Selecting a Canister Stove for Cold Weather Backpacking
Part II: Commercially Available Canister Stove Systems

Quick update: just tore apart a valve mechanism, and should be able to accomplish the thermal testing on the “blue” and “red” plastic today or tomorrow. This should give us a good indication of the temperatures they withstand before softening. Will keep you posted. (for the technically minded, the technique we are using is TMA – Thermal Mechanical Analysis, and we are looking to determine Tg, “glass transition temperature”.

Roger – I hear your request for graphs…will try to comply!

The data are in! (and thank you to an interested and generous co-worker). Looks like the blue and red pieces are both the same material. Unfortunately, it looks like there is not a really, really, clear-cut answer for us: this material exhibits a very, very broad softening range….starting at about 50C, and ending (rather catastrophically) at about 175C. Between those temperatures, it progressively softens. A raw crunching of the numbers puts would indicate that we would really start being concerned at a temperature above about 100C (for the techies…this would be the indicated Tg measured from loss-modulus).

I think the bottom line would be as follows: below 50C is just fine. Between 50C and 100C there is some risk to warping the red pin and shutting down the gas flow. Above 100C you are running on borrowed time, and the “shutoff mechanism” just mentioned by Roger (i.e. “mushrooming” of the pin) is much more likely to occur.

Roger – I know you want plots: I am trying to paste the appropriate plots into this message…but can not figure out how to do it (the paste option does not work). Can someone enlighten me please?

Ok…lets try this. Here is an image of the data from the thermal analysis on the red piece from the stove. The top line is the elastic modulus (read “stiffness”) versus temperature (in degrees C). You can see that the modulus starts decreasing at 50C, and does so continuously until about 170C, at which point the part just plain fails. For our purposes, what we REALLY want to look at is the bottom line (the one with the “hump” in it. This is what is called the “loss modulus”…and is more of an indication of when the part really starts “flowing”..i.e. irreversibly deforming. This curve is very broad, with a peak at 106C (most materials have a much more sharply defined peak). Usually we would define the peak as the so-called “glass transition point”…but with this material with such a broad peak it is thus hard to define a specific temperature at which the material starts to “flow”. But we CAN say the following: below 50C (122F) the material will not flow. Between 50C and perhaps 80-90C we are probably ok…but there is some risk of the stove throttling down the gas flow. Above 106C you are REALLY running a risk of the valve shutting off…and forget it if you reach a temp above 170C.

Sorry the results did not turn out clearer…but at least this result is consistent with the observation that “not-too-hot-to-touch” is “ok”. Here is the link to the plot. (did I put this in correctly Robert…or can I improve on the way the image is posted? How did you actually get your pictures imbedded into your posting?)

Thank you Douglas for giving the info on displaying the image….Figured out how to get the image larger….hopefully not too large now.

This might have been commented on before – but one thing that could be done is to use temperature sensitive stick-ons. See URL below….

http://www.tempil.com/pdf/Tempilabel.pdf

While I was looking into the canister fitting for the burst test, it ocurred to me that the metal might perform differently at 75 or 100 degrees (=burst temp?) than in my cold-water hydrotest tank. Thoughts? I’m going to have to make the fitting: is this a waste of time?

Also, I am looking forward to part III of the article! I do love technique and theory articles (and trip reports!) more than gear reviews…here’s hoping for more like this stove series.

Brian

I recommend reading “The nature and behaviour of mixtures of fuels” by Roger Caffin for more reading on this topic. I found this enlightening when I first read it a while ago.

http://www.bushwalking.org.au/FAQ/FAQ_Mixtures.htm

I just found this article and discussion in my search for winter camping cooking. Just curious, why don't they use higher mixes of propane in the gas canisters? Forgive me if this has been answered already and I missed it.

Thanks for the comparisons and boil times. I'm in the market now for a winter stove and this has given me the direction I need.

>Why wouldn't we just use propane which has a low boiling point and would vaporize on it's own down to an extreme minus 40 degrees, why introduce Butane into the mix?

To quote Rick Dreher, above:
>> A pure propane canister is a heavy duty affair for this reason (e.g., the green steel Coleman propane canisters).

A green steel Coleman propane canister holds 16.4 oz of fuel and weighs 31.4 oz–a canister weight of 15.0 oz. Two Coleman 70/30 butane/propane 220g canisters hold 15.5 oz of fuel and weigh 27.2 oz. Scaling the 70/30 canisters by the fuel weight ratio gives an equivalent canister weight of 12.4 oz per 16.4 oz of fuel, or about 2.5 oz more canister per pound of fuel for pure propane instead of 70/30. (I hate English units.)

I have a remote-canister liquid-fed stove (Coleman Exponent Xtreme) so I'm not too worried about my stove's cold-weather performance. But for winter camping, the extra weight of pure-propane canisters doesn't seem too much of a burden for the performance. Propane stoves, in addition to the big clunkers, are available in low-priced top-burner and remote-canister configurations, although I'm not sure how much they weigh. My personal experience with a dual-burner propane stove at merely freezing temps was not positive, so there may be other problems with using gas-feed propane (although I haven't had any problems with my gas grill at temps down to +10F).

Roger – I think this article has missed a trick – with a remote canister stove you are no longer restricted to the squat canisters used for canister-top stoves. Any canister with a threaded Lindal valve can be used, eg

These are lighter (~85g empty) and cheaper than the squat 100g canisters but contain 170g or 175g gas (30% or 35% propane).
They are widely available in hardware stores (they are meant for small blow torches) and come in various brands – even CampingGaz now, their only threaded canister!

Roger – I'm pretty sure there is no internal tube.
They are very easy to use – raise the 'base' end by supporting it with something: boot/rock/snow/whatever so the canister is lying at an angle of 30* to 45*. There is no need to invert it vertically unless it is almost empty.

The Eta Packlite has a pre heat loop, so you should be able to run it in inverted mode without a problem. I think Roger has reviewed it elsewhere on the site and that the review was favorable.

Generally, manufacturers recommend that the stove be started with the canister upright, the pre-heat tube allowed to warm, and then that the canister be inverted. In practice many of us have found that starting a stove with the canister inverted is no big deal if you go easy on the gas at first.

HJ