- The Filtration Society - http://www.filtsoc.org -

Volume 12, Issue 3

NEW LABORATORY DEVELOPMENTS FOR BELT THICKENER OPTIMIZATION
Pascal Ginisty and Christophe Peuchot (pages 154-158)

Gravity belt thickening is commonly used for sludge volume reduction. In many applications it is preceded by chemical conditioning to enhance water release and solids capture. Laboratory tests are usually performed to select the best polymer and optimal dosage for full scale application. Drainage kinetics are very sensitive to mixing conditions during flocculation. Thus, a specific device called the ‘boo-test’ was developed to control these parameters and guarantee the repeatability of conditioning, necessary for a reliable comparison of products. Torque rheometer and laser diffraction granulometer measurements give complementary data for describing the flocculation process.

SINGLE PARAMETER SIEVE CALIBRATION USING PRECISION GLASS MICROSPHERES
Graham Rideal and Jamie Storey (pages 159-162)

Sieving is the oldest method of particle size analysis but unfortunately its reputation has suffered from bad practice, by both the manufacturer and user. A combination of using poor quality mesh or even putting the wrong mesh in a frame, combined with casual analytical practices has reflected very badly on the method. International standards have endeavoured to specify acceptable tolerances using microscopy but the resulting 13 parameter definitions are difficult to replicate when trying to match sieves. Also, only wire spacings rather than actual apertures are measured. This study investigates the use of precision glass microspheres for the NIST traceable certification of sieves. A new image analysis software program was developed to analyse the internal parameters of each aperture. Excellent agreement was seen between the optical techniques and the glass microsphere calibration method when analysing a 63 micron sieve.

REDUCING OPERATING COSTS OF BAG FILTERS USING A FILTER EXPERT SYSTEM
Gunnar-Marcel Klein, Peng Bai, Tim Neuhaus and Theo Schrooten (pages 163-169)

This article addresses the optimization, including upgrades, of bag filters to reduce the life cycle cost (LCC) of an industrial bag filter installation. Energy consumption, mainly by fan motors and by the consumption of compressed air, is the biggest portion of the LCC of a bag filter. Today, improved jet-pulse cleaning systems enable the use of very long bags, leading to a reduction of both, investment and operating costs. A systematic analysis of the process parameter ‘air tank pressure’ was carried out for bag lengths up to 12 m. The results are summarized in a decision matrix, which allows the cement producer to decide the maximum possible bag length in filter upgrades, as well as for new installations. The test series are performed using special high speed pressure transducers to precisely determine the interior pressure in filter bags over their length and for different operation parameters.

The second part of the paper focuses on a newly developed tool to determine the optimum operational set points of jet-pulse bag filters. The filter expert system ProExpertise, by Intensiv-Filter, is able to calculate the differential pressure, the consumption of pressurized air and the energy demand for different operating parameters. The program presented is able to predict the optimum cycle time, at which the sum of the energy consumption for the fan and for the jet-pulse cleaning has a minimum value. The results are dependent on operation parameters such as volume flow, raw gas dust load, filter media type, bag length, among others, and provide even experienced filter experts with valuable information to reduce operating costs.

ProExpertise is based on the well-known filter equations for cake filtration and contains experimentally determined specific cake resistances and residual pressure loss values for Pural SB dust, limestone (taken after the raw meal mill in cement production) and ground cement. The expert system can be used to reduce the energy demands of existing and new filtering installations in various industrial applications.

OIL REPELLENT NANO-COATINGS FOR INCREASED FILTRATION PERFORMANCE
Stephen Coulson and D.R Evans (pages 170-171)

Consumer and industrial products are manufactured from a range of materials that are selected for specific bulk properties, cost and/or ‘look and feel’. However, many materials chosen in this way do not display the optimum surface properties. This presents an opportunity for surface modifications to apply desirable properties such as fire retardancy, anti-microbial, protein resistance and water/oil repellency. It is critical that these modifications do not alter the bulk properties of the product and retain desirable physical attributes. Additionally, they should be ultra-thin and well adhered. For commercial success, the desired effect needs to be a cost-effective and robust industrial process.

CROSSFLOW MICROFILTRATION OF OIL FROM SYNTHETIC EFFLUENT WATER
Yousef Alanezi and Richard J. Wakeman (pages 172-183)

Crossflow microfiltration of oil from water was studied experimentally under various operating conditions using a multi-channel ceramic membrane. Crossflow velocities, oil concentrations, and ionic strength effects on equilibrium permeate flux were investigated. An increase in crossflow velocity for oil emulsions from 1.14 to 2.28 m/s caused an increase in the equilibrium permeate flux. In contrast, as feed oil concentrations increased from 300 to 2400 ppm, equilibrium permeate fluxes were decreased. Likewise, when the ionic strength for the feed emulsions was increased, the permeate flux declined. These different observations are discussed in term of the hydrodynamics and particle interactions in relation to the filtration process.

For the modelling of experimental results in this work, the applications of back transport models (such as torque balance, inertial lift and shear-induced models) in the area of liquid-liquid separations contribute new knowledge. For the experimental critical flux results the shear-induced diffusion model showed a better prediction in comparison with other back transport models when particle size was used as the fitting parameter. From the particle size distribution analysis, the number frequency of these fine droplets was less than 5% in the polydisperse emulsions. Hence, the smaller particles are causing fouling, which is in agreement with the findings of previous studies.

SIMULATION OF PRESSURE DROP AND CAPACITY FOR PLEATED AIR FILTERS LOADED WITH DUST
Philipp Hettkamp, Gerhard Kasper and Jörg Meyer (pages 183-192)

Air intake filters for combustion engines are commonly pleated and made of thin filter paper. At high dust concentrations these media function as surface filters. The properties of the dust cake and its distribution within the pleat then become crucial to pressure drop evolution and filter lifetime. The dust distribution in turn depends on the flow field in the pleat and particle properties. In this paper the Navier-Stokes equations are solved numerically for a domain representing one model pleat and then calculate particle trajectories and local growth rates of the dust cake. Continuous cake growth is approximated by assuming stationary flow during the deposition of an incremental amount of dust. Using such an approach the influence of pleat geometry, particle inertia and volume flow is investigated. Additionally, simple analytical models for the loading of idealized pleats are presented.

U- and V- shaped pleats exhibit distinctly different loading behaviour. The pressure drop slope of U-shaped pleats is initially linear. V-shaped pleats show a non-linear rise of the pressure drop. Therefore, rectangular pleats will generally yield higher dust capacities than tapered pleats. Particle inertia can have a pronounced effect on the loading curve of U-shaped pleats, while it is less important for V-shaped pleats. It is shown how dust capacity varies with pleat width. An optimum width exists because decreasing pleat width increases filtration surface while it diminishes the volume available for dust deposition. The results of numerical simulation and analytical model agree almost perfectly for the V-shaped pleat, and match qualitatively for the U-shaped pleat.