How to Read a Roofing Slate Test Report: The Six Rows That Carry the Whole Verdict

Part of the Roofing Slate Guide's Quality, Testing & Standards section. It follows our EN 12326 explainer and the ASTM C406 explainer — those pages decode the standards; this one decodes the piece of paper you actually get in your inbox.

Picture the scene: a container arrives in Manchester, and the buyer walks around the opened door holding a three-page PDF printed on A4. The pallets are shrink-wrapped, the strapping is tight, the tiles look exactly like the sample he approved. He hands the report to his colleague, who squints at it and says the same sentence every importer has said at least once: "so… is this good?"

The honest answer is: the report is never good or bad by itself. It is good or bad compared to what you ordered, and compared to what the rest of the report says. I read these documents for a living, and I have seen a real lab report condemn a container just as surely as a forged one — not because the test failed, but because the buyer didn't know which numbers to compare. This page fixes that.

Row one: the header — read the batch before you read anything

Before any number, check the frame. A legitimate report names five things in its header block, and the absence of any of them is a reason to slow down: the testing laboratory (with its name and reference), the standard tested to (EN 12326-1, ASTM C406, or both), the sample description (slate type, size, thickness), the test date, and — the one most buyers skip — the batch or lot number.

That last line is the thread that ties the paper to the stone. A report without a batch number is a statement about a slate; a report with your batch number is a statement about your slate. Every serious dispute I have ever been called into — and I have been called into a few — started with a report that was real, honest and entirely irrelevant, because it described a different season's production.

Rows two and three: water absorption, and why it outranks everything

The absorption row tells you how much water a dry slate drinks after a soak, as a percentage of its own weight. It is the closest thing the slate trade has to a single health number, because absorption decides how much ice a winter has to work with inside the stone. Our own explainer on water absorption goes into the full two-day lab routine; the row you want on the report is the apparent absorption value in %.

As a rule of thumb that has served me across thirty years of reports: below 0.6% is the top EN class (code A1); below 0.25% clears the American S1 gate; and the stone we quarry in Lushan sits around 0.2%, which is the same story in both alphabets. A number above 1% is a slate that will drink, freeze and crack from the inside — whatever the label on the cover says.

Rows four, five and six: strength — and the thickness that changes its meaning

Strength appears on a report under a few names — breaking load, flexural strength, modulus of rupture. They are cousins, not twins: breaking load is the raw force in newtons or pounds that snaps the sample; flexural strength and modulus of rupture re-state that force per unit of size and thickness. Different standards slice the same event differently, which is why our flexural strength page treats them as two numbers asking one question: will it stay whole when it has to?

Here is the reading trick that catches more wrong reports than any other: strength only means something in the context of thickness. The modulus formula squares the thickness — so a 6 mm slate tests dramatically stronger than a 4 mm slate of the same stone, and a report that doesn't state the sample thickness is a report you can't interpret. Our own Lushan material tests at roughly 60 MPa dry and above 45 MPa still after full water saturation — several times above the S1 floor — but that number only means anything because the report also says which thickness was on the rig.

The freeze–thaw row: what to notice before the number

Freeze–thaw is the test that decides a cold-climate roof. The lab saturates samples with water and cycles them between deep freeze and thaw, then weighs them looking for loss. Two figures matter on this row: the weight loss (you want 0.0%, or near it) and the cycle count (industry practice is around 40 cycles on the American side and around 50 on the European side). A report that says "freeze-thaw tested" without a cycle count is a report that hasn't finished its sentence — and our full freeze-thaw explainer has a ten-minute home test you can run yourself on any sample.

For scale, our own stone has been cycled to 120 cycles with zero measurable loss, which is how a roof that sits in frozen climates for decades can come back with no frost claims. That margin — 120 cycles against a 50-cycle ask — is exactly the shape of a number that doesn't just pass, but lives comfortably in the grade.

The chemistry rows: carbonate, sulphur, iron — the quiet arithmetic

Some reports include a chemical block, often near the back: carbonate (CaO), sulphur (SO₃), iron oxide (Fe₂O₃) and friends. These numbers are boring until the day they aren't. Carbonate is the mineral acid rain eats first; a stone that fizzes under a drop of dilute acid is a stone whose edge will erode over decades, and the chemistry behind it is unpacked in our carbonate explainer. The practical rule European specifiers work with is below 5%; our Lushan slate reports CaO around 0.56% — far enough from the danger line that the roof in question sits quietly in acid-rain cities for decades.

The last row on the page: thickness tolerance — where CNC hides

Most buyers never glance at this block, and it is one of the biggest tell of a professional producer. The report will give the nominal thickness and a measured range. The tighter the range, the more controlled the process: a CNC-dressed slate holds around ±0.5 mm, while hand-split material drifts by a millimetre or more. A tight thickness row is the print the quality of the rest of the operation — it says the factory measures what it makes. It is also the row that, on site, protects the installer from a wall that was pitched for 6 mm and received a mixed bag from 4 to 8.

The part nobody warns you about: four ways the numbers don't add up

A good report is internally consistent. A suspicious one trips over its own arithmetic. After a decade and a half of reading these files, here are the four disconnects I look for first:

• 1. A grade that doesn't match its own absorption. The report's summary line says "S1" but the absorption row shows 0.9%. The standard never says this — the two columns contradict each other. Either the summary is a sticker or the sample was swapped. The fix is easy: keep reading past the verdict, at least to the numbers that justify it.
• 2. Strength without thickness. The modulus row shows a healthy number but no sample thickness anywhere in the document. That is not a measurement — it is a fact that refuses to be checked. Always match the strength to the thickness row before you believe it.
• 3. "Tested" without a cycle count. A freeze-thaw row that says "passed" but omits how many cycles — a common short-hand that means the stone was sent on one winter, not fifty. The cycle count is the honest currency; demand the currency.
• 4. A date that makes the paper irrelevant. The report is beautiful, the batch number is present — but the test date is three years old, or the batch number doesn't match the number on the packing list. Paper tells a story, but a report is only a fact if the story is your story.

The three-minute reading you can do right now

Forget the lab jargon. When the next report lands in your inbox, run the six-line pass:

1. Header — laboratory, standard, sample, date, and a batch number you can trace to your order.
2. Absorption — the % itself. Below 0.6% EN top class, below 0.25% the US S1 gate, our Lushan around 0.2%.
3. Strength — flexural or modulus, and the thickness of the sample on the same line.
4. Freeze-thaw — weight loss plus a cycle count that tells you the stone was really cycled, not waved at.
5. Chemistry — carbonate CaO under 5% if the report carries the block; a fizz test in the yard to confirm.
6. Verdict vs numbers — cross-check every rating against its own rows. If they disagree, the paper is a sales sheet.

That pass takes three minutes and it will outlast any vendor's PowerPoint. And when you find a producer whose report is consistent six rows deep — whose test results are public, whose batches are numbered, whose numbers you can follow from the quarry shed to a pallet — you have found the supplier everyone else is looking for. That's what a quality evaluation eventually reduces to: trust, but verify — and verification is nothing but reading the numbers in order.

Questions buyers ask me most

Keep Reading

• Understanding EN 12326 Roofing Slate Standards — the standard behind the report: what Part 1 tests, what Part 2 classifies, and how to read the certificate.
• Understanding ASTM C406 Roofing Slate — the American grading cousin, S1/S2/S3 and the two tests that gate them.
• Roofing Slate Water Absorption Explained — the row that matters most, tested and explained in plain terms.
• Flexural Strength and Breaking Load — how the strength rows are measured, and what the numbers mean on site.
• Jiujiang Slate Test Results — our own public test data, batch-tied, for the stone behind the rows above.