Does anybody know (and could maybe post) how the efficiencies are computed in the Polytests wood stove tests, or similar tests in general? I see the combustion efficiency, heat transfer efficiency and overall efficiency columns in the spreadsheets in the report, but the procedures in the report seem to only specify the emissions calculations. I don't see anything about efficiency.
I'm guessing that the combustion efficiency (about 95%) might be computed from the CO content (incomplete combustion), but the heat transfer efficiency (about 75%) really dominates the overall result. The stove under test is not sitting in any kind of calorimeter (it is sitting on a scale to determine the fuel burn rate), and ambient temperature in the (rather large) room does not seem to change much. They do measure the stove surface temperatures on all sides. Are they just assuming some heat transfer coefficient to air? But how could they, as they apparently have the attached blower going (transfer coefficient is a strong function of air speed)? Flue gas temperature is monitored, but they are only monitoring the sampling volume flow, unless I misunderstand their setup (the flow meters are in the sampling lines, not the main flue). Even if they knew the flue gas flow, are they assuming all flue gas energy is lost?
Anyways, I would appreciate a comment, if anybody knows how the efficiency measurement is done.
Answering my own question, and posting, in case anybody besides me is interested... :
During the CSA BS415.1-10 test particulate emissions are sampled in a dilution tunnel, which does draw in additional air, but flue gas temperature, CO and CO2 content (as % of exhaust gas) are monitored in the flue (stove) pipe directly attached to the stove, without any additional admixture or dilution. Excess air is computed from the difference between measured CO & CO2 content and a theoretical CO2 number derived from complete burning of an assumed wood composition. From the continuously monitored weight of the stove + stove pipe + wood one knows the amount of fuel burned and thus the theoretical heat input, and from the exhaust gas temperature and composition and the excess air computed from the measured CO & CO2 content one computes the heat loss through the flue. I think the heat stored in the stove (from the surface temperatures and stove mass) is also computed, but that probably doesn't change much once the stove is running. From the computed heat loss through the flue (and the heat stored in the stove mass) and the weight of the burned wood the net efficiency is computed continuously.
This efficiency is summed/averaged over the burn time, averaged over different burn loads (maybe weighted according to observed North American wood stove heat use profiles?), and that is the HHV efficiency listed on the EPA web site.
In calculating the energy content of the wood the moisture content is subtracted (i.e. the energy is computed from the dry wood mass), but not subtracted is the heat loss due to vaporization of that water, and of the water resulting from the burning itself - this is what using the 'high heating value' HHV for the wood means. If I understand the measurement procedure correctly, condensation of water in the stove is not allowed, the flue gas has to be above 115°C, so this water vapor energy loss is an additional de facto loss for every stove measured according to this procedure.
Varying moisture content wood in the test load will therefore change the efficiency, but the effect is relatively small, 0.15% efficiency change per 1% moisture content change. There is no efficiency correction for that, but the fuel has to be in the range allowed by CSA BS415.1 (18% to 25% dry weight basis).
That's it, in a nutshell, or at least my understanding of it. Please post corrections and clarifications.
Added: an interesting paper online describing a 2016 research effort by chemical engineering undergrads to implement an EPA/CSA BS 415.1-10 equivalent alternative test setup:
https://scholarworks.uark.edu/cgi/viewcontent.cgi?article=1083&context=cheguht . I think it nicely describes the essentials of the measurement method.