Recently I felt it was time to re-evaluate my attitude toward canister-top gas stoves, and also to perhaps re-evaluate the stoves’ performances too.
My one big obstacle with this type if stove has always been the possible danger of to closely encasing the canister while trying to gain maximum efficiency from the gas burned. I have recently looked at the Primus canister-top windshield but have, despite it good write-ups and proponents, dismissed it for my purposes.
My experiences with Meths stoves have convinced me the close wind shielding which encompasses almost the whole stove and cooking pot (as far up the pot as possible – even to the lid or further) makes for efficient utilisation of the heat output from whatever the stove or fuel in question, it also eliminates some heat loss from the pot to cold air passing the pot.
The Primus canister-top windshield just does not do that so I set about trying to come up with something cheap, simple and easy to make at home.
My first criterion was obviously canister safety. Keep the canister cool at all costs. So, don’t enclose the canister. From there the problem was more or less defined and simplified. Take the canister out of the equation and produce an efficient wind shield for the burner and pot.
After quite a few false starts I came up with the advanced prototype demonstrated in the following video. In the process of development I also crystallised a few other thoughts and ideas and the cooking kit arrived too. None of this is fixed in stone so anyone who would like to continue further – please be my guest.
The base plate and shielding in the video are 0.3mm aluminium sheet bought from HERE . I deliberately folded ridges into the base plate to act as supports, but I’m not sure that they, or in fact the aluminium base plate are really necessary. The prototype base plate is shown here…..

- Thermo Wrap prototype base plate
…..was simply cut out of ‘thermo wrap’ insulation material which I buy by from Focus Do It All, or B & Q in 10 metre rolls. I also use this material for pot cosies etc. That prototype was quite sufficient but I wanted to try to advance the design slightly. The former design is by far the easiest to make and slightly easier to pack. Talking of packing brings me to the origins of the ‘cooking kit’. I hate to have to fold aluminium sheeting as it quickly work hardens the fold line which then cracks. So, having established the windshield I wanted I sought around for some means of packaging it, preferable in a dual-purpose system. I had recently purchased some spare ‘Caldera Cone Caddy’s from HERE and one of them seemed ideal. It would double up as a carrier/holdall for my spoon, cloth, stove and (possibly) a ‘100’ gas cartridge. Add to this a multi part cosy system and I have either one large food container for ‘cooking’ rehydrated food or two small pots for similar purposes; the larger ‘half’ for food, the small ‘half’ for a drink.
The pot shielding obviously needs to be made of aluminium so I established the maximum height of any pots I was likely to use and cut a strip to this size. For length I simply measured the circumference of the largest pot I am likely to use (MSR Titan Kettle or Tibetan Ti-900) and that gave me a size. This size also accommodates Tibetan Ti-700 and 550 pots as well as the Alpkit MyTi Mug. This is where I wanted to be; a single multi-purpose set-up that was lightweight (Not UL I hasten to add).and effective.
This system finally put together I started gathering test data……

- Some of the results before compilation
I don’t propose to detail some 74 tests but will précis the main points.
I used three different pots – Tibetan Ti-900, Tibetan Ti-700 and Tibetan Ti-550.
I used three different stoves – Snow Peak Giga Power Titanium, Optimus Crox (folding burner model) and a Vargo Jet-Ti.
The windshield was as seen in the video but the base was firstly Thermo Wrap (prototype above) and then the aluminium foil seen in the video.
All the tests were carried out outside, on differing days and in differing conditions of wind and ambient temperature.
Results:- within very narrow margins all the tests gave similar results. Neither type of pot, type of stove nor weather created any major differences. The fastest boil time was 2min 46secs using the Tibetan Ti-900 pot containing 300 ml of water, the Giga Power stove and 6 grams of gas. The slowest boil time was 4min 26secs using the Tibetan Ti-700 pot containing 300 ml of water, the Giga Power stove and 6 grams of gas.
All of the remaining 72 test results were within these parameters.
Conclusions:- Stove types made little difference although I did have one or two issues with the Optimus Crux. This is a beautifully designed stove with the innovative idea of a folding burner head to facilitate easy packing. My concern with this stove (or at least mine in particular) has always been the lack of rigidity of the head. When flipped into the working position there is a slight wobble on the head, nothing major, but it does mean that pots on this stove can tilt slightly. Weight for weight with the other two stoves in my tests it is comparable, but both alternatives do give a firmer pot support. I especially like my Snow Peak GST – 100 Giga Power Titanium. It is, in my estimation, the Rolls Royce of canister-top units.
Now a little extrapolation:-
Given a gas usage (estimated on the higher side of my results) of 6 grams to boil 300 ml of water then a 220 gram (net) canister should provide 36 boils. Thirty-six units of 300 ml are equal to 10.8 litres. In imperial that’s 18 pints or 2.25 gallons.
From my experience with meths stoves I would reckon that to boil the same quantity with an efficient meths burner would require 540 ml of fuel (15 ml per 300 ml of water = 15*36 = 540)
Taking weight into account then this is how the two fuels stack-up given the above scenario…….
Meths – Container + 540 ml fuel = 496 grams
Stove (say E-K burner with case) = 93 grams
Total……..589 grams
Gas – Canister = 220 grams fuel = 354 grams
Stove (say SP Giga Power with case) = 97 grams
Total……..451 grams
Cooking Kit (as in video) – Total……..183 grams (excluding pot)
Tibetan Ti-700 pot = 110 grams.
So, cooking kit, Tibetan Ti-700 and gas stove and fuel weighs in at 774 grams and cooking kit, Tibetan Ti-700 and meths stove and fuel weighs in at 882 grams.
I offer this as food for thought………………….
When enclosing any burner you need to be careful about starving the burner of oxygen and producing carbon monoxide. Most of the time in the outdoors this doesn’t really matter, but if you are enclosed in a tent, especially if there is not much ventilation you need to take some care. Roger Caffin of backpackinglight.com has written some very good articles on this. The best canister stove, producing low levels of carbon monoxide was…..the Snow Peak GST100. As you say it is the Rolls Royce of canister stoves.
Hi Robin, Yes I agree, using any form of cooking equipment inside a tent has its pitfalls. So far as my system is concerned I have tried to ensure an adequate supply of oxygen, firstly with the punched holes around the base and secondly with leaving a ‘front’ opening.
I’m so impressed that you can keep the camera still…AND TALK…whilst holding the camera iin your mouth Robin. No mean feat!
Anyway …when you light your thermo-nuclear device exactly how many square miles will be laid waste? I couldn’t quite see that in your calculations.
Great to see you back in action after your Ireland trip.
Insurgency training camp visit perchance?
The square mileage laid waste depends on wind and the toxicity of the fallout.
Great to be back in action (full stop).
I take my hat off to you sir for such dedication in pursuit of your research! Detailed and informative. i dont think I would have the patience!