Pot Size and stove efficiency

When looking at the efficiency of a stove, we have to include in our discussion not only the stove and the fuel but the size of the cooking pot which is being heated. We have to consider how the heat is transferred from the flames to this pot. Imagine tha

Pot Size and stove efficiency

When looking at the efficiency of a stove, we have to include in our discussion not only the stove and the fuel but the size of the cooking pot which is being heated. We have to consider how the heat is transferred from the flames to this pot. Imagine that we have a little pot sitting on a great big roaring stove. Most of the heat is going to go straight up past the pot and be wasted. This is illustrated in the picture to the right of a metho stove running hard . You can see the blue flames going up the sides. Now imagine a larger pot with a smaller flame. More of the flame will be hitting the pot, and so more of the available heat will get into the pot. So if you want greater efficiency, turn the stove down below 'flat out'. In fact, the stove usually has to be out before heat losses from a shiny sheltered pot become significant - provided you have a good lid on it. But make sure the flame does not have orange tips.

The possibilities of improving efficiency were brought home to me on one trip when I had miscalculated the amount of fuel needed. I had not taken enough fuel, and it looked as though we were going to be forced to cook over a wood fire (and mess up my nice shiny stainless steel pot). However, I realised the problem early in the trip, and decided to try to stretch out what I had. I turned the stove right down for every meal: dinner took (maybe) an extra 5 minutes to cook. I let dinner cook with the stove off, rather than leaving it running low. But instead of using 40 g per day of gas, I found I was using about 25 g per day. Yes, we ate the same meals. So it can be done.


Pot Lids and stove efficiency

But there is still more to the efficiency of fuel use. Let's assume that what you are really doing is cooking dinner (rather than trying to run a physics class), and this means heating a pot of water. Obviously, if you are trying to fry something the following will be less relevant. While you are heating your pot of water up, the water is doing its best to cool down - by evaporation. This can be quite significant, and I have measured the rates of cooling under different conditions. The results of one experiment are shown to the right. The red line shows the rate of cooling of a shiny stainless steel 1.5 Litre pot of water that had a lid on it. There was some room for evaporation at the edges and through the hole in the lid where the thermometer went. The blue line shows the rate of cooling when everything else was the same but there was no lid. The horizontal axis is minutes; the vertical axis degrees Centigrade. Clearly, the escaping steam is cooling down the pot very significantly. If I really seal the pot up the rate of cooling is even less than this.

This has two consequences. The first is obvious: if you are cooking dinner without a lid on the pot you are going to be using a lot of fuel just to counter the effects of evaporation. You have to carry that fuel. The second consequence comes when you try to 'dutch oven' your dinner - that is bring it to the boil then let it sog for 5 - 10 minutes. For this to be successful the pot has to stay hot, and it is not going to do that for very long if evaporation is allowed. By taking just a little care with the lid on my pot, I regularly manage to Dutch Oven our dinners after about 15-30 seconds of boiling. The pot might cool down about 5 - 8 degrees in the 10 minutes needed: this is not a worry. (Food at 90 C is still far to hot to eat directly!) This saves a huge amount of fuel of course. It also means I have virtually no risk of burning the dinner.


Pot Metals and stove efficiency

Does which of the available metals you choose make a difference? Arguments rage: each has its advocates, although close examination of the arguments suggests each pushes a different advantage: weight, cost etc. Most of the arguments have one thing in common: a near-total ignorance of engineering facts. I am going to ignore the show-off factor here: if you want a bright red titanium pot, so be it.

Let us assume that you are going to heat one litre of water from 15 C to 95 C in six minutes. First we list the properties of four different metal pots. For the engineering purist let me add that I am making quite a few approximations here. For example, since I don't know exactly what alloy of stainless steel is used in pots, I cannot give its properties exactly. If you know better, let me know. I also list the properties of representative pots of each metal, as best as I have been able to ascertain them.

Metal Density
Heat Cap 
Thermal Cond
Pot weight
Wall Thick
Aluminium pot #1 2.7 910 222 250 1.5
Aluminium pot #2 2.7 910 222 125 0.75
Titanium pot 4.5 540 19 140 0.5
Stainless Steel pot 7.7 490 12 195 0.4
Water contents 1.0 4180   1000  

I must emphasise here that some of these parameters are rough. In particular, the wall thicknesses are very approximate. This is a pity, as it will turn out to be important. And some of the thermal properties of stainless steel are very suspect as they vary significantly with the alloy. But the figures are close enough.

First we work out how much energy (heat) will be required. We assume that the pots are all the same size and shape (most unlikely) and then work out the energy flux through the bottom of the pot to get this energy transfer in the six minutes allowed. From the flux and the pot thickness we can then work out the temperature gradient across the pot wall. That will tell us how each pot might behave. For this I am going to assume that the bottom of the pot is about 180 mm diameter: there are many pots of all three metals of about this size on the market. It's a good generous pot for two people. The area of the base will then be 25,447 mm2. I am also going to assume that the efficiency of heat transfer from the flame to the base of the pot is 100%. It will never be this high in practice (33% might be possible), but this really just means I am going to treat each pot the same. My calculations are as follows: feel free to correct me if I am wrong.

  Pot mass
Pot Heat
Water heat
Total Heat
Aluminium pot #1 .25 18,200 334,400 352,600 979 0.26
Aluminium pot #2 .125 8,954 334,400 343,354 954 0.13
Titanium pot .14 6,048 334,400 340,448 946 0.96
Stainless pot .195 7,644 334,400 342,044 950 1.23

What does all this maths tell us? Several rather interesting things, as follows.

  • The tiny thermal mass of the pot is trivial compared to the huge thermal mass of the water, no matter what metal is used.
  • The power required in theory is about 1kW: in practice you will probably need about 2.5 kW to get this performance. That's fine as most small stoves (but not alcohol or solid fuel) can easily do this and more.
  • The temperature gradient across the pot wall is very small: well under two degrees. This is not going to matter one iota: you couldn't measure it in the field!

Clearly, the bottom line is that the power required for all pots is about the same. What pot you use is not really going to change how much fuel you need. Far more significant will be whether you use a windshield and a lid on your pot as these things can halve the amount of fuel you use.

However, this is not the whole story. In the field, as opposed to on the test bench, you wil find that these four pots have seriously different behaviours. Both stainless steel and titanium pots have a bit of a reputation for be able to burn your dinner. The problem is that the wall thickness is so small there is little chance for the heat to spread out sideways, and hot spots will occur. The aluminium pot #2 might be expected to have similar problems nto the Ti and SS, except that the vastly superior conductivity is going to help spread the heat much better. Mind you, the thicker aluminium pot #1 is going to spread the heat out even better still. My #1 pot has a Teflon coating inside, and it cooks wonderfully (and cleans with a rinse). This is yet another case where 'high tech' may not mean 'better', not withstanding the marketing hype. You might be well advised to make your buying decision on the quality of cooking rather than the high-tech wow factor (and Ti pots are awfully expensive!). 
[Footnote, 2006: I have been advised by one major supplier of Ti cookware that they are seriously thinking of abandoning the Ti market: too dear and limited sales.]

Some interesting tests were reported to the author recently by Tony Beasley at the ANU. He had tried running this experiment with three real pots and a stove. He did his best to keep the stove rate constant for all three pots of course. The pots were all about the same size, but he found slightly different heating rates for each one. But when I looked more closely at the results and adjusted them for the small differences in pot cross section, the heating rates came out roughly the same - allowing for experimental error. I understand Tony is pursuing this work and will report again with more results.