Dear Ron
You are right – nothing can be predicted. It is important at this time in the evolution of stove testing protocols to recognize that single tests run to a rigid protocol are not nearly as helpful in assessing stove performance as tradition has led us to believe. A WBT is a two power level test so if the stages are reported separately, that is a beginning step in the right direction. Using two or three pot sizes is important. With and without lids is important. Different power levels is the basic test by any engineer assessing the performance of a thermal system. Thus we advocate heterogeneous protocols that produce performance curves. Doing this is not difficult. It is far better to test at three different power levels than to test three times at one. The reason is that if three identical tests are within 10% of each other, nothing was learned. If three non-similar tests are conducted, you have the basis of a performance curve. Something to keep in mind is whether or not the test is of the protocol or the product. The EPA test for wood stoves is a really good example of the testing of a protocol. As the test is so far from the conditions of actual use, the results are not relevant to the emissions that will come out of the product. So people developing stoves for the US market all get a chance to see how the protocol treats their product. It is not a very good test of how the product burns wood in a home (which as you recall, was the original purpose of the test). We are entering a new era in stove testing. Stoves are starting to be treated as normal combustors or boilers or heaters and there is no reason not to. When you buy a heat exchanger you expect and get a performance chart with variables on the axes. When I bought the vacuum and pressure pumps for the lab I looked at performance charts for the products and selected the ones that suited my purpose. The same should apply to stove performance. Yesterday we did the first fully instrumented stove test! It was a shakedown process in which we found a couple of bugs and glitches as one might expect, but we did get real time gases (O2, CO, CO2i, H2, NO, NO2), the total system mass, PM 1.0, 2,5, 4,0 and 10. We did not get temperatures because it was not important but we have a 20 channel Agilent 34972A multiplexer to satisfy the most demanding stove or building evaluator. Today we tested again using a different coal with lower CxHy content in order to determine why we are getting gas measurements when we are supposed to be getting air at the instrument (inadequate pipe size on the waste lines) and why we got very low CO2 on the diluter (pulling air into the outlet because the gas flow was too low). Looks like today we got everything right by twiddling knobs only. We can get a dilution of 4:1 at 150 litres per hour in the diluter and 55:1 at 250l/hr. That is right on spec and we are delighted. We can run it much higher if we want but at the moment the PM is being kept below 150 mg/m^3 so it is in range. We have already blocked a few filters so we will look at what to do about increasing the area of them. Both tests were of a traditional stove with a closed outlet hole – reduced to 65mm diameter to limit the total air mass passing out of the stove. The fire is lit immediately next to the chimney outlet. That is the only change. We expect an 80% reduction in particulate and CO emissions. Today we proved what we suspected for some time which is that particulate emissions are very low during coal combustion as well as in the dying fire phase (we were never sure). The PM measurements are corrected for the dilution in the diluter and the Excess Air in the stack. The multiplier this afternoon was 40.6. The TSP (PM10 and under) was about 130 microgrammes per cubic metre in the dying fire phase. That is quite a bit less than outdoor air in November in Ulaanbaatar. Seriously. You are better off breathing what comes out of an ELCD chimney than the open air in UB in November. Amazing, neh?! We were joking that we can clean up the air outside by running it through a well run coal stove. So, all very exciting. We are still seriously short of instruments I would love to have but having got this whole thing running for about 5% of what we were told it would cost, we can’t be more pleased. For those who need to develop some sort of lab don’t hesitate to get in touch. I am committed as you know to spreading knowledge around until it falls off the table. John D, do look up that little NDIR CO2 sensor. Looks promising. Regards Crispin From: [email protected] [mailto:[email protected]] Sent: Friday, 06 August, 2010 11:15 AM To: Crispin Pemberton-Pigott Cc: Discussion of biomass Subject: Re: [Stoves] Coal stove testing in Ulaanbaatar - 2 pictures Crispin and list: Thanks for today's additional information on testing, including the nice 8 page report. I read your 8 pager moderately closely, but not real closely because it was all on paraffin and ethanol stoves - and I am only interested in stoves that might make charcoal. One surprise is that there are commercial stoves where you think you are turning up or down and the reverse happens. I don't think that could possibly happen with a TLUD or BLDD with primary air control. Any explanation? I am surprised that they are not cleaner, and are so variable on where they operate best - apparently nothing can be predicted? I think a summary sheet on the 100 or more stoves tested so far would be very helpful. Am looking forward to those that burn or (better) pyrolyze wood. Still a few questions below (with truncations). <snip> >Late last year we trained 20 people from 7 universities in Africa how to >conduct basic tests and how to use the spreadsheet. They are being provided >some test equipment including a hood. [RWL2: Fantastic. Who is paying for this really important addition?] That was funded by PRoBEC. [RWL2B: Not surprised. The Germans are really doing great things on all aspects of renewables and climate. >…It looked like a good burn but the CO level was quite high and the designer >could not go further without emissions testing. Using only the controllers >provided for primary and secondary air, it was possible to reduce the CO/CO to >0.5%. [RWL3: Sounds like either TLUD or BLDD?] Side draft. >…We then constructed a slightly modified combustion area which produced a best >figure of an additional order of magnitude reduction to 0.02%. [RWL: I would have no idea how to make this big a difference with only "slightly modified". Can you say a bit more? Actually that was discussed in detail – the messages about the depth of the fuel bed. The depth was increased by 10mm from 75 to 85. The air supply was more than cut in half and the secondary air nearly eliminated. That should confuse the biomass burners! J >In other cases we have located good potential in stoves that looked average or >‘bad’ using this approach. [RWL: Same question - what sort of changes? ] I was referring to locating performance issues that are detected by running the stove at various powers and with various pots and operating procedures. Regards Crispin
_______________________________________________ Stoves mailing list [email protected] http://listserv.repp.org/mailman/listinfo/stoves_listserv.repp.org http://stoves.bioenergylists.org http://info.bioenergylists.org
