[Stoves] WBT test results: metal grate with rock bed and conical pot skirt

Wed Jan 22 17:30:32 CST 2020

Dear Friends

This testing by the SNV lab has been done using their Simplified Water Boiling Test (SWBT).  It is not the same as the ISO test in terms of calculations, and it is also not the same as the WBT in certain respects, being closer to the latter than the former.

The important difference is that you will note in the report that was attached to Kevin’s message the use of the term “Wood fuel equivalent consumed”.  This is the bugbear of that WBT – it does not report the wood consumed to operate the stove, it reports the dry wood fuel equivalent of the energy presumed to have been released from missing fuel mass, with a crude compensation made for the charcoal remaining in the ashes under the fire.  The more char created, the bigger the deduction.

The problem with this approach (which dates back to 1985) is that the method was designed for stoves that all burn pretty much the same way, leaving about the same amount of char (as a % of total fuel burned).  Kevin’s modification changes this % of fuel appearing as char at the end.  Direct comparisons between stoves, in this case the same stove with a change, are not valid if one is purporting to report the fuel needed to light a fire and cook a meal. The WBT doesn’t do that.

This issue came to light in 2007 when I made a big noise about it and some efforts were made to propose suitable corrections to the method.  An ETHOS technical Committee as formed to correct the test, but did so under secret instruction not to have the resulting change invalidate a body of work that had already been completed.  So that effort failed even though the problem was obvious, more so with stoves that produce a relatively large amount of char.

Ron Larson did his best to assert that the remaining char should be treated as unburned fuel that could be used in a different fire. He then planned to bury the char for carbon sequestration.  His position was that the stove should be made to look as if it didn’t use much fuel so it would get a high “energy efficiency rating” and then have the char to sequester and that such a rating was “needed”.

A very long drawn out argument ensued in the halls of this channel and the ISO meetings about whether it was legitimate to report the “energy released value converted to an equivalent dry mas of fuel” or the actual fuel mass needed to run the fire.

Now, for Kevin this matters a lot.  The advantage of the rock bed and the metal grate on rocks with or without rocks on top is to burn the char instead of having it left over.  It reduced emissions and reduces the fuel needed to cook.  When SNV reports that there was a set of SWBT’s run and the difference between the three stove fire and the rock grate was a fuel saving of 35%, and the metric used was Wood fuel equivalent consumed, the answer given is actually high compared with a “real test” and low compared with a similar test conducted on a three stone fire.  For example in one test the reported value was 38% but the actual value was 34% (a difference of more than 10% of value). The technical change saves more fuel than that, if the energy in the char mass remaining was deducted from the energy released from the fuel actually consumed for both stoves.  If anyone needs further clarification, I can provide it.

If the test was a WTB or a char-deducted SWBT, the error is the same: the stove uses more fuel than the “Wood fuel equivalent consumed” indicates.  However, the char remaining with the grate will be much less so the error (from true value) is smaller.  This means there is a mismatch between the errors in the comparison report.  The rocks+grate will have a smaller error than the three stove fire because the mass of char (as a %) is smaller.

If you have the spreadsheet that was used to create a test report, set the value of char remaining to zero for all three sections of the test and the mass of fuel needed to operate the stove throughout the test will be reported “reasonably correctly” if it was done near sea level.

When the adaptation or improvement to the stove results in a drop in the mass of char produced, a WBT in any form under-reports the level of improvement.  The improvement is not “conservative”, it is “wrong”.  In fact both are wrong, but one is not as wrong.

It is worth noting that the fixing of this error in the WBT has been strongly resisted by: Berkeley (LBNL), EPA, ANSI, Colorado State U, Illinois State U, PCIA, Berkeley Air, Aprovecho, GACC (currently the official custodian of the WBT) and Envirofit.  The error (which in general over-reports performance, especially of Rocket-type stoves with no grate) is bundled into the calculations of hundreds of CDM, VCS, GS and VERRA carbon-trading projects.

Envirofit and GERES both produced their own versions of the WBT but did not correct this fundamental error in reporting the fuel consumption. The former added two more errors, and GERES corrected one.

The new ISO standard lab test permits the reporting of a corrected version and the original erroneous version, and the wording is such that it is easy to be deceived into thinking one is the other.  Worse, the Field Test mentions as a test method the Berkeley CCT 2.0 (2003) which has the same fuel consumption error, so that is a fail too.

In all cases, try to be clear about which test was used, which metric is being reported to you, and which errors have been included in the result. The difference can easily push your stove up or down one or in the case of char making stoves, two tiers.  These test results are not interchangeable.

If you want to know what to use instead, try the EAP CSI Test used in Indonesia 2012-2017, later fully developed into the CSI test for stoves, heating stove and low pressure boilers.  It is used as the default test method at the following labs:

SEET Lab, Ulaanbaatar
BEST Lab, China Agricultural University, Beijing
YDD Lab, Yogyakarta, Indonesia
Ministry of Energy Lab, Jakarta, Indonesia
SeTAR Centre, University of Johannesburg
Stove Testing Lab, Department of Geography and Environmental Sciences, North-West University, Mafikeng Campus, Potchefstroom, South Africa

Given that Kevin has achieved such significant results with very inexpensive methods, and that there is a real possibility the relative performance is being under-reported, it is perhaps wise to hold a session at ETHOS this year to discuss the implications and possible retroactive corrective measures that might be taken in light of the persistent refusal for years to correct these fundamental problems in performance assessment.

Regards
Crispin

From: Stoves <stoves-bounces at lists.bioenergylists.org> On Behalf Of K McLean
Sent: Wednesday, January 22, 2020 11:22
To: Stoves and Biofuels Network <stoves at lists.bioenergylists.org>
Subject: [Stoves] WBT test results: metal grate with rock bed and conical pot skirt

SNV just released the attached test results that I find very exciting.

Compared to an open-fire cookstove without modifications:
46% less firewood - metal grate under rock bed
42% less firewood - conical pot skirt.

SNV performed Simplified Water Boil Tests (SWBT) which only tested thermal efficiency, not emissions.

Metal Grate and Rock Bed
The first test was a modification of rock beds alone.  Last year, SNV's SWBT showed a 31% improvement in fuel usage by adding a rock bed to an unmodified open-fire cookstove.  In this new test, SNV added a metal grate under the rock bed and elevated the grate by a few very small rocks.  With the metal grate and rock bed, fuel saving jumped to 46%, compared to the unmodified stove.  This is a simple modification that can be made and sold for well under 1 USD.

Conical Pot Skirt
A conical pot skirt directs the hot gases of an open-fire cookstove to the cookpot.  It is easy to make and inexpensive to make, probably around 1 USD.  SNV's SWBT found that it reduced fuel usage by 42%.

The conical pot skirt has limitations.  It may not work well with three-stone.  (We are working on other pot skirt designs for three-stone.)  In SNV's configuration, the skirt was custom made to the pot with a 1cm gap.  Further testing is needed to determine the amount of efficiency lost when used with a smaller pot, and therefore a bigger gap.

[Drawing - Conical Skirt on Metal Bar Cookstove.jpg]

We next need to test with rock bed, metal grate and pot skirt.  I hope we'll find the fuel efficiency is improved by more than 60%.

Thank you,
Kevin

Kevin McLean, President
Sun24
https://sun24.solar<https://eur04.safelinks.protection.outlook.com/?url=https%3A%2F%2Fsun24.solar&data=02%7C01%7C%7C68b4b09031ec4d4c877f08d79f5791b6%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637153071090435200&sdata=4bKyzbMuEFhVtVe8tT%2F3FQGhnw9gF4i4sB6cP2HG83I%3D&reserved=0>  Sun24 Cookstoves Overview<https://eur04.safelinks.protection.outlook.com/?url=https%3A%2F%2Fdocs.google.com%2Fdocument%2Fd%2F1ZryS7gQ1q3zKLZPM2KcXdtIHbOYQp4PbloPqMvrlZ5Y%2Fedit%3Fusp%3Dsharing&data=02%7C01%7C%7C68b4b09031ec4d4c877f08d79f5791b6%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637153071090445208&sdata=zf7y5lEcWWtn8%2Bb7JW7wTpbPh8n941soE3VWx4%2B9XPg%3D&reserved=0>
Tampa, Florida, USA
+1 (813) 505-3340

             [https://docs.google.com/uc?export=download&id=13teZQbnv_6ek6JkEy-4n0yyUVTrl3M2W&revid=0B8zNcWsuiuj3N2RuVFFYL3EvdElxR1lDcm9QQVlaL0F4VnVFPQ]         [https://docs.google.com/uc?export=download&id=11k6wcaMcCbykOoLu-Qo4n6Kbdhplk_er&revid=0B8zNcWsuiuj3ejdsRjRXQWU2OW4rVUpwMUx0Q0U0eXgxcStBPQ]
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.bioenergylists.org/pipermail/stoves_lists.bioenergylists.org/attachments/20200122/f25eb506/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image002.jpg
Type: image/jpeg
Size: 11127 bytes
Desc: image002.jpg
URL: <http://lists.bioenergylists.org/pipermail/stoves_lists.bioenergylists.org/attachments/20200122/f25eb506/attachment.jpg>