Why charcoal batches vary

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.

Two batches of the same grade from the same factory will not return identical numbers, and a supplier who tells you otherwise is describing a sales position rather than a process. What follows is why that happens, and — the part nobody publishes — how much it moved on ours.

The short answer

Charcoal is an agricultural product processed by heat. The shells arrive from different trees, harvested at different ages, in different weather; carbonization runs a temperature profile rather than a set point; the binder is a proportion of a mix; and drying takes days. Each of those is a lever, and a batch is what happens when all of them land slightly differently on the same day. The useful question is not whether a supplier's batches vary — they do — but whether the supplier measures the variation and will show it to you.

Our own spread, three batches, six weeks

These are the three certificates we hold, in date order. They are not three views of one good batch — they are three batches, tested by one laboratory between June and July 2026, and the movement between them is the point of the table.

Fixed carbon moved by almost four points across six weeks. Moisture more than halved. That is the honest picture of a process running normally, and it is published here as what three samples measured — not as a range we will hold your next container to. Three certificates are three certificates. The contractual number is the ash ceiling in your grade, and it is the batch COA for your own shipment that proves it.

The levers, and who established them

The causal work here was done in the literature, not by suppliers. Three findings are worth having, each with the material it was measured on — because the levers transfer between feedstocks and the values do not.

Binder proportion is the dominant lever

A 2018 study of processing conditions found that "all briquettes properties were predominantly affected by amount of molasses used", while compaction pressure between 50 and 150 MPa had "negligible influences on all briquette properties", and drying temperature "did not have major influence on briquette properties except moisture content" (International Journal of Energy and Environmental Engineering, online 17 May 2018). Read the substrate before transferring the number: that work briquetted eucalyptus charcoal dust with sugarcane molasses, not coconut shell with tapioca. What carries across is the ranking — the binder share moves more than the press does.

Ours is tapioca at a typical 1.5–3% of the mix. Binder is also the reason a volatile-matter figure is not purely a statement about the charcoal: part of what a laboratory measures as volatile matter on a finished briquette came from the binder rather than from the shell.

Carbonization decides what there is to bind

The US Environmental Protection Agency's charcoal process document puts it plainly: "product constituents and the distribution of these constituents vary, depending on raw materials and carbonization parameters" (AP-42 §10.7, September 1995 revision). That sentence sits in a section about what carbonization yields — charcoal plus gases, pyroacids, tars and water — which is exactly the right level for this page and is not a statement that a published specification varies.

In our own ovens the run is a temperature profile across stages, not a single set point, which is the structural reason this parameter is a source of variation rather than a constant: two batches can hit the same endpoint by slightly different routes. The same document is a useful reminder that other markets do this differently — it describes US barbecue briquettes bound with 9–10% starch solutions and dried at about 135 °C for three to four hours to roughly 5% moisture, against our tapioca and our 160 °C drying stage of 75 to 85 hours. Neither is wrong. They are different products with different economics.

Drying moves moisture, and mostly only moisture

This is the finding most often overstated. Drying is not a general quality lever — in the 2018 work it touched moisture and left the other properties largely alone. That matters for reading a certificate: a moisture difference between two batches is the least alarming difference on the sheet, and an ash difference is the most, because ash comes from the shell and the process rather than from how long something sat in an oven.

Where the factory looks for it

Variation is caught where it is created, which is why sampling is not a single gate at the end. Ours is taken at random, at least five pieces per 50 kg, at 6 stages — carbonization, screening, grinding, mixing, extrusion, drying — recording weight, ash content, moisture, ash colour. Those 6 stages map onto the levers above almost one-to-one: carbonization and screening for what the shell became, grinding and mixing for the binder, extrusion and drying for what happens to the formed piece.

What that produces per batch is a file holding the laboratory COA and the factory’s own results for the batch — ash content, moisture, burning time, drop test, cracks, temperature — reachable from the code printed on your carton. Comparing two batches is therefore a thing a buyer can actually do, and the audit protocol is how.

A note on standards, and what we are not claiming

Indonesia has a national standard for charcoal briquettes — SNI 01-6235-2000, in force, established 30 June 2000, eight pages. Its title, in Indonesian, is "Briket arang kayu": wood charcoal briquettes. That is the single most useful thing about it for a coconut-shell buyer, and it is invisible if you search for it in English. We have not read it — it is a priced print-only document — so we quote no requirement from it and no density figure attributed to it. We name it because you should know it exists and know what it covers.

And the claim we are not making: nothing here says our variation is smaller than anyone else's. We have three certificates. Three is enough to show that batches move and enough to show which way; it is not enough to characterise a process, and we are not going to compute a capability index from it and present the result as a standard of consistency.