SunProtect.me

Methodology

Every rule this calculator applies is below, with the number it uses and the source it comes from. Nothing is a black box, and nothing on this page was typed by hand — each figure is read from the same module the calculator runs, so this page cannot drift from the method it describes.

What “Method v1” means

The protection guidance is at version 1 and the sunscreen quantity at version 1. A version changes when the arithmetic or a threshold changes — not when wording is edited or a page is restyled. It is printed beside the result so that a figure you wrote down last summer can be told apart from the same figure today.

Estimating the phototype

The slider runs 0–100 and starts at 42. Its position maps to a first estimate:

  • 0–14 — Type I
  • 15–30 — Type II
  • 31–47 — Type III
  • 48–64 — Type IV
  • 65–82 — Type V
  • 83–100 — Type VI

These boundaries are a user-interface heuristic, not a validated diagnostic scale. Nothing in dermatology maps a colour to a phototype. What does map to one is behaviour in sunlight, which is why “Improve estimate” asks 6 burning options and 6 tanning options, and why answering either one replaces the tone estimate rather than being averaged with it.

With both answered: if they agree or differ by one, the result is the average, rounded down — toward the more sun-sensitive of the two. If they differ by more than one, the result is the lower of the two and the page says the answers do not map neatly to one phototype. Rounding down is a deliberate departure from the brief this site was built to, which specified ordinary rounding: ordinary rounding sends a half-step toward the less sensitive answer, which is the opposite of the direction the conflicting-answer rule already takes.

Turning the UV Index into guidance

The scale is accepted from 0 to 15 and grouped as the World Health Organization groups it:

  • 0–2 — Low
  • 3–5 — Moderate
  • 6–7 — High
  • 8–10 — Very High
  • 11+ — Extreme

At UV 3 and above, the calculator recommends broad-spectrum SPF 30 or higher, reapplied about every 2 hours outdoors and after swimming or heavy sweating, plus a hat, sunglasses, covering clothing and shade. At UV 8 and above it adds a line about keeping midday exposure short. Below UV 3 it mandates nothing, while saying protection is still worth it for long stretches outdoors.

The estimated phototype does not enter any of that. The function that produces the result is not given it — it changes only the sentence of context printed alongside. That is enforced rather than intended: a test asserts that at every UV Index, on every activity, all six phototypes receive an identical result object.

Estimating how much sunscreen

The quantity is a product, not a model:

  • 30 ml per adult, per application, for full exposed-body coverage
  • 18 ml per child — a planning estimate, not a measurement
  • 9 ml per adult and 6 ml per child for face, neck and arms only
  • multiplied by applications per day, multiplied by days
  • divided by the bottle size and rounded up, with no margin added on top

The adult figure is the American Academy of Dermatology's. The child figure is the middle of a 15–20 ml planning band: a child can be any size between an infant and an adult, so there is no exact number to give and the page says so wherever it appears. The face/neck/arms figures come from the “teaspoon rule” — roughly half a teaspoon each to the face and neck and to each arm.

Worked through, the default on the homepage — 2 adults, 1 child, 7 days, 3 applications a day, 200 ml bottles — gives 1,638 ml and 9 bottles. That is the same example the test suite pins, so the figure on the homepage, the figure in the tests and the figure here are one number in one place.

The same amount, in four units

One adult application is written differently depending on who is writing it, and none of the four is wrong:

  • 30 ml — what this calculator uses, and what a European bottle is measured in
  • about 1 US fl oz — the American Academy of Dermatology's “one ounce, or a shot glass”
  • 6 teaspoons — the British and Australian framing, at a metric teaspoon of 5 ml
  • 2 mg per cm² — the density sunscreen is tested at in a laboratory, which is where every other figure comes from

Two cautions the sources rarely give. A US fluid ounce is 29.5735295625 ml, not 30 — the familiar “1 oz = 30 ml” comes from American food labelling, where the fluid ounce is defined as exactly 30 ml as a simplification. And an imperial fluid ounce is 28.41 ml, about four per cent smaller, which is why every ounce figure on this site says “US”. Teaspoons are the loosest of the four: metric 5 ml, American 4.93 ml, and the British Association of Dermatologists' own figures imply nearer 6 ml.

Why the total is a floor, not a target

The 30 ml above is the American figure, and it is the lowest of the mainstream ones. The NHS puts a whole-body adult application at 6 to 8 teaspoons — 30 to 40 ml — and Cancer Council Australia at about 7 teaspoons, or 35 ml. Nobody is wrong; there is simply no agreed point value, because how much skin is exposed varies with the person and the swimwear.

This calculator uses the low end because brief and source agree on it, and because a figure that can be checked against a named authority is worth more than an average nobody publishes. The consequence is worth stating plainly: treat the total as a minimum. A visitor packing to the NHS figure would want roughly a fifth more than this page suggests.

Reading a bottle outside the United States

The calculator says “broad-spectrum”, which is the wording used in the United States and Australia. A bottle bought elsewhere may not carry it:

  • Europe — a UVA mark inside a circle. Under Commission Recommendation 2006/647/EC it means UVA protection is at least a third of the SPF, with a critical wavelength of at least 370 nm.
  • Japan, Korea and much of Asia — a PA grade, from PA+ to PA++++. Four levels; more plus signs is more UVA protection.

All three describe protection against UVA, which the SPF number does not measure — SPF is a UVB figure. A PA grade is a grade and nothing more: it is often explained as a multiple of how long skin takes to darken, and this site does not describe protection that way for the reason in the next section.

What it refuses to compute, and why

There is no safe-exposure time anywhere on this site. The usual formula multiplies a burning time by the SPF, which assumes the standard test thickness is what ends up on skin. It is not: measured in real use, people apply about 20 to 50 per cent of it. A number built on the thickness nobody applies is optimistic in exactly the conditions where being wrong matters most, and presenting it turns a bottle into permission to stay out.

This is enforced, not merely intended. One build script checks every source file for the identifiers such logic would need and every sentence the calculator can produce — about four thousand of them, enumerated rather than sampled — against a list of the phrasings that would say it. A second runs the same list over the rendered text of every published page.

The calculator also makes no diagnosis, predicts no risk, and recommends no product.

Sources

Guidance this site implements:

Also cited, for figures this site reports rather than implements:

Where the two UV scales come from

The five bands above — Low, Moderate, High, Very high, Extreme — are the exposure categories from Global Solar UV Index: A Practical Guide, published in 2002 by the World Health Organization with the WMO, UNEP and ICNIRP. They are what weather forecasts colour-code.

WHO's own question-and-answer page, and the US Environmental Protection Agency, present a coarser three-band action framework instead: 0–2, 3–7, and 8 and above. The two are not in conflict, and the relationship is worth stating because it is what this calculator is built on: its two thresholds are exactly the action framework's boundaries — protection from 3, the midday line from 8 — while the five bands it displays are the Practical Guide's. Both are cited below, because the site takes a different thing from each.

Where this departs from a source, and why

Three places. Each is deliberate, and each errs toward more caution rather than less.

  • Type VI does not say “never burns”. DermNet's table reads “Never burns, always tans darkly”. That is the conventional wording and this site will not use it: deeply pigmented skin burns less readily, not never, and a sentence saying otherwise would undo everything else the page says about protection at every phototype. A build gate rejects the phrasing.
  • The burn/tan tie rounds down. Where the two sun-response answers differ by one, the estimate resolves to the more sun-sensitive of them rather than to the arithmetic average.
  • Below UV 3 the page does not say “no protection needed”. The EPA's wording for the lowest band is “No protection needed”; this calculator says it mandates nothing for brief exposure, and that protection is still worth it for long stretches outdoors or sensitive skin.

How this is checked

A build that fails any of its gates does not publish. They cover the arithmetic (64 unit tests, including exhaustive checks over every phototype, UV Index and activity combination), the safety rules above, colour contrast in both palettes, and a post-build audit of the published HTML that re-runs the calculations and fails if any figure on a page disagrees with the function that produced it.

What that gives is internal consistency, which is not the same as being right. So the rules are also checked against the documents they are attributed to.

That check was last done on 22 August 2026 and is due again by 22 August 2027 — after which the build fails on its own, with no commit, rather than letting the date quietly go stale. It found six discrepancies, all now closed: the five-band UV scale was cited to a WHO page that shows three; “water resistant” was missing from the AAD's recommendation, which is three things and not one; “heavy sweating” was weaker than the AAD's “sweating”; the departure from DermNet on Type VI was undocumented; a sentence overstated what DermNet emphasises; and the statement about later diagnosis in darker skin had no source behind it.

That is a check against documents, not a clinical opinion — which is why the footer says “checked against its sources” rather than “reviewed”. See about for who maintains it and how to report a result that looks wrong.

Updated 22 August 2026.