clean stove findings

the clean stove team studied some general principles concerning stove and burning efficiency.  some are obvious, others more technical.  clearly, efficiency depends in part on the prep of the wood, including using smaller pieces with more surface area, and better drying with coverings, heat, wind, etc.  many fluid dynamics principles are also key such as controlling heat as it travels to the point of heat exchange, reducing time that the heat takes to get from the source to the food, and improving the flow of exhaust fumes out the flue.  turbulence of the air/fuel/flame mixture will help with increased temperatures and complete burning.  in general, insulating the stove to keep heat in and warming (and speeding up) intake air also helps, as does creating a good draft of exhaust.  further, a two part combustion process allows complete burning of the wood material and elements, producing only carbon dioxide and water and producing almost no smoke.  primary combustion happens through the production of coals and embers, and secondary combustion creates a gas burn and actual flame from the emissions coming off the coals/embers.  understanding all of this and a process called wood pyrolysis were important in developing a cook stove concept.

the team organized their benchmarking by four main concept areas.  for the most part, it is possible to mix and match from each category to make different stove systems. 

·    fuel type was limited to wood, which is commonly used and widely available in Guatemala and other developing areas. 

·    combustion chamber was by far the most complex area for consideration.  five main types were investigated and will be discussed below.  

·    heat transfer methods, ways by which the heat is brought to the pot/food, included griddles, grates, open flame, pot skirts, radiant heat, and forced air.  open flame is the current method with a three stone fire; griddles and grates are clear; radiant heat is like using an oven; and forced air pushes air through a device with a fan.  pot skirts are a cool idea that optimize the cooking interface.  basically, a skirt goes around and up the sides of a pot, holding heat against it more tightly and longer, transferring much more heat to the food.  unfortunately, that requires exact skirts for each pot.

·    finally, three flue designs were considered:  straight, balanced, and valved.  straight flues – simple pipes that lead exhaust air out of the stove -- are the easiest and cheapest types of flues.  balanced flues take in cool air from outside, allowing it to be heated by the exhaust air in a central pipe, and use that pre-warmed air as oxygen input to the fire.  this improves the efficiency of the fire but is more expensive.  an open air fire uses no flue at all.

the team narrowed down their main concepts to six stoves (combustion chambers):  three stone open air (baseline), clay enclosed, rocket stove, TLUD stove, downdraft J chamber, and charcoal combustion chamber. 

·    a three stone fire (see photo from the original design challenge post) cooks with an open flame and a pot balanced on three stones over the fire.  because heat escapes in all directions, it is highly inefficient and allows much of the fire’s energy to dissipate in ambient heat and smoke, rather than high temperature delivered to the cooking surface.

·    a clay enclosed stove simply contains the fire within walls.  this allows the heat to be more focused, but does not appreciably increase efficiency.

·    a rocket stove uses an L-shaped chamber where small pieces of wood are loaded into the bottom, resulting in the ends of the sticks burning.  concentrated heat goes up into the chamber to a relatively small cooking surface.  this design allows very little wood  to create quite a lot of focused heat. 

·    a top-lit updraft stove (TLUD) produces a secondary burn area where much of the exhaust particulates are combusted, producing much less smoke and less waste of energy.  once again, small pieces of wood are used, but they are placed into the top of the combustion chamber.  as the wood pyrolyzes at the bottom of the stove chamber, gases flow up the sides of the chamber and are re-ignited near the top, creating an even hotter flame and more completely burning excess particulates. 

·    a downdraft J chamber stove acts similarly to a rocket stove, but the fuel is loaded from the top and heat flows sideways before going up to the cooking surface.

·    a charcoal combustion chamber stove is said to be similar to other stove designs, but uses a reflective surface inside to reflect heat back into the chamber.

the team used decision matrices to examine their options for combustion chambers, heat transfer methods, and flue designs.  a decision matrix takes the original design specifications and weights relative to one another for importance.  each concept is then evaluated as to how well it meets each criterion, and the score for each is determined by adding the weighted evaluations.  using this method, the best two concepts for combustion chamber were the rocket stove and the TLUD.  for heat transfer, a griddle stood out far and above the rest.  for flue design, especially since cost and easy manufacturability are important, a straight flue beat out a balanced flue.  the combustion chamber decision matrix is included here in case you are interested.

having thus figured out their best design options, the students went about building simple prototypes of a rocket and a TLUD stove.  they used metal paint cans as the main component of their combustion chambers, and assembled other parts from easily available products (sheet metal, other cans, rocks, etc.)  time and temperature were collected from a boil test with each stove.  the rocket stove took 21 minutes to boil four cups of water while the TLUD took only 12, a 43% improvement.  (though it should be noted that there were difficulties in keeping the rocket stove lit at the beginning, possibly biasing the data.)  the second test determined the amount of time each stove could maintain a cooking temperature with a set amount of fuel.  this test appeared to also favor the TLUD stove.  the team also measured outside skin temperature of the stoves and found the same results from both stoves:  too hot to be touched.

next up:  the design solution!

team's TLUD stove prototype