Experiment Results

[Experiment Results] Fiji Fuel Dec 18 – Buka Stove Comparison

Brian Hu
Brian Hu | 2 months ago | in Sustainable Fuel Consulting

Lean Phase: Solutions

Assumption: Buka Stove 5.0 is more efficient than Buka Stove 4.0 and three stone fire


Link to previous post:

The 3-stone fire, Buka 4.0 and Buka 5.0 were trialled three times each where the temperature was measured every minute and the time when the water reached 100⁰C was recorded.

For strength testing the 4.0 and 5.0 were loaded progressively until they exhibited notable bending. The 4.0 could hold up to 90kg and the 5.0 could hold up to 20kg. The experiment results can be found in the following spreadsheet:

Please read the following experiment report for the full results:

The following table is extracted from the experiment report.


Average time to boil

Degrees per min

Average wood consumed (g)

Average wood consumption rate (g/min)

Degrees per gram


31m 8s





Buka 4.0

17m 53s





Buka 5.0

26m 18s






In reference to our success matrix, the 5.0 was lower than the 4.0 across all metrics and is clearly in the Red Light point.

Validated Learning:

Our assumption that the Buka 5.0 is more efficient than the 4.0 was completely contradicted in this experiment. Though the results cannot be wholly defined as quantitative, the differences in performance between the 5.0 and 4.0 is clear enough to indicate the 5.0 is not competitive with 4.0 in efficiency which necessitates design improvements. Inefficient combustion was identified as a contributor to the poor performance of the 5.0.

Next Move:

Following the Red Light, the December Fuel Team has designed a modified 5.1 design which aims to address the issues of the 5.0 while incorporating customer feedback and have contacted a manufacturer. It is recommended to conduct another experiment with the 5.1 and compare it to the 4.0 and 3-stone fire again in an experiment similar to this. Ideally, partnerships with universities should be secured so that fire testing can be done in an environment with controlled conditions or if this is not possible, more trials could be done to mitigate the variance between trials.

Haziq Ahmed 2 months ago

Status label added: Experiment Results

Reply 0