The burn test, as a method

By Mohamad Sinno, Charcoal Expert · published 7 September 2026. Not reviewed by the Head of QC — the reviewer byline on our older quality pages carries the date those pages were read, and this one has not been read yet.

Everyone selling charcoal will tell you how long it burns. Almost nobody will tell you what they did to find out. This page is the second thing rather than the first: the variables a burn test has to fix before its result is a fact rather than an impression, and the procedure to produce your own number on your own bench.

The short answer

There is no published standard for charcoal burn testing, so the conditions you state are the method. Fix seven things — mass, setting, ignition, airflow, ambient, endpoint, and what you record — then run the same test on every sample you are comparing. A duration quoted without those seven is not a measurement of anything, whoever published it.

The seven variables

1. Fix the sample and weigh it. One piece, one shape, one size, weighed before you start. Mass is the variable that moves the result most and the one most often left unstated.
2. Fix the setting. An open crucible or stand, a bowl with foil, or a heat-management device. These are three different tests and they do not produce comparable numbers.
3. Fix the ignition. State the source, the duration and the state you stop at — glowing on all sides is a state you can judge; "lit" is not.
4. Fix the air. Still bench air, a draught, or drawn air at a stated rate. If the piece is in a bowl being smoked, the airflow is the session and it must be described.
5. Fix the ambient and write it down. Room temperature and rough humidity. They are easy to record and they are the first thing an argument about a differing result will come back to.
6. Define the endpoint before you start. Self-extinction, no visible glow, or a temperature falling below a threshold. Deciding afterwards is how a result becomes whatever the tester wanted.
7. Record more than the duration. Time to full ignition, whether the ember stayed even, smoke and odour, and the mass and appearance of the residue. The duration alone answers almost nothing.

Fixing them is not fussiness. Every one of the seven can move a result by more than the difference between two suppliers, which is why a comparison run under two different setups tells you about the setups.

The one published supplier procedure

Across every supplier page captured for this cluster, exactly one publishes a burn procedure with its conditions attached rather than an outcome. Oxicharcoal — trading under PT Bumi Energi Sejahtera — sets it out in four steps: "Use a single standardized cube sample (customer-provided size; typical weight ≈ 15 g)." "Place the cube on a metal crucible or heat-resistant stand on a stable surface." "Ignite the cube with a standardized flame for 2 minutes, then allow it to burn freely until self-extinguished." "Record burn time, residual mass (ash %), smoke level and odor; photograph the cube at key stages." The page fixes the shape too — "All burning tests are performed using cube samples only" (Oxicharcoal, undated — the page carries no date field).

That is a real method and it is worth borrowing: a stated mass, a stated ignition duration, a stated setting, a stated endpoint and a record of four things rather than one. Note what it still leaves open — airflow and ambient — which is why two labs following it exactly could still disagree.

One caution, and it is the reason this page exists. The same page also publishes a panel of typical burn time and typical ash. Those are the seller's expectations, not outputs of the procedure above, and we do not reproduce them here. A procedure and a claim printed on one page are not the same document.

The only stated airflow anywhere in the evidence

Airflow is the variable the whole trade omits, and the only place a figure for it turns up is smoking research rather than charcoal testing. A clinical study using a research-grade waterpipe recorded puffing continuously and reports mean flow rates of 12.11, 11.66, 11.81 and 11.24 litres per minute across its four arms, with puff volumes of 0.63–0.74 L, puff durations of 3.04–3.62 s and interpuff intervals of 21.0–27.9 s (ClinicalTrials.gov NCT05453773, results posted 31 December 2024; the study is completed).

That is not a charcoal standard and we are not proposing it as one. It is cited for a single purpose: it is the only source in this whole evidence base that says what the air was doing. If you want a bench number to bear on a real session, this is the order of magnitude the session actually operates at — and it is measured in litres per minute, which no supplier burn claim states at all.

The bench log we cannot show you

A worked example from one of our own documented batches belongs here, and we do not have one to publish. The drying log and the container loading log are not published, and no sample page of either has been supplied. What the factory records per batch is stated rather than shown: ash content, moisture, burning time, drop test, cracks, temperature.

And there is a second, sharper gap, on our own side. Burning time is one of those recorded parameters — but the conditions under which the factory measures it have never been written down for publication. No mass, no ignition duration, no setting, no endpoint. So we publish no burn duration at all, and we would rather carry that as a visible hole than quote a number the record does not support. It is an open item on our data sheet, and when it closes it will appear here with its conditions attached.

What we can state is the ignition state the factory works to, because a different test depends on it: for the drop test, a piece is heated until glowing red and held five minutes after heating before it is dropped. The piece is heated until glowing red, held five minutes after heating, then dropped five times from 1.2 m onto a hard tiled surface. It must not break. That is a stated ignition condition — just for a durability test rather than a duration one.

Turning a result into a buying decision

Run the test on two suppliers under one setup and the comparison is meaningful even though neither number is comparable to anything published. That is the useful asymmetry: your own bench, applied consistently, beats every quoted figure on the internet for the only question you actually have, which is which of these two.

Every row points back at a measured value on a certificate, which is the point: a bench test generates a question, and the document answers it. Neither works alone.