[Stoves] Wonderwerk stove, more information

Crispin Pemberton-Pigott crispinpigott at outlook.com
Sun Jul 24 23:35:45 CDT 2016


Dear Kirk

 

I have been looking into the math of the results you obtained. They were prepared by LBNL using a set of calculations coded in Python that mimic the calculations described in the WBT 4.2.3 but do not exactly follow the calculations in the accompanying spreadsheet, where the sheet differs from the protocol. In other words, they are preparing the numbers according to normal engineering standards and then re-expressing them according to the WBT theory.

 

In fact they do not normally present WBT calculations and metrics unless you ask for them because, as most labs are aware, they answers are not all that useful for real work – and I am serious about that. They prepare formal performance numbers but if someone insists on a “WBT result”, they are provided according to the WBT method.

 

Not only are they correcting the numbers, LBNL also does not use the standard EPA test apparatus having found they get better answers using other approaches. They have a SeTAR type diluter – I knew about that because I provided the drawings though I don’t know when it is used. They are also using dry CO2-free air for dilution as per the SeTAR/SEET/BST labs. I was most interested to hear they are already recording the humidity in the stack (in the diluter and via dilution number). This is important because the real time determination of the fuel analysis (what just burned) can be calculated if the humidity is known. Watch this space, testers. 

 

So, if you were to ask them for the performance numbers based on regular criteria such as the fuel efficiency based on fuel consumed per replication and work performed (no ‘credit’ for char) you get the energy efficiency. That number can be used to directly compare stoves and their fuel consumption. Obviously you would have to have it for the other stoves as well to make a fuel consumption comparison. Some number are available from the CSI Indonesia approved stoves list as that is one of the metrics if you want to check.

 

The emissions per net MJ delivered to the pot is a common one, and the emissions per minute. No emissions number “per litre” are valid so don’t accept/claim those unless you are sure it was for high power and it was based on the initial water volume.

 

The unfortunate news is that as they do not provide a ‘WBT spreadsheet’ filled in so it is not possible to work backwards and recover low power metrics that are valid (because you would have to multiply by the number of litres remaining to get them, and you don’t know how much there was). 

 

For those following this thread, have a look through the stove test result posted on the GACC website and you will notice that some stoves are providing a ‘spreadsheet’ of outputs and others provide just the numbers on a form. One that I noticed has a very long list of outputs that is copied from the Results page of a PEMS/LEMS. Please note that the results calculated by that sheet (assuming you have used the updated version) does not give the same answers as the WBT spreadsheet 4.2.3 or 4.2.4 because Nordica corrected some of the calculations on the LEMS ‘Test’ sheet. Compare with caution.

 

> When and if the test you support is available the Wonderwerk will do well when tested by it, as will the other four stoves Aprovecho has developed, and other good stoves everywhere.  The CO and the PM are measured directly with precise instruments.  There are no questions here and good stoves will show their quality. 

 

That is what we always expect. The issue is not that these stoves are ‘not good’ it is that in order to accept a stove into a program it has to be tested and reported using valid metrics. I have tried very hard to find your raw data so that I can produce for you valid reporting metrics. The WBT low power metrics are not, so we can’t use your results. Anyone interested can read the literature and find out what the problem is. The EPA reproduced the experiments concerned, validating the claims in Zhang et al, 2014 that there is no validity in a low power metric where the measurements have been divided by the number of litres remaining in the pot, so that pretty much settles it.

 

If, during the coming stoves camps, stove producers ask for the actual test results on the original spreadsheet, there are some calculations that we can make to extract valid performance numbers that are good enough to make recommendations for further investigation. It would be helpful if the whole community of stovers insisted on this because in spite of long haul efforts trying to get these things corrected, we are still faced with results presented using invalid or incorrectly calculated metrics. Dean Still has confirmed to the US-DOE that I have been trying to get these corrected for 10 years. Imagine!

 

>…Where is your test?

 

While I am flattered that you once again call it ‘my’ test. It is a team effort. You can get a copy from the list of protocols at the GACC website http://cleancookstoves.org/technology-and-fuels/testing/protocols.html

 

At the same location there is a slightly old version of the SeTAR HTP v1.05. When it comes to calculating the thermal performance it is very similar in some ways to the Indian CIS 13152 (just underneath). That has been updated to become the CSI-Indonesia test.  It is a water heating test, and uses a monitored heat exchanger in the pot so the water never boils. It can report the heat flux into the pot in real time as well as emissions at different power levels, as well as overall numbers. This will be combined with the ‘decombustion’ formulas sometime this year to provide the most accurate real time reporting theoretically possible. It will be of great interest to stovers producing char while cooking. 

 

The space heating version of the HTP v1.05 was validated by SGS in the Netherlands in 2011 for use in the Mongolian Clean Air Project. It is now their National Standard and includes additional (perhaps, obscure) metrics recognizable to power station and combustor design engineers. One, for example, reports power per unit volume. Kind of interesting.

 

The CSI Technical Test document is below that. 

 

A test that I did create alone is the UCT 2011 from the University of Johannesburg, further down the page. That was validated by DGIS in 2008, if I recall correctly, for their Access to Modern Energy programme implemented by (then) GTZ. It is similar in many ways to the method used by Omar Masera in Mexico City although they were developed quite independently.

 

Happy testing everyone. Check the fine print.

Crispin

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.bioenergylists.org/pipermail/stoves_lists.bioenergylists.org/attachments/20160725/895bd368/attachment.html>


More information about the Stoves mailing list