[Stoves] Heterogeneous testing protocols

[CEDESOL Foundation]

Interesting comments Crispin.  Did no one listen to you in the Hague?

On 4/25/12, Crispin Pemberton-Pigott <crispinpigott at gmail.com> wrote:
> Dear Testing Friends and Appealing to Peter Verhaart for comment (because
> you were there!)
>
>
>
> I have been looking over the history of stove testing and it is compelling
> that the HTP was used for reporting stove performance as early as 1982-3.
> There are a number of references to it in
> http://www.cookstove.net/others/fuel-economy.html with a number of
> citations
> from Prasad and Verhaart as well as Piet Visser.
>
>
>
> There are some Sankey diagrammes and also inherent heat flows. These are
> useful for determining stove performance in a way that allows useful
> predictions of field performance to be made based on laboratory tests.
> Alex,
> See Fig 12 and 13. Is that what you have in mind for system analysis? Is
> there software or a function in Excel that can produce that automatically?
> It could be built into a stove design tool.
>
>
>
> I am attaching some performance curves produced created from a set of more
> than 80 tests of three stoves. The relationship between power and pot size
> is clearly visible for each stove. If you want to know how it will cook
> your
> favourite 'burn cycle' you can make a calculation. I am suggesting this is
> a
> useful extension of the good works done in Eindhoven.
>
>
>
> The advantage we have these days is that getting precise emissions
> measurement is much easier, as is logging mutiple sets of data in real
> time.
> The resulting displays are easily generated and optimal conditions
> identified quickly.
>
>
>
> So my question is how did we go from conducting really good test and stove
> analysis with really valuable heterogeneous characterisations in the early
> 1980's to a simplistic water boiling test in 2012? It is clear from the
> works pouring out of Iowa State for the past 5 years that WBTs are not
> helping us much in providing useful information - nothing like what we used
> to get before and what we an get now from heterogeneous tests.
>
>
>
> So my question to Peter Verhaart is, what do you think about the current
> state of affairs? How do we recover from it?
>
>
>
> Grant B-T, if you have time, what are we going to do to revive testing to
> the point that a programme manager can make an informed choice about stoves
> before rolling something out? I am sure you know the systematic error level
> in the WBT's is over 30%.
>
>
>
> Dr Nate Johnson, you have beat your head against this wall and surely have
> some advice to take us forward. The industry is getting stuck even as the
> funding available for stove projects grows by the millions. Never has so
> much stove money been directed by so little meaningful information on the
> performance of the products being promoted. This has to change. In fact it
> has to be restored to its former glory using the new tools and
> understanding
> that has developed in the intervening years.
>
>
>
> Interestingly there is a reference at the very end of the article at
> http://www.cookstove.net/others/fuel-economy.html to the carbon balance
> method of testing. This is the method used at the SeTAR Centre and it is
> not
> an accident. Most EPA-like methods use a hood (or cabinet) and mass (or
> whole)-flow calculations, though EPA does use carbon balance methods for
> vehicles.
>
>
>
> The issue is precision and accuracy.  The systematic errors together with
> the experimental errors have to be less than 7% to place a stove securely
> on
> performance tiers 20% apart. That is basic math. In order to do this in
> affordable labs, I suggest that it will require a set of single purpose
> tests performed in an HPT manner. The attached chart is one way to provide
> that information.
>
>
>
> Testing costs a lot.  Money is changing hands. There is little point in a
> lab providing test results that can be challenged if the stoves turn out
> not
> to perform in the band claimed. The more complicated and imprecise the
> assessment task, the greater the potential for failure to perform.
> Reputations are on the line. We have to provide better quality
> characterisations of performance (including durability, safety and social
> acceptance).
>
>
>
> Serious stuff.
>
>
>
> Regards to all
>
> Crispin
>
>
>
>

-- 
Sent from my mobile device

David Whitfield V.
Executive Director
CEDESOL Foundation