Specifications — every value with its method, basis and report number

A specification figure without its basis is not a specification. Ash at 2.3% means one thing on a dry basis and something lower as-received — and most sheets do not say which.

So this page starts with how to read a number rather than with our numbers. Every value below carries its basis, its test method and the report it came from, and §4 is the set of checks that makes any other sheet on your desk legible — including against this one.

§1 · Basis, before any number

A proximate analysis can be reported on several bases: dry, as-received, air-dried, or dry and ash-free. They describe the same material and they produce different numbers. On an as-received basis the moisture is still in the denominator, so every other fraction is diluted and reads lower. Ash and volatile matter both fall. Nothing about the charcoal has changed.

That is why two sheets are not comparable until you know the basis of each, and why a lower ash figure on a rival quotation is not evidence of cleaner material. It may be the same material reported differently. Most sheets in this trade do not state a basis, and a number that cannot be compared is not doing the job a specification exists to do.

We publish ours on a dry basis, which is the harder of the two to flatter, and we say so in the table rather than in a note underneath it. If you are holding a sheet that quotes a lower figure than ours with no basis printed on it, §4 is how you find out what you are actually looking at.

§2 · The specification

Two reports, shown as two reports. The proximate values come from one and the calorific value from another; both were issued by the same laboratory, both are dated, and both cover the same lot. Showing them separately rather than merging them into a single undated column is the point — each one can be checked with the laboratory by its number, which a merged table would make impossible.

The calorific row is net rather than gross. That is the same kind of choice as dry against as-received: a gross figure on the same material reads higher, because it counts the heat carried off in water vapour as well as the heat you get to use. A sheet that quotes a calorific value without saying which of the two it is leaves you the same gap as a sheet quoting ash without a basis.

The last row deliberately breaks the pattern. Moisture at packing is our own measurement on the line rather than a laboratory result, and it answers a different question from the laboratory moisture above it. One is what a laboratory found in the sample it was sent; the other is what goes into the box and starts taking water back out of the air on the way to you. Both are true, and a sheet showing only one of them is not lying — it is just not telling you which question it answered. What happens to the second figure after the container is discharged is on why your charcoal degrades in storage.

Burn duration is not on this table. It is not a laboratory result and there is no standard method behind it, so it does not belong in a column beside things that have both. It has its own section, its own protocol and its own conditions: §5.

§3 · What each parameter does to a bakhoor burn

A table without this section is a list of facts you cannot act on. Each entry below is what the parameter changes about a coal under a mabkhara, which is not always what it changes about a coal in a calorimeter.

Ash

How much non-combustible residue is left behind. Less of it means less material accumulating between the coal and the bakhoor as a session runs. But for this application the percentage is the less interesting half: how the ash forms, whether it holds together and when it starts insulating the resin are what a buyer actually notices, and none of that is in a proximate analysis. Why your charcoal ashes badly separates the two.

Volatile matter

The fraction that burns off as gas rather than glowing as carbon. For bakhoor this is the parameter that matters most, because volatiles are what produce smoke and odour at ignition — the minutes when the resin is going on and the room is deciding what it thinks. Ours is the row on this page that needs its own section, and it has one: §6.

Fixed carbon

The glowing-combustion fraction, and the one most often sold as a grade. More of it means hotter and longer. For bakhoor that is a variable to control rather than to maximise, which is our position and is argued in full, with its weakest link marked, on which charcoal is correct for bakhoor. We do not restate the argument here; this page gives you the number it applies to.

Moisture

Water in the material. It drives cracking and popping at ignition, because trapped water flashes to steam inside a cube — why your charcoal cracks or sparks covers the mechanism and the test that tells a factory fault from a storage one. It also comes back: porous carbon reabsorbs moisture from humid air after packing, which is why the figure at packing is a starting point rather than a promise about month four.

Calorific value

Total energy per unit mass. It is useful for comparing one material against another and weakly predictive of how a cube behaves on a mabkhara, which is a session of controlled heat rather than an exercise in maximum output. We publish it because it is on a report we hold, not because we think it is the figure to choose on.

Binder

Ours is native tapioca starch, with the paste held at 80 °C for 30 minutes. We publish no ratio: the sources disagree, the published research spans a far wider range than any of them, and the ratio is not the variable that decides whether the binder was cooked through. The temperature is, which is why that is the figure we commit to — why your charcoal smells sour sets out what turns on it, why a paste held below the gelatinisation conclusion leaves raw starch in the cube however long it is held, and what our own figure does and does not establish.

§4 · How to read any supplier's sheet

Four checks. Run them on this page as readily as on the quotation beside it — the value of a check you can only apply to other people is nil.

1. Does it state a basis? Dry, as-received, air-dried, dry and ash-free. If the sheet does not say, the numbers on it are not comparable to anything, including to each other across two revisions of the same sheet.
2. Does it name a method? ASTM D1762 is the charcoal-specific proximate method; ISO 1171, 18123 and 18134 are the biofuel equivalents for ash, volatile matter and moisture. A figure with no method beside it is an assertion rather than a measurement, whoever is making it.
3. Is there a report number and a laboratory you can call? An accredited laboratory will confirm a report it issued. Carsurin, Sucofindo, SGS and Intertek all will. A sheet with a laboratory's name but no report number is a sheet that cannot be checked.
4. How old is it? A dated report from this quarter indicates routine testing. An undated one, or the same one reused for years, indicates a single flattering result that has been doing a lot of work.

The conversion you will want, and why we are not giving you a formula

As-received ash and volatile matter are lower than the same material reported dry, because the moisture is still in the denominator. That relationship is all you need to know that a lower figure on an undated sheet with no basis may be nothing but a reporting choice. The exact arithmetic needs that sheet's moisture figure, which is usually the number they have not given you — and if they do give it to you, ask which basis that was on before you use it.

Send those four questions to whoever is quoting you. The answers sort suppliers faster than any number does: a factory that tests routinely can answer all four from the filing cabinet, and a trader who bought a container on description cannot answer any of them.

§5 · Burn duration — the protocol, then the results

There is no ISO, ASTM or EN standardised burn-duration test for incense charcoal. Not a disputed one, not a difficult one — none. So every burn-time figure in this trade, ours included, is unstandardised unless it states its conditions, and a number printed without them tells you nothing you can check. Ours are printed first.

The protocol — our proposed method

1. Ignite one piece on an electric burner rated 600 W, and run the burner for 10 minutes.
2. Switch the burner off. Everything timed after this point is the charcoal sustaining itself.
3. Time from ignition to complete ash with no residual heat — not to mostly ash, and not to the moment the glow stops being easy to see in a lit room.
4. One piece at a time, one format at a time, from the same lot.

The burner ignites and is then switched off. It supplies no heat during the measurement.

That sentence is the whole credibility of the numbers below. A burn time measured with the element left on is a measurement of the burner, not of the charcoal, and it can be extended indefinitely by turning the dial up. If a supplier quotes you a burn time, the question that settles what it is worth is whether their heat source was still on.

The results

One dimension belongs beside that table rather than in it, because it is the only geometry a buyer in this trade specifies by number. Discs are pressed at 33 mm and 40 mm standard; other sizes on request — the two sizes the Gulf shelf already sells, which is why we press them, and why the format argument on the quick-light comparison can be made at all. One grade of charcoal throughout: the diameters differ in geometry and not in formulation.

Which of the two the disc figure above was measured on, we do not know. The diameter was not recorded at the time and we are not going to reconstruct it from memory, so the row stays one row and this paragraph is the caveat rather than a footnote somewhere quieter. It matters because the other property we measure by format does differ between the two: time to full white-ash coverage is five minutes apart across them, which is published per diameter on the page that owns it. Burn duration may well differ the same way. Until it is measured per diameter, treat the figure above as one disc rather than as both, and the safe reading for a purchase order is the shorter expectation.

All three are the same material — the lot covered by both reports in §2 — and the spread between them is most of a factor of two. That is the argument for stating a format: a single burn-duration figure covering a whole product line is either the best of three or an average of things that do not average, and either way it is not the number for the format you are buying.

What this does and does not tell you

It is a total, not a profile. A coal that runs to the end of a line in that table is not thereby holding a useful temperature for the whole of it, and for bakhoor the window that matters is the plateau rather than the total — that argument, and the window we build to, are on the page that owns them. Total burn time tells you how often a customer reaches for the next piece. It does not tell you what the resin is doing while they do not.

The protocol is ours rather than anyone's standard, so treat these as a baseline for comparison rather than as a specification. Run it on your current supplier's cube on the same burner, with the same ignition period and the same end-point, and the comparison is worth more than either number alone. That is the only use we would make of it ourselves.

§6 · Volatile matter, and where ours sits

This is the row that does not flatter us, and it is the parameter this whole site argues bakhoor cares about above the others. So it gets a section rather than a row.

Ours is 13.7% on a dry basis. The ceiling in the national briquette standard at our origin is 15%, and the range usually quoted for well-carbonised coconut shell is 8–12%. Ours is inside the ceiling and above the typical band. Both of those are true at once and we would rather print both than the one that reads better.

Why it sits there

Our carbonisation runs 400–650 °C, and that is a process choice rather than a ceiling we run into. Volatile matter at this level is a consequence of that range. Carbonising harder drives volatiles down and fixed carbon up — that relationship is not in dispute, and a supplier who wants a lower number on this row knows exactly how to get one.

The reason we have not is that the same move raises fixed carbon, and a higher fixed-carbon coal runs hotter and longer. That is the direction the rest of this site argues against for a resin: heat that climbs past the band where oud volatilises does not make the bakhoor stronger, it makes it briefer. The argument, with its weakest link marked, is the page under all of this.

What we are not claiming

That this figure is the optimum for bakhoor is our position, and it is unmeasured.

We have not tested what volatile matter does to what a buyer smells in the first minutes of a burn, and we have not found anyone who has. What we have is a process choice, a reason for it, and a figure that follows from it — which is an argument rather than a finding, and it is exactly as strong as the fixed-carbon argument it rests on. If that argument is wrong, this row should be lower and we would want to know.

What you can do about it meanwhile: volatile matter is the parameter behind an acrid, smoky note at ignition, so the test that matters to you is olfactory rather than numeric. Light one cube with nothing on it and smell what comes off in the first minutes — that is the protocol on why your charcoal smells sour, and it will tell you more about whether this material suits your blend than a comparison between two percentages will.

§7 · What these reports do not cover

Between them the two reports give a proximate analysis and a calorific value. That is a description of what the material is made of. It is not a description of how it behaves in the only place that matters to a bakhoor buyer, and no laboratory we could send it to would produce one.

• Odour at ignition. No standardised method exists. The sealed-jar and coal-alone tests on the diagnostic pages are ours, proposed as methods rather than published as standards.
• Ash behaviour over a session. The percentage is measured; how the layer forms, holds and insulates is not, and that is the half buyers describe when they complain.
• Burn duration. Not in either report, because there is no laboratory method for it. §5 is ours instead, with its conditions printed above its numbers.
• Anything about a named competitor's material. We have not tested it and we do not publish comparisons against suppliers by name. The checks in §4 are what we offer instead, and they work without us.

The gap between what a report measures and what a buyer experiences is the reason this site carries buyer-executable tests at all. Run them: the fault diagnostics index lists each one with what it can and cannot tell you.

§8 · Questions buyers actually ask

Another supplier quotes lower ash than you. Is their charcoal cleaner?

Check the basis before you conclude anything. If their sheet is as-received and ours is dry, the two numbers are answering different questions and the gap between them may be moisture rather than ash. If their sheet states a basis, a method and a report number and still comes in lower, then it is a real difference and worth asking us about.

Can I see the certificate of analysis?

Ask and we will send it. We do not publish the document itself on this site: the copy we hold carries a buyer's name and a lot number that belong to somebody else's shipment. The report numbers are in the table above, and an accredited laboratory will confirm a report it issued to whoever holds the number.

Are these the numbers my container will have?

No, and no honest sheet promises that. These are measurements of particular material. The figures that govern your shipment are the ones on the certificate of analysis issued for your lot, which is a document you should be asking every supplier for by name rather than accepting a web page in place of.

Why is fixed carbon calculated rather than measured?

Because that is what the proximate method does: ash, volatile matter and moisture are measured, and fixed carbon is what is left when you subtract them. It is stated in the basis column rather than tucked underneath the table, since a calculated figure and a measured one are different kinds of evidence and a buyer is entitled to know which one they are reading.

Can I use your burn times to compare against another supplier's?

Only if theirs came from the same conditions, which is unlikely enough that the honest answer is no. Run §5 yourself on both cubes instead. Two numbers you produced in one afternoon on one burner are comparable to each other, which is more than can be said for two numbers from two sheets.

Reviewed by Mohamed Noor, Director — 23 September 2026

§9 · What supports each value on this page

The table in §2 is third-party data with its sources named, which makes most of this ledger short. The rows worth reading are the ones where we stop reporting and start reasoning: the burn-duration protocol is ours, and so is the claim that our volatile matter is where it ought to be.

What this page does not publish: a binder ratio, for the reason in §3; a figure for what volatile matter ought to be, because we have not measured it; and any comparison against a supplier we have not tested.

§10 · Test it against the sheet you already hold

Put this page beside the specification you were sent last week and run §4 down both. If the other sheet survives it, you have a supplier worth keeping and you have lost nothing by checking. If it does not, you have learned something that no number on either page would have told you.

Then do the part a specification cannot do for you, which is burn the material. The figures above describe what our cube is made of; only a burner tells you what it does with a resin on top of it.

Request the 1 kg sample, and check our numbers against your own bench