How to read a cooling spec sheet

Cooling products carry more acronyms than almost any other category of home appliance, and the acronyms are not equally trustworthy. Some are the output of a government test procedure. Some are the output of a marketing department.

Here is all of it, in one place.

A BTU per hour figure is a rate of heat removal, and on its own it is meaningless without knowing which test produced it.

ASHRAE-128 is the older standard and produces the larger number. It measures the unit’s raw cooling output.

SACC — Seasonally Adjusted Cooling Capacity — is the DOE-mandated figure for portable air conditioners. It subtracts the heat radiating off the hot exhaust duct inside your room, and the cooling load created by hot outdoor air infiltrating to replace the room air the unit blows outside.

The gap is not small and it is not consistent. Across the units we have checked it runs from 14.3 percent to 50 percent. Two units both sold as “14,000 BTU” can differ by 2,500 BTU of actual rated capacity.

The full explanation of what SACC subtracts and why.

CEER, SEER2 and EER2 are all “BTU per watt-hour”, which makes them look comparable. They are not.

• CEER rates room air conditioners at a single rating condition, plus standby power.
• EER2 rates mini-splits at a steady-state condition. It is the closest thing to an apples-to-apples comparison with CEER.
• SEER2 rates mini-splits across a modeled cooling season, including part-load operation — which is why the numbers are so much larger.

So a SEER2 35 mini-split is not twice as efficient as a CEER 16 window unit. The honest comparison is EER2 against CEER, where the certified maxima are 19.1 and 16 respectively.

Worked through in full here.

This is the most consequential misreading in the whole category, because the two scales invite opposite intuitions.

Sones are linear. Two sones sounds twice as loud as one sone. Four sounds four times as loud. Certified ventilating fans use this scale.

Decibels are logarithmic. A 3 dB difference is a small perceived change; roughly 10 dB is a doubling. Portable fans and portable air conditioners use this scale, and never as a certified figure.

So 0.7 sones against 1.4 is a genuine halving, while 46 dBA against 48.7 dBA is barely noticeable. Reading the sone figure with dB intuition makes you underrate a real difference; reading dBA with sone intuition makes you overrate a trivial one.

And note the coverage gap: window air conditioners publish no noise figure of any kind in the certified dataset, and manufacturer dBA claims on fans and portables are not independently tested.

Every category on this site carries some version of “cools rooms up to X square feet”, and none of them come from a test.

We found the arithmetic. Across Honeywell’s evaporative cooler range — spanning a thirteenfold difference in airflow — every published coverage claim equals the CFM divided by 1.64. The CL25AE at 764 CFM claims 466 sq ft. The CO60PM at 2,697 CFM claims 1,643 sq ft. Both are the same division.

A formula applied to airflow cannot know your ceiling height, your insulation, your glazing, your climate or how many people are in the room. It is a marketing figure with the shape of a specification.

Notably, the certified datasets contain no coverage field at all. The manufacturers who decline to publish one — Hessaire, for instance — are being more honest than the ones who do.

The full working on that finding.

1. Identify the category, because it determines which metrics are certified and which are claims.
2. Find the certified efficiency figure — CEER, CFM per watt, or SEER2/EER2. If there is not one, you are looking at a category with no certification program.
3. Find the capacity figure and check which standard produced it. On a portable, that means finding the SACC number and ignoring the ASHRAE one.
4. Find the wattage or annual kWh. A product with no published power draw cannot be compared with anything on running cost. That absence is itself information.
5. Ignore the square footage. Size from the room instead. ENERGY STAR’s sizing table does the job the coverage claim pretends to.