Compensating Bridge Circuit

The invention relates to compensating bridge circuits, particularly to bridge circuits used for measuring mechanical quantities.

When mechanical quantities are measured by electrical means, the measured quantity is converted by a transducer to a corresponding electrical quantity. Transducers are functionally usually based on the change in the value of a resistance, an inductance, or a capacitance, resulting from a displacement caused by the measured quantity. Since such displacements are mostly very small, the resultant change in magnitude of the electrical output quantity is generally also small. This applies for instance to the measurement of mechanical quantities by strain gauges in which the changes in resistance that require detection and measurement are in the order of 10.sup..sup.-4 to about 10.sup..sup.-8. Apart from a few special cases, the required measurement is mostly performed by causing the change in the electrical quantity to affect the balance of a Wheatstone bridge and by either indicating the output voltage of the unbalanced bridge by a pointer on a scale (deflection method) or rebalancing the bridge by adjustment of a resistor (null method). Whereas the deflection method is preferred for solving the majority of measuring problems since the apparatus requirement is low, the null method is better for effecting very precise measurements because it eliminates the effect of fluctuations in the supply voltage and generally contains fewer parameters which affect the degree of precision.

Many different kinds of compensating circuits for measuring physical quantities are known. One such circuit is shown in FIG. 1. A voltage V.sub.b obtained from an input bridge A comprising strain gauges S is compensated by the output voltage V.sub.bb of a second balancing bridge B containing a measuring potentiometer Mp. In this compensating circuit it is a drawback that even the slightest drift in the value of the resistors R in the balancing bridge may falsely indicate that the input bridge is out of balance, thus adversely affecting the precision of the measurement in a manner that is particularly objectionable in cases in which the magnitude of a measured quantity is to be observed for a longer period in relation to a previously fixed reference value. Another drawback arises when the measured quantity is required not in terms of the degree of imbalance of the bridge but as an absolute value. In such a case, the measuring potentiometer would have to be provided with a scale of its own depending upon the type of the input transducers and the measuring range. Naturally this would cause major complications.

Another known circuit arrangement is that shown in FIG. 2. Here the balancing bridge is fed with a voltage reduced by a potential divider R.sub.d1, R.sub.d2, R.sub.d3. This leads to better null stability and at the same time the scale division of the measuring potentiometer Mp can be calibrated by varying a resistor R.sub.d1 in the potential divider. The somewhat large number of precision components in this arrangement is an undesirable feature. This applies more specifically when balancing is to be effected digitally with the aid of a stepped resistor network in which the resistors are attributed values according to a predetermined scheme. It will be appreciated that a linear potentiometer circuit contains twice as many components as a variable liner resistor.

Yet another compensating circuit is diagrammatically shown in FIG. 3. Here the potential divider has been replaced by a variable resistor R.sub.k. The circuit relies on a high ohmic differential amplifier with the gain g amplifies the input voltage V.sub.b. When the voltage g .sup.. V.sub.b will be compensated, the voltage across the null type indicator will be zero and the current i.sub.m flowing through the measuring resistor R.sub.m will be ##EQU1##

On the other hand, the relationship ##EQU2## holds for the compensating branch to which the supply voltage U.sub.S is applied.

It follows that ##EQU3##

The reading of the measured quantity therefore depends linearly upon the conductance G.sub.k of the balancing resistor R.sub.k. The measuring range can be easily varied by R.sub.m. On the other hand, an unfavorable factor is the incorporation of the amplifier in the measuring circuit. Its properties substantially affect the precision that can be achieved. For instance, the input resistance must be sufficiently high to keep the load on the transducer bridge A negligibly low. Moreover, the output resistance and the amplification must not be liable to drift. The only remedy is a high degree of feedback in the amplifier. Under the most unfavorable circumstances, the output voltage will then be determined only by the quality of the feedback network. This therefore also determines the precision that can be achieved, besides the resistors R.sub.m and R.sub.k. Another drawback, particularly when the signals are very weak is the background noise of the measuring amplifier which is superimposed on the measuring signal.

According to the invention, these drawbacks are avoided by compensating the output voltage V.sub.b of the Wheatstone bridge A including the transducer and being unbalanced by the measured input quantity x, by a second voltage V.sub.d obtained from an ohmic potential divider comprising a balancing resistor R(y) having a conductance y.DELTA.G.sub.y or a resistance y.DELTA.R.sub.y, wherein the supply voltage of the potential divider is composed of the input voltage V.sub.o feeding the bridge and the divisional voltage V.sub.d in such a way that when balance is obtained there is a linear relationship between the input quantity x and the balancing quantity y, i.e. so that y = const. .sup.. x.

Other objects of this invention will appear in the following description and appended claims, reference being made to the accompanying drawings forming a part of the specification wherein like reference characters designate corresponding parts in the several views.

FIGS. 1-3 are schematic circuit diagrams showing prior art circuitry related to the invention;

FIG. 4 is a simplified circuit diagram showing a first embodiment of the circuitry in accordance with this invention;

FIG. 5 is a schematic circuit diagram of an embodiment of circuitry used for an A.C. power supply; and

FIG. 6 is a further circuit diagram showing an embodiment of the circuitry of this invention useful with a D.C. power supply.

In order that the principle which underlies the invention may be better understood reference will now be made to FIG. 4. When the system is balanced, the output voltage V.sub.b of the input bridge which may first be assumed to be linearly proportional to the input quantity x will be exactly compensated by a voltage U.sub.T obtained from a potential divider, and determined by the conductances G.sub.o and y.DELTA.G.sub.y.

Introducing a transducer coefficient a the following holds: ##STR1##

The current i which now flows through the divider is driven by the sum of the supply voltage V.sub.s which may differ from the bridge supply voltage V.sub.o by any factor, and the divisional voltage V.sub.d amplified by a gain factor g. In other words ##EQU4##

Substituting (1) in (2) results in: ##EQU5##

It is apparent that for g = 1 equation (3) becomes ##EQU6##

This means that irrespectively of the constant a the input quantity x will be directly indicated by the balancing quantity y if ##EQU7##

These conditions can be observed even if the balancing resistor has a fixed graduation .DELTA.G.sub.y, because the choice of V.sub.s and G.sub.o still offers two degrees of freedom. Special advantages are thus obtained in connection with the rated values of the components because, given a suitable choice of V.sub.s, the value of G.sub.o can be chosen from within a range that can be implemented with minimum tolerations and optimum stability. The precision of the circuit arrangement -- apart from the sensitivity of the null indicating instrument -- is in practice determined exclusively by the precision of the values of G.sub.o and .DELTA.G.sub.y because the bridge output voltage V.sub.b and the partial voltage V.sub.d will generally be much lower than the supply voltage V.sub.s.

The properties of the amplifier, such as fluctuations in gain and background noise will affect the result of the measurement to a correspondingly minor degree.

In some instances the output voltage of the input bridge cannot be regarded as being linearly proportional to the quantity x. This will be the case for example if the input bridge contains only one resistor which changes linearly with the quantity x (in a so-called quarter bridge circuit). In such a case the output voltage of the bridge, assuming a supply voltage V.sub.o, will be ##EQU8##

The bridge transmission coefficient being assumed to be 0.5 and .DELTA.R/R.sub.o representing the relative change in resistance. When the system is in balance, referring to equation (3), it therefore follows that ##EQU9##

After rearranging the terms ##EQU10##

The solution for x yields ##EQU11##

It will be seen that a linear relationship between the value x and the balancing quantity y will exist if ##EQU12##

It therefore transpires that irrespectively of the voltage V.sub.S feeding the potentiometer, the amplification can be so adjusted that there is in fact a linear relationship between the measured quantity and the indicated quantity. Conversely, some non-linearity of the transducer can be compensated. This is an arrangement which may be desirable for instance in a temperature measuring bridge which contains a platinum resistance type temperature feeler having a non-linear temperature-resistance function.

An embodiment of the circuit corresponding to the invention as described with respect to FIG. 4, is shown in FIG. 5. The circuit as shown contains the receiving bridge which is fed with a voltage of V.sub.o and produces a condition dependent output voltage V.sub.b. Furthermore and exactly as in FIG. 4, the circuit contains a zero detector and a potential divider with the conductive values G.sub.o and y.sup.. .DELTA.G.sub.y as well as the amplifier A with the amplification +g.

The voltage source is shown in FIG. 5 and is realized with a transformer T which is connected with the supply voltage source V.sub.o of the receiver bridge. With the help of the reverse polarity switch SW, the voltage supply for the potential divider can be reversed in its polarity. As a result of that, both positive as well as negative output voltages of the receiver bridge can be made over into zero. Thus, respective positive and negative measuring values can be determined. As it is evident from the drawings the circuit shown in FIG. 5 is intended only for AC current use because of the transformer used.

Another embodiment is shown in FIG. 6 wherein the voltage supply for the potential divider required corresponds to the explanations to FIG. 4. In this embodiment, an operational amplifier is used as shown. The circuit contains the receiver bridge fed with the voltage V.sub.o, the zero detector, the potential divider with the conductances G.sub.o and y.sup. . .DELTA.G.sub.y and the change over switch SW for changing over the meter. The voltage source for the supply of the potential divider is realized with an operational amplifier OP wired in a known manner with the resistors R.sub.1, R.sub.2, R.sub.3, R.sub.4. The circuit sketched in FIG. 6, of which still other variations are possible, clearly may be operated with AC or DC voltage as shown.

The invention may be embodied in other specific forms without departing from the spirit or essential characteristics thereof. The embodiments are therefore to be considered in all respects as illustrative and not restrictive.