US3446228A - Opposed jet pure fluid amplifier - Google Patents

Description

May 27, 1969 STOUFFER ETAL v 3,446,228

OPPOSED JET PURE FLUID AMPLIFIER Filed QCC. 19, 1966 Sheet 014 INVENTORS.

RONALD D. STOUFFER BY aomfymai i ATTORNEY May 27, 1969 R, sTQUF F-ER ET AL 3,446,228

OPPOSED JET PURE FLUID AMPLIFIER Filed Oct. 19, 1966 Sheet 2 of 4 INVENTORS.

RONALD D. STOUFFER BY JOHN bVGEFZBER ATTORNEY May 27, 1969 STQUFFER ET AL 3,446,228

OPPOSED .JET PURE FLUID AMPLIFIER Sheet Filed Oct. 19, 1966 FIG. 5

Time

l lm+ AIU+Q FIG. 6

INVENTORS.

RONALD D. STOU FFER BY JOHN yBEg ATTORNEY R. D. STOUFFER T 3,445,228

OPPOSED JET PURE FLUID AMPLIFIER May 27, 1969 Sheet Filed Oct. 19, 1966 INVENTORS.

RONALD D. STOUFFER BY JOHN WMGEZBE E ATTORNEY United States Patent 3,446,228 OPPOSED JET PURE FLUID AMPLIFIER Ronald D. Stoutfer, Bel Air, and John W. Gerber, Baltimore, Md., assignors to Martin-Marietta Corporation, New York, N.Y., a corporation of Maryland Filed Oct. 19, 1966, Ser. No. 587,886 Int. Cl. Fc 1/10, 1/12 US. Cl. 137-815 8 Claims ABSTRACT OF THE DISCLOSURE Opposed, coaxial power streams are emitted from opposite ends of an hourglass shaped interaction chamber. Two sets of outlet ports straddle the power stream inlets, one set at each end of the chamber, and are formed in part by extensions of the curved chamber wall. At least one pair of control stream passages open into the chamber, and the particular outlets into which the impacting power streams are directed are selectively controlled by the control streams.

This invention relates to fluid logic devices, and more particularly to those of a multi-stable variety characterized in having opposed and mutually impacting fluid power streams which may be selectively manipulated to perform fluid amplification and switching functions.

Fluid amplification elements and their related systems currently are the subject of increasing interest on the part of industry. The elements, having operational characteristics based upon the phenomena of fluid dynamics, may be selectively arranged and interconnected to perform numerous logic functions heretofore generated by electronic, mechanical and electromechanical systems. Various configurations of the pure fluid elements have been combined to create analog and digital computing devices, while other applications selectively arrange the elements to provide control functions for industrial processes, tooling and the like.

Active pure fluid elements have been fabricated to reflect several fluid dynamics design approaches. Generally they are configured to perform either an analog or a digital function. For the most part, the configurations have incorporated a singular power stream or jet of gaseous fluid under relatively high pressure along with two or more outlet ducts into which the power stream is selectively caused to fiow. The presence of the entire or a portion of the stream in a particular duct provides an ultimately controlling power function. Diversion of the power flow into selected output power ducts is accomplished by the energization of one or more transverselyoriented control streams which intercept and react with the power streams in various manners. The control jets are generally of a lesser power magnitude; consequently, a gain may be realized with the elements.

Diversion of the power streams is also influenced by the configuration of the walls of the pure fluid element through which it is caused to flow. Particularly where pure fluid switching functions are desired, the fluid logic elements are generally designed to make use of variations of wall attachment or boundary layer effects where the power flow will divert to and tend to stabilize along one particular element wall unless manipulated by the influence of a control stream. This biased characteristic of stream path or direction of flow readily lends itself to bistable element design for use in digital devices and the like.

Pure fluid elements utilizing wall attachment or boundary layer phenomena along with certain of those using simple power stream deflecting techniques are in many instances adaptable to fabrication within planar units.

3,446,228 Patented May 27, 1969 Such planar units advantageously may be produced by higher volume manufacturing techniques, one of the more popular being a variation of typical photoetching processes. When thusly fabricated, however, the fluid elements selected for use preferably should be operable in thin and relatively small overall dimension.

Assembled to form a control device, the pure fluid systems offer the attributes of low cost and relatively high reliability. Their inherent structural integrity and mode of operation provides a performance exhibiting resilience to vibration, shock, high temperature and high radiation environments. Further, their adaptability in some instances to manufacture using printed circuit techniques makes available desirable high density overlap packaging designs.

While the fluid digital devices have found increasing favor as a desirable substitute for equivalent electronic systems, their development has encountered limitations. For instance, fluidic element operating rates are presently much slower than the operating rates of comparable electronic devices. A network of logic elements for sequencing, computing, and control system application must be designed to complete a data cycle within a certain time interval parameter as is established by intended system use. Where individual element operating rates are comparatively slow or protracted, the number of logic elements which may be included within a network will be limited by the allowable data cycle elapsed time interval. The number of logic elements incorporated within a network is directly relatable to the numerical accuracy and functional capability of a digital device. It follows, then, that a reduction in the number of individual elements and in operating rates is of importance to pure fluid component development. Response times of typical pure fluid elements are found to be relatable to a number of operational or design factors, the more apparent being that of overall size or dimension. Fluid element designs which may be scaled down are available to a much wider spectrum of operating rates and attendant uses. The smaller elements are also operable under lower power supplies.

Another factor affecting response times rates lies in element configuration. The switching of fluidic devices utilizing a wall attachment or Coanda elfect involves an undesirable time-consuming period required to break down the pressure differentials developing the mutual atfinity between a power stream and an adjacent sidewall.

Generally, the establishment of a Coanda or boundary layer elfect requires an initial turbulent power stream flow. This form of flow is difiicult to attain in miniature nozzles due to their high ratio of surface to cross section. Consequently, conventional fluid elements are not readily scaled down to meet miniaturization requirements, and their resultant larger size, in turn, necessitates higher power input requirements. This added power demand is not desirable where a larger number of logic elements must be interconnected within an integrated circuitry.

Studies have determined that there exists a minimum Reynolds number below which wall attachment devices and the like will not function. Generally, these boundary effect configurations must operate at Reynolds number values above 1000. Inasmuch as the number represents a ratio of fluid inertial to viscous forces, it will be apparent to those skilled in the art that the above limitational value will, in turn, render the performance of wall effect elements sensitive or dependent upon temperature conditions.

A further hindrance to the development of pure fluid logic devices lies in the presence of spurious pressure deviations extant within the interconnected dormant conduits of a typical operating system. Those power and control ports or ducts within a logic network which are not receiving fluid will often be subjected to random, extraneous and usually negative pressure variations. aberrations may, in turn, aflect related elements, thereby establishing undesirable impedances and the like throughout the interconnecting channels of a logic system. Measures taken to correct the discrepancies generally invoke undesirable circuit additions, which, in turn, detract from otherwise available supplies of power.

OBJECTS The inventive pure fluid element now presented offers a relatively broadened scope of desirable circuit operational parameters to those engaged in the fabrication of fluid circuitry. The new fluid element configuration utilizes a novel arrangement of mutually opposed, transiently impacting power streams to evolve multistable characteristics.

The element has been found to function under powerjet Reynolds number values suggesting an initial laminar fluid flow. As such the device may be readily scaled down to meet miniaturization and reduced input power requirements. Of particular advantage, the inventive element has been found to operate at Reynolds number values below 10. In view of this relative independence of Reynolds number limitation, the element enjoys a wide performance range at elevated temperatures. The active fluid element, in addition to its advantageous miniaturization capability, is fabricable within a conventional logic plane. As a result, it readily may be incorporated within high density integrated circuitry packaging schemes.

The inventive element is also characterized in having a very fast reaction time while retaining the desired memory characteristics of slower wall effect devices. In the element, fluid stream directional flow is in part stabilized by virtue of a juxtaposed and mutually reinforcing orientation of its two power streams. As a consequence, no appreciable time lapse is encountered in breaking down or negating the pressure differential of a boundary layer effect with a lesser powered control stream at the instant of switching.

As another object, a significant operational advantage derived from the new logic element configuration lies in the isolated status of its dormant or inactive output ports or ducts. During operation, these dormant channels are found to remain unaffected by negative or other spurious pressures. In logic network arrangements having high sensitivity, this isolation characteristic may be utilized to considerable advantage.

A further object of the invention is to provide an active pure fluid element having more than one mode of operation. By appropriate adjustment of opposing power and control stream inputs, the inventive element may operate with more than two different stable power stream orientations. This characteristic will allow, in numerous fluid logic applications, for a significant reduction in the number of individual elements required to perform a given function.

Still another object of the invention is to provide an active pure fluid logic element which may incorporate an increased number of distinct control stream ports, thereby permitting enhanced circuit design flexibility.

These and other objects of the invention are further described and illustrated in the following discussion and related drawings in which:

FIGURE 1 is a perspective view of fluid element shim fabricated in accordance with our invention;

FIGURE 2 is a perspective view of an alternate configuration of a fluid element in accordance with the invention;

FIGURE 3 is a planar view of a fluid element in accordance with the invention, illustrating the relative positions of its power streams at the instant of their aligned interception and impaction;

FIGURE 4 is a planar view of a fluid element in accordance with the invention illustrating one of its available stable modes of power stream flow;

FIGURE 5 is a planar view of a fluid element in accordance with the invention illustrating another of its available stable modes of power stream flow;

FIGURE 6 is the graphic representation of a stepped function which may be generated utilizing bistable pure fluid circuitry;

FIGURE 7 is a schematic drawing of a pure fluid circuit for generating a typical stepped function equivalent to that which may be generated by the present fluid element.

FIGURE 8 is a line schematic representation of modes 10 of operation of the instant inventive fluid element for evolving the stepped function discussed in connection with FIGURES 6 and 7.

STRUCTURE When fabricated in accordance with conventional manufacturing techniques, fluid element profiles are etched or cut from sheets of metallic or plastic material. The sheets, sometimes referred to as shims, may be aligned and stacked between outer facing sheets retaining fluid input and outlet ducts and ports. Appropriately interconnected, the laminar assembly forms an integrated fluid logic circuitry capable of performing a preselected logic function.

Referring to FIGURE 1, a portion of a planar shim formed from a metallic sheet 10 is depicted. From sheet 10, the profile of an embodiment of the fluid device of the present invention has been etched. The element is characterized in having at least two mutually opposed primary or power stream input passages such as those shown at 12 and 14. Passages 12 and 14 terminate forming nozzles at respective recessed areas or notches and 22 disposed within the endwalls and 32 of a power stream interaction chamber 24. The chamber 24, situate 35 between the power stream nozzles, is defined by the endwalls 30 and 32 and sidewalls 26 and 28. Symmetrically aligned and having a convex curvature as viewed from within the chamber, these sidewalls are seen to impart an hourglass shape to chamber 24. Outlet passages 34, 36, 38 and 40 extend outwardly in fluid communication from the corners of the interaction chamber and are seen to diverge slightly forming diffusers for pressure recovery purposes. As may be evidenced from the drawing, the outward wall of each of the outlet passages is fashioned having a contour representing a smooth transitional extension of the nearest interaction chamber sidewall.

Control stream passages as shown at 42 and 44 are positioned within the element at each side of the primary input passage 12. In the design configuration of FIGURE 1, the passages are fashioned to introduce a control signal stream having an angular orientation with respect to the outlet direction of the power stream. As is seen in the drawing, this orientation may be other than 90 degrees. Thusly positioned, a control stream will inject to react 5 with the power stream within or near the recess '20.

Similarly, control stream ducts 46 and 48 are positioned along both sides of power stream passage 14 in a manner providing for fluid flow interception within or near the outlet of recess 22. 1 a

An alternate design embodiment of the fluid device is portrayed in FIGURE 2 wherein a larger scale element profile within shim 50 is enclosed between the facing planar inward surfaces of plastic blocks 52 and 54. The blocks are secured together by bolt and nut assemblies as at 56. Portions of the fluid element profile configurations within shim 50 are identified by numbers corresponding to those utilized in conjunction with the element profile description of FIGURE 1 'where no structural or functional alteration is encountered.

As. in the case of FIGURE 1, the device of FIGURE 2 incorporates an hourglass-shaped interaction chamber 24 having concave sidewalls 26 and 28 into which lead power stream passages 12 and 14, which, in turn, are mutually aligned to face each other along a common axis.

Conduits 17 and 19, respectively, interconnect input compartments 16 and 18 with exterior power fluid supply means. Similarly, conduits 35, 37, 39 and 41 are adapted to interconnect the corresponding output passages 34, 36, 38 and 40 with appropriate external tubular leads. The arrangement of the device of FIGURE 2 differs in the positioning of its control passages. Particularly, control stream passages 58, 60, 62 and 64 are deployed outwardly from sidewalls 26 and 28. Under this alternative design, a fluid control system control stream issuing 'from any of the passages will intercept a contiguous power stream within chamber 24 and at a more normal angular relaitonship. A greater or lesser number of control ports situate as shown in either or both of FIGURES 1 and 2 may be incorporated with the fluid device to provide control stream switching functions.

OPERATION Basic flow FIGURES 3, 4 and 5 are provided for the purpose of illustrating the flow patterns or sequence of fluid movements encountered during the basic switching operations of the device. At the instant at which fluid is injected at relatively equal pressures through both power ducts 12 and 14, the respective resulting fluid power streams 12a and 14a will directly impact at some location within the constricted central portion of the interaction chamber 24. This instantaneous stream condition, as is pictorially represented in FIGURE 3, is one in which the streams or jets are not in equilibrium and will convert rapidly to a stable tangential flow pattern such as that illustrated in FIGURE 4. In that figure, the stream 12b from passage 12 is shown crossing chamber 24 to exit from opposite outlet 36, while jet 14b from passage 14 crosses to exit from the diagonally opposite outlet passage 38. The streams pass each other at about the center of chamber 24. Under the stable flow pattern shown, it is observed that fluid vortexes as are depicted at 11a, 11b and 110 are developed tending to enhance or reinforce flow pattern stability. Where no control pulses or pressure variations are imposed upon the impacting stream configuration of FIGURE 3, the power jets may also assume an alternate stable flow pattern wherein the stream from passage 12 will exit from outlet passage 34 and the stream from passage 14 will exit from diagonally opposite outlet passage 40. Unless acted upon by a suitable control function such as a fluid pulse, either of the abovedescribed flow patterns will remain stable. While such mutually opposed and impacting power jets can be in equilibrium while head-on in a free air or less confining environment, the confining, restrictive geometry of the constrictive portion of the hourglass-shaped chamber 24 will prevent such a flow status. However, should the point of power stream mutual impaction be located beyond the constrictive portion of chamber 24, an equilibrium status will be established.

SWITCHING From a stable orientation wherein the power stream from passage 12 exits from outlet passage 34 and the power stream from passage 14 exits from outlet passage 40, the alternate stable orientation illustrated in FIGURE 4 may be effected by the presence of a control pulse or stream from either or both of control passages 42 and 48.- Such control streams are depicted respectively at 42b and 48b. The power streams having been thusly switched, the output pressure is high at passages 36 and 38 and low at ports emanating from passages 34 and 40. Conversely, a switching from the flow mode of FIGURE 4 may be effected by the presence of a control pulse or sfieam at either or both control passages 44 and 46. The switched fluid flow mode will then provide a high output pressure at passages 34 and 40 and a low pressure at passages 36 and 38.

Precise positioning of the control passages has not been found critical to the effective design of the element. For instance, identical power stream flow patterns and switching techniques may be accomplished with the alternate configuration of the element shown in FIGURE 2. In either arrangement, the control streams need not intercept the power streams at a angle.

Inasmuch as only a relatively minor fluid flow or pressure is required from any of the control passages to deflect and switch the power stream orientations, an advantageous gain is realized with the device. Once mutual impact of the opposed power streams is initiated, the higher energy power streams, themselves reacting with each other, within the constrictive portion of chamber 24, apparently will emphasize or accelerate fluid path reorientation. As a consequence, elapsed switching time is shortened and gain is enhanced. Further, no boundary wall or Coanda effect must be overcome by a control stream in initially deflecting a power stream to impaction. As a result, a higher switching speed and sensitivity of the element is realized.

Within designs utilizing load sensitive, unvented diffusers, the basic flow configuration of the element may also be switched by blocking either or both of the outlet passages or diffusers receiving a power fluid fiow. For example, the flow mode depicted by streams 12b and 14b in FIGURE 4 may be switched to the alternate mode by blocking either or both of outlet passages 36 and 38.

The basic configuration as described above has been found to be relatively insensitive to output loads or impedances. This desired characteristic is thought to be a result of the somewhat independent status of the power streams in any of their stable configurations or modes.

Turning to FIGURE 5, another state of operation and fluid flow pattern available with the device is illustrated. By imposing a higher pressure upon one supply stream, so as to cause impaction beyond the restricted portion of chamber 24, the element will function as a stable, directimpact device. As shown in the figure, power stream is established under a suitably higher pressure head than oppositely disposed power stream 140, thereby causing the point of impaction to be positioned within the relatively wide diverging portion of chamber 24 near to output passages 34 and 36. Under this condition, all fluid outflow has been found to egress evenly through each of the nearest output passages 34 and 36.

This same output fluid flow pattern may also be realized by the simultaneous, equal and continuous actuation of both control jets at one end of the element. For instance, the presence of a continuous control tflllld flow at each of the control passages 42 and 44, both reacting with power stream 12c, will produce stable, mutual power stream impaction beyond the constrictive portion of chamber 24 and an output pressure at outlet passages 34 and 36. Conversely, the presence of simultaneous and continuous control fluid flow at each of the control passages 46 and 48 will produce stable mutual power stream impaction and a corresponding output pressure at outlet passages 38 and 40.

PERFORMANCE The relatively broad functional capability of the present element as a multistable switching device and pure fluid amplifier is illustrated in connection with FIGURES 6 through 8. In a number of fluid logic applications, it is desirable to generate a stepped function in which any one of four states may occur at a point in time as a result of an input logic. A truth table for such function is as follows:

where A, B, C and D are fluid input signals and a, {3, y and 6 are output signals representing four discrete levels, for instance of fluid pressure over a time At. The function thusly generated is typically represented by the graph of FIGURE 6. conventionally, the stepped function is evolved through a fluid logic circuitry the design of which is illustrated schematically by the diagram of FIGURE 7, which will be seen to function as the equivalent of one program of use for the present pure fluid element. In the figure, four fluid AND gates 7 1, 12, if): and 24 are appropriately interconnected to control signal input conduits 76, 77, 78 and 79, which respectively are adapted to convey fluid input signals A, B, C and [D representing those set forth in the above-described truth table. Signal A is connected by conduit 76 to gates n and 7 3 signal B is connected by conduit 77 to gates 12 and 18, signal C is connected by conduit 78 to gates 22 and E, and signal D is connected by conduit 79 to gates 7 1 and B. The AND gates E to 1% are vented respectively at outlets 81 to 84 and power input fluid is supplied the elements respectively at ports 91 to 94. As is known in the art, the programmed output signals a, 5, 'y and will, when approriately reacting to the dual input signals, be selectively present respectively at output ports 101, 102, 193 and 104 to avail a stepped function as shown in FIGURE 6.

The above-described conventional circuitry and its related step function operation represents the equivalent of a logic program which may be performed by the singular pure fluid element of the present invention. Referring to the line schematic of the element in FIGURE 8, control signals A, B, C and D corresponding to the input signals of the truth table and equivalent circuit are selectively in troduced respectively along control passages 48, 46, 44 and 42 of the element. Power input fluid corresponding to the power inputs 91 to 94 of the equivalent circuit are present at input passages 12 and 14. Responding as earlier described to appropriate energization of the control signals A to D, resultant output signals E, F, G and H will appear respectively at the outlet passages 36, 40, 34 and 38. The resultant equivalent truth table evolving the stepped function of FIGURE 6 from the element will be as follows:

In addition to performing the identical stepped function task of the equivalent circuit with one rather than four elements, the present element additionally incorporates a desirable memory characteristic in connection with the a and 5 functions. For instance, returning to FIGURE 7, if the presence of output a is desired for a time interval At from the equivalent circuit, a continuous associated signal A+D must appear at element 71 for time At. In comparison, only a short interval control pulse need be present at either or both control passages 48 and 42 of the present element to hold the power stream output a at passages 36 and 38 until a new input signal is imposed. It will be apparent that the bistable characteristic of the element may similarly be utilized to evolve the ,8 function. In order to develop similar memory characteristics for the equivalent circuit of FIGURE 7, additional fluid memory elements would be required.

As many may be evidenced from the foregoing discussion, the pure fluid logic element now presented lends it- 8 self to a broadened flexibility in fluid circuit design, and accordingly will be found to have advantageous application to a wide variety of logic circuit requirements.

We claim:

1. A fluid logic device comprising:

(a) at least two power stream input passages aligned along a common axis and disposed oppositely so as to provide for the mutual impaction of fluid power streams issuing therefrom;

(b) an interaction chamber in fluid communication between said input passages and having wall means configured to provide a volumetrically restrictive region intermediate said passages so as to render the fluid flow path orientation of said streams unstable during their mutual impaction within said region;

(c) at least four fluid outlet ducts communicating with said chamber and situate to receive said power streams following their mutual impaction; and

(d) means providing at least two control fluid passages for selectively activating fluid control streams that will interact with the fluid flow of at least one of said power streams to selectively control the fluid outlet ducts receiving said power streams following their mutual impaction.

2. The fluid device of claim 1 wherein at least two of said control fluid passages are positioned adjacent to and respectively upon opposite sides of at least one of said power streams so as to selectively provide control stream fluid flow reactable therewith.

3. The fluid logic device of claim 1 wherein said interaction chamber wall means are formed having a con vex curvature as viewed from within said chamber, intermediate said oppositely disposed power stream input passages, thereby establishing said volumetrically restrictive region.

4. The fluid device of claim 3 wherein said fluid outlet ducts are formed in part by smooth extensions of said curved interaction chamber wall means.

5. The fluid device of claim 4 in which the profiles of said power stream input passages, said interaction chamber and said fluid outlet ducts are incorporated within a planar shim.

6. The fluid logic device of claim 3 wherein at least two of said control fluid passages are positioned adjacent to and respectively upon oposite sides of at least one of said power streams so as to selectively provide control stream fluid flow reactable therewith.

7. The fluid device of claim 6 wherein said fluid outlet ducts are angularly positioned from said power stream passage axis and spaced therefrom and are formed in part by smooth extensions of said curved interaction chamber Wall means.

8. The fluid device of claim 7 in which the profiles of said power stream input passages, said interaction chamber, said control fluid passages and said fluid outlet ducts are formed within a planar shim.

References Cited UNITED STATES PATENTS 3,272,212 9/1966 Bowles 137-81.5 3,272,215 9/1966 Bjornsen et al 137--81.5 3,295,543 1/ 1967 Zalmanzon 13781.5 3,323,532 6/1967 Campagnuolo 137-81.5

SAMUEL SCOTT, Primary Examiner.

Images