Brief Description of the Drawing
The above and other features of my invention will now be described in detail with reference to the accompanying drawing in which:
FIG. 1 is an overall circuit diagram of a proximity sensor according to the invention, substantially as disclosed in my prior applications Ser. Nos. 80,016 and 290,866;
FIGS. 2, 3 and 4 are fragmentary circuit diagrams showing several modifications of a component of the system of FIG. 1;
FIG. 5 is a graph showing an input voltage developed across the proximity sensor of FIG. 1; and
FIG. 6 is a graph similar to FIG. 5 but relating to a modified system according to FIGS. 2, 3 or 4.
Specific Description
In FIG. 1 I have illustrated a sensing circuit 1, responsive to the proximity of a metal part (not shown), which is connected by a two-wire line 2, 3 to an alternating-current source 4 in series with a load 5 in the form of a current-responsive indicator. Member 5 may be a relay whose contacts can be reversed to signal the attainment of a predetermined spacing of an external metallic element from the sensor 1.
The contactless distance sensor 1 comprises an oscillator 6 which may be of the type described in my copending applications Ser. Nos. 290,866, 290,867 and 290,868 all filed 21 Sept. 1972, now U.S. Pat. Nos. 3,747,010, 3,747,011 and 3,747,012 respectively. In the illustrated embodiment, the oscillator 6 comprises an NPN transistor 6a whose collector circuit includes a parallel-resonant network 6b consisting of a capacitor 6b' and an inductor 6b". A feedback inductor 6c is connected between the base of the transistor 6a and a common terminal 6d of a pair of resistances 6e, 6f forming a voltage-divider network; the two coils 6b", 6c are inductively coupled as diagrammatically indicated in the drawing. Resistance 6e is bridged by a shunt capacitor 6g. A resistance 6h is connected between the emitter of transistor 6a and a negative bus bar 15 also tied to the resistance 6f. This oscillator generates an output of a frequency determined by the tuned or tank circuit 6b and at a level depending, in a manner known per se, on the damping induced by the proximity of metal parts to the oscillator (specifically to its tank circuit 6b) which lowers the Q of circuit 6b and therefore reduces the effective collector resistance of transistor 6a along with the amplification factor V so as to attenuate the oscillator output.
I also prefer to provide a snap-action or bistable amplification stage 9 triggerable by the output of the oscillator 6 when the loop gain KV of the amplifier 6a thereof makes the transition between values greater and less than unity. The trigger stage 9 comprises a transistor 9a of the NPN type whose base is tied to the collector of transistor 6a by a d.c.-blocking coupling capacitor 9b. The base of transistor 9a is biased positively by a transistor 9c connected as a diode to the negative bus bar 15 of the circuit. The output of transistor 9a, whose amplitude decreases upon the approach of a metallic element as described above, is applied by an emitter impedance, in the form of an R/C network consisting of resistors 9e and 9f bridged by a storage capacitor 9g, to the base of an NPN switching transistor 23 included in a component 10 which serves in part as a voltage-generating network and also includes a binary electronic switch shown as a thyristor 12. The voltage-generating network of this component includes a storage capacitor 18 connected between the negative bus bar 15 and a positive lead 13 supplying the transistors 6a and 9a of the preceding stages; capacitor 18 is shunted by a Zener diode 16 in series with a rectifying diode 19, the latter lying in a charging circuit for the capacitor which includes a high-ohmic ancillary resistor 17 in parallel with the anode/cathode path and associated leads 14 and 26 of thyristor 12. The gate of the thyristor is connected through a lead 20 to the collector of transistor 23 and thence via the collector/emitter resistance 23 thereof to negative bus bar 15. The collector of transistor 23 is further connected through a gate resistor 21, which in this instance is also of the high-ohmic type, to a positive bus bar 22 directly tied to the anode lead 26 of thyristor 12; the voltage-generating network consisting of ancillary resistor 17 and the constant-voltage impedance 16, 18, 19 is bridged across the two bus bars 13 and 15. These bus bars, in turn, are energized with pulsating direct current from a rectifier bridge 25 connected across the wires 2, 3 downstream of load 5. The alternating voltage developed across the input diagonal of bridge 25 has been indicated at V.sub.i.
Switching transistor 23 conducts as long as a sufficiently positive charge is accumulated on capacitor 9g, i.e. as long as transistor 9a of trigger stage 9 is turned on by a biasing potential on its base corresponding to a relatively high amplitude of the oscillations generated by transistor 6a in detector unit 6. Upon a substantial reduction in the amplitude level, trigger transistor 9a and switching transistor 23 are cut off whereby the positive-going potential of lead 20 energizes the gate of thyristor 12 sufficiently to drive a firing current throut the gate/cathode path 27 of the thyristor with resulting short-circuiting of the high-ohmic ancillary resistor 17. That operation sharply lowers the potential of bus bar 22 from the value it had attained just prior to the firing of the thyristor; this has been illustrated in FIG. 5 where the square wave V.sub.i represents the a-c counterpart of the raw-rectified pulsating voltage developed beyond bridge 25 across bus bars 15 and 22. It will be seen that each zero crossing of this input voltage V.sub.i is followed by a sharp spike S which appears as a positive pulse on bus bar 22 and is required to overcome the gate/cathode resistance 27 of the thyristor before the latter can conduct to reduce the input voltage to a predetermined amplitude whose absolute value is only slightly higher than the breakdown level of Zener diode 16. (It will be understood that the flanks of the square-wave pulses are not perfectly vertical, as diagrammatically illustrated, so that the pulsations through thyristor 12 are separated by brief periods of nonconductivity.)
Independently of the state of conductivity of thyristor 12, Zener diode 16 is traversed by a current drawn either through resistor 17 or through the parallel combination of that resistor and the main current path of the thyristor. Capacitor 18 thus remains charged, via diode 19, to substantially the Zener level to keep the power-input lead 13 of oscillator 6 and trigger 9 energized. Detector unit 6 is therefore continuously maintained operative, even if the oscillations cease completely, through the steady energization of lead 13.
At the beginning of operations, i.e. before power supply 4, 25 is connected to the bus bars 15 and 22 of proximity sensor 1, capacitor 18 is of course discharged so that the cathode lead 14 of thyristor 12 is at nearly the potential of negative bus bar 15 when the circuit is closed. It may therefore happen that the gate lead 20 is initially driven more positive than the cathode lead 14 to such an extent that, regardless of the absence of a metallic element whose proximity is to be detected, the thyristor fires and actuates the relay 5. This drawback can be avoided by connecting a large capacitor across the main path 23 of transistor 24 to hold down the initial gate voltage, yet a simpler and more dependable solution is provided by the improved components 10a, 10b and 10c shown in FIGS. 2 - 4 which also eliminate the spikes S of FIG. 5.
In FIGS. 2 - 4 I have employed, as far as applicable, the same numerals as in FIG. 1 to designate analogous elements. It is to be understood, however, that in the modifications of these Figures the switching transistor 24 must saturate (or at least conduct) when the thyristor 12 is to be fired, i.e. when the load 5 is to be energized, whereas the transistor is cut off in the alternate condition in which the input resistance of sensing circuit 1 is to be high. Thus, contrary to the arrangement of FIG. 1, these modified systems will indicate the approach of a metallic element by the release of relay 5 unless trigger stage 9 is altered (e.g. by the introduction of an inverter) to drive the base of transistor 24 negative when the oscillation amplitudes in the output of detector 6 are high.
In FIG. 2 the cathode lead 14 of thyristor 12, forming part of the component 10a replacing the component 10 of FIG. 1, is connected to a junction J representing a point on a voltage divider constituted by the ancillary resistor 17 and a relatively low-ohmic cathode resistor 28, this voltage divider being bridged across bus bars 15 and 22 in series with transistor 24 whose collector is joined to its resistor 28. Gate lead 20 is connected through another high-ohmic resistor 21 to the power-input lead 13 for components 6 and 9 (FIG. 1) originating, as before, at a common terminal of capacitor 18 and its charging diode 19. The common terminal of this charging diode with Zener diode 16 is connected to junction point J through a further rectifying diode 29 in cascade with diode 19, the cascaded diodes being so poled as to oppose the flow of discharge current from capacitor 18 to junction J.
In the operation of the system of FIG. 2, thyristor 12 remains nonconductive as long as transistor 24 is cut off. Junction J, and therefore the cathode lead 14, is then maintained at a positive potential which exceeds the Zener potential, present at the point Z to which the three diodes 16, 19 and 29 are joined, by the small voltage drop developed across the forward resistance of diode 29; gate lead 20, on the other hand, is essentially at the charging potential of capacitor 18 which falls short of the Zener level by the voltage drop across the forward resistance of diode 19. Under these conditions, therefore, thyristor 12 acts as an open switch and the input resistance of circuit 1 remains high so that voltage V.sub.i varies sinusoidally with that of source 4 and the current flow through relay 50 is insufficient to actuate same.
As soon as transistor 24 saturates, the cathode voltage of thyristor 12 drops to a low level only slightly above that of bus bar 15. Since the gate voltage of the thyristor is not affected by the conduction of transistor 24, gate current now flows through the path 27 so that the thyristor fires and draws current through its anode 26 from bus bar 22. The firing point occurs early in each half-cycle, as soon as bus bar 22 has gone sufficiently positive to sustain conduction. The potential of this bus bar therefore never rises above that established by the Zener potential and the voltage drops across diode 29 and thyristor 12 so that voltage V.sub.i is a square wave as illustrated in FIG. 6, lacking the spikes of FIG. 5.
It will also be noted that upon the initial connection of component 10a to its power supply, i.e. with capacitor 18 discharged, gate lead 20 is at low voltage so that thyristor 12 will not fire, as it might in the system of FIG. 1 with transistor 24 cut off. Also, whereas in that system the gate resistor 21 is connected in series with switching transistor 23 between bus bars 15 and 22 so as to have virtually the entire supply voltage (e.g. of 220 V) developed thereacross, this resistor is traversed in FIG. 2 by only a small current flow and is therefore readily realizable by monolithically integrated circuitry.
If the series diode 29 were omitted, the system would function in essentially the same manner but only over a restricted range of supply voltages as compared with that of FIG. 1. With a low-voltage source the reverse flow through diode 19 alone could drain the capacitor 18 sufficiently to prevent an adequate energization of the components fed via lead 13. By the provision of diode 29, in a system designed for a supply voltage of 220 V, the same can be made operative over a range down to about 40 V, similar to that realizable with the system of FIG. 1.
The system of FIG. 2 has a minor drawback in that the load current passing through thyristor 12 and Zener diode 16 must also traverse the diode 29, with added energy dissipation and the need for carrying off the developing heat. The presence of this diode also militates against a realization of component 10a by monolithically integrated circuitry.
In FIG. 3 I have shown a further modified component 10b in which the Zener diode 16 is connected directly across the storage capacitor 18 and to the conductor 13, with diode 19 relocated into the connection between this conductor and the gate lead 20 of thyristor 12; diode 19 is shunted by the gate resistor 21. A second Zener diode 30 lies directly in series with the cathode lead 14 of thyristor 12 and in parallel with the combination of switching transistor 24 and low-ohmic resistor 28, the latter being again connected to the collector of this transistor. In this embodiment the gate voltage of the thyristor exceeds by only the forward drop of diode 19 the voltage of conductor 13 substantially corresponding to the Zener potential at the junction Z' of elements 16, 17, 18, 19 and 21; the cathode voltage, on the other hand, equals the Zener potential of diode 30 as developed at the junction Z" of that diode with cathode resistor 28. Thus, in order to prevent any untimely firing of the thyristor 12 with switching transistor 24 cut off, the Zener potential at Z" must be at least equal to that at Z'. Upon the saturation of transistor 24, a gate current limited by resistor 21 flows through the path 27 with resulting closure of the electronic switch represented by the thyristor.
In FIG. 4 I have shown still another modified component 10c in which the relationship of the Zener levels of diodes 16 and 30 is made noncritical by the provision of an inverter stage 31 connected across transistor 24, the inverter being energized from bus bar 22 and having an output lead 33 forming part of an ancillary energizing circuit from the gate of thyristor 12. This ancillary circuit includes an emitter-follower NPN transistor 32 with a collector tied to lead 13, a base joined to lead 33 (and returned to the collector via gate resistor 21) and an emitter connected by way of a low-ohmic resistor 34 to gate lead 20. The system is otherwise identical with that of FIG. 3.
As long as switching transistor 24 is cut off, output lead 33 of inverter 31 carries a low voltage close to that of bus bar 15 which also holds down the gate voltage of the thyristor 12, the latter being thus prevented from firing. If, however, transistor 24 becomes conductive, inverter 31 virtually insulates the lead 33 from bus bar 15 so that the more positive potential of lead 13 is now applied to the base of transistor 32 while the voltage on cathode lead 14 is sharply reduced. The firing current through the gate/cathode path 27 of the thyristor is drawn through transistor 32 and resistor 34 without loading the resistor 21 which can therefore have a relatively high value. In this way, the current flow via resistors 17 and 21 over lead 33 into inverter 31 during cut-off of transistor 24 is minimized.
I shall now give, by way of example, representative values for the principal impedance elements of a system according to my invention designed for a supply voltage of 220 V: ancillary resistor 17 150 k.OMEGA. gate resistor 21 in FIG. 1 220 k.OMEGA. gate resistor 21 in FIGS. 2 and 3 5.6 k.OMEGA. gate resistor 21 in FIG. 4 68 k.OMEGA. cathode resistor 1 k.OMEGA. storage capacitor 18 4.7 .mu.F Zener potential of diode 16 in FIGS. 1 and 2 5.6 V Zener potential of diode 16 in FIGS. 3 and 4 4.7 V Zener potential of diode 30 in FIG. 3 5.6 V Zener potential of diode 30 in FIG. 4 4.7 V
It is to be understood that compatible features of different embodiments may be combined with one another; thus, for example, the inverter 31 of FIG. 4 (with or without its emitter-follower transistor 32) could also be used in an ancillary energizing circuit for the gate of thyristor 12 in the system of FIG. 2.