Convert g to t

Grams are the unit of recipes and postage outside the United States, and the unit nutrition labels use nearly everywhere.

A tonne is 1,000 kilograms, and it is not the same as either the US short ton (907 kg) or the UK long ton (1,016 kg). Where the difference matters, this one is sometimes written "metric ton".

Going from grams to tonnes moves the decimal point 6 places to the left and changes nothing else. There is no factor to remember and no rounding to decide: the digits stay in the same order and only their position changes.

1,234 g is 0.001234 t — same digits, moved along. That is worth knowing because it is the one kind of conversion you can check at a glance: if the digits of the answer are not the digits you started with, something other than the conversion has happened to them.

This page uses the metric tonne, which here is 1000000 g. The US short ton is 907184.74 g, 9.3 % lighter; the UK long ton is 1016046.9088 g, 1.6 % heavier. Three units, one letter between two of their names.

Where the figure came from usually settles which one it is: tonnes for shipping, freight and emissions nearly everywhere, short tons in American industry, long tons in older British sources. Write "metric ton" where it has to be unambiguous — of the three it is the only name that cannot be read as one of the others.

Grams to tonnes is two thousands, not one, and the failure mode is doing only one of them. A figure that should be 1.4 t comes out as 1,400 t or as 0.0014, and both are wrong by a factor of a thousand in a direction that is hard to spot because the digits are right. It is the same class of error as a misplaced megabyte, and it survives review for the same reason: nothing about the number looks damaged.

The defence is to write the kilogram down. Grams to kilograms, kilograms to tonnes, each step a visible division by a thousand, is slower and self-checking; a single remembered constant of a million is faster and gives no signal when it has been misremembered. Where a spreadsheet does it, the intermediate column costs nothing and makes the exponent inspectable.

Vehicle emissions are the case almost everybody meets. A rating of 95 g of CO2 per kilometre is a gram figure by design — it keeps the number in double digits where a reader can compare models — and every regulatory and corporate total built on it is in tonnes. Fifteen thousand kilometres at that rating is 1,425,000 g, which is 1.425 t in a year.

The reason the two units differ is that they are answering different questions. The gram figure is a rate and is compared against other rates; the tonne figure is a quantity and is added to other quantities. Converting the rate to tonnes per kilometre would give 0.000095, which is arithmetically identical and useless for the comparison the rating exists to support.

Manufacturing meets the same conversion from the other side. A moulded component weighs 12 g, and the question the purchasing and the packaging-waste sides both ask is how much material that is across a production run. A million units at 12 g is 12 t of polymer, and the same arithmetic run against the packaging weight gives the figure a producer-responsibility return wants.

What makes this worth care is that the per-part figure is usually rounded before it is multiplied. A part recorded as 12 g might be 11.6 or 12.4, a spread of about three per cent, and three per cent of 12 t is 360 kg. Where the total will be reported or paid against, keep the measured weight to the precision the balance gave and round once, at the end.

Environmental figures arrive as grams per litre or milligrams per litre, and the permit or the annual return is in tonnes. The conversion has two multiplications in it — concentration by volume, then grams to tonnes — and the volume is usually the larger source of uncertainty by a wide margin. A load of 0.4 g per litre across ten million litres is 4,000,000 g, which is 4 t.

The unit chain is worth writing out in full when the numbers are that large, because a litre and a cubic metre differ by a thousand as well. Grams per litre into a volume in cubic metres, with nobody noticing the mismatch, produces exactly the factor-of-a-thousand error this whole page keeps circling — and at that scale the answer is still plausible enough to be signed off.

A tonne is the wrong unit for most gram figures. 250 g is 0.00025 t; 12 g is 0.000012 t. Numbers in that shape cannot be compared at a glance, cannot be checked for a misplaced digit, and are the reason a reporting template that insists on tonnes for everything produces worse data than one that allows the writer to pick.

The rule that works is the same one the metric system uses everywhere: choose the unit that puts the number between one and a thousand. Below a kilogram, use grams; between a kilogram and a tonne, use kilograms; above that, tonnes. Where a form fixes the unit, convert into it and keep the original figure alongside — the audit trail is the grams, not the tonnes.

Multiplying a two-digit figure by a million gives a seven-digit figure and two digits of knowledge. A car rated at 95 g/km emits about 1.4 t a year, not 1.425 — the third and fourth digits came from the distance and the exponent, not from the measurement. Reporting them makes a rounded rating look like a metered total.

This is the one substantive thing to be careful about in a conversion with no arithmetic in it. The decimal shift is exact and preserves every digit it is given; what it cannot do is tell you which of those digits were ever meaningful. Round the reported tonnage to the precision the worst input had, and say what that input was.

Anybody building a total from published coefficients meets all three metric mass units in one afternoon. Fuel factors come in kilograms of CO2 per litre, electricity factors in grams per kilowatt-hour, material factors in tonnes per tonne. Each source picked the unit that kept its own number readable, and none of them expected to be added to the others.

So the first task in any such calculation is normalising the units, not doing the arithmetic. Convert every factor to a single mass unit before multiplying anything, write that unit at the head of the column, and keep the original figure and its source next to it. Most errors in this kind of work are not arithmetic errors at all — they are two correct numbers in different units, added together.

The reported figure is almost never the end of the exercise. It gets divided: tonnes per unit produced, per employee, per square metre, per unit of revenue. Those intensity figures are what a target or a benchmark is actually set against, and they inherit every rounding and every unit assumption made on the way up from grams.

That is an argument for keeping the working. A total of 1,425 t means little; 1,425 t from 15,000 units at 95 kg each means something, and it can be checked, corrected and compared with a competitor who counted differently. Where a figure will be published, publish the divisor and the unit alongside it — a tonnage with no denominator is a number nobody can argue with and nobody can use.