Stove Development

The testing and development of the loose biomass stove from this past two weeks have yielded many important and interesting results. We had many moments of utter frustration when facing the problem of power output, but after some small, crucial changes, we are tremendously excited about our current product.

As mentioned in the previous development blog post, the current design is based largely on the design of the Kisangani Stove Group (KSG) in Njombe, Tanzania.  Testing showed that the design produces a long, steady burn that did not require tending for at least three hours. However, the stove lacked the power needed to boil water in a reasonable amount of time. Following the Aprovecho Water Boil Test (WBT) protocol and using five liters of water, we were only able to heat the water to 90 degrees Celsius, peaking at the two hour mark of the burn. Additionally, the first thirty minutes of every burn were always extremely dirty and did not produce a lot of heat. 

A new lid design seemed to solve some of these issues. Instead of the assorted concentric concentric cylinders and rings of metal rod in the KSG inspired lid, the new lid is simply a disk with an opening and three triangles of sheet metal to support the pot. Despite its simplicity, our new lid produced positive results.

With this new lid, the burn was clean within five minutes and the stove was able to boil the Aprovecho prescribed five liters of water (The local boiling point in Arusha is 95.7 degrees Celsius).  It seemed as though the innermost cylinder on the KSG lid was actually restricting airflow in the first thirty minutes of the test burns, thus producing inefficient combustion and visible smoke. With just a simple opening, the new lid design allowed more airflow for cleaner combustion. Furthmore, the smaller pot stand on the new lid lowered the pot, allowing more heat to be transfered to the pot.

Nevertheless, the new lid still did not solve a crucial problem: boil time. Once again using the Aprovecho standard of five liters, the stove still  required two hours (126 minutes) to reach a rolling boil. After conducting the WBT and performing some rough calculations, we found that our stove had less than 1/10th the firepower of other improved cook stoves currently being sold. Even without these calculations, it was clear that power output was a huge problem. Mamas using three-stone fires only a short walk away were able to boil our five litres of water in just twenty-five minutes.

Thinking that updraft and the height between the heat at the bottom of the stove and the pot were part of the power problem, we built a version 70% as tall as our original prototype. We hoped that the shorter chimney would provide less resistance to the primary air for hotter combustion and the smaller design might become a cheaper alternative to our earlier design. Unfortunately, the smaller stove was extremely difficult to light and even when lit, could not sustain a burn.

At this point, we seemed to be at an impasse. Though we could achieve a clean burn for over three hours using a cheap alternative to firewood on the KSG inspired design, we still could not complete the Aprovecho Water Boil Test in a reasonable amount of time. This suggested that our design lacked the power that would be needed to cook foods as well or as fast as traditional, less clean stoves. Knowing that our stove would ultimately be judged by the cooks of Tanzania, we were unsure of how to progress with the design.

However, we stumbled upon a solution to the power output problem the next day in the AISE workshop. Instead of just one pipe, we packed our stove with two pipes for the chimney simply out of curiousity. The difference was astounding. Just by changing the chimney from a single cylinder to a double cylinder ‘8’ shape, we were able to boil water in 18 minutes. Though the start was slightly dirtier, the burn became clean after only a few minutes. Furthermore, this configuration produces more complete combustion- much more of the fuel was burned to white ash when peering into the fuel canister after the burn. Subsequent tests have been able to duplicate this remarkable result and sustain burns between ninety mintues and one-hundred minutes. Eureka!

Since the stove group had been planning to visit EARD-CI and the stove demonstrated such strong results, we all brought the stove to EARD-CI headquarters yesterday. As we explained to them the progress of the stove and our plans to conduct a test using rice husk, Edith suggested we try to cook a meal on our stove. It was such a treat to enjoy a delicious lunch of fresh corn with avocado cooked on our own stove! Despite all of this good news and good food, it does bear note that the stove was unable to burn rice husk in the later burn.

These past few weeks of development have yielded truly exciting results. With enough power achieve a rolling boil in a time comprable to similar improved cook stoves, the efficiency to sustain the burn for over ninety minutes, and the consistency of each burn, we feel that we are close to having a mature product. However, we still have a lot of work to do. Next steps include:
1) Completing a full WBT as well as using the IAP meter to accurately quantify power and  air pollution from the stove.
2) Resume testing with coffee-husk and even explore other loose biomasses so communities with different resources can still benefit from the stove.
3) Polishing the design. Namely, designing stronger handles, adding a gate to regulate primary air, and a more sophisticated pot stand
4) Investigating distributors and possible distribution channels