Methodology

Sizing work fails quietly. An oversized mini split does not break, it stops modulating and leaves the room cold and damp. A trailer within its tow rating can still be hundreds of pounds over on payload. A tankless heater rated for five gallons a minute delivers under three in a northern winter with nothing wrong with it. In every case the equipment works and the answer was wrong, which is why the reasoning behind a number matters as much as the number.

Every result names its source

Each calculator lists the formula, standard or code section behind it, and each result row states which one produced that specific number. Where a figure comes from a published standard — ASHRAE 62.2, NEC 690.7, IRC M1502.4.5 — it is cited by section rather than described vaguely. Across this site that is 91 distinct cited sources.

This is not decoration. A front axle result that says “278 lb lighter than unhitched, from tongue weight × overhang ÷ wheelbase” can be checked on a public scale, argued with, or used to decide where the load goes. One that says “within limits” and nothing else cannot be checked at all, which is the state most calculators leave you in.

Calculations are code, and code is tested

Every calculator is a small program with an automated test suite that runs before anything is published, and the tests do more than confirm the arithmetic. A subset of them assert the specific behaviour that makes each tool worth using: that tongue weight reduces front axle load rather than adding to it, that a twenty gallon pressure tank delivers about six, that a solar string sized on nameplate voltage exceeds the inverter limit on a cold morning.

Those assertions are deliberately brittle. If a change to an assumption breaks one, what changed is not a number but what the tool claims. That is the point of writing them that way.

Published data is checked against its own definition

Some reference values are not looked up at all — they are defined, and can be recalculated from first principles. Wire gauges follow a geometric standard, circular mils are a definition, conduit areas are a circle, nominal lumber sizes follow a documented rule, and a sheet area is a multiplication. Every value of that kind on this site is verified against its own definition by an automated check, not transcribed and hoped over.

One source of truth

Reference data lives in one place and feeds both the calculators and the published charts. The pump head calculator imports its friction loss from the pipe size calculator rather than keeping its own copy, and the mini split and furnace pages take their load model from the same function the room BTU page uses. Two pages on this site cannot give different answers for the same question, because there is only one answer in the code.

What these tools do not do

Every calculator carries a disclaimer naming what it does not model, and those are worth reading. Cooling and heating load estimates are rules of thumb, not ACCA Manual J. Pump sizing covers total dynamic head but not net positive suction head. Mini split modulation limits and cold-weather output are typical published behaviour and vary widely between models — the submittal for your equipment governs.

Building and electrical codes also vary by jurisdiction and change between editions. Local amendments are frequently stricter than the model codes these tools reference, and your authority having jurisdiction decides what is acceptable where you are.

Independent review

The calculations here are compiled from published standards and established engineering practice, and are checked by the automated test suite described above. They have not been reviewed by an independent licensed professional.

That is the reason tools whose output would be safety-critical if it were wrong are not published on this site. Conductor and overcurrent sizing, structural spans and bearing, stairs and guards, and roof snow loading all depend on code tables where a single wrong cell produces a confident and dangerous answer — and a table that is wrong in a consistent way passes every test that can be written for it. Those tools are held back until their tables have been checked against a verified current code copy.

What is published is geometry, material takeoff, and sizing that works from figures you read off your own equipment. Verify anything you intend to build, wire, or rely on against your current code book or a licensed professional.

Corrections

We would rather be corrected than be wrong. If you can point at the table or code section we have misapplied, write to hello@sizingcalcs.com with the inputs you used, the result you got and the result you expected. Reports that identify a specific source get tested against, and fixed.

We do not answer project-specific technical questions or review plans. See the contact page for what we can and cannot help with.