LOW-ENERGY DYNAMIC FILTER

Abstract

A means to exploit the Dean Vortices for dynamic filtering on a macro scale intended for application in utility and industrial processes is disclosed. This method relies on an apparatus of computed construction to optimize the centripetal force and minimize the effect of gravity on the separation and effectiveness of the Dean Vortices. The method is also supported by an apparatus of construction which results in an optimized elliptical flow channel that enhances the formation and persistence of the Dean Vortices.

Claims

1. A method to exploit Dean Vortices and to assist in the separation of suspended particles and/or a non-uniformly distributed solute in a dynamic fluid flow where such a fluidic system may exhibit non-uniform characteristics (viscosity and density) relative to the force of gravity, by orienting the flow path in a plane parallel to gravity such that the force of gravity is counteracted using centripetal forces. This method comprises the following procedures. a. Placing the major flow circuit to be in a plane parallel to the force of gravity. b. Manipulating the centripetal force vector at each increment of arc length around the major flow circuit to compensate for the force of gravity. This is achieved by altering the acceleration component of the centripetal force vector, in one or both of the following two ways. i. Configuring the radius of the major flow channel at each increment of arc length to create a toroidal flow channel, or a toroidal spiral flow channel, as a means of configuring the circular acceleration component of the Centripetal Force to compensate for the gravitational component of the forces acting to collapse the separation of the Dean Vortices, such that the sum of the aligned gravitational force and the resulting centripetal force is consistent throughout the full toroidal circumference. The appropriate radii are calculated using the Total a (acceleration) equation, solved for r (the toroidal radii), where is the rotational angle between r and the plane perpendicular to the gravity force vector F.sub.g, and v is the flow velocity v: $r=\frac{v^2}{\left(\mathrm{Total}.Math..Math.a\right)+\left(\sin .Math..Math.*F_g\right)}$ ii. Alternatively, configuring the cross-sectional area at each increment of arc length of a constant volumetric flow rate circular or spiral flow channel as a means of controlling the fluid flow velocity and thus the acceleration component of the Centripetal Force to compensate for the gravitational acceleration component of the forces acting to collapse the separation of the Dean Vortices and providing a non-varying resultant total centripetal force in support of the Dean Vortices.

2. A method to enhance the early formation of Dean Vortices in a flow channel subjected to centripetal acceleration and to extend the persistence of the Dean Vortices over a broader range of flow velocities than typical of a rectangular flow channel to propagate Dean Vortices and eliminate the turbulent discontinuities in the corners of a rectangular cross-sectional flow channel with a width-to-height ratio large enough to satisfy the start-up parameters to form Dean Vortices. a. The method claimed is to compress a circular channel in a controlled, uniform manner at a fixed fluid flow velocity until the formation of Dean Vortices is detected. b. When the state of Dean Vortices is achieved, the compression height is noted. c. The process is iterated for each combination of fluid viscosity, density and the dynamic effects of the solute fluid system.

3. A method to enable the selection of suspended particles or certain unsaturated solute in a fluid flow for inclusion, no matter where along the zero net force line between the Dean Forces of hydrodynamic force and centripetal force the particles may find equilibrium, as contrasted to the established practice of bifurcating the flow at the midpoint of the major flow. This method is to position a tunable edge into the flow, collecting the two output streams for analysis, and adjusting, or tuning, the edge angle and position until the desired separation is achieved. a. Where the method of particle selection is either a manual process, removing samples from each of the divided flows for laboratory or instrument analysis, or an automated process using in situ instrumentation to monitor in real-time, the Total Suspended Solids (and in some cases Total Dissolved Solids) and Turbidity in each of the flows is analyzed and graded. The tunable edge which separates the two flows is positioned either manually or by a control computer in response to the strength of signal or measured relative quantities of Total Suspended Solids and Turbidity in the two flows. b. Where the manual or automated positioning of the tunable edge separating the principal flow into two flows by positioning the edge to sweep across the flow and creating a table for each of the two flows relating Total Suspended Solids (and in some cases Total Dissolved Solids) and Turbidity vs. position and angle of the tunable edge and finally positioning the edge to provide the largest difference in the measured values between the two flows. c. Where during continuous operation, the operator or control computer varies the position and angle of the tunable edge from the measured optimum position until a change in the measured values of Total Suspended Solids (and in some cases Total Dissolved Solids) and Turbidity is noted and the tunable edge is then moved in the opposite direction until a similar change in the measured values is noted and this process is continued verifying that the positioning of the edge is at a local optimum.

4. An apparatus in support of the Method of claim 1, which is implemented using a continuous fluid flow channel constructed such that the fluid channels are arranged to affect a toroidal flow of varying radii to meet the requirements of the method to compensate for the effects of gravity. The fluid channels are uniquely stacked in parallel rows that maintain the effective radii. The apparatus has the following characteristics. a. The major flow axis of a toroidal flow (toroidal radii) lies in a plane which is oriented parallel to the force of gravity. b. The shape and radii of the major toroidal flow channel are consistent with the calculations made in the above method (claim 1) to compensate for the gravitational component of the forces acting to collapse the separation of the Dean Vortices and provide a total centripetal force in support of the Dean Vortices.

5. The Apparatus of claim 4 with refinements and enhancements consisting of: a. The circular cross section of the flow channel of claim 4, compressed as claimed in claim 2, to enhance the propagation of Dean Vortices in a flow channel by altering the circular cross section creating a smooth, precise elliptical cross sectional shape with a significant ratio between the long and short diameters of the ellipse. The apparatus has a pair of rigid plates to uniformly compress the circular channel to the pre-determined height to form an elliptical shape.

6. The apparatus of claim 4 with enhancement to refine the separation of particles by using a moveable ledge within the flow, which uses the method described in claim 3 to alter said position according to the application. This apparatus has installed in it a movable or positionable thin smooth, sharp edge extended from the bifurcated meeting of the two exit fluid channels and into the major flow channel supporting the Dean Vortices of concentrated particles and reduced particle populations distributed across the zero net force line between the Dean hydrodynamic and centripetal forces such that the movable edge can be indexed across the major flow channel with the effect of positioning the edge to divert more or less of the flow and capturing more or less of the particles distributed orthogonally to the main flow by the Dean Forces.

7. The apparatus of claim 4 extended by stacking multiple apparatuses of claim 4, where the eternal cross sectional areas of the flow channels of each stacked claim 4 apparatuses are equal and the effect of stacking extends the persistence time of the Dean Vortices to enable the stabilization of the Dean hydrodynamic and centripetal forces to drive the particle selection process.

8. The Apparatus of claim 4 extended by stacking multiple apparatuses of claim 4, where the eternal cross sectional areas of the flow channels of each stacked claim 4 apparatus range from larger to smaller and the selection process discharges the collected particles after each stack segment and then transfers the remaining flow to the smaller cross sectional area of the next stack, thereby enabling successive separate particle collection actions effected serially using the same principal flow with the effect of reducing the population of particles and the average size of the remaining particles after each stack segment separation and removal from the continuing flow.

Description

BRIEF DESCRIPTION OF THE SEVERAL VIEWS OF THE DRAWING

[0011] FIG. 1. View 1A: Illustrates the relative-flow velocity distributions of (current state of the practice) example round straight flow [0012] View 1B: Illustrates the relative flow velocity distributions of (current state of the practice) round curved flow channels [0013] View 1C: Illustrates 3 views of rectangular curved flow channel with the gravitational forces orthogonal to the Dean Flow. The top left illustration shows the single flow when velocity is insufficient to create Dean Vortices; the top right illustration shows the flow changed to Dean Vortices as the velocity increases; the bottom illustration shows the apparent centripetal force, the hydrodynamic force, the flow and the gravitational force vectors inside the curved rectangular channel of the top right illustration. View 1D: Illustrates the novel elliptical flow channel with the gravitational forces orthogonal to the Dean Flow tending to collapse the separation of the two competing Dean Flow forces.

[0014] FIG. 2. Illustrates gravitational forces on rectangular (View 2A) and elliptical (Views 2E, 2F) channels with major flow axis oriented parallel or in plane with gravity; while gravitational forces shown orthogonal to flow in both rectangular and oval shaped flow (Views 2B, 2C and 2D) illustrate forces relationships avoiding the collapse of the cross-flow Deans Vortices.

[0015] FIG. 3. Illustrates the physical configuration of the variable-radius toroidal stack of the Low-Energy Dynamic Filter, showing in View 3A a single stack and in View 3B a configuration of 9 plates, making a four-layer stack where each single stack shares an end plate with the neighboring stack.

[0016] FIG. 4. Illustrates the adjustable separation trimming edge of the Low-Energy Dynamic Filter. View 4A shows the trimming edge plane view; View 4B shows the trimming edge in position where the flow is directed equally between the product and waste collection channels; View 4C shows the trimming edge in position to divert most of the flow to the waste collection channel.

[0017] FIG. 5. View 5A is a polar plot representation of the results for the variable radius calculation in which the value for acceleration (a) is set to 2 g's (standard gravitational units) and a constant velocity of 2.5 m/s. The polar plot shows data at intervals of 5 angular degrees and describes a plot forming a Cartesian Oval. View 5B is the actual data used for the plot, shown in meters, and annotated to indicate the values at 0, 85, 90, 180, 265, and 270. View 5C shows an example Cartesian Oval with force and velocity vectors shown at =0 and at =90 for illustrative purposes. The equations used to calculate the values for the radii r and to derive the vectors are as follows (equations are numbered here for reference later):

F.sub.g=F.sub.Gsin() (e.sub.1) [0018] where F.sub.G is the standard gravitational force in mks units and is the angle between r and the centripetal force F.sub.C, as shown in 5C

Total Force=F.sub.C(sin()F.sub.G) (e.sub.2)

Force=ma where m is mass and a is acceleration (e.sub.3)

Total a=(v.sup.2/r)(sin()F.sub.G) (e.sub.4) [0019] where v is the axial velocity for a specific r at angle

r=v.sup.2/((Total a)+(sin()F.sub.G) (e.sub.5)

[0020] Using the specific values stated above (a=2 g, v=2.5 m/s, and F.sub.G=9.8 m/s.sup.2)

r=(2.5).sup.2/((29.8)+(sin()9.8)) (e.sub.6)

[0021] Using values of at 5 degree intervals yields the r values in the table of View 5B, and the polar plot of View 5A.

REFERENCE

[0022]

TABLE-US-00001 Marking Description r Represents the radius of curvature of the channel. The angle of rotation around the toroidal flow (angle ) F.sub.g0 Gravity Force vector (F.sub.g) when the radius r is chosen at angle = 0 F.sub.g90 Gravity Force vector (F.sub.g) when the radius r is chosen at angle = 90 F.sub.C0 Centripetal Force Vector (F.sub.c) when the radius r is chosen at angle = 0 F.sub.C90 Centripetal Force Vector (F.sub.c) when the radius r is chosen at angle = 90 v.sub.0 Represents the axial velocity (v) when the radius r is chosen at angle = 0 v.sub.90 Represents the axial velocity (v) when the radius r is chosen at angle = 90 43 Input channel. 44 Stack leaf compression plate cover, top. 45 Stack leaf support plate with integral cross over. 46 Stack leaf compression plate cover, bottom. 47 Outlet channel, filtered. 48 Outlet channel, contaminant stream, for single stack 52. 48-1 Largest diameter channel stack contaminant output stream in a four-layer stack 52A. 48-2 First reduced diameter channel stack contaminant output stream in a 4-layer stack 52A. 48-3 Second reduced diameter channel stack contaminant output stream in a 4-layer stack 52A. 49-4 Third reduced diameter channel stack contaminant output stream in a 4-layer stack 52A. 49 Flow channels, side one. 50 Flow channels, side two. 51 Integrated crossover flow channel between side one and side two. 52 A single stack, layer, including flow channels one and two, support plate with crossover, top plate and a bottom plate. 52A A four-layer stack, where each single stack shares an end plate with the neighboring stack. 53 Callout showing stack input channel, illustrating the movable selection edge. 54 Compression bolts. 60 Inner stack crossover fittings 61 Largest diameter channels side one and side two. 62 First reduced diameter channels side one and side two. 63 Second reduced diameter channels side one and side two. 64 Smallest diameter channels side one and side two. 65 The geometric center of the rotational toroid. 66 The minimum radius point of the computational distribution of the variable radius. 67 The maximum radius point of the computational distribution of the variable radius. 112 The flow velocity distribution into a straight tube. 114 Straight tube section. 116 The Reynolds or eddy effects of flow distribution from the drag of the boundary layer against the tubing walls to the center flow of a straight circular tube. 122 Flow velocity distribution of a curved tube 124 A curved tube showing internal flow distribution. 126 The Reynolds or eddy effects of flow distribution from the drag of the boundary layer against the tubing walls to the center flow of a curved circular tube. 131 The distorted flow in a rectangular cross-section channel with insufficient velocity v to form Dean Vortices. 132 Illustrates the Dean Vortices formed inside the rectangular curved channel with sufficient velocity v. 134 The major fluid flow vector. 136 Gravity force vector (F.sub.G). 140 Centripetal Force vector within the flow chamber (F.sub.c). 142 Hydrodynamic Force vectors. 210 The diverter edge that can be adjusted to separate the flow across the flow channel 50 from inside edge to outside edge. 220 Indicates advancing the edge of diverter 210 to the outside wall moving more of the cross sectional flow to the inside or waste collection channel. 230 Indicates advancing the edge of diverter 210 to the inside wall moving more of the cross sectional flow to the outside or product collection channel.

DETAILED DESCRIPTION OF THE INVENTION

[0023] The invention detailed herein addresses two issues seen in current state-of-the-practice apparatus through evolution and adaptation of the Dean Flow analysis used to supplement the well-known Reynolds analysis of flow as applied to toroidal systems, composed of relatively small diameter flow channels d. The Dean Flow analysis relating the Dean Number, the diameter of the flow channel and the radius of curvature of the flow channel, when (d<<r) as applied here, is stated as follows:

[00001]$\begin{array}{c}{D}_e=.Math.\frac{\left(\mathrm{Centripetal}.Math..Math.\mathrm{forces}\right).Math.\left(\mathrm{inertial}.Math..Math.\mathrm{forces}\right)}{\mathrm{Viscous}.Math..Math.\mathrm{forces}}=\\ =.Math.\frac{\sqrt{\left(.Math.\frac{r}{d}.Math.\frac{v^2}{{\left(r/d\right)}^2}\right).Math.\left(.Math..Math.v^2\right)}}{\left(v/d\right)}\\ =.Math.\sqrt{\frac{d}{2.Math.r}}.Math.\frac{.Math..Math.\mathrm{vd}}{}\\ =.Math.R_e.Math.\sqrt{\frac{d}{2.Math.r}}\end{array}$

[0024] Where

[0025] D.sub.e is the Dean's number

[0026] is the density of the fluid

[0027] is the dynamic viscosity

[0028] v is the axial velocity

[0029] d is the diameter of the flow channel

[0030] r is the radius of curvature of the channel

[0031] R.sub.e is the Reynold's number

[0032] As shown in FIGS. 1 and 2, current state of the practice creates apparatuses that do not address issues of the dynamic flow of a fluidic system where the suspended and/or neutrally buoyant particles and the fluid under the influence of gravity can no longer be considered to have uniform viscosity and density and thus enable the gravitational distortion and collapse of the Dean Vortices. The first issue is that these apparatus present the Dean Vortices in a plane orthogonal to the gravitational forces, allowing for the deformation of the Dean Vortices and leading to their collapse. The second issue is that the shape of the channels used in these apparatuses allows distortion in the corners. FIG. 1 View 1A (top) shows a cross sectional view of a straight section of a circular flow channel and the distribution of velocity vector rings 112 that illustrate typical slow boundary layer flow against the walls of the flow channel as depicted by the velocity vector distribution 112 for a straight conduit; 116 is an illustration of the circular rolling eddy currents created in a segment 114 by the distribution of velocity as illustrated by vectors 112 caused by flow adhesion at the boundary layer near the walls of the conduit.

[0033] In a curved section 124 of a flow channel, the compressed velocity vectors rings 122 and the compressed and non-uniform distribution of the velocity vectors 122 are the result of centripetal forces in the curved conduit. The rolling eddy currents 126 bending in their flow are the result of the centripetal forces introduced by the acceleration of the curve in the conduit while still exhibiting the flow rate distribution 122 as a result of boundary layer adhesion. But the round conduit does not allow sufficient laminar flow velocity v to support stabilized Dean Vortices.

[0034] The evolution of a curved rectangular cross-section distorted flow 131 to stabilized rectangular cross-section distortion 132 is predicted by the Dean Number, D.sub.e, when the Dean Number, as characterized by the ratio of the diameter of the conduit d to the radius of curvature of the circular or spiral path r, is much smaller than unity and there is sufficient flow velocity v. A rectangular flow channel with a significant variance between the lengths of the long and short sides enhances the onset of the Dean Vortices and aids in their persistence over a larger range of velocities.

[0035] The formation of Dean's Vortices orthogonal to the flow 134 is the result of the contrasting effects of centripetal force 140 and hydrodynamic forces 142. In the current practice, this phenomenon has been exploited to create micro-channel membraneless filtering to isolate and collect suspended or neutrally buoyant particles where the hydrodynamic forces are balanced by the centripetal forces of the relative density and hydrodynamic drag, or resistance, for particular particles. Of important note is that in the current practice, as shown in view 1C in FIG. 1, the gravitational force (F.sub.g) 136 is orthogonal to the major flow vector 134 and orthogonal to the centripetal forces (F.sub.c) 140 vectors and the hydrodynamic force vectors 142. Thus the gravitational force is compressing the short radius of the Dean Vortices. But the current practice micro-channel implementations of membraneless filtering rely on the uniform viscosity and density of the fluid system, including the suspended or neutrally buoyant particles and the fluid to limit the effects of gravity on the propagating Dean Vortices.

[0036] Gravitational forces in this orientation 1C are not countered by the flow vector 134 or the centripetal force 140. As such, the centripetal force and the hydrodynamic force of the two counter-rotating Dean Vortices are subject to compression of the system of fluid and suspended or neutrally buoyant particles, if that composite system of fluid and particles does not move with the fluidity of a homogeneous fluid. This is the normal orientation of similar apparatuses found in the field of both research and application today and as such there are no co-aligned forces that can be manipulated to counter the disrupting effects of gravity acting with significant leverage against the modest forces of centripetal acceleration and hydrodynamic flow.

[0037] It is this inventor's contention that gravitational effects have prevented the large scale adoption of dynamic filtering for utility and industrial applications. The unique method detailed herein benefits from real world empirical analysis of trial and error along with analytical support in the areas that are usually taken as assumptions in both Reynolds R.sub.e and Dean De analytical expressions. The method described herein rotates the major spiral or circular flow 134 to operate in the gravitational plane.

[0038] These typical working assumptions are that the fluidic density p and the fluid viscosity are both constant throughout the flow. Such assumptions limit the adaptation of the Dean Flow in application to particle separation to a limited class of suspended or neutrally buoyant particles that are not affected by the force of gravity in a dynamic flow when captured by the competing centripetal and hydrodynamic forces of the Dean Vortices.

[0039] The method present herein specifically extends the previous method of application of the Dean Vortices to the simultaneous application of fluids with suspended or neutrally buoyant particles that are affected by the force of gravity in a dynamic flow when captured between the centripetal and hydrodynamic forces without sacrificing the effectiveness of segregating those particles that are not affected by gravity in a dynamic flow by rotating the circular or spiral axis to be in the same plane as the gravitational vector 136.

[0040] The technique of this method is the control of the forces of implementation of the Dean Vortices to counter the effects of gravity. The effects available to control are the result of the centripetal force and the fluid velocity v, which in the case of incompressible fluids could be manipulated directly or indirectly by building flow channels of variable cross-sectional area; thus with constant fluid flow volumes it is possible to directly control fluid velocity.

[0041] Alternatively, the effects of velocity could be manipulated by controlling the centripetal acceleration, which can be realized as the angular rate, which for a constant velocity of flow is actually just a function modifying the radius of the circular or spiral path r to create a toroidal path.

[0042] Foundational to this method is the re-orientation of the toroidal flow channel (all views in FIG. 2) to be in the vertical gravitational plane aligned with the gravity force vector 136 such that the force of gravity is aligned with the centripetal force vector 140. Additionally, because the Dean Vortices are orthogonal to the major flow 134 as the major flow moves around the toroid, the major flow vector 134 will align with the gravitational force vector 136. Thus, the gravitational force vector is minimally disruptive because it is orthogonal to the counter-flow paths of the two counter-rotating Dean Vortices 2E, 2F.

[0043] In these instances when the principal toroidal flow 134 is vertical either up 2F or down 2E (angle is either zero or 180 degrees) and operating either against 2F or with 2E gravity, the principal flow and the force of gravity combine to push the forces of the Dean Vortices either up 2F with the principal flow against gravity or down 2E with the force of gravity. In either direction the forces of the Dean Vortices are not forced towards collapse. Although the vortices may be pushed to elongate their long axis and bow up or down, they are not pushed to collapse their short axis or the separation between the two counter-rotating vortices.

[0044] At every other point along the toroidal flow path where the sine of is not zero, this method computes the radius r of the toroidal flow channels as a function of the angle between the radius and the plane normal to the gravitational plane while the flow velocity remains constant.

[0045] The equation for Total Force (e.sub.2) forms the basis of this method. The centripetal force can be varied such that the gravitational force component 136 that is orthogonal to the long axis or working axis of the hydrodynamic 142 and centripetal forces 140 that drive the Dean Vortices with the result of particle separation can be countered at all angles except zero and 180 degrees when the sine of is zero. In this example, for a given velocity (2.5 m/sec) and total centripetal force (2 g's) a table 5B of radii r can be computed where the vertical component of gravity acting to influence the centripetal force can be countered in the plane of interest by calculating radius r by equation (e.sub.5) as executed using specific values (e.sub.6).

[0046] While at the horizontal extremes ( is zero or 180 degrees) where the centripetal forces 140 cannot be used to counter the force of gravity 136, the forces of gravity are operating directly against or with the force of the major toroidal flow 134. If the fluid is not compressible, nor the flow channel significantly deformable, then the flow velocity v will remain nearly independent of the gravitational force pushing or pulling the flow velocity vector and thus have a minimal effect on the elongation of the Dean Vortices cross-flow characteristics.

[0047] FIG. 5 includes a table (View 5B) of r values, expressed in meters, computed for a velocity of 2.5 m/sec and a continuous net centripetal force of 2 g's. As shown in the radial plot (View 5A), the center of the rotation of the toroid 65 is offset and supports radii from the minimum 66 to the maximum 67 as shown on the radial graph illustrating radii from a minimum of about 0.21258 meters to a maximum of about 0.6377 meters creating a graphical shape known as a Cartesian Oval (66, 67).

[0048] The other issue addressed by this disclosure is the shape of the flow channel. Whereas the state of the practice employs rectangular channels to enable the onset of the Dean Vortices, this disclosure improves upon the rectangular cross-section channel and the inherit discontinuities presented by the corners of the channel to smooth flow through the implementation of elliptical-shaped flow channels. Elliptical-shaped flow channels with significant variance between the long and short diameters enable the formation of Dean Vortices over a much broader range of flow velocities than the rectangular cross-section channel.

[0049] The preferred embodiment of this method is presented in FIG. 3 in the unique apparatus 52 of stacked plates 44, 45 and 46 and deformable tubing 43, 47, 48, 49. The deformable tubing in this implementation supports an enhancement that aids the formation of Dean Vortices by creating a non-circular flow channel 1D, 2C, 2D, 2E, 2F. This technique avoids the turbulence-inducing right-angle corners of a rectangular flow channel. Instead, this apparatus features a compression-limiting crossover duct fixture 51 that serves to limit the compression of the round deformable flow channel to become an elliptical flow channel 1D, 2C, 2D, 2E, 2F with a significant variance between the long and short diameters. This configuration enhances the onset of the Dean Vortices and aids in their persistence over a larger range of velocities.

[0050] A significant aspect of the apparatus is the optimization of the toroidal loops with nearly ideal radial dimensions 49, 50 by stacking them in the horizontal dimension between the compression plates 44, 45, 46 and making use of the compression-limiting crossover duct 51 to seamlessly route the flow 49 to another toroidal construct 50 with identical radial dimensions.

[0051] Also, apparatus 53 is implemented as a tunable feature of the dynamic filter employing a movable edge 210 to select particle collections for inclusion or exclusion that may be distributed along the zero net force path between the opposing Dean Forces 140, 142 of the centripetal force pushing the mass of the particles to the outside and the hydrodynamic force pushing the hydrodynamic resistance to the inside. The mass of the particle is pushed to the outside by the centripetal force while the resistance to hydrodynamic flow causes the suspended particles to be pushed to the inside of the channel. A particle's susceptibility to centripetal vs. hydrodynamic flow is related to the particle's density and physical construct as well as the principal flow velocity. The zero net force point is the position between the outer wall and the inner wall where a particular particle's mass is pushed outward by centripetal force to exactly match the inward hydrodynamic pressure from the counter hydrodynamic flow.

[0052] Adjusting the tunable edge incrementally towards the outside 220 or the inside 230 of the output flow channel and variation of the principal flow velocity enable this apparatus to sort or separate an extended range of mass vs hydrodynamic drag particles.

[0053] The assembled apparatus 52A can consist of any number of stacks 52. As an example, 52A presents an assembly of 9 plates capturing 4 pairs of spiral channels 61, 62, 63, and 64, each pair configured to optimum radius r for a particular channel diameter d.

[0054] Bolts 54 are used to compress the flow channels to an optimum elliptical shape 1D, 2C, 2D, 2E, 2F. Typical input fluid channels 43 feed at desired fluid flow velocity v. The dynamic filter of each stack 61, 62, 63 or 64 of selected diameter and fluid flow velocity will separate suspended particles to discharge through fluid channels 48-1, 48-2, 48-3 and 48-4 with the remaining flow channel 47 free of an extended range of suspended particles separated by this configuration.

[0055] Crossover compression-limiting device 51 serves to connect the two toroidal flows of each stack 52 and the stack cross-connecting flow channels 60 serve to provide connections between the co-mounted stacks 52 into structure 52A.