Patent Application
20070277490
The mesh used in to make the pads of this invention is made with a conventional knitting machine that knits wire instead of thread. Although it is preferred in this invention to use a thinner wire, such as 0.006″ as compared with the 0.011″ wire conventionally used in the industry, the advantages obtained by alternating high and low density meshes to make a multilayer pad provides improved capacity and a lower pressure drop. The wire must be chosen to be inert in the environment of use, and type 304SS or 304L is typically used when droplets are removed from an aqueous or organic stream; 316L, Alloy 20, and polytetrafluoroethylene-based polymers (TEFLON) are used for sulfuric acid; 410SS and glass are used for mild chemicals; Monel alloy is used for corrosive chemicals, nickel is used for caustic; and the like as is known in the industry.
Multilayer pads in the prior art have a density gradient increasing in the direction of flow; for example, a lower density pad upstream and a higher density pad downstream. Surprising it has been found that simply varying the density between adjacent layers of a multilayer pad provides improved capacity and a lower pressure drop. Thus, alternating from high to low to high, or low to high to low, provides improvements. As shown below in Example 4, in comparison with a pad having the same total thickness using meshes of lower density (5 lb/ft3 and 9 lb/ft3) than the embodiment of the invention in Example 4 (9 lb/ft3 and 12 lb/ft3 mesh densities), the present invention provides a lower pressure drop and has a higher capacity.
The pads of this invention are first made by knitting a wire, such as 0.006″ type 304SS wire, on a conventional knitting machine to produce a knitted tube or sock. The knitted tube is first flattened by being run between roller. The flattened pad is then crimped by being run between two rollers, at least one having a pattern (like an embossing pattern) thereon so that the mesh tube is flattened and crimped. The crimped mesh tube may optionally then be crimped in a different direction or orientation between a second pair of rollers, having the same or a different pattern, the rollers oriented differently with respect to the mesh than the first pair. For example, if the first pair of crimping rollers is disposed horizontally, the mesh is fed horizontally; then, without changing the orientation of the mesh, it is run between a second pair of crimping rollers disposed vertically. The two pairs of crimping rollers need not be orthogonal to each other, although that is preferred; it is enough that the orientation of the second pair is different than the orientation of the first pair with respect to a given tube orientation. The resulting tube has a lower density than the original knitted and uncrimped tube. By varying the number of wires used to knit the tube and the number of layers of mesh used to make a given layer in the mist collector, as well as the number of crimping operations, a layer having a desired density can be produced. Increased crimping lowers the final density of the mesh. In the examples following, one mesh style (3BF) is made from 0.006″ wire into a mesh layer having a density of about 7.2 lb/ft3 (120 ft2/ft3; 98.6% voids), and another mesh style (3BA) is made with the same wire diameter into a mesh layer having a density of 12.0 lb/ft3 (200 ft2/ft3; 97.6% voids).
The mesh layers are then layered or stacked in a desired order on a frame or grid that both supports the multilayer pad and connects it to the process vessel (and holds it in place against the gas flow). The grid is secured and the mist collector is then installed into the process equipment.
A sample multilayer pad according to this invention was compared with a conventional multilayer pad on a testing apparatus. In the testing apparatus, a mixture of air and water was used as the test stream, using a 15 HP radial blade blower with an inlet damper for air flow control, a 12 inch., Sch. 20, 16 foot long exit pipe from the blower, and using a Dwyer model DS400-12 multi-orifice flow sensor for air flow measurement, all for supplying a vertical test chamber. The outlet from the test chamber had a 40 inch long Sch. 20 pipe including a FilterSense Model LM-70 Mist Gauge for entrainment measurement. The pressure drop was measured with an inclined manometer (measurement in inches of water column), the test system temperature having been measured with a Weksler Instruments dial thermometer with 1° F. gradations. The inlet loading was applied with either full cone water (Bete SCM9SQ), but more preferably, as in these examples, using a two-fluid (Spraying Systems ½J+SU 79) air-water spray nozzle. The water was recycled to the inlet using a 2 HP Teel centrifugal pump and metered using a zero to five GPM rotameter. The multilayer pad devices were tested to determine their capacity to eliminate water drops from the air stream. Capacity was determined by the air velocity at which breakthrough started to occur. The amount of water in the exit stream was determined using an electric induction probe (model LM 30, ProFlow brand series, from Impolit Environmental Control Corp., Beverly, Mass.). The outputs of the probe were used to calculate entrainment ENTR as galwater/mmSCFair (gallons of water per million standard cubic feet of air).
The sample pad (designated below as “LDP”) was constructing by layering the aforementioned style 3BA and 3BF meshes as follows, all being made with 0.006″ 304SS wire, to make the pad designated as “LDP” herein, all of the 3BF layers having been crimped twice and the 3BA layers having been crimped once. A “layer” in the pad construction is a single crimped knit tube. The sample LDP had the following construction (the direction of gas flow being bottom to top):
The LDP mist collector pad was tested against a six inch thick style 4CA pad, made with 0.011″ wire, having a density of 9 lb/ft3, 85 ft2/ft3 surface area, and 98.2% voids, layered to provide a thickness of six inches. As shown in
An eight inch comparison device was made (upstream to downstream) using five inches of a conventional vane separator (VH12), one inch of 7CA mesh, and two inches of 3BF style mesh. The style 7CA mesh is made from 0.011″ wire, has a density of 5.0 lb/ft3, a surface of 45 ft2/ft3, and 99.0% voids. This VH127CA3BF pad was tested against the LDP pad of this invention. As shown in
A six inch pad (2″7CA4″4CA) was made with two inches of 7CA style mesh and four inches of style 4CA mesh to produce a six inch pad. This was tested against the six inch LDP pad. As shown in
The present six inch LDP pad was tested against eight inches of standard undulating vanes with a spacing of one-half inch between vanes. As shown in
The present LDP pad was tested against an eight inch pad made using (upstream to downstream) one inch of the 7CA style mesh, two inches of 3BF style mesh, and five inches of the conventional vane (similar to Example 3 but reordered).
These examples show that a multilayer pad having more layers than in the art provides a lower pressure drop, higher capacity, or both, in comparison with a separation device having a lower density and/or greater thickness. Only a thicker set of conventional vanes provided a lower pressure drop along much of the velocity range, yet breakthrough occurred at a lower velocity for the vanes than for the novel LDP pad.
The foregoing description is meant to be illustrative and not limiting. Various changes, modifications, and additions may become apparent to the skilled artisan upon a perusal of this specification, and such are meant to be within the scope and spirit of the invention as defined by the claims.
