DETAILED DESCRIPTION OF THE INVENTION

[0054] FIG. 1 illustrates a launcher 100 which launches a table tennis ball B. The ball B is held within a basket 105, which contains perforations 106, which reduce aerodynamic drag and back-pressure upon the ball during launching. Logic 108 controls a solenoid 25, whose plunger, or hammer, 115 strikes the ball B upon actuation of the solenoid 108. As indicated by arrow 118, the logic 108 performs three functions.

[0055] In block 125, the logic 108 inquires whether sensor 120 detects that the ball is present within the basket 105. If not, the NO branch is taken, and the logic idles in loop 126 until the sensor 120 detects a ball.

[0056] When a ball is detected, the logic exits loop 126, and reaches block 130, which imposes a delay. As will be explained below, the delay is adjustable, and can range from a length of zero to a few dozens of seconds in duration. Next, block 135 is reached, wherein the solenoid 110 is energized, thereby ejecting the ball from the basket.

[0057] The operation just described allows a pair of launchers 100 to play the game of “catch” as shown in FIG. 2. In FIG. 2A, launcher 100A launches the ball B to launcher 100B, which catches ball B in its basket 105B. The sensor 120 (not shown) in launcher 100B detects the presence of the ball. In FIG. 2B, the delay of block 130 in FIG. 1 is imposed. In FIG. 2C, launcher 100B returns ball B to launcher 10A, and the sequence of FIGS. 2A, 2B, and 2C continues.

[0058] Details concerning the construction of launcher 100 in FIG. 1 can be found in U.S. Pat. No. 5,100,103, issued on Mar. 31, 1992, and filed on Feb. 20, 1990, Ser. No. 482,035, and in U.S. Pat. No. 5,125,668, issued Jun. 30, 1992, filed on Apr. 24, 1990, Ser. No. 513,928. The inventor in both of these patents is Gregory A. Welte, a co-inventor herein. Both of these patents are hereby incorporated by reference in their entireties, including definitions stated therein.

[0059] The logic 108 can be implemented in either digital or analog format. FIG. 3, which contains a partial, annotated, copy of FIG. 4 of the ′103 patent identified above, illustrates analog circuitry which can implement the logic 108 . The overall action is to trigger the ONE SHOT into driving Darlington 52 temporarily into conduction, to apply a pulse to solenoid 25.

[0060] In executing this action, sensor 120 in FIG. 1 closes switch 40 in FIG. 3 momentarily, when the ball B is detected. Switch 40 can take the form of a transistor. Upon closure, capacitor 42 in FIG. 3 charges almost immediately. When switch 40 is opened, capacitor 42 begins to discharge through resistor 44. This discharge causes voltage Vcap to drop.

[0061] When Vcap falls below the voltage on line 48, comparator 46 trips, triggering the ONE SHOT, which issues pulse 51. The time interval between release of switch 40 and the tripping of comparator 46 corresponds to the delay of block 130 in FIG. 1. Pulse 51 momentarily turns on Darlington 52, causing the hammer H of solenoid 25 to strike the ball 29.

[0062] The launcher 100 in FIG. 3 is equipped with two knobs 140 and 145. Knob 140 controls the power applied to the solenoid 25, and thus controls the force with which the hammer H strikes the ball, thereby controlling the distance which the ball 29 travels. One approach to controlling this power is to adjust the duration of pulse 51. This adjustment can be made by adjusting the RC time constant of the appropriate resistor and capacitor used by the ONE SHOT, as indicated by arrow 150. This adjustment allows the distance between launchers 100A and 100B in FIG. 2 to be changed. Another approach is to adjust the voltage applied to the solenoid 25. FIG. 2 shows ten volts being applied; that voltage can be changed.

[0063] Still another approach is independent of knob 140. The speed with which the hammer H of the solenoid strikes the ball 29 in FIG. 3 depends upon how far the hammer H is withdrawn from the solenoid at the time current is applied to the solenoid. If the hammer H is held at a slightly withdrawn position, the final speed will be small, compared with that occurring if the hammer H is withdrawn farther. A set screw SET can adjust the resting position of the hammer H.

[0064] Knob 145 in FIG. 3 controls the length of the delay of block 130 in FIG. 1, by changing the RC time constant of resistor 44 and capacitor 42 in FIG. 3, as indicated by arrow 155. Control over this delay will be important in other embodiments, discussed later.

[0065] In FIG. 3, the basket 105 can pivot about pivot P, thereby changing angle A, to thereby allow an additional adjustment in the range of the launcher 100.

[0066] In FIG. 4, a motorized miniature truck 200 carries a basket 205. The truck 200 follows a path 210, which can represent a guide rail which constrains the truck, or can represent the path taken by the truck because of either the truck's design or programming. The truck moves from the FIRST position to the SECOND position. Then, when the truck reaches the THIRD position in FIG. 5, the launcher 100, of the type shown in FIG. 1, launches a ball B. If the launching is timed properly, the ball B will be captured by the basket 205, as indicated in the FOURTH position.

[0067] The truck 200 continues along path 210, as indicated by the FIFTH position in FIG. 6, carrying the ball, and reaches the SIXTH position. The player (not shown) removes the ball B, loads it into the launcher 100, and the truck continues to the FIRST position shown in FIG. 1.

[0068] During the truck's travel from the FIRST position to the SIXTH position, the firing of the launcher 100 can be undertaken in several ways. In one approach, the time delay is appropriately adjusted by knob 145 in FIG. 3. Switch 40 in FIG. 3 is momentarily closed when the truck is located at the FIRST position in FIG. 4. The truck proceeds to the SECOND position, and the launcher 100 launches the ball B when the truck is at the THIRD position in FIG. 5, when the time delay expires, thereby assuring that the basket 205 will capture the ball as indicated in the FOURTH position.

[0069] In another approach, the firing of the launcher is under direct control of the player. For example, the RC time constant of resistor 44 and capacitor 42 in FIG. 3 can be very small, so that launch of the ball B occurs immediately upon closure of switch 40. The player controls switch 40, and causes launch when the truck is located at the THIRD position in FIG. 5.

[0070] In a third approach, a computer controls the launch, as discussed later.

[0071] In a fourth approach, a sensor 207 in FIG. 5 detects the presence of the truck 200, and issues a signal on line 73. Line 73 connects to diode 141 in FIG. 3, and triggers the ONE SHOT.

[0072] FIG. 7 illustrates a variation of the apparatus just described. The truck 200 is equipped with a launcher 100D. After the truck reaches the position shown in FIG. 7A, the launcher 100 fires. A net 250 within the truck 100 catches the ball B, and feeds the ball B to the truck's launcher 100D. The net 250 can take the form of basket 105 in FIG. 3, which is part of launcher 100.

[0073] In FIG. 7B, after a delay, the truck's launcher 100D launches the ball B to a stationary launcher 100C, which catches the ball B, as indicated in FIG. 8A. Then, after another delay, stationary launcher 100C launches the ball B to the first launcher 100, as indicated in FIG. 8B. If the delays of the proper lengths are selected, the sequence just described will continue.

[0074] FIGS. 17 and 18 illustrate a variation. In FIG. 17A, truck 200 drives along path 520. In FIG. 17B, launcher 100F launches a ball B into the truck. In FIG. 18A, the truck 200 continues its travel. In FIG. 18B, the truck 200 launches the ball B to the launcher 100F, and the process continues.

[0075] In another embodiment, shown in the sequence of FIG. 9, truck 200 carries a mobile launcher 100E, and drives away from a stationary launcher 100A. The latter launchers ball B, which is caught by the mobile launcher 100E. The mobile launcher 100E then launches the ball B to the stationary launcher 100A. The appeal of the apparatus of FIG. 9 to a player is that care must be taken to set the delays and power settings of both launchers properly.

[0076] As shown in FIG. 10, the truck 200 need not follow a straight-line path, but may follow a serpentine path. The truck may be programmable as to path. The patents incorporated above describe programming approaches. Also, programmable vehicles are known in the art.

[0077] FIG. 11A illustrates another type of launcher 300. A barrel 302 contains a launching station 305, and utilizes the principles of FIGS. 1 and 3 to launch a ball. A magazine of balls 307 may be provided, which holds a supply of balls. A net 315 catches incoming balls, and delivers them to the magazine.

[0078] In operation, the launcher 300 can launch a ball toward a wall 310, as indicated in FIG. 11B. The wall deflects the ball, as in FIG. 12A, and the net 315 catches the deflected ball, as in FIG. 12B. This process repeats.

[0079] In FIG. 13, a wall is not used, but a human (not shown) catches and returns the ball B.

[0080] FIG. 14 illustrates construction details of one type of launcher. In FIG. 14A, a barrel 400 contains a launching station 405. Balls are delivered to the launching station 405 by a funnel 410, through aperture 412. A sleeve-type net 415 delivers balls to the funnel 410. A goal-type net 420 catches incoming balls, and delivers them to the sleeve-net 415. A solenoid 425 launches the ball (not shown).

[0081] A base 430 supports a mast 435, which carries a U-shaped bracket, which supports the barrel 400. FIG. 14B illustrates the apparatus in assembled form.

[0082] A significant feature of FIG. 14B is that the funnel 410 does not contact the barrel 400. Further, the components are configured such that, when the barrel 400 is rotated about either axis 450 or axis 460, the aperture 412 remains in a position which enables funnel 410 to deliver balls to the launching station.

[0083] This relative fixity of position of funnel 410, with respect to aperture 412, allows the barrel 400 to be positioned without moving goal-net 420. This can be important when two launchers are used in the game shown in FIG. 2. First, the goal-net 420 of the second launcher is positioned, and the first launcher 400 is adjusted to shoot a ball into that net. When this adjustment is accomplished, the second launcher is then adjusted to fire a ball into the goal-net of the first launcher.

[0084] However, if the adjustment of the second launcher required movement of its goal-net, then the first launcher would require additional adjustment, in order to strike the newly positioned goal-net of the second launcher. The separation between barrel 400 and funnel 410 in FIG. 14 eliminates this problem.

[0085] FIG. 15 is a cross-sectional view of part of FIG. 14, showing barrel 400, launching station 405, solenoid 425, ball B, and funnel 410.

[0086] FIG. 16 is a cross-sectional view of an apparatus of the type shown in FIG. 14, but with some modifications. The sleevenet 415 of FIG. 14A has been eliminated. FIG. 16 shows that the axis of rotation 460 of barrel 400 is concentric with funnel 410. Barrel 400 can also rotate about pivot P, as indicated by arrows 470. A sensor 500 detects the presence of ball B, and delivers a signal so indicating to control 510, which is discussed later.

[0087] Apparatus which controls the events described above is shown in FIGS. 19-21. FIG. 19 is a top view of FIG. 7A, with components added. Two sensors, SENSOR 1 and SENSOR 2, are shown. Such sensors are described in the incorporated patents. These sensors detect the presence of the truck 200 in FIG. 17, and issue signals which will be called “vehicle-present” signals.

[0088] A computer is indicated as a receiver of the “vehicle present” signals. When a vehicle-present signal is received from SENSOR 1, the computer orders the launcher 100F (corresponding to launcher 100 in FIG. 7A) to launch a ball. This order can take the form of a signal to diode 141 in FIG. 3, which triggers the ONE SHOT. The precise time required for a proper launch will depend upon many factors, so that a trial-and-error process will be required to determine when the launch signal should be issued, and to program the computer accordingly

[0089] If the launch is successful, the truck 200 in FIG. 7B will receive the ball. Then, the truck will launch the ball, as in FIG. 8A. Launcher 100C will catch it. Then, as in FIG. 8B, launcher 100C launches the ball to launcher 100, which then awaits another signal from the computer.

[0090] In the preceding scenario, the truck 200 in FIGS. 7 and 8 launched the ball under its own control. In another embodiment, the truck can be controlled by the computer. In FIG. 19, the presence of the truck 200 at SENSOR 2 can be detected by the computer, and the computer orders the truck 200 to launch the ball at an appropriate time afterward, in order to deliver the ball to launcher 100F as in FIG. 18B.

[0091] The computer 550 can communicate with the truck using an RF, or infrared, link, as indicated in FIG. 22. Alternately, a “hard-wired” link can be used: path 520 in FIG. 19 can take the form of a model railroad track. Truck 200 can take the form of model train 200A in FIG. 23. The computer communicates with the train 200A through the rails of track 520 in FIG. 19.

[0092] In another embodiment, a model train 200A in FIG. 23 is used, rather than a truck, and the train 200A runs along railroad tracks. The train 200A carries two, or more, nets 580. The computer causes a launcher to launch two balls in rapid succession, each to strike one net.

[0093] The launchers can be adjusted in position by the computer. FIG. 24A shows a servo-mechanism, commonly called a servo. Such mechanisms are commercially available, and are used to control radio-controlled model aircraft. FIG. 24B illustrates a linkage used to rotate a launcher 100H left-and-right, as indicated by arrows 213, through rotation of crank 610. FIG. 24C illustrates a linkage used to move cannon 620 up-and-down.

[0094] It may be desirable to control the exact instant of firing the ball from a vehicle by reference to a station on the ground. FIG. 19A illustrates a disc 535 carried by a mast 536 supported by a base 530. The base is positioned upon the railroad track 539 as indicated in FIG. 20. The train 200A in FIG. 21 carries a sensor 540, which detects the proximity of the disc 535.

[0095] In operation, the disc 535 provides an approximation of the location for shooting the ball. The delay, controlled by knob 145 in FIG. 3, “fine tunes” the actual time of shooting. Thus, for example, the disc 530 of FIG. 19A would be placed at the position of SENSOR 1 in FIG. 19. After fine-tuning the delay, the train 200A of FIG. 21 will successfully shoot the ball into launcher 100C in FIG. 8A.

[0096] FIGS. 24D and 24E illustrate another embodiment. A launcher 100M is concealed within a case designed to resemble office furniture. An actuator 570, through linkage L1, raises a LID, or otherwise exposes the launcher, as shown in FIG. 24. A remote control 590, known in the art, allows an office worker to open the LID, and actuate the launcher 100M, in order to play a clandestine game of catch.

[0097] Preferably, the remote control 590 is capable of using servos of the type shown in FIG. 24 to adjust the launcher 100M, so that the worker can remotely adjust the direction, and range, of firing by the launcher. In addition, the remote control 590 allows adjustment of the delay and the power, thus, in effect, allowing remote control of knobs 140 and 145 in FIG. 3.

[0098] FIG. 25 illustrates another variation, wherein launcher 100N is fed balls by a miniature basketball basket 620. This embodiment has the advantage of eliminating long, slender objects, which may injure children.

[0099] FIG. 26 illustrates another embodiment, wherein launcher 100 is mounted to a clothespin-like clamp 700. A NET mounts to the launcher 100. FIG. 27 illustrates this embodiment fastened to a bracket 730 supported by base 430. A baffle 720 may be added to funnel 410, to guide the ball B into the basket 105 when caught.

[0100] An advantage of the embodiment of FIG. 26 is that it can be used in a stationary mode, as in FIG. 27, or can be clamped to a vehicle, such as truck 200 in FIG. 7A.

[0101] Many of the FIGS. above show a CONTROL which controls shooting of the ball. FIG. 28 illustrates one architecture for a control 800. Sensor 120 detects the presence of ball B, and issues a signal on line 810. Block 820 indicates that nothing happens, at least not automatically, until this signal is received.

[0102] A switch 830 is provided. If the user has positioned the switch in a position calling for continuous, or automatic, shooting of the ball B. block 840 detects this fact, and starts the delay 850. If the switch is not so set, then automatic shooting does not occur, but may occur for other reasons, as will be seen.

[0103] The user can adjust the delay 850, as indicated. After the delay, block 870 actuates solenoid 425, thereby shooting ball B. As indicated, the user can control the power with which the solenoid 425 strikes the ball B. If ball B returns to the launching station 890, as when another launcher returns it, or a human tosses it into funnel 410, the steps just described are repeated.

[0104] In addition to the automatic shooting just described, shooting can be triggered in other ways. A remote signal received by block 880 can induce shooting, such as signals received from a computer (or other logic), a sensor, of from a switch actuated by a user, all as indicated.

[0105] FIG. 29 illustrates a somewhat more hardware-oriented description of the control 800. OR gate 900 receives signals from the ball sensor 120, from a switch, from a computer, or from a sensor. When any of these signals is received, delay 910 is triggered. The length of delay is adjustable, as indicated by knob 920.

[0106] When the delay expires, a signal is applied to line 925. A second OR gate 930 receives this signal, together with signals from a switch, a computer, or a sensor. (These latter signals can be applied when shooting of the ball is desired without the delay. If a delay is desired, these signals would be applied to OR gate 900 instead.)

[0107] OR gate 930 triggers a switch 940, which connects a power source 950 withg the solenoid 425. As indicated by knob 960, the user can adjust the power source 950, to control the force with which the solenoid 425 strikes the ball. Circuits which implement the blocks of FIG. 29 will now be discussed.

[0108] In FIG. 30A, delay 910 can be implemented by two SCHMITT triggers 1010 and 1020, connected through resistor R and capacitor C. FIG. 30B illustrates the steps involved. When V_IN−1 goes LOW, as indicated, V_OUT−1 goes HI. Capacitor C charges through resistor R, producing the exponential rise of V_IN−2. When V_IN−2 crosses the trigger point of SCHMITT 1020, V_OUT−2 goes LOW, as indicated.

[0109] FIG. 30C summarizes the preceding events in a timing diagram. The DELAY is indicated. As indicated in FIG. 30B, resistor R can be adjustable, thereby allowing adjustment of the delay, as described earlier.

[0110] FIG. 31 illustrates a circuit for implementing the switch 940 and power source 950 of FIG. 29. Basically, resistor R holds V_IN−4 at a LOW state (or a HI state, if R is connected to a high voltage), thereby holding V_OUT−4 in a LOW state (or HI state). It will be assumed that V_IN−4 is held at a LOW state.

[0111] But when V_OUT−3 goes HI, V_IN−4 is temporarily pulled HI, but then exponentially decays, as capacitor C charges. When the rising V_IN−4 crosses the TRIP point for SCHMITT trigger 1040, the output of the latter goes LOW, as indicated. Then, when the decaying V_IN−4 again crosses the TRIP point, V_OUT−4 goes HI again. (Because of the hysteresis inherent in a SCHMITT trigger, the two TRIP points are not identical, but that detail is ignored here.) While V_OUT−4 is LOW, the field-effect transistor, FET, is triggered into conduction. This triggering energizes the coil 1050 of the double-pole, double throw RELAY. This energization causes reed 1060 to change to the dashed position, thereby connecting 24 volts across the solenoid 425, to shoot the ball (not shown).

[0112] As indicated, resistor R is adjustable, to adjust the length of the PULSE applied to the FET, to thereby control the amount of time the solenoid 425 is energized, to thereby control the amount of energy delivered to the ball (not shown). Also, the one-shot which is shown can be replaced by a properly connected 555 timer, as known in the art.

[0113] FIG. 32 illustrates another approach to controlling the energy delivered to the ball. One of power resistors R1-R4 is selectively placed in parallel with solenoid 425, byb adjusting the wiper W of rotary switch 1100. If no resistor is to be placed in parallel, the wiper W is connected to the NC terminal.

[0114] The resistors are of different values. For example, if resistor R1 equals the resistance of the solenoid 425, then placing that resistor in parallel with the solenoid 425 will cut the power absorbed by the solenoid in half. Of course, this approach wastes power, but the waste may be tolerated, in the name of simplicity.

[0115] FIG. 33 illustrates another approach to firing the solenoid 425. A large capacitor C1 is connected as shown. FIG. 33 illustrates a type of voltage doubler. Other voltage multipliers are known in the art. FIGS. 34A and 34B illustrate the operation, with non-relevant lines eliminated.

[0116] When the RELAY is in its non-powered state, battery BAT charges capacitor C1, as indicated in FIG. 34A. When the RELAY is powered, as in FIG. 34B, the capacitor C1 is placed in series with the battery BAT, thereby doubling the voltage applied to the solenoid 425.

[0117] The apparatus of FIG. 33 allows a higher voltage to be applied to the solenoid 425 than is available from battery BAT. In other types of voltage multiplier, two, or more, capacitors are charged through diodes, and then placed in series. For example, if ten capacitors are charged to 9 volts each, when they are placed in series they provide 90 volts of potential.

[0118] FIG. 35 illustrates a perspective view and a side view of a target T containing a sensor S, as described in the incorporated patents. The sensor S detects a ball-strike by a table tennis ball and produces a signal. When a ball arrives, as in FIG. 36, the signal is fed to a CONTROL, which fires solenoid 425. A linkage LI causes the target T to pivot, as in FIG. 37, thereby ejecting the ball.

[0119] A close analysis of the sequence will illustrate an interesting fact. It may be thought that the instant at which the target pivots is critical, but such is not believed to be the case. The reason is that the ejection of the ball can be divided into two events: (1) the bounce of the ball from the target, which occurs whether or not the target pivots, and (2) the “swat” issued by the pivoting target.

[0120] By conservation of energy principles, and the principle of superposition in linear systems, it can be shown that it does not matter whether the bounce and the swat are simultaneous, or whether the bounce occurs first. (The third situation, where the bounce occurs after the swat, is clearly impossible, because the bounce causes the swat.) The bounce-before-swat situation is somewhat more likely, due to the processing delay required for the signal issued by the sensor S to become transformed into a power signal reaching the solenoid 425.

[0121] FIG. 38 illustrates another embodiment. Launcher 100 launches ball B to a basket BAS. Basket BAS catches the ball, and then drops it, through a hole (not shown) onto a serpentine TRACK. The ball rolls down the track, as in FIG. 40, and then becomes airborne at the END, whereupon it jumps into net 420. The control circuitry described above senses the return of the ball, and causes the process to repeat.

[0122] FIGS. 39, 41, and 42 illustrate another embodiment. Two launchers 100 are shown. It is emphasized that the NET, as seen by each launcher, is on the right side of each launcher. However, because the launchers face each other, the cannon C of each faces the NET of the other.

[0123] With this arrangement, the launchers can juggle, if properly configured. For example, in FIG. 39, launcher 100X launches ball B. Just before the ball B reaches its intended NET, launcher 100Y launches ball BB. Then, launcher 100X launches a third ball B3, when the first ball B is within NET 1 and ball BB is still in flight. This process continues.

[0124] Figure 42 illustrates the juggling in greater detail. Frame “A” indicates the initial situation, and the box BB indicates symbols for three balls: a hollow ball, a solid ball, and a cross-ball (on the left). In frame B, the hollow ball is launched by the left-hand launcher 100K, and the cross-ball is loaded immediately into that launcher.

[0125] Then, in frame C, when the hollow ball reaches the 1 o'clock position, the right-hand launcher 100M launches the solid ball. In frame D, the hollow ball and the solid ball meet at the 2 o'clock position. Next, in frame E, the solid ball is caught by the right-hand launcher. In frame F, which is the mirror-image of frame C, when the solid ball reaches the 11 o'clock position, the left-hand launcher launches the cross-ball.

[0126] Then, in frame G, which is the mirror-image of frame D, the solid ball and the cross-ball meet at the 10 o'clock position. Finally, frame H is reached, which is the same as frame B as to position of balls, and the sequence repeats.

[0127] A single launcher can juggle by itself, as shown in FIG. 43. Details of those particular launchers have been omitted, for simplicity.

[0128] FIG. 44 illustrates one embodiment. A ball B is positioned at a launching station, adjacent solenoid SOL. Ball B closes a cat's whisker switch SW, which pulls line L1 HIGH.

[0129] A digital CONTROL, comprising a microprocessor, such as that sold under the name BASIC STAMP, by Parallax Computing, receives the HIGH signal on line L1. Logic executed by the CONTROL is illustrated in FIG. 45.

[0130] The logic idles in block 1600 until switch SW is closed. Then, in blocks 1605 and 1608, the RC time constants #1 and #2 in FIG. 44 are read by the CONTROL, using instructions programmed into the CONTROL. (Actually, the RC time constant, multiplied by a constant, is read.) Reading these RC time constants allows the user to provide analog inputs to the CONTROL, by adjusting resistors RA and RB.

[0131] These analog inputs indicate the amount of the delay, and the power with which the ball B should be shot. In block 1610 in FIG. 45, the delay is computed. For example, capacitor CA may be 0.1 microFarad, and resistor RA may be adjusted to 5,000 ohms. The CONTROL, in reading the RC time constant, may return a number of 300 as indicating the RC time constant. (Again, this number is not exactly the time constant, but a number related to the time constant.) Block 1610 may divide this number by 30, to produce a delay of 10 seconds.

[0132] In a similar manner, block 1625, using RC #2, computes a closure interval. This closure interval is the length of time, in milliseconds, for example, which RELAY in FIG. 44 is switched. Closure for a smaller time applies less of the energy of capacitor CC to solenoid SOL.

[0133] Then, in block 1620, the CONTROL waits for the delay computet and then, in block 1625, closes the relay for the computed closure interval. The CONTROL executes the closure by applying current to the COIL of the RELAY, through opto-isolator OPTO-ISO.

[0134] A diode D has charged capacitor CC, which is about 3,000 microFarads, to about 90 volts, from wall current indicated as 120 volts AC. Applying current to the COIL causes the REED to connect to terminal T2, thereby discharging the capacitor CC into the solenoid SOL.

[0135] In the apparatus of FIG. 44, the solenoid SOL was taken from a standard household door chime, and has a resistance of about 6 ohms. It launched the ball B for a horizontal distance of 15 to 20 feet.

[0136] A USER SWITCH can be provided in FIG. 44. The optional path indicated in FIG. 45 can then be taken, when a closure of the USER SWITCH is detected When the closure is detected, the ball B is shot, through the action of block 1635.

[0137] The USER SWITCH is indicated as blocks 1640 in FIG. 39, and allows the user to manually control the juggling of the balls. That is, without actually tossing and catching the balls, the user controls the juggling by controlling the timing of the respective launches.

[0138] A ball dispenser 1670 may be provided, which is controlled by a switch SW3. The user utilizes this dispenser to deposit the cross-ball indicated in frame B, FIG. 42, immediately after the hollow ball is launched. The incorporated patents describe ball dispensers. Also, a ball dispenser can be constructed by a tube and trap-door, which is actuated by a solenoid, through switch SW3.

[0139] The program contained within the CONTROL of FIG. 44 is generated within a microcomputer, such as computer 550 in FIG. 22, and then downloaded into the CONTROL by using line L5 in FIG. 22. The program is written using an interface available from Parallax Computing for the BASIC STAMP.

[0140] FIG. 46 is a front view of a basket suitable for use with several of the foregoing embodiments. FIG. 47 is a side view of the basket of FIG. 46. FIG. 48 is a plan view of a cardboard sheet (8.5 inches by 11 inches) which may be manipulated into the basket of FIG. 46.

[0141] The sheet of FIG. 48 is transformed into the basket of FIG. 46 by cutting along the solid lines, and lightly scoring along the dashed lines. The type of dashed line shown between sections I and H designates an interior corner (surface I is folded toward surface H), and the type of dashed line shown between sections H and G designates an exterior corner (surface H is folded away from surface G). For scaling purposes, these two lines should be two inches long, and the square around the circular hole should be one and five-eighths inches by one and five-eighths inches.

[0142] After making the cuts and score lines, fold along opposite edges of the square so that surface E faces toward surface F. Next, fold along other opposite edges of the square so that surfaces B face away from one another. Next, fold between sections B and A so that surfaces A face away from the circular hole and portions of sections A overlap one another. Next, staple sections A to one another to define a square opposite the circular hole. The resulting square and the flaps on sections B form a “base” to be disposed about the launcher.

[0143] After the “base” is complete, fold along lines between sections D and section E so that surfaces D face toward one another. Next, fold along lines between sections C and D so that surfaces C face toward surface E and portions of sections C overlap both one another and section F. Next, dispose tape about the resulting “box” or “tube”.

[0144] After the “box” is complete, and the outermost portions of the sections G have been removed, fold each surface K toward surface J and then back again. Next, fold each surface G away from its adjacent surface H; fold each surface H toward its adjacent surface I; and fold each surface I toward surface J. On one side at a time, insert section G inside the box so that the fold line between sections G and H coincides with the top of the box. Staple section G to section D and then repeat for other side. Finally, staple each section K to adjacent section H.

[0145] FIG. 49 shows a flow chart suitable for controlling the launching of balls between two juggling stations constructed according to the principles of the present invention.

Additional Considerations

[0146] A table tennis ball can be used as the projectile 8 in FIG. 1. Such balls weight about 0.1 ounce.

[0147] A “capture cross-section” of the basket 105 in FIG. 1 can be defined. This term refers to the area of the “inlet” or “mouth” of the basket, through which the ball B passes while being captured. In one embodiment, a capture cross section of 150 square inches is contemplated. In another, the capture cross section is 15 times the cross-sectional area of the ball.

[0148] The capture cross section can also be defined in terms of angles, analogous to the spherical coordinates used in trigonometry. These angles bracket the paths of the incoming ball. For example, if the basket can catch balls, from those travelling horizontally, to those travelling vertically, the angle would range from zero (horizontal) to ninety (vertical). A similar approach can be applied in the horizontal direction. For example, if the basket can catch balls incoming from the east, and the north, and all angles in-between, it would capture angles spanning 90 degrees.

[0149] In one embodiment, the basket is sturdy enough to capture a table tennis ball, but not to capture a golf ball. The mass of a golf ball is a defined quantity. One reason is to limit play to harmless balls.

[0150] In one embodiment, the basket is effective to capture balls which have been airborne for 15 feet, and cannot capture ground-borne balls, such as rolling balls.

[0151] Numerous substitutions and modifications can be undertaken without departing from the true spirit and scope of the invention. What is desired to be secured by Letters Patent is the invention as defined in the following claims.