Patent Application
20070102337
This instant invention is generally related to deep bed filtration systems as described in U.S. Pat. Nos. 3,814,247, 3,900,395, 4,048,068, 4,197,205 and 4,197,208, all to Hirs.
The instant invention relates generally to an improved deep-bed filter used for filtering water, sewage or other aqueous liquids, and specifically to an improved apparatus and method for liquid filtration utilizing a polymeric filtration media layer for enhanced contaminant filtration with minimum pressure drop through the filter.
Typical deep bed filters often have difficulty handling peak or emergency loads without over-design of the filter or the addition of an auxiliary filter to handle the extra contaminant load. Simply adding layers of filtration media to existing deep bed filters to solve the aforementioned problems is ineffective. Furthermore, backwashing many prior art filters having media of varying particle sizes but identical specific gravities results in a reversal of the media grading order, i.e., from small to large.
This reverse gradation problem has been solved to some degree by using media materials having differing specific gravities. However, even when using materials of such different specific gravities as anthracite and sand, if the granules of coal are large enough they will stratify at lower levels within the filter bed. These aforementioned dual media filters are generally used to handle increased turbidity loads and will provide longer periods of filter operation between backwashing. However, when turbidity gets very high and coagulants must be used these filters are still subject to surface binding, thereby requiring frequent backwashing.
The primary problem with known in the art dual media deep bed filters is that large coagulated particles and floc that are larger than the voids in the top layers of media are captured at the surface instead of passing into the depth of the media. This buildup of surface contaminants on the filter causes pressure buildup on the filter surface, thereby restricting flow of high turbidity liquids and preventing effective use of adequate chemical flocculating agents. Furthermore, during filter backwashing media classification takes place and the coal fines settle at the surface of the filter, thereby closing off any voids and further restricting fluid flow through the filter. Increased relative turbidity is often the result.
The instant invention utilizes an additional filtration media layer that is both larger and lighter than sand disposed below. This upper media layer mixes with the sand to enhance and maintain interstitial sites at the filter surface and throughout the mixed strata. Furthermore, this media is comprised of a plurality of cylindrical polymeric particles, each cylinder having a carefully selected diameter to maximize mixing with the finer sand particles subsequent to filter backwashing. This uppermost media strata prevents excessive buildup of contaminants on the surface of the deep bed filter, thereby enhancing fluid flow therethrough, even during periods of high contaminant loading. Other uses and advantages of the instant invention will become apparent from the detailed description of the preferred embodiments below in conjunction with the accompanying drawing Figures.
Referring now to
The vessel 20 contains a plurality of layers of filter media including a layer of sand 50 over which is placed a layer of polymeric particles 70. The filter media typically contains large granular particles for example, below the filter sand 50. The polymeric particles 70 are cylindrical in shape, having a diameter of about 1.5 mm and a length of about 6 to 12 mm. Furthermore, in one embodiment, the polymeric particles 70 may have a specific gravity of approximately 1.15. The cylindrical shape and specific gravity of the particles 70 are selected to allow settling thereof at a rate which permits mixing with the finer sand particles 50 as the filter media layers settle subsequent to backwashing.
The cylindrical polymeric particles 70 mix well with the lower layer of sand particles 50 to form an intermediate layer 60 containing polymeric media and sand particles. This feature of the instant invention provides a significant improvement over prior art deep bed filtration systems.
Since the upper portion of the sand particles 50 tend to be relatively small, filter 10 surface loading often occurs around this area of the filter when high concentrations of organic contaminants or chemical flocs are present in the fluid to be filtered. The mixing action of the cylindrical polymeric particles 70 with the finer sand particles 50 creates a plurality of voids or interstices forming the intermediate layer 60. These voids permit contaminant penetration into the filter 10, thereby greatly enhancing filter efficiency due to significantly decreased surface loading.
The selection of size, shape, and specific gravity of the polymeric particles 70 is crucial to proper filter operation. The polymeric particle 70 characteristics are selected to provide a desired amount of mixing with the sand 50 to facilitate filter penetration. For example, prior art spherical polymeric particles are less desirable for use in deep bed filters because the rate of settling of a sphere is too great, thereby causing the particles to settle to the filter bottom. In one embodiment, the layer 70 ranges from 6 to 12 inches in the filter.
Since a cylinder has a higher area to volume ratio than a corresponding spherical particle, and therefore settles at a much slower rate, cylindrical particles are superior for use as a top media layer. Furthermore, polymeric particles having a specific gravity of 1.15 have an ideal settling rate to facilitate mixing with the sand particles 50, wherein exemplary sand particles range from 45-55 mm. Specifically, nylon or polypropylene composite rods having a specific gravity of 1.15 are commercially available and are readily cut into cylindrical particles 70 of appropriate size.
In one embodiment, a plurality of sand layers may be formed within the filter 10. As shown in
As shown in the table given below, an experimental filter formed in accordance with the present invention was compared to a plant filter formed as known in the art. The experimental filter indicated in
As shown in the table given above, an experimental filter formed in accordance with the present invention was compared to a plant filter formed as known in the art. The experimental filter indicated in
With regard to clarity as shown in the table, “60% less than 0.02 NTU” or “25% less than 0.02 NTU” is defined as meaning that 60% of the total volume of water has clarity better than 0.02 NTU, or, that 25% of the total volume of water had clarity better than 0.02 NTU, respectively. The results
The plant filter indicated in Table 1 was sized at about eight feet by about 20 feet. The layer of anthracite was about 24 inches in depth. The filter sand was about 21 inches in depth and sized at about 45-55 mm. The flow rate through the plant filter was about 2.5 gallons per square foot. The cycle time of filtration was about 100 hours. The clarity was about 25% less than or better than 0.02 NTU.
The experimental filter was operated parallel to the plant filter and samples taken simultaneously. Water was supplied from ground wells having a relatively high hardness content of 300 ppm. Lime treatment was used to reduce the hardness. The city utilizing the plant filter had approximately 100,000 water users. For comparative purposes, an extensive run time of 1200 hours was employed. The average experimental filter run or cycle time was 82 hours before backwashing. The plant filter ran for 100 hours before backwashing. The object of the experiment was to determine if the effluent quality could be improved while concurrently raising the flow rate. It will be appreciated that in general, effluent quality is compromised as flow rates are increased unless more sophisticated filtration is used.
The plant flow rate was maintained at about 2.5 gallons per minute per square foot while the experimental filter was maintained at about 7.5 gallons per minute per square foot. The total amount filtered by the experimental filter was about 950,000 gallons. As shown in
It can therefore be concluded that as indicated in the present test and as compared to the plant filter, the present or experimental filter provided about 250% or more improvement in the water quality as measured by turbidity. It should further be emphasized that the substantial improvement in turbidity was realized while increasing the experimental filter flow rate to about 300% of the plant filter flow rate. This is certainly unexpected, and counterintuitive to what one of ordinary skill would expect. Typically, improved turbidity is most often achieved by reducing flow rates, increasing chemical treatment, and/or providing additional filtration subsequent to the deep bed filter. All of these options, however, translate into greater operating costs. The present system provides improved water quality at higher flow rates and at relatively lower costs.
While the preferred embodiments of the invention have been described in detail, it will be appreciated by one of ordinary skill in the art that the instant invention is susceptible of various modifications without departing from the scope of the following claims.
