Automatic Polarity Switching Output Circuit

This invention relates to a method for the automatic identification or recognition of the polarity of the electrical connection of a load and the automatic association of the appropriate current source or drain with the load, as well as to an output circuit in the nature of an interface between two circuits for carrying out the method.

Power-switching output circuits are generally designed with transistors. In the context of the present invention, output circuits are involved which, in actual fact, switch, i.e., carry a switching current. As a function of the use of the transistors, they can be the npn or pnp type, which have opposite current directions, i.e., switch to a positive or a negative potential. The circuits carrying the switching current are designed accordingly and therefore this also applies to the circuits switched with the switching current.

In order to have greater freedom with increased circuit complexity, use is made of output circuits which an carrying a switching current to both the negative and positive potential, an example being a push-pull circuit. Using this circuit as a basis, there are user-related modifications with corresponding advantages and disadvantages.

Reference is made to position sensors as a typical user problem. These can be position sensors which operate inductively, ultrasonically, optically or capacitively to sense the position for a movable article. It is clear that a multiplicity of circuits are possible, all of which have an output circuit for "switching" any user. Such position circuits either have an output switching to positive potential, or an output switching to negative potential, with optionally the possibility for both, although it is not possible with prior art circuits to simultaneously use the latter possibility, e.g., the push-pull output circuit. The switching function can be such that through the physical sensor function closing of the switch is brought about (normally open contact, NOC) or withdrawn (normally closed contact, NCC). On connection, the use must either adopt a positive normally open contact and therefore a negative normally closed contact, or a negative normally open contact and therefore a positive normally closed contact. Further developed variants in the direction of a universal output circuit must always be rewired in order to obtain the desired switching polarity.

An object of the present invention is therefore to provide a method for the automatic identification of the polarity of the electrical connection of a load and the automatic association of the corresponding current source or, conversely, the corresponding drain, with the load, together with an output circuit in the sense of an interface between two circuits for carrying out the method and in particular an output circuit for switching networks, which do not have these disadvantages and instead automatically set the desired switching polarity without any wiring change or line interchange. In addition, this output circuit should have minimum circuit complexity, so that a broad range of uses is not hindered by constructional and economical considerations.

The invention is described in greater detail hereinafter with reference to preferred embodiments shown in the attached drawings, wherein:

FIGS. 1A, 1B and 1C are schematic circuit diagrams of three forms of circuit outputs according to the prior art, 1A to a positive potential, 1B to a negative potential and 1C in the form of a push-pull circuit against one or other potential;

FIG. 2 is a schematic diagram of a basic circuit of an embodiment of the invention with current sensing;

FIG. 3 is a schematic diagram of a basic circuit of an embodiment of the invention with voltage sensing;

FIG. 4 is a schematic diagram of a circuit usable as a sensor in the embodiment of FIG. 2; and

FIG. 5 is a truth table illustrating the states of a bistable device in the embodiment of FIG. 2.

An essential part of the inventive output circuit is a relatively uncomplicated measurement circuit. The invention is based on the idea of generating an indicator current in the output circuit. This indicator current can be made to flow across a blocking element in much the same way as through a valve, through the external wiring, i.e., the user wiring, or in the case of reversed polarity of the valve it flows minimally or does not flow at all. The measurement circuit can consequently detect whether the load at the terminals (the user circuit) is connected to a negative or positive potential.

In place of the sensing of an indicator current, it is possible to connect a balanced voltage to the user circuit. As a function of the polarity, the user circuit voltage unbalances the symmetry toward one or other polarity and causes the circuit to carry out a counter-reaction eliminating the asymmetry again, which indicates whether the load is connected to a negative or a positive potential. Thus, in the case of either current or voltage sensing, the output circuit connects to the load an npn or pnp driver stage, both already being provided in the output circuit. Thus, there is no need for polarity reversal, rewiring or the like, which naturally helps to prevent incorrect connections during assembly. Thus, e.g., a proximity switch with the output circuit according to the invention has connections, which appear to be two "neutral" connections, which are connected to the load and everything else is carried out by the output circuit. The other important effect is the greatly reduced inventory since it is no longer necessary to store different output stage types (pnp or npn, normally open or closed contacts).

It is clear that the transistor outputs according to the prior art shown in FIGS. 1A, 1B and 1C do not have these advantages, because they operate in either the pnp or npn mode and, in response to the corresponding signal at the base, connect through in the only possible current direction. In the case of the push-pull circuit shown at FIG. 1C the situation is the same, because it either operates in the npn or in the pnp mode, as a function of the terminals between which the load is switched.

FIGS. 2 and 3 show the completely automatic switching of the output stage to one or other driver type.

The basic principle of the output circuit of the invention is as follows.

The basic principle is that a measurement circuit in the output circuit of the sensor measures, at the user circuit or load connected to its terminals, the voltage polarity to which it is connected. The detection of one or other of the two states establishes the state of a storing element M and switches the correct driver to the output terminals. This state is maintained for as long as there is no change to the polarity of the output terminals. However, if there is an interchange of terminals, e.g., due to the replacement of a defective device, and consequently there is no specification as to how the connection is to take place, then the measurement circuit correctly restores the stored state, so that the correct driver is associated with the user circuit.

Fundamentally the measurement circuit can comprise either of two indicators, either for current sensing or voltage sensing, an embodiment for each of the two possibilities being shown hereinafter.

A current I.sub.c is provided from a current source I to a diode D1. If a load is connected between a common point P at the cathode of the diode and a lower potential, as is the case with R.sub.L, this establishes a complete circuit so that current I.sub.c can flow through the diode and load R.sub.L. This current flow is detected and amplified by means of a current reflector S the input of which is connected to the anode of diode D1 and the output of which is connected to the input of a transimpedance amplifier A, which supplies at its output a voltage proportional to the detected current. This voltage is evaluated and stored in a following circuit M for activating one or other driver T1 or T2. As an example, circuit M can include a flip-flop circuit to act as a memory of the state last detected by the current sensor. An output of the flip-flop is connected to the bases of pnp and npn transistors T1 and T2 the emitters of which are connected to the positive and negative lines, respectively, and the collectors of which are connected to a common point P at the cathode of the diode between the possible load locations.

However, if the load is connected to a higher potential, i.e., if load R.sub.H is connected in the circuit rather than R.sub.L, then little or no current can flow through the diode. This is naturally noticed by the current reflector S which indicates to the series-connected amplifier A "no current", producing at the amplifier output a corresponding voltage signal.

Thus, at the output of amplifier A it is possible to read off the two states indicating to which potential, i.e., high or low, the attached load is connected. The detected one of the two measured states can then be used for the particular switching state to be retained by a storing circuit and on the basis thereof the appropriate driver stage is switched on.

The current reflector S, shown as a block in FIG. 2, can be replaced by any of several circuits such as those known as Wilson current mirror circuits, one of which is shown in FIG. 4. This specific example is a Series TL011 circuit made by Texas Instruments Incorporated.

The embodiment shown in FIG. 3 uses "voltage information" in place of "current information" and in that circuit a voltage caused in the load by the indicator current is sensed. The supply voltage V.sub.cc is symmetrically divided by a voltage divider including impedances R1 and R2 and the two half-V.sub.cc voltages are compared with a reference potential. When an external load R.sub.H or R.sub.L is connected between the common point P and the positive or negative lead, there is an asymmetry toward one or other polarity, which leads to compensation of the circuit, i.e., the driving of one of the two transistors T1 or T2 of a push-pull circuit. This current signal is used for setting the storage element, e.g., a flip-flop, as described in connection with FIG. 2.

Automatic load detection preferably takes place in a starting phase during the time of a starting pulse suppression of the proximity sensor. Starting pulse suppression lasts 20 to 40 ms. This time is long enough for measuring and positioning the storage element. By the time the suppression has ended, then the interface and therefore the sensor connected thereto has the correct polarity relative to the user circuit, i.e., the load. Thus, during starting, the time interval is used up entirely at the start, so as to be able to determine the load with the aid of the indicator current. The result, i.e., the state, is stored in a storage circuit such as a flip-flop and the current direction released by the load. This is either in the direction of a positive source through an npn driver transistor, or in the direction of a drain through the pnp driver transistor, as a function of the flip-flop output setting. This ensures that the two drivers are not simultaneously switched on.

The measurement circuit in the output circuit of the sensor measures, with the user circuit applied to its terminals, to which voltage polarity it is connected. The detection of one of the two states is used for fixing that state in a storing element and for connecting the correct driver to the output terminals. This state is maintained for as long as there is no change in the polarity of the output terminals. If e.g., through the replacement of a defective relay the terminals are interchanged, then the measurement circuit directly restores in correct manner the stored state, so that the correct driver is always associated with the user circuit. However, if e.g., through the replacement of a control card or system, positive instead of negative switching sensors are now required, unlike hitherto, there is no need to replace the equipment. The equipment is instead initialized by the invention when the voltage is switched on, setting the state corresponding to the new circumstances.