[Stoves] PCIA den Haag

Crispin Pemberton-Pigott crispinpigott at gmail.com
Fri Mar 2 13:03:56 CST 2012


Dear Frank

 

First, thanks for the offer to do fundamental research that will provide
input to the development of better test methods. If you have the means to
have them published, that will be best.

 

My immediate suggestion is that you look at the heat content of typical
charcoal residues from known wood and known stoves. For example a Rocket
stove burning Douglas Fir as is common at Aprovecho. That is a particular
stove type and the fuel is apparently constant. The question arises: what is
the heat content of residue char and does it vary with the initial moisture
level?

 

We need to know the Carbon and Hydrogen content to determine the heat
content without having to resort to a bomb calorimeter which is time
consuming even if inexpensive. The heat potential can be calculated pretty
well if you have the H2 and C content.

 

>From the gasifiers we know that the amount of char left varies greatly with
moisture  content. It is very likely that the heat content per g is also
variable with the initial moisture content because 'there is something else
going on'. Rice hull char has a very low energy value, but what is it? Lots
of post-grad work in there for the willing.

 

Dr Penn Taylor (we can call him that now) outlined one method for
determining the accuracy of two WBT's. There were several steps or points
that led to the total. Each point is an error that needs to be constrained.
If a test method has fewer error points, the final result is more precise. 

 

WBT's are not inaccurate because the measurements are not made precisely,
but because the task given is complex and depends a great deal on the skill
of the operator. One issue is the determination of heating efficiency of a
stove. It could be the % of heat that goes from the fire to the pot (or the
room if space heating) or it could be the % of heat that comes from the fire
and is retained by the pot so some work is done (like heating the water).
Those numbers are not the same. It could be the average efficiency including
the ignition of the tinder and kindling all the way to the boiling point.
That is a very different number because the stove body might absorb a lot of
the heat initially. Also, the skill of the lighter of the fire (the cook)
might be widely variable, introducing a systematic error. Similarly with
emissions. If the skill of the cook is high, the fire will start cleanly in
some stoves. In others it makes nearly no difference. Should dumb cooks be
penalised? Or should we penalise the stove?

 

One of the reasons to report the net heat transfer efficiency (the
efficiency of the whole system) at different power levels is that not only
is it a very useful metric for creating a performance curve of efficiency
v.s. power, but it can be done quite precisely with simple equipment. You
could easily make a determination within 3% in your lab right now. But you
can't light a stove within 3% each time on different days using a variety of
fuels.

 

The heat transfer efficiency is largely an indication of the performance of
the stove body with that pot(s) on it. It is a very valuable metric because
if you make a change, you can see how the efficiency was affected. If that
is determined under 'constant conditions', meaning a fire that is not
currently changing in power level, the result can be really accurate,
probably within 1% using a thermocouple that is accurate to 0.2%. You get
the idea? So when reporting how a stove 'performs' one can generate accurate
numbers of great value.

 

Determining the emissions from a stove during an 'ignition phase' (lighting
to stable, full power) is a clearly defined section of a cooking event.
Again it can be determined accurately by making sure the systematic errors
are minimised and repeating the procedure a number of times, reducing
experimental error (like fuel variation). Using this approach one can build
up a characterisation of the stove that is precise and useful.

 

Dr Nat Johnson (we can almost call him that) showed us a chart where he had
the efficiency of a group of stoves plotted twice: once with the charcoal
subtraction invoked and once by pretending the char remaining was tossed.
The coefficient of variance (COV) with the (inaccurate) char heat content
included was huge. When the char was ignored the COV dropped by 50%, the
results were obviously more accurate (looking at the graph) and he wondered
why we introduce such large errors into an otherwise pretty good test
section. We may have to consider char, but how it is handled can be
discussed to limit these large error bars.

 

The best way to start this process following the watershed IWAgreement is to
ask first what the buyer (often a Donor) wants. They probably want to know
if the stove they are selecting will perform as advertised. That is a
starting point for selecting methods. If the method does not address the end
goal, it has no value to anyone. If a designer wants to know what efficiency
the user will experience, the test (or that section of it) has to provide
the required information. You lead the rest of your life like that, why not
stoves too??

 

Regards

Crispin

 

 

From: stoves-bounces at lists.bioenergylists.org
[mailto:stoves-bounces at lists.bioenergylists.org] On Behalf Of Frank Shields
Sent: Friday, March 02, 2012 6:50 PM
To: 'Discussion of biomass cooking stoves'
Subject: Re: [Stoves] PCIA den Haag

 

Dear Crispin,

 

Perhaps lengthy but very well written and states the problems in rating
stoves. All the work in the past (WBT etc.) were useful measures of a small
part of the big picture. Starting with test methods that are simple and work
on small units of the process are key. Then they need to work when used in
the bigger lab scale of testing stoves. Then verified they represent what
happen in the field.

 

What would be useful is a list of the unit tests that need to be worked on.
Then have labs suggest ideas for research and have money available for the
research. Determining energy in ash-char mixes, carbon energy in
particulates leaving the combustion box, and location of formation of these
particles are a few I can think of. If we can develop simple methods that
can be used by most labs that have been verified to work when correlated
using costly lab equipment is a start.

 

So much to do. Getting organized as to what needs to be done by posting a
flow chart or tree. Then have a way to include all that want to contribute
with suggestions of procedure etc. by e-mail directed to a sub-number of the
tree.  The ones making decisions can look at the list of all suggestions and
move appropriately. But I am sure they have their own process. 

 

 

Thanks

 

Frank

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.bioenergylists.org/pipermail/stoves_lists.bioenergylists.org/attachments/20120302/ea500b0e/attachment.html>


More information about the Stoves mailing list