Finally it's snowing in California so I'm thinking of winter trips. Bringing me back to deciding whether I 'need' a new winter stove.
So I did some tests to try and approach a decision on upright canister vs. inverted, and if upright, do I keep my existing Jetboil or get something with more heat.
First, over simplifying some of the basics as I comprehend them: (My apologies for any confusion due to my mixing of units). If I got any of this horribly wrong do tell and I'll edit (with attribution of course) to reduce confusion.
Upright canister stoves use a pressurized flammable gas (liquid under pressure, gas in use). They shoot it through a small orifice, mix it with air, then burn it. Pressure differential between the two sides of the orifice determines flow velocity through the orifice. The orifice size (cross sectional area) times the flow velocity determines the gas flow volume. A valve regulates the pressure at the orifice from 0 (valve off) up to a maximum of the pressure in the fuel can (presuming that the valve and fuel lines are oversized compared to the orifice.) Normally the air inlets are non-adjustable.
Pressure in the fuel can is based on the boiling point of the gas(ses) and the temperature. Pressure outside the fuel can is... whatever the local atmospheric pressure is, and varies mostly by altitude, slightly by weather/temperature.
Commercially available stoves are designed to not explode or flame out at the upper range of possible environmental conditions. If too much gas is delivered it won't mix with enough air before reaching the burner and the flame will "lift off" or blow out, resulting in a dangerous condition of how stove and unburned, heavier than air, fuel. The way manufacturers generally do this is by sizing the orifice so that at the upper pressure range it doesn't allow an excessive amount of gas to flow.
Another method, standard on all modern propane stoves, is to use a pressure regulator. In the US, domestic and RV propane systems use either a single or two-stage regulator. The first drops from tank pressure to about 10 psi (69 kPa), the second drops to "11 water column inches" or 0.4 psi (2.7 kPa). Camping stoves and BBQ's generally have a single stage "high pressure" regulator which drops pressure to about 10 psi. The regulator uses a spring loaded diaphram to limit pressure on the downstream side.
Welding gas and air compressor regulators are somewhat similar but are adjustable.
A few butane backpacking stoves use a regulator- MSR Reactor, Soto Regulator, and Jetboil Sol are three. With a stable pressure differential between the valve and atmosphere, the designer of the stove doesn't have to compromise the designed orifice size/air inlet/burner configuration based on the maximum working canister pressure.
A final point on regulators- it's not a pump so it can't increase pressure from the can to the valve. It won't make isobutane jump up and dance at -10.
Bla bla bla.
What it comes down to is delivering butane gas and air to the burner. At low temperatures it means low pressure gas which means you need a bigger orifice to get much heat.
At 68F/20C a 30/70 propane/isobutane mix sits at about 4 atmospheres (bars) of pressure (per http://www.bushwalking.org.au/FAQ/FAQ_Mixtures.htm read Roger Caffin's whole article if you haven't already) At 32F/0C it drops to 2.5 bar. Under cold use the propane boils off much faster than the isobutane, so we need to see the pressure of pure isobutane or maybe a 10/90 formula. Pure isobutane is at 1.6 bar, 10/90 is at 1.8. Then of course at 12F/-11C the isobutane is at 1 atmosphere and happy to take a nap in an open container at sea level.
Remember, it's not the absolute pressure that's important, rather the differential between the fuel and atmospheric pressure. At sea level, room temperature there is 4-1= 3 atmospheres or roughly 45 psi pressure differential. At freezing (32F/0C) there's only 0.6 atmosphere or 9 psi pushing that isobutane out. Get up in altitude to 3,000 meters and the basic barometric pressure drops to 70% so you get 0.9 bar 13 psi.
OK. On to the D&#* Jetboil PCS. It was designed to be super efficient. The burner output is low so it doesn't waste energy up the sides of the pot. Also because of the built-in wind screening effect of the fins and base plate, too much gas would probably flame out. Finally the neoprene cozy would get burned up if too much heat blows past.
I'm not sure of the jet size, but the jet does have "21" stamped on top so I think it may be 0.21 mm. This is based on the Jetboil website mentioning that the Helios jet is 0.25 mm, and the Helios is close to double the PCS output.
All this means that the rated power level is low. Originally advertised as 5200 btu, I seem to recall one of the ads for the Sol mentioned that the PCS is 4500 btu. And it is probably rated at STP (70F, sea level), with a fresh can of "Jetpower" Isobutane/propane mix. That means the rating is based on having 3 bars, or 44 psi pressure across the jet orifice. At freezing (32F/0C), with a partly depleted canister the pressure differential drops to 0.6 bar or 9 psi. Basically 1/5 of the rated amount.
At least that's my theory. Time for a test. I don't think air temperature is a critical part of the test. Most important are canister temperature and cooking water temperature/volume.
I had a mostly used Snopeak GigaPower 110 gram fuel can (24 grams fuel remaining). According to the Snowpeak website this fuel is 85% isobutane and 15% propane, though this can is a couple years old and some older reports had it at 80/20.
Conditions of test:
55F garage, no wind.
500 ml of ice cold water (32F/0C) per run, strained from pitcher filled with ice cubes and water.. PCS pot pre-chilled with ice cubes and water between runs.
I measured the time to raise temperature to 80C, using a dial coffee steaming thermometer.
Fuel consumption amounts using an empty fuel canister to set my scale tare. My scale is probably really only accurate to about 2 grams.
First test was mostly to attempt to boil off as much propane as possible from the canister. I used a Brunton Stove Stand to more easily maintain canister temperature.
Fuel Can attached to stove stand and a bowl of ice cubes was stored overnight in a -10F/-23C freezer, then wrapped in a towel for the test to maintain cold condition.
14 minutes 80C° temperature rise
4 g fuel used ending with 20g remaining
The flame was seriously weak! If outdoors any tiny bit of wind would have blown most of the heat away. Extrapolating from this test, going from snow to 0.5 liter of boiling water would take more than 1/2 hour!
Next test to see how the can would perform right at freezing, now that most of the propane is gone.
Cold 0C canister in ice and water bath
6:04 minutes 80C° temperature rise
5 g fuel used with 15 g remaining
Way better, but still weak. The canister might not have fully warmed up to 0C at the start. It did have a thin coating of ice on the bottom and one side after the test.
Next test with stove upright (no Brunton Stove Stand)
Cold 0° canister in ice and water bath
6:10 minutes with 80C° temperature rise
5 g fuel used with 10 g remaining
Next test to see if under warm conditions such an empty can would still deliver
Warm 25C° water bath
3:05 minutes with 80° temperature rise, 4:00 to full boil, canister ran out at 8:32
Final test with a half full canister (63 grams at start) in icewater bath
3:40 minutes for 80° temperature rise, boil at 4:30 minutes (My timing might be off on this run)
Back to my theory- that the pressure at 0C with only isobutane would be 1/5 of room temperature on a full can. I think this is off a bit as my 0C tests resulted in only double the heating time as the 25C test (though the 25 C test was on an almost empty can.) I may re-do the "standard temperature & pressure" test with a full can.
I do think that an upright canister could be a good performer in cold weather if you can keep the canister in a liquid water bath- even if that water is on the verge of freezing. The water acts as a big heat sink to keep the canister at a stable temperature.
In order to do this, one needs a bowl sized for the canister of choice. It seems that it would be best to have an insulating cozy, flat bottom, and perhaps three small feet so it isn't wobbly.
The questions I'm left with- Would replacing the jet with a larger orifice (maybe 0.25mm?) significantly boost winter performance? Some of the small canister stoves are rated at 9,000 btu or double the PCS. My PCS valve body has Primus stamped on the side. Do all Primus jets have the same threading? Is there a catalog somewhere?