|
Environmental Project No. 1139, 2006 Comparison of measuring results between solid fuel stoves tested in accordance with EN 13240 and NS 3058Contents1. Description of the Assignment PrefaceA considerable share of particles pollution in Denmark comes from residential woodburning stoves and boilers. This pollution is mainly caused by incorrect manual stoking. But the design of the stoves also plays an important role, as moderne woodburning appliances emit a lot fewer particles than do older models. Residential woodburning stoves are essentially construction materials and regulated by the EU construction products directive. This directive places a number of requirements on the construction of the appliance, including requirements regarding fire safety, human health and environment. The requirements are specified in a harmonized European standard (EN 13240), which, however, specifies neither requirements nor method of measuring emissions of particles. The Danish Environmental Protection Agency wishes, with this project, to examine, if there is a correlation between particles and CO-emissions, that in the absence of a European standard for particles, would allow a reduction of particles emissions by setting stricter limits to CO-emissions from residential woodburning stoves. Sammenfatning og konklusionerFormålet med udredningen er at undersøge, om der kan påvises en sammenhæng mellem mid¬delværdien af CO-emission i %, målt under EN-afprøvningen og det vægtede gennemsnit af partikelemissionen, målt under de fire delforløb under NS- afprøvningen, opgivet som gram udledte partikler pr. kg indfyret tørstof. Undersøgelsen baserer sig på målinger på i alt 26 brændeovne over perioden august 2001 til november 2005. Ud af de 26 brændeovne foreligger der tillige måling af THC/OGC på de 20. Det må konkluderes, at man ikke på baggrund af denne udredning kan drage tydelige paral¬lel¬ler mellem CO-værdier opnået under EN-afprøvning, og værdier for partikel¬emission op¬nået under NS-afprøvning. Det vil derfor umiddelbart ikke være muligt på det foreliggende grundlag at substituere krav til partikelemission med krav til CO-værdi. Summary and conclusionsThe purpose of this report is to find out whether there is a connection between the mean value of CO emission in %, measured during the EN test, and the weighted average of the particle emission, measured in the four partial courses during the NS test, stated as gram of released particles per kg of fired dry matters. The examination is based on measurements of a total of 26 solid fuel stoves during the period from August 2001 to November 2005. Measurements of THC/OGC have also been carried out on 20 of the 26 solid fuel stoves. It must be concluded that it is not possible on the basis of this report to make clear parallels between CO values achieved during EN testing and values for particle emission obtained during NS testing. Therefore, as a matter of course it will not be possible on the existing basis to substitute requirements on particle emission with requirements on the CO value. 1. Description of the AssignmentThe purpose of this report is to find out whether there is a connection between the mean value of CO emission in %, measured during the EN test, and the weighted average of the particle emission, measured in the four partial courses during the NS test, stated as gram of released particles per kg of fired dry matters. Thus, this is an attempt to compare data from the same solid fuel stove, but originating from two different tests under different test conditions. To keep the record straight we hereby give a summary of the most important differences of the two measurements: Flue draught During the EN test a regulated flue draught of 12 Pa is applied, whereas the NS test is carried out by natural chimney draught of the test chimney, which typically is higher. Damper settings During the EN test a damper setting is used, which optimizes the duration to at least 45 minutes and generates low CO and OGC values and a high CO2 value. During the four burn rate categories of the NS test, different damper settings are applied, from almost closed to fully open, in an attempt to obtain the four required combustion rates in an interval lower than 1.25 and up to more than 2.8 kg of dry matter per hour (Class 2). Applied fuel Hardwood tree is applied in the EN test, and the amount is stated by the manufacturer. During the NS test conifer wood of 50 x 50 mm is used, and the amount is calculated as a function of the combustion chamber’s volume. Furthermore, it is closely defined how the NS test fuel must be cobbled up and placed, whereas it is the manufacturer who determines the shape and the placement of the EN test fuel. In comparison, 2 kg of wood is required for a random stove tested according to EN, whereas 2.7 kg of wood is the amount of wood required for a random stove tested according to NS. 2. ReferencesEN 13240:2001 with the supplement DS/EN 13240/A2:2004, in which the requirements on CO are laid down to be inferior or equal to the information stated by the manufacturer (however, max. 1%) and the requirement on efficiency is laid down to be superior or equal to the information stated by the manufacturer (however, at least 50%). NS 3058-1, in which the combustion rate is measured as kilogram of combusted dry matter per hour at the four burn rate categories (Class 2), is defined as:
NS 3059, in which the maximum allowable particle release for solid fuel stoves is stated as:
3. Terms and DefinitionsThe result of the examination is available as diagrams, in which the particle emission is shown as a function of CO value and OGC value, respectively. For each graph, a linear regression has been marked with the purpose of illustrating the tendency. An R² value for the linear regression has been stated. Definition of R², cf. "Complete Business Statistics", Irwin/McGraw-Hill 1999: The coefficient of determination R² is a descriptive measure of the strength of the regression relationship, a measure of how well the regression line fits the data. - and furthermore, about the size of R²: An R² value of 0.9 is very god, a value greater than 0.8 is good, and a value of 0.6 or more may be satisfactory for some applications. When R² value is 0.5 or less, the regression explains only 50% or less of the variation in the data. If we are interested only in understanding the relationship between the variables, lower values of R² may be acceptable, as long as we understand that the model does not explain much. 4. Provision of DataThe examination is based on measurements of a total of 26 solid fuel stoves during the period from August 2001 to November 2005. Measurements of THC/OGC have also been carried out on 20 of the 26 solid fuel stoves. Please refer to the complete data list in appendix 1. The Danish Technological Institute will like to thank the following manufacturers of solid fuel stoves for having provided data for the examination: HWAM Heat Design A/S 5. TheoryIt is expected that there will be a direct connection between the particle emission and the CO value. After the stoking in a test charge there will be a period of time before the flames will arise, and before the combustion seriously gets going. Notoriously there will be a high CO value in this period of time during the EN test and a correspondingly high particle emission during the NS test. In the end of a test period, when the fire goes out, there will also be a relatively high CO value, whereas the particle emission presumably is insignificant. However, the amount of CO is from the start-up normally higher than from the end, so totally there ought to be a connection between high CO values and high particle emissions. A similar connection exists between THC/OGC (tar material – measured according to FID measuring method) and CO. 6. ResultsConnection between particle emission and CO value (appendix 2 and 3) Appendix 2 shows the connection between the total particle emission of the NS test (i.e. the weighted average of all four burn rate categories) and the CO value of the EN test. The graph shows a slightly increasing particle emission as a function of increasing CO value. However, R² = 0.1 is so low, that it will be uncertain to accept the comparison for a connection between CO value and particle mass. A more obvious connection is seen if an isolated comparison is made of the NS burn rate category 1 and the CO measurement of EN (appendix 3). Burn rate category 1 is the lowest category in the NS measurement – in Class 2 an amount of dry matter inferior to 1.25 kg per hour must be combusted. Notoriously, category 1 is the category in which the solid fuel stove has most difficulties to meet the requirements in NS3058 (weighted mean value during all 4 burn rate categories, maximum 10 g/kg – and the highest value during one category, maximum 20 g/kg). Appendix 3 shows a more clear connection between high CO value and high particle emission, even though the results are somewhat sporadic and the R² value of 0.36, cf. the definition above, hardly justifies a comparison between requirements on CO value and particle emission. Connection between particle emission and THC/OGC values (appendix 4 and 5) Even though THC/OGC is not a suitable requirement according to EN13240, we have chosen to include the results in this examination, since both parameters obviously are related to contamination. The pattern is almost similar to the CO dependence above, because of the increasing tendency in the diagrams, both in appendix 4 and 5. However, the R² values (0.20 for NS, total, and 0.39 for NS, category 1, isolated) do not permit a firm comparison of the requirements for THC/OGC and the particle emission. 7. ConclusionIt must be concluded that it is not possible on the basis of this report to make clear parallels between CO values achieved during EN testing and values for particle emission obtained during NS testing. Therefore, as a matter of course it will not be possible on the existing basis to substitute requirements on particle emission with requirements on the CO value. The reason for the absence of the expected connection between the EN test’s CO value and the NS test’s particle emission is probably due to different test conditions during the two tests and the fact that the CO emission and the particle emission during a test course do not necessarily take place simultaneously (please refer to a more detailed description of these differences under paragraph 2 – formulation of the assignment). To eliminate these elements it would be necessary to combine the two test methods and register all desired parameters in an overall test course. Bilag 1-5Click here to see Tables and Figures.
|