[Stoves] Error in the recommended Shell Foundation HEH Project Water Boiling Test http://ehs.sph.berkeley.edu/hem/hem/protocols/WBT_data-calculation_sheet.xls

[Crispin Pemberton-Pigott]

Dear Friends

I would appreciate it if a couple of people could check out what Nigel and I
have found and to confirm our calculations relating to the Shell Foundation
HEH Project Water Boiling Test evaluation sheet.  It is in the file:

http://ehs.sph.berkeley.edu/hem/hem/protocols/WBT_data-calculation_sheet.xls


which is available from 

http://ehs.sph.berkeley.edu/hem/page.asp?id=42

The version on the site appears to contain two calculation errors in the
cell showing the net heat content of wood, compensated for its moisture
content.  The error is large and this file should not be used for evaluating
the performance of stoves unless it is corrected.  I believe this
spreadsheet was used for the establishment of a series of performance
benchmarks among many other things.

Specifically, the formula for calculation of the heat energy available from
a sample of wood is in the Test-1 to Test-4 sheets in cell E22 which is
titled "Effective calorific value (accounting for moisture in the wood)"

E22 contains the formula

=(E20-(E21*((N22-E17)*4.2/1000+2.26)))/(1+E21)

Where 
E20 = net calorific value of dry wood (LHV)
E21 = moisture content (wet weight basis)
N22 = local boiling point
E17 = ambient air temperature
4.2 = specific heat of water
2.26 = latent heat of boiling water

It should read

=(E20*(1-E21)-(E21*((N22-E17)*4.2/1000+2.26)))

Or more accurately:

=(E20*(1-E21)-(E21*((N22-E17)*4.186/1000+2.257)))

The formula difference causes an error that varies with moisture content:

Wood at 18414 KJ/Kg LHV

Moisture   Original    Corrected  Difference
  0%        18414        18414        +0.00%
10%        16749        16313        +2.61%
20%        15345        14212        +7.96%
30%        14164        12112        +16.94%
40%        13152        10011        +31.37%
50%        12275          7911        +55.17%

The effect of the error in the heat available from the fuel figure is so
large that anyone using that spreadsheet should consider recalculate any
stove tests they have made using it.  The effect of the error is to greatly
under estimate the performance of a stove that was tested using damp wood.
It does not overestimate the performance of a stove using dry wood.


Specific Fuel Consumption

This same spreadsheet also has the calculation error for the specific fuel
consumption that was discussed on this group some time ago.  The issue is
that the evaluation method calls for the water remaining in a pot at the end
of the boil and the end of the simmering phases to be used in the
calculation of how much wood it took to boil and simmer a litre of water.

The choice of using the initial or final water mass is not a matter opinion.
The consequence is real.  Using the final water mass instead of the initial
mass rewards less efficient stoves with a higher apparent efficiency, and
rates a more efficient stove with a higher quantity of fuel used per litre
than should be the case.  This was corrected in the ProBEC stove testing
protocol.
 
Some people using HEH test evaluation sheet have been perplexed by the
difference between the lab and field tests.  The SFC error can explain some
of this but now it is clear that the "Effective calorific value available"
error has a much greater effect.

The spreadsheet is locked with a password so it will have to be corrected at
Berkeley.  The test is also used in the Indian and Chinese improved stove
programmes which should consider re-evaluating the work done so far.  If
this discovery is confirmed, it would be best if both the SFC and the
effective calorific value calculations were corrected at the same time and
the site updated.

Regards
Crispin and Nigel Pemberton-Pigott
Waterloo
Ontario
Canada