Nine checkpoints from shell to seal

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

Every charcoal factory publishes a step count. Almost none publishes what is measured at each step, and none of the ones we looked at publishes what is written down. That second thing is the only part a buyer can ever check, so it is what this page is.

The nine checkpoints

They run from the gate of the carbonization plant — before the material is ours — to a random audit of cartons that are already packed. Six of the nine are also formal sampling stages, marked in the last column: at those, pieces are pulled at random, at least five pieces per 50 kg and weight, ash content, moisture, ash colour are recorded per sample.

Checkpoint one is worth pausing on because it is a rule about buying rather than about testing: no copra flesh, no young coconut, mature on-tree shell only. Most defects further down this list are cheaper to prevent there than to detect at seven.

Checkpoint-by-checkpoint criteria, sampling and what a failure does are set out on the in-line checkpoint reference.

One number we use and do not publish

What is written down, and what is not

Per batch the file holds the laboratory COA and the factory’s own results for the batch: ash content, moisture, burning time, drop test, cracks, temperature. Two of those have written methods you can repeat: the drop test — 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. Hexagonal: three drops from 50 cm. — and the in-house moisture check, drying oven at 120 °c for 5–10 minutes, which is a quick check rather than the constant-mass method a laboratory uses.

The file is reachable from the goods: The batch code identifies the production batch whose QC file holds the laboratory COA and the factory’s own results. The code itself is a four-digit batch production code, two letters identifying the packing team, and one check digit.

And here is what we cannot show you. The drying log and the container loading log are not published, and no sample page of either has been supplied. The natural thing for a page like this would be one batch's complete QC file, walked page by page with the commercial details redacted. There is no such published file, and producing a redacted-looking one would be the fabricated evidence this whole section argues against. The absence is stated instead of filled.

What is published, and checkable against a named laboratory, is the certificate for each batch — 3 of them are on the certificates page, with the routes to verify one.

The other thing that survives a batch is physical rather than written: boxes are drawn from every production tonne and kept, so a shipment can be compared against what it was made from years later. What that archive can settle and what it cannot — the two figures that keep in open storage and the one that does not — is at retained samples and per-batch COAs.

When a batch fails

The rule is one sentence and it is deliberately narrow: when an in-process check fails — moisture, drop test, cracks, odour — the batch, or the failing part of it, is rejected and does not ship. We claim no re-drying-and-retest route, because we do not run one.

That is stricter than the market convention, and the difference is worth naming rather than boasting about. Suppliers commonly describe reprocessing a failed batch — adjusting parameters and running it again — and one publishes a drop-test pass threshold at a stated fraction of the sample rather than a whole-sample criterion. Those are their stated practices, dated, and they are not ours. A route we do not operate is not a route we will describe.

What this system is not

It is not a certification. This site holds no quality-management certificate and claims none. That matters on this particular search, because certification claims are common here and they are not all the same claim: one source we read describes its factory as operating under standards-guided procedures in one paragraph and as certified in another. Guided-by and certified are different words with different evidence behind them, and a page that uses them interchangeably has told you something about its own care.

It is not a step count competing with other step counts. Nine is what P22 enumerates, from a gate we do not own to cartons already sealed. It is not larger or smaller than anyone else's number in a way that means anything, because nobody counts the same things. What is comparable is the middle column of the table above — and that is the column the rest of this category leaves blank.

It is not an audit. A system described by the party operating it is a description. If you want it verified, the routes are a third-party inspection at loading, a laboratory of your own choosing on a sample you drew, or a visit — we host both site visits and video walk-throughs of the line.

What it means for a reorder

A checkpoint system produces evidence, and evidence is only useful if you can tie it to the goods in front of you. The practical sequence is short: read the code off the carton, ask for the file that code opens, compare it against the certificate, and repeat the two checks you can run yourself. That is set out step by step in the batch audit protocol, and why two batches of one grade differ at all is traced here.