Abstract

Aim: This paper presents the experimental study and digital analysis of flammable dust granularity performed in the Combustion Processes and Explosion Lab at the Scientific and Research Centre for Fire Protection (CNBOP-PIB). It contains experimental results of sieve analysis of several dusts such as aluminum, cocoa, lycopodium and breadcrumbs. It notices some problems identified during the screening of dust in order to examine its granularity.

Methodology: Except some experimental work during the study the authors also used the laboratory microscope, camera with adapter for microscope and a computer in order to develop an alternative test method for granularity measurement. Due to a large variety of material properties it is not possible to nominate only one sieve analysis method that can be applied for all the combustible dust types. These dusts have large variety of properties both physical and chemical, such as density, fragility, grain shape, cohesive or magnetic properties. The main problem arising with these properties is suitable selection of parameters and sieving technique. For each material there is a need to determine individually the optimal sieving time, shaking amplitude or the frequency of manual tapping, which sometimes turns out to be necessary. Parameter selection needs to be made empirically, which sometimes can be very time consuming. The alternative method is based on the analysis of digital photography using the open source Image-J software. In this method, the image was processed in order to obtain a binary image with a specific threshold.

Results: Granularity testing using only a digital method proved to be much more time consuming. The main reason was the large discrepancies in the results due to errors in all phases of work, such as the complicated process of preparation of the sample shots to facilitate further processing and graphics processing process itself. The processing of each of the photo requires the separation of accumulated dust particles or to remove problematic fragments, which in case of several tens of pictures takes a lot of man-hours.

Conclusions: On the basis of tests results, this method cannot be considered as effective and alternative to the sieving analysis. The reason could be some discrepancies of the diameter of the sieve and the apparent diameter - two different ways of determining granularity of dust in this two methods. However, this method allows improving the sieving analysis process and estimating the granularity of the pre-dust, choose the right set and bulk of dust.

Keywords: dust granularity, microscopy, digital analysis, sieving

Type of article: original scientific article