55F is a code line before it is a winemaking choice

In most wineries the barrel room is a warehouse whose temperature is held with cool night air, and the 2005 guidebook wrote that practice down as the California norm. In Temecula that night air does not arrive cold enough, and the design-day low is 62F. So the free half of the old assumption is gone, and the winemaker reaches for mechanical refrigeration to make up the difference. A barrel room's temperature stays a winemaking preference only until it reaches 55F, where it turns into a building classification the winemaker chose without meaning to. Below 55F, Title 24 Part 6 reads a nonresidential room as a process space. At 55F or less, served by mechanical refrigeration, in a building or space of 3,000 square feet or more, that same room is a refrigerated warehouse. That is three conditions, not one, and a commercial barrel room usually has the refrigeration and the floor area before anyone touches the thermostat. So the temperature is the condition a winemaker actually sets, and it is the one people get backwards.

The threshold is 55F, and "less than" is not "at or below"

Read the definitions the way a plan checker reads them. A process space is thermostatically controlled to hold a temperature less than 55F. A refrigerated warehouse holds its space at 55F or less, by mechanical refrigeration, at 3,000 square feet or more. So a room held at exactly 55F falls out of the process-space definition, which stops below 55, and with the refrigeration and the floor area it is a refrigerated warehouse instead. Below 55, both temperature conditions are met at once, so there it is the refrigeration and the 3,000 square feet, not the thermostat, that decide which name applies. Both temperature definitions read verbatim identical in the 2022 and the 2025 editions, so this is not an artifact of one code cycle.

The consequence is not cosmetic. A refrigerated warehouse of 3,000 square feet or more, served by mechanical refrigeration, picks up Section 120.6(a) of the 2025 Standards. That means R-28 walls and roof for a cooler under Table 120.6-A, plus an infiltration barrier on the passageways to nonrefrigerated space and evaporator fan controls, all under that same section. Call it the difference between an insulation value a designer picks and one the code sets. A freezer in the same table owes R-40 on the roof, so the cooler's R-28 is the lighter end of the requirement, not a way out of it.

Under 3,000 square feet the room drops out of Section 120.6(a) and into the Appliance Efficiency Regulations at CCR Title 20, sections 1601 through 1608. A winemaker who settles at 58 to 60F rather than 55F meets neither definition and owes neither path. So the number on the mechanical schedule, not the number on the wine list, decides which code the building answers to. Below 55F, the process-space path is not envelope-free either: the code's applicability table marks the mandatory envelope section and the performance path as not applicable for process spaces, then points them to the prescriptive daylighting requirement for large enclosed spaces and to the alterations provision. The requirement moves; it does not vanish.

None of this is winery-specific, because the code has no wineries in it. A full-text scan of the 2025 Standards turns up the word wine exactly once, in the lighting definition of a bar. There is no winery provision and no winery exemption. A barrel room is classified on its temperature, its floor area, and whether mechanical refrigeration serves it, and on nothing about what is in the barrels.

65 to 75%: the only humidity range with a reason

So the temperature line is sharp; the humidity target the trade designs to is not. The only bounded relative-humidity range published by a wine research institute that this research could reach is the Australian Wine Research Institute's 65 to 75%, and it is the only one that states a reason at both ends: lower humidity drives faster evaporation, higher humidity encourages spoilage microorganisms. A CEC-funded guidebook, LBNL/PUB-3184, puts cave humidity at 70 to 90% in one passage and at 80 to 90% in another, disagreeing with itself inside a single report. The research body's ceiling, 75%, sits below both of the guidebook's numbers. Nobody reconciles them.

The temperature figure rests on the same thin footing. No viticulture or enology research source reachable here publishes a barrel-room temperature setpoint at all. The numbers in circulation, 52 to 55F for aging and 58 to 60F after malolactic fermentation, trace to that 2005 guidebook rather than to a research recommendation. A whole mechanical design leans on a setpoint whose provenance is one report now two decades old.

And the two targets pull against each other. In a room held at 59F and 80% relative humidity, moisture does not condense until it meets a surface below 53F, a margin of about six degrees. Push the same room to 90% and that margin narrows to under three degrees, so vapor condenses on anything 56F or cooler. Which means the harder you chase the high end of the humidity band, the smaller the error you can afford on any cold surface in the room, and the cooling coil holding the room at 55F is the coldest surface in it.

At 4 gallons a barrel, the wine will not humidify the room

There is a reflex to assume a room full of evaporating barrels makes its own humidity. Run the number instead. Take LBNL's worksheet input of 4 gallons lost per barrel per year, which is the generous end of the published evaporation figures and is stated for Northern California, not Temecula. Four gallons of water is 33.4 pounds a year, or 0.0038 pounds an hour. A 1,000-barrel room, and 1,000 is an illustrative multiplier here rather than a count of any real building, gives off about 3.8 pounds of vapor an hour. Call it 4,000 Btu/h of latent load, roughly a third of a ton of refrigeration.



Against a room that has to hold 65 to 75% RH while outdoor air works against it, a third of a ton is nothing. The angel's share does not humidify the barrel room; the water has to be put in on purpose, which is why the humidifier is there and why it runs. It is also why a service tech gets called back in August to a room reading well below its target with the nozzles going full time: the coil holding 55F condenses moisture out faster than the nozzles add it back.

The design day never falls below 62F

Here is the assumption worth checking. The 2005 guidebook describes the California norm plainly: in most wineries the cellar is a warehouse whose temperature is held with cool night air, or with artificial cooling that supplements the night air on the hot days. Cool night air is the free half of that sentence, and in Temecula it does not arrive cold enough. The design conditions come from Appendix JA2 of the 2025 Reference Appendices, the edition a 2026 build falls under, and for Temecula the 0.5% cooling dry bulb is 96F against an outdoor daily range of 34F. Subtract the range from the peak and the design-day low is 62F. That subtraction is a modeling approximation, not a published value: JA2 Table 2-3 has no minimum-temperature column, so the 62F is derived rather than read off the table. At the rarer 0.1% condition, a 101F peak, the derived low is 67F.



Sixty-two degrees is above every target in the record: it runs seven degrees over the 55F aging ceiling, and it is above the 58 to 60F band the room holds after malolactic. So on a design day the night air cannot pull the room down to any setpoint the guidebook names, and the free cooling runs out at the exact hour the load peaks. That is the ordinary way a climate-driven assumption fails: it holds on an average night and is gone on the design night.

The same appendix carries a second gap. Compliance energy calculations must use one of the Commission's sixteen official hourly weather files, stated flatly at the top of JA2. Temecula's ZIP codes fall in Climate Zone 10, whose representative city and compliance file is Riverside, at 840 feet. Temecula sits at 1,006 feet. The compliance model and the site are 166 feet and a city apart before the sizing even starts. And the data underneath is old: JA2 points the reader to an ASHRAE dataset published in 1982, with a supplement from 1994, for the full California design-location listing. A barrel room going up in 2026 is sized against design conditions assembled decades before it.

So the barrel room is three arguments at once, and each turns on a number the trade treats as soft. 55F is a code line with R-28 behind it. The humidity setpoint has two authorities and a contradiction. The night-air cooling the guidebook assumes gives out at 62F on a design day. A contractor bidding the refrigeration and humidification on a Temecula winery is bidding against all three, and the winemaker fixed the first of them by naming a temperature.

The mechanical permits for that work post on the same county portals that carry every other trade, before the equipment is even bought. We track more than 50 of those portals across Southern California, 18 trades from HVAC and refrigeration to Solar & Battery, and it is priced month to month, so a contractor chasing wine-country jobs is not signing a year to learn whether the leads are there.