The note is filed here because it is the only document in the archive dated the same day as the Corrales Comment; see the Corrales Comment page for its context.