How do you tell how much gas is left in a canister?

Weigh the canister and subtract the weight of the empty can. On a kitchen scale that gives the gas left to the gram. The empty can weighs 98 to 103 g for the 100 to 110 g size, 144 to 160 g for the 230 g size depending on the brand, and about 194 to 202 g for the 450 g size. With no scale, float it in still water: the water line moves about 6 g per millimetre on a small can and about 9 to 10 g per millimetre on the larger ones.

How much does an empty gas canister weigh?#

Manufacturers publish the net content (the gas) and, on most product pages, a gross weight for the full can. The difference is the empty can: steel shell, valve and the vapour space. That is the number you subtract.

The small size is consistent: every brand here lands within 5 g of 100 g. The 230 g size is not. A Primus can carries 16 g more steel than an MSR one, and Jetboil sits between them. The 450 g size differs by about 8 g between the two brands that publish it.

Two cautions about this table. The full weights are rounded published figures, not measurements of individual cans, and none of these makers states a fill tolerance on the pages we read. And the table covers only the brands whose full weights could be read from a fetchable page. For any other can, or to remove both uncertainties, weigh one empty can of that product yourself, once, and write the figure down. If your canister prints its own empty or tare weight, that beats any table.

How to weigh a canister and read the gas left#

The method is the same one the cold weather canister guide uses for planning winter trips, and it takes ten seconds.

1. Unscrew the stove. Weigh the canister alone, with its plastic cap if you will carry it, on a kitchen scale reading to 1 g.
2. Subtract the empty can from the table, or your own measured figure. Add the cap's weight to the empty figure if you weighed with the cap on (the published full weights above do not say whether they include it).
3. Write the result on the can in marker, with the date.
4. Convert to water boiled. The stove fuel planning guide derives 18 g of gas per litre boiled for a canister stove with a lid and a windscreen, rising to 37 g in wind with no screen. 60 g left is about 3.3 litres in good conditions and 1.6 in bad.

A worked example. A Primus 230 g can reads 248 g on the scale. 248 minus 160 is 88 g of gas: about 4.9 litres boiled at 18 g per litre, or four person-days at the 1.1 litres a day of the fuel planning guide's example menu. The same reading on an MSR 227 g can means 104 g, almost a litre more.

The stove fuel calculator turns a menu into grams, so the comparison becomes one subtraction: grams the trip needs against grams in the can.

Does the float test actually work?#

Yes, and the reason is Archimedes, not folklore. A floating object sinks until it has pushed aside its own mass of water. A canister is close to a straight-sided cylinder below its shoulder, so every gram of gas burned lets it rise by the same small amount. The water line moves in a straight line between the full mark and the empty mark, and a line halfway between them means half the gas is left.

What matters is how far the line moves per gram, which depends only on the can's footprint. Using the published diameters and full and empty weights, and treating each can as a flat-bottomed cylinder in fresh water:

Three things fall out of that table.

Precision is about 6 to 10 g. If you can read the water line to the nearest millimetre on a can that bobs slightly, the small can is good to about 6 g and the larger ones to about 9 to 10 g. A kitchen scale is ten times better, but 10 g is about half a litre of boiling water with a lid and windscreen, which is close enough to decide whether a can comes on the trip.

Water temperature does not matter. The USGS density table runs from 0.99987 g per cubic centimetre at 0 degrees C to 0.99669 at 26.7 degrees C, a difference of 0.3 percent. On a 371 g can that moves the line by about a tenth of a millimetre. Salt water is a different case: it is denser, so a can floats higher in it and reads emptier than it is. Do the test in fresh water.

Another brand's gauge reads wrong. The empty line on a Primus 230 g can sits about 2.5 mm deeper than on an MSR 227 g can, because of the extra 16 g of steel on a slightly narrower base. A printed gauge, or marks you made on one brand, only belong on that product.

How to do the float test#

1. Use still, fresh water deeper than the can is tall: a sink, a pot, a lake margin out of the current.
2. Remove the stove and cap. Lower the can in upright.
3. Tip it slightly as it goes in to let out any air trapped under the domed base. A trapped bubble adds buoyancy: 5 cubic centimetres of air floats the can as if it held 5 g less gas.
4. Let it settle and read the line at eye level. On MSR's 110 g and 227 g cans, the printed float gauge does the reading for you; MSR says the 450 g can has no gauge.
5. Dry the can and the valve threads before refitting the stove.

On a can with no gauge, make your own scale once, at home. Float a new can and mark the water line with a paint pen. Float an empty can of the same product and mark that line, then copy both marks onto each new can of that product at the same height above the base. Halfway between the marks is half the gas; every millimetre down from the full mark is the gram figure in the table above.

What about shaking the can, or a dedicated gauge?#

Shaking tells you that liquid is present. It does not tell you whether that is 30 g or 130 g, and the difference between those is a two-day trip.

Dedicated canister gauges exist. Jetboil's JetGauge, for instance, weighs 85 g and has settings for its 100 g, 230 g and 450 g cans; the listing calls it compatible with all isobutane and propane backpacking canisters. Any gauge of this kind does the subtraction above with an empty weight built in. Given the table, a setting matched to one brand's 230 g can will be out by several grams on another brand's, 6 g between Jetboil and MSR and 10 g between Jetboil and Primus on the published figures. A kitchen scale and the right empty weight does the same job without the uncertainty, and the scale stays at home.

Is a part-used canister worth carrying?#

This is where the empty weight earns its keep. A larger can's steel is heavier than a smaller can's: 44 g more for MSR (144 against 100), 51 g for Jetboil (150 against 99) and 62 g for Primus (160 against 98). So:

• A part-used 230 g size can weighs less than a fresh small can of the same brand only while it holds under 38 g (Primus), 49 g (Jetboil) or 66 g (MSR) of gas. If the trip needs 80 g, a fresh 100 g can weighs 198 to 213 g. A 230 g size can holding exactly 80 g weighs 224 to 240 g. The part-used can carries the same gas in more steel.
• Two small cans instead of one 230 g size can carry 36 to 56 g of extra steel for roughly the same gas, depending on brand. Two 100 g Primus cans are 196 g of steel against 160 g for one 230 g can; two MSR 110 g cans are 200 g against 144 g.
• Part-used cans are warm weather cans. The cold weather guide explains why: propane leaves first, so a can two thirds burned is close to pure isobutane and gives little pressure near freezing. Burn the part-used cans in summer and take a full one in winter.

When is a canister actually empty?#

"Empty" on the table above means the can with no liquid left. In practice the stove gives up before that in the cold, because the remaining liquid is too cold or too isobutane-rich to push gas out at a useful rate. The site's working figure is to treat the last 10 to 15 g as unavailable in cold conditions. In warm weather you can burn a can until the flame dies.

A can that weighs at or a few grams above its empty figure is finished. One that weighs 15 to 30 g over is still useful for a summer brew or a test burn. One that the stove will not light in the cold but that weighs 30 g or more over is not empty, it is cold: warm it in a jacket or a pan of cool water, as the cold weather guide describes, and it will run again.

Check your canisters before a trip#

A five minute routine at home, with a kitchen scale:

1. Gather every part-used can. Weigh each and subtract the empty figure for that product. Write the grams and the date on the can.
2. Match cans to trips. Run the menu through the fuel calculator. Choose the can, or combination, with the least steel that covers the grams plus 15 g in cold weather.
3. Label and store. Keep part-used cans labelled and somewhere cool, as storing gear between trips sets out, and never in a parked car in the sun, for the reasons in leaving fuel canisters in a hot car.

Common mistakes#

Subtracting the net content from the gross weight on a used can. The net figure printed on the can is the gas it held when new. What you need is the can's empty weight, which the label usually does not state and the table above does.

Using one brand's empty weight for every can. On the 230 g size it varies by 16 g between brands on the published figures, nearly a litre of boiling water at 18 g per litre.

Floating the can with air trapped underneath. The domed base holds a bubble if the can goes in flat. It reads emptier than it is, by a gram for every cubic centimetre of air.

Throwing out a can that will not light in the cold. Weigh it first. If it is 30 g or more over its empty figure, it is cold, not empty.

Carrying three part-used cans on one trip. Each one carries its own 100 to 160 g of steel. One fresh can of the right size is lighter almost every time, and the stove types guide covers when a different fuel makes more sense.

Frequently asked questions#

How much does an empty 100 g gas canister weigh?#

About 98 to 103 g. Working from published full weights, a Primus 100 g can comes to 98 g empty, a Jetboil 100 g can 99 g, an MSR 110 g can 100 g and a Snow Peak 110 can 103 g. Subtract that from what your scale reads to get the gas left. Weigh one empty can yourself for the exact figure for your product.

How much does an empty 230 g canister weigh?#

Between 144 and 160 g depending on the brand, on published full weights: MSR's 227 g can is 144 g empty, Jetboil's 230 g can 150 g and Primus's 230 g can 160 g. The 16 g spread is why one brand's empty figure should not be used for another's can. The 450 g size is about 194 to 202 g empty.

How do I check the fuel level in a Jetboil canister?#

Weigh it and subtract about 99 g for the 100 g can or 150 g for the 230 g can, which are the empty weights implied by Jetboil's published full weights of 199 g and 380 g. Without a scale, float it in still fresh water: each millimetre the water line drops from the full mark is about 6 g of gas on the small can and about 9 or 10 g on the larger one.

How accurate is the canister float test?#

About 6 g on a 100 to 110 g can and 9 to 10 g on 230 g and 450 g cans, if you read the water line to the nearest millimetre. The line moves by one millimetre for every 6 to 10 g of gas because a floating can sinks by its own weight. Use fresh water and let trapped air out from under the base, or the can will read emptier than it is.

How many days of cooking are left in my canister?#

Divide the grams of gas left by 18 for a canister stove with a lid and a windscreen, or by up to 37 in wind with no screen. The answer is litres of water you can boil. 60 g left is about 3.3 litres with a lid and screen, or 1.6 in wind, which at about 1.1 litres per person per day for dehydrated meals and hot drinks is one and a half to three person-days.

Why does my canister feel full but the stove won't light?#

Almost always cold, not empty. Below freezing the liquid cannot produce enough pressure, especially in a part-used can that has lost most of its propane. Weigh it: if it is 30 g or more above its empty weight there is gas inside. Warm it in a jacket or stand it in a pan of cool liquid water, never near a flame.