Cold weather experience w/ a canister stove

[Mark Ferwerda]

After reading Roger Caffin's series on how well canister stoves can work in the winter I thought I'd try it out. This past weekend a friend and I headed to Canaan Valley in WV to do a 3 day backpacking trip. I bought a MSR Windpro and a new Coleman Cannister cartrige (70% butane/ 30% propane), First night was mid/upper 20s and I rigged the cannister for liquid feed (attached some legs to the cannister so it would be upside down), and proceeded to heat some water for supper. It worked well. I left the stove together for the night and the temps dropped to @ 10 degrees. For breakfast, I light it up and it working great, for the 1st pot of water (pot is the infoamous Wal-Mart Grease Pot). I also needed to melt some snow to have water for the day so I leave the stove on and proceed to melt another potful of water. By now the stove is definitely slowing down. I open the valve a bit to get more flame. I get enough water to fill my 1st bottle with water and proceeded to try to melt enough snow for 1 more bottle of water. By the end there was barely a flame with the valve wide open. I checked the cartridge and there was plenty of fuel left. Once I got home I checked the stove with the same canister and it worked great. Is there a minimum temperature that this setup will work at? I checked Roger's article and I couldn't find a minimum temperature rating, just that it will work in "really cold" weather. Anybody have an idea what happened or have a similar experience? Thanks!
Mark

[Brian James]

I had a Primus 1lb canister sputter out when it was 2/3rds full — sitting on my kitchen table! This blew my mind — after all the only time I would carry a huge 1lb canister is if I was very isolated or in very cold conditions or both! I would have been screwed. The canister came from MEC and it had a spot of rust on the threads, making me think it was old.

I chalked it up to a valve problem. I could hear lots of liquid inside but I never broke it open to see what was going on for obvious reasons.

This question makes me wonder even more why Andrew Skurka didn't carry a canister stove on his "ultralight in the icebox" trip. Canisters are lighter and ostensibly *less* prone to failure than a complicated pump system. What gives?

How many other people here have had canisters fail with liquid left in them either a) in warm conditions, or b) inverted in cold condtions?

Brian

[douglas ray]

It sounds to me like their might have been a drop of moisture in the system somewhere that froze and plugged something. Moisture in the valve mechanism perhapse? The other theory would be that parts of the valve are made of dissimilar materials that contract at different rates and at some point don't fit together well enough to work.

[Brian James]

I think the huge question here is whether the canister worked when you got home?

After all, it's not like the canister worked fine for every use and then one time it was screwed on and nothing came out. It stayed connected (i.e. valve in the open position) and sealed, and yet the fuel supply just "petered out." That doesn't sound like a materials or blockage problem. That sounds like a problem with the performance of the fuel itself.

[jim bailey]

Have been using a MSR Pocket Rocket for the past few winter’s with a great deal of success. Usually only out for 1 or 2 nights. Have found that using a homemade canister cozy makes a huge difference also making sure fuel canisters are buried in pack and insulated from the cold. Have seen others run into the same problems described by others and it is usually due to canister being exposed to cold temps for prolonged periods. Have also witnessed canisters come back to life on cold trips by being warmed inside an insulated jacket for about 20 minutes. This might be due to fuel mixture as well, have had a lot of success with iso/butane mixture from MSR. Would definitely give the canister cozy a big thumbs up and try that.

[Roger Caffin]

Several comments, which I hope may help.

The Coleman fuel is 70% butane and 30% propane. This is fine for warmer weather, and will work in cold weather , but I would recommend using a fuel which was 70% ISO-butane and 30% propane. Many other brands offer this.

Reason: Butane boils at -0.5 C (~31 F). Iso-butane boils at -12 C (~11 F). The lower boiling point helps. But propane boils at -40 C (-40 F) or maybe a little lower even, so a liquid-feed canister stove can work down to nearly -24 C as the canister will be pressurised. Warm the canister up a bit and it is fine at even lower ambients of course. Me, I might not be so fine at -40 F though … :-)

Added by edit: why won't the canister work down to -40 F? Because the higher vapour pressure of the propane is diluted by the presence of the butane. Both the Powermax and the Kovea 30%propane/70%isobutane will kark at around -24 C (about -10 F) when the internal pressure drops to about one atmosphere.

I have used the MSR WindPro myself in the snow with an inverted canister, and I found that the valve sometimes blocked up exactly as you described. Annoying. I was able to clear it every time by opening and shutting the valve at least half a turn each way. Cook, wiggle, cook, wiggle …

What is causing this valve problem? I don't know. The canister does contain a small amount of a smelly chemical to warn you of leaks (often a mercaptan), and possibly this was gunging up at the valve, or even freezing around the valve. The Fyrestorm and the Xtreme do not do this in my experience – no idea why not!

Note: this problem is completely different from the problem people have trying to use an upright canister stove in the snow. In that case the butane stops boiling and the propane gets used up fast, eventually leaving almost pure butane.

You are welcome to look at http://www.bushwalking.org.au/FAQ/FAQ_GasStoves.htm
where I have a lot more info about gas stoves and so on. Not part of BPL – I wrote these pages before I joined BPL, but maybe they will help too.

cheers

[Mark Ferwerda]

I'm posting this as a follow up to my 1st posting.

Back home in Maryland we are having an unusually cold spell here so I thought I'd try the same set-up again (same stove, same cannister). Last night was about 10 degrees (Fahrenheit) and I left the stove out for about 3 hours before I fired it up. I put a pot of water on it and checked it every 5 minutes or so for 30 minutes. It worked like a charm, never had to adjust it at all. I left the stove out overnight and fired it up again this morning (5 degrees now). Let it run for another 30 mins and it works great again.

So now I'm trying to figure out what is the difference? Cannister is less full than before but I wouldn't think that would be a big deal. I do remember that on the trip when I put the 1st pot of snow on the stove there was snow on the outside of the pot and when it melted the water dripped onto the stove and put the flame out, but it started right up again with no problem. Maybe this was a one time deal but something did cause a problem.

Anyways, thanks for all the comments and ideas that you provided. This was my first time backpacking in snow and I learned alot from it, and look forward to my next trip!
Mark

[s e]

Roger is right on! It is not likely the cylinder is faulty, it just cools as it is used just like hairspray or spray paint can. I posted this on 11/15/06 for propane:

You must remember how propane works. The pressure in your tank/cylinder is directly related to the temperature of the propane. At 60F, the pressure is about 102psi, at 20F about 47psi and at -20F it is about 13.5psi. Propane at atmospheric pressure instantly goes to -44F and stops boiling hence no more vapor to build pressure. If you consume pressure faster than the propane boils, you will eventually fall below the regulator output pressure required to run the unit (about 10-14"wc or 1/2psi). This is why you see condensation or even frost on a propane tank/cylinder under a large useage load. Propane would be used for refrigeration if it was not so unstable and flamable.

What do you do when you need it to work? First thing is make sure the tank/cylinder is open to air and not in snow or insulated in anyway. The propane needs to be a higher temp than -44F to produce vapor pressure. DO NOT TRY TO HEAT A TANK/CYLINDER! PRESSURE BUILDS FAST (@100F, it is at 220psi) AND OPEN FLAME AROUND IT IS A BAD IDEA! THINK SAFELY PLEASE! My grandfather always told me that those that did not respect electricity were self eliminating problems. There are many many energy sources we can say the same thing about and propane is no different.

Try to decrease your consumption rate and if the unit stops functioning properly, the propane must find some safe warmth like the sun or just a break to warm up to the outdoor temp. Even if it is -20F, once the propane gets back up to -20F, it will have some pressure back in it.

Edited by smilingbear at 11/15/2006 08:16:54 MST.

[Roger Caffin]

> the propane must find some safe warmth like the sun or just a break to warm up to the outdoor temp.

I usually allow the canister to get some radiation from the stove to keep it warm. Now, this may sound dangerous but it is not IF you follow this rule scrupulously:
Monitor your stove at all times, and monitor the temperature of the canister at all times. If you can happily touch the canister it is safe.
You start to get an 'ouch' response when the surface you're touching gets much over 40 C (104 F). The canisters are safe to considerably higher temp than that – they have to be by law.
I used to think that the only temp to worry about was at the high end, but of course you also have to worry about the low end. If your skin starts to stick to the metal, then the canister is too cold! Yeah, well, problems there too!

But I repeat: the valve on the WindPro is not designed as an expansion valve the way the Xtreme valve is, and it seems to have some problems with blocking up. You have to watch out for this.

Now, why didn't Mark have a problem at home later on? It may be that the canister was so COLD at home that the valve did not act as an expansion valve. The boiling may have been happening downstream nearer to the stove. I have seen this happen too. In this case whatever had been gunging up the valve before would now be happening somewhere else, and possibly was swept away safely into the flame. Dunno yet.

[Tony Beasley]

For my interest I recently wanted to get an idea of what pressures were approximately inside a canister at cold temperatures and I thought that the results might beof some interest in this debate.
I found an old psi (gauge pressure) pressure gauge (not calibrated) made an adaptor for the gauge to go on a canister, purchased two new 8oz/227g canisters one 20% Propane 80% Iso-Butane the other 30% Propane 70% Iso-Butane, I then placed them in the different temperature compartments of my new fridge/freezer which has digital temperature readouts. The freezer is at -20 degrees C and the fridge is at 4 degrees C, the fridge also has a 0 degrees C compartment. (for interest the refrigerant used in the fridge is Iso-butane). The testing was conducted at 600m (2000ft) elevation.

Results

-20C/-4F: 20/80 0 psi/0 kPa on gauge
30/70 5 psi/34.7kPa

0C/32F: 20/80 8psi/55.16kPa
30/70 19psi/131kpa

4C/39.2F: 20/80 11psi/75.84kPa
30/70 26psi /179.3kPa

Tony

[Michael Martin]

Tony-

That's great information! If you ever have a chance, I'd love to see how a canister with n-Butane rather than Iso in the mix fares.

btw, please double check your mixture percentages — I think you might have reversed the propane fractions.

Cheers,

-Mike

[Tony Beasley]

Thanks Michael for pointing out the error very careless of me, I have corrected the mistake. I will look at doing the test with n-Butane when I can.

Tony

[Michael Martin]

Sorry…I didn't mean to nitpick!

Really, that's a great experiment, Tony. Thanks for sharing your results with the forum.

Cheers,

-Mike

[Roger Caffin]

Tony makes a very good point here:
> one 20% Propane 80% Iso-Butane the other 30% Propane 70% Iso-Butane
The canisters with the low percantages of propane will die much faster in the cold. I checked the figures for -20 C and 20/80, and yep, that canister dies.

This is why I love the Xtreme stove and Powermax canisters: the fuel mix is 40% propane and 60% butane. The butane won't be doing very much at all in the way of pressure at -20 C (it boils at ~0 C), but the high propane fraction means it will drive fuel out of the canister at quite low temperatures, even below -20 C. Of course, once the fuel gets to the stove, the butane burns very nicely.

It does not hurt to keep any fuel canister 'warm' inside your jacket in the winter, where 'warm' is a relative term of course. :-)

[Alec Muthig]

Rather than speak from the science side of things, I'll speak only from experience.

My old Primus canisters would fail once it hit freezing. They were not inverted so may have been much more reliable if they were. To get them to work, even on some Rocky Mountain summer nights, I'd have to nurse it along.

My Coleman Xtreme canisters were a huge step up and worked to temps below 0 (f), although moisture sometimes got into the connector and made connection difficult when it froze up. I love this stove and fuel but no longer trust it at really cold temps (-10 or colder).

I just got back from the Arrowhead race in Northern Minnesota – during that nasty little cold snap and am very glad to have had my XGK. It was colder than -30 over -45 with the wind chill. Propane stoves were failing on the trail.

Out in the wilderness in the extereme cold, I don't want a fuel that I have to nurse, or keep in my coat, but rather something that will fire right up and keep me from losing my toes (or my life).

[Brian James]

"Out in the wilderness in the extereme cold, I don't want a fuel that I have to nurse, or keep in my coat, but rather something that will fire right up and keep me from losing my toes (or my life)."

Amen! If full canisters are reading 0 or 5 PSI at 0F/-20C, what about a half-full canister that's been evaporatively cooling for 10 minutes at 20F/-7C? I bet it would get down close to the 0 PSI mark too if you didn't get some heat into it.

Could you run more tests? For instance, test a 1/2 empty canister? 3/4 empty one?

Very informative.

[Roger Caffin]

> My Coleman Xtreme canisters were a huge step up and worked to temps below 0 (f), although moisture sometimes got into the connector and made connection difficult when it froze up. I love this stove and fuel but no longer trust it at really cold temps (-10 or colder).

You are right: the gauge pressure inside either a Powermax canister or a Kovea 30%propane/70%isobutane canister will drop to zero around -10 F.

Pure propane (LPG) would function down to about -35 to -40 F, but the bottles are heavier. Pity.

Mind you, priming an XGK at -40 F is going to be fun – just getting a flame at that temp is not that easy for start!
Cold!

[Tony Beasley]

Hi Brian,

I think that you have a very good point with pressures in partly full canisters at cold temperatures but it is not as straight forward as just putting a partially full canister into a fridge/freezer and then testing the pressure. If the canister is very cold most of the Propane boils off first but in the warmer uses both gasses are used. Roger Caffin is the authority on this subject he has written many excellent articles in BPL and on his http://www.bushwalking.org site.

I do have some partially full canisters which I will put into the fridge/freezer and then test but I do not keep a register of use of my canisters and I don’t know if the canisters have been used in the cold or warm or both. I will post the results soon anyway.

I would like to test a 40% Propane 60% iso-butane if I can find where to buy them here in Australia

I have gone away from upright canister stoves in cold weather to upside down canisters feeding liquid fuel to a pre-heating tube and then to the burner jet for reasons explained in Roger Caffin BPL articles.

Selecting a Canister Stove for Cold Weather Backpacking
Part I: Stove and Fuel Fundamentals
Selecting a Canister Stove for Cold Weather Backpacking
Part II: Commercially Available Canister Stove Systems

I hope to post pictures of my latest liquid feed canister stove tomorrow.

I still remember my MSR whisperlight jet becoming clogged on a very cold winter night in the Snowy Mountains here in Australia and the battle with very cold hands to clean it but when I finally got the stove going it performed beautifully even though it ueses a lot of fuel.

Tony

[Douglas Frick]

>You are right: the gauge pressure inside either a Powermax canister or a Kovea 30%propane/70%isobutane canister will drop to zero around -10 F.

I'm trying to figure out why. Wouldn't the propane in the Powermax canister still vaporize down to -35F/-40F even if the Butane doesn't? (The liquid-phase components would still come out mixed, as desired.) Now I wish I'd run some tests on my Xtreme when it was -25F outside a few weeks ago.

[Brian James]

Tony: I'm very familiar with Dr. Caffin's excellent articles. I am also very keen to hear about real-world experiences in this domain as I have so far been too chicken to winter camp using only canisters on my old Primus varifuel stove.

Basically the upside-down-canister theory is simple and obvious: unless it's -40 out, your propane will still boil and that will push butane into your stove.

In the real world, I suspect that a lot or most of your propane stays suspended in your liquid mixture. Thus, as you use the canister you are burning up much of that precious propane.

So after you've burned most of the fuel in a canister in the cold, I think you've probably burned most of the propane *even if the canister was inverted.*

Once you have little propane left in your canister to create pressure, you have to contend with the effect of the canister cooling itself while it runs. (I don't know how much it cools in winter conditions, but I know my canisters drop 40 degrees F in a very short time at room temperature.)

I suspect that, due to these two factors, the pressure goes to zero before the air is at the theoretical liquefaction temperature point of propane.

I think that a valuable test would be to run a canister upside down *below the liquefaction point of isobutane* until it's only about 1/3 or 1/4 full. Then measure the pressure at that temperature, and at lower temperatures.

I would postulate that the canister would become useless before it reached -40 — long before. I'd love to know more!

Brian

[Brian James]

PS: Even in the very cold, canisters make a fantastic preheater for a liquid fuel stove. I've carried a 4-ounce canister for my Varifuel as a lazy man's preheat and what a treat that's been.

Just attach the canister and run for about 15-30 seconds. this gets the combustion bell glowing cherry red and the preheat tube nice and hot. Then attach the liquid fuel bottle and slowly turn it on — nothing but gas comes out! Light immediately and you will never have to deal with the "fireball effect" again. Much faster, simpler, and less leaky than carrying a bottle of methyl hydrate.

[Tony Beasley]

Hi Brian,

Although I am interested in what happens inside a canister in cold conditions it is not my main area of interest or expertise and I only have very limited resources (time and money which competes with family, work and bushwalking, trout fishing, long distance running/mountain running) I prefer to put most of my spare resources into my main area of interest (stove/pot efficiency, stove burner design and upside down canister stoves). I will do some more cold temperature pressure testing of caniaters with what I have and I will post the results when I have them.

Maybe one of BPL stove editors could do some work on the problems that you have suggested.

Tony

[Roger Caffin]

> >You are right: the gauge pressure inside either a Powermax canister or a Kovea 30%propane/70%isobutane canister will drop to zero around -10 F.
> I'm trying to figure out why. Wouldn't the propane in the Powermax canister still vaporize down to -35F/-40F even if the Butane doesn't? (The liquid-phase components would still come out mixed, as desired.)

Yes, the propane does continue to boil at -35 C.
The problem is all about mixtures. If you have half propane and half isobutane, then the final pressure is half of the propane pressure and half of the isobutane pressure. And so on. When you have only 30% propane, its high vapour pressure gets diluted down by the lower VP of the isobutane, such that the total VP falls to 14.7 psi at about -26 C.
Yeah, mumble. Annoying.

More in the next post about liquid feed.

[Roger Caffin]

> Basically the upside-down-canister theory is simple and obvious: unless it's -40 out, your propane will still boil and that will push butane into your stove.

If I have published that -40 C figure somewhere, my apologies. Because we are dealing with mixtures the figure should be -26 C for 30% propane / 70% isobutane. Other mixtures are generally not as good.

> So after you've burned most of the fuel in a canister in the cold, I think you've probably burned most of the propane *even if the canister was inverted.*
I am not 100% sure what you mean here, so let me put it this way. The amount of propane needed to pressurise an inverted canister is negligable. So the fuel coming out of the fuel line will be the same as what is in the canister. If you have a 30% propane / 70% isobutane fuel, that is what you will get out of the fuel line, and that is what will be left in the canister. As long as you only use a liquid feed of course.

> I suspect that, due to these two factors, the pressure goes to zero before the air is at the theoretical liquefaction temperature point of propane.
You have to distinguish here between absolute pressure (psia) and gauge pressure (psig). Absolute pressure is … well, absolute. Gauge pressure is the difference between the absolute pressure and one atmosphere (=14.7 psia). So if you have 14.7 psia inside the canister at sea level, you have pressure in there, but the external pressure is the same so the gauge pressure (differential) will be zero.

Cheers

[Tony Beasley]

I threw my collection of 8 full and partially used canisters in the fridge in turn at 4 degrees C (39F), 0 degrees C (32F)and -20 degrees C (-4F) and then measured the pressure in each of the canister. Two of the canisters are 20% propane/80% iso-butane mix and the other six are the other 30% propane/70% iso-butane. All of the used canisters have been used in the normal upright position at some time some have been used in the upside down liquid feed configuration, some have been used in cold temperatures(at least for Australia) and some only in warm temperatures. I have not kept a record of use so I cannot relate the pressures/temperatures to any specific use.

% full by weight for 20% propane/80% iso-butane

100% full 4C/39F: 11psi, 0C/32F: 9psi, -20/-4F C: 0psi

21% full 4C/39F: 11psi, 0C/32F: 3psi, -20C/-4F: 0psi

% full by weight for 30% propane/70% iso-butane

100% full 4C/39F: 26psi, 0C/32F: 20psi, -20C/-4F: 5psi

70.5% full 4C/39F: 11psi, 0C/32F: 8psi, -20C/-4F: 0psi

37% full 4C/39F: 11psi, 0C/32F: 3psi, -20C/-4F: 0psi

35.4% full 4C/39F: 9psi, 0C/32F: 4psi, -20C/-4F: 0psi

17% full 4C/39F: 11psi, 0C/32F: 10psi, -20C/-4F: 0psi

2.8% full 4C/39F: 11psi, 0C/32F: 4psi, -20C/-4F: 0psi

It could be possible by the figures above to say that even in moderate temperatures that the Propane boils off faster than the Iso-butane.

It should also be possible if enough P/T data can be collected to determine how much propane is left in the canister by pressure /temperature reading.

Tony

[Author]

Posts