[Stoves] briquette air feed

[Crispin Pemberton-Pigott]

Dear Frank

>A part of the testing I would like to see changed is going from testing the
stove as 'we think it will be used'  to 'a continuous run for hours' 

I am all in favour of that. The point of running any test is to get useful
information. After a while, no new information is gained so continued
testing is pointless.

>...to lower the percent errors when determining energy of remains fuel and
time a temperature reads simmer to boiling.  

If the point of a test is to find out what the thermal efficiency is at low
power, run the stove at low power and put on a pot of cold water. Between
30-70 degrees C the thermal efficiency will be obvious, especially at the
lower end of that range.

>These errors in their measurements will be the same if we run the stove for
an hour or for 20 hours but it will be a larger percentage of the one hour
and a much small percentage a 20 hour run. 

The error is not caused by the duration of the test, it is from measuring
only a small portion of the energy that passes through the pot and declaring
that to be the total. Only a small percentage of heat passing into a pot at
low power disappears as evaporated water, even less if the lid is off.

>And the increase in fuel burned will more represent an average of the fuel
and changing conditions when loading. We can then do a simmer test for 20
hours and keep the results separate.

This repeats the error. If the losses from the water (radiation) the pot
(radiation, convection, conduction and the lid (ditto) are large compared
with the heat lost from water evaporation then the calculated result is in
(large) error.

>I realize 20 hours is too long ...

We are doing some very long tests in UB to see what there is to learn.
Basically nothing. We have reduced the fuel burned from 95% to 90% of that
loaded for two reasons: there is nothing of interest to learn and because
too many stoves are unable to burn 95% of the fuel in a reasonable time (say
12 hours). Most stoves can burn 90% of a fuel load in a few hours, including
a refuelling episode. Let's say 6 or 7. That is a reasonably long and
accurate test yielding 1500-2500 data points. The mass corrected thermal
efficiency and the emissions (similarly corrected) are the results of
interest. The accuracy is high and the result meaningful when expressed in
emissions per unit of heat potentially produced.

We did some tests for 20 or more hours and learned nothing at all. We did
several to prove that there was no new information to be gained in relation
to emissions and efficiency, then stopped. We can pretty accurately predict
daily and monthly emissions and the fuel savings/use of a variety of stoves
with tests lasting no more than 7 or 8 hours. As UB is primarily a PM
reduction project, that is enough.

Regards
Crispin