[Stoves] Sets of tests and the CoV of results - various methods

[Crispin Pemberton-Pigott]

Dear Friends

Due to the interest in the reproducibility of a test result, and  because I earlier posted something about the difference between WBT result variability and the minimum expectation for validation of a CSI protocol equipment, I am forwarding here a set of results from a stove test conducted using the current Indian national standard (chula / cooking stove test).

[cid:image002.jpg at 01D27951.0EF206A0]

Notice that all the output metrics are below 10% CoV. If this had been the result of a test for inclusion in the Central Java Pilot it would have received 1 star (just) for fuel efficiency, 3 stars for CO, 2 stars for PM (total 6) and come with a written warning that it was unlikely to be accepted on two counts: lack of fire power and a low heat flux (average heat flow into the pot expressed in Watts per cm2).

In other words it would be acceptable to the project (which would approve it being made available if a marketer wanted to try to sell it). The inclusion of market information and consumer behaviour, which is what  comes out of the work Cecil was including in a 'testable' and 'rateable' form, combined with the technical metric (Heat flux or 'heat flow rate' - HFR) where Cecil and I worked together gives a lab the ability to make a more holistic prediction of how the product will fare in use.

The point of the ratings was to provide a marketer with additional information beyond the usual PM/CO/efficiency triumvirate.

We don't tell marketers what they should or should not try to promote, but we do give warnings arising from the field study that reflect the difference between performance and expectations. We have not done a comprehensive scan of all the things we could do with a lab evaluation because the time for this was short, but the concept is being continuously discussed.

The 'innovation' was to include technical conceptual thinking with social conceptual thinking and arrive at field observations or even tests that can be conducted by the anthropologists and brought 'home'. Through lengthy conceptual consultations certain testable truths were unearthed that related to the usability of a stove, as while being able to retain the standards metrics in as relevant a form as possible (predicting performance in use).

This exploratory work is of course very difficult to get funded.  We were most fortunate to have support form the WB Stove pilot team with additional support from the SE Asian gender studies team (who provided Cecil and two others).

Not that it got much attention at the time, I did report here the development and importance of the heat flow rate as a market-relevant. A 'heat flux' is a heat flow rate per unit area but I found the word 'flux' is not commonly understood.  Further, in a large fraction of the market footprint, they do not express the term in the same manner: in the West it is Joules per second per unit area (m2 or cm2). In Asia almost all countries use Joules per unit area per second and is called the 'density of heat flow' (all Russian-speaking former Soviet Republics + China).

The importance is therefore the understand that this technical expression of socially-assessed performance, is the opinion of the users as to what constitutes 'adequate cooking power' without making reference to the firepower - the only thing reported by the WBT that even hints at social acceptability.

Let's then consider another CoV: this time the CoV of what people in the community expect. If what constitutes an acceptable HFR has a high variation within a target market, one has to pick the higher end or the stove will be rejected by all requirements above the stove's capacity.

Reminder: if you know the HFR for a stove you can calculate the time needed to boil 5 litres of water (or any other quantity) with surprising accuracy (±30 seconds) obviating the need to ever conduct such a test.

Regards
Crispin

-------------- next part --------------
A non-text attachment was scrubbed...
Name: winmail.dat
Type: application/ms-tnef
Size: 79027 bytes
Desc: not available
URL: <http://lists.bioenergylists.org/pipermail/stoves_lists.bioenergylists.org/attachments/20170128/c585436d/attachment.bin>