Background Art Ultrasonic devices of the type mentioned by way of introduction are used in the art in various applications.

They may be used, for example, in thickness-gauging or in testing of materials such as metal, in which cases the transducer is applied to one of the material sur- faces, whereupon a short ultrasonic pulse is transmitted. The ultrasonic signal will be reflected in the form of an echo in the boundary between the material and the sur- roundings or in the place of a defect in the material.

By detecting the echo it is possible to discover a defect in the material that is not visible to the eye, and by measuring the time elapsed between transmission of the ultrasonic pulse and reception of the reflected echo it is also possible to calculate the distance to the defect and/or the opposite side of the material.

Another field of application for ultrasonic devices of this type is flow measurements in a flowing gas or liquid. In this case, use is made of at least two trans- ducers oriented towards each other and mounted along an

axis that at least partly extends in the direction of flow of the fluid. By alternately transmitting the ultra- sonic pulses from one transducer and then from the other, measuring the time required for the sound to travel to the opposite transducer and comparing the time difference in the two directions, it is possible to obtain a very accurate measure of the flow rate of the fluid since the time difference is proportional to the flow rate of the fluid. One embodiment of such a flow measurement device is shown, for example, in SE 8803925 (US 5,214,966).

In the two systems described, a transducer is thus operated both in the transmitting mode and in the re- ceiving mode, and to achieve optimal operation both the transducer and the transmitting stage and the transducer and the receiving stage must be electrically matched as far as possible. This is particularly important in the receiving mode, and an incorrect matching in this mode will reduce the sensitivity and may cause interfering signals to appear resulting in reduced accuracy of measurement or erroneous measurement results. Such a situation may arise if the transmitting stage has not been shielded or separated from the transducer in the receiving mode. In this case, the transmitting stage wi t as an electric load on the transducer, attenu- ating or completely extinguishing the received and often weak signals. One obvious alternative embodiment would be to use separate transducers for transmission and reception. This would make it easy to achieve complete separation of the transmitting and receiving stages, but it would also be a considerably more expensive solution.

It is therefore desirable to provide simple and efficient shielding or separation of the transducer from the transmitting stage in the receiving mode.

Various solutions have already been presented, for instance in the book Ultrasonic Density Sensor for Liquids (ISBN 3-8265-4614-8) by Alf Putter, Shaker Verlag GmbH, Jan. 1999, in which, on pages 90-91 under

the heading"5. 2 Transmit/receive switch", three dif- ferent embodiments of a separating circuit for an ultra- sonic device are described. According to the embodiments described, separation can be obtained by using a) an active switch, such as a transistor, b) two counter- acting diodes connected between the transmitting stage and the transducer, and c) a resistor connected between the transmitting stage and the transducer. However, each of these alternative embodiments have drawbacks.

For instance, the embodiment according to alterna- tive a) is difficult, or at least expensive, to implement in systems operated at high voltages, as is the case, for example, when using so-called micromechanical trans- ducers. In addition, a switch will constitute both a re- sistive load and a capacitive load on the transducer in the receiving mode, which means that the load will vary with varying temperatures and frequencies as will the accuracy of measurement.

In the embodiment according to alternative b) ex- tremely poor separation of the receiving and transmitting stages is obtained because the diodes are oriented in such manner that the forward direction of one is opposite to the forward direction of the other. In addition, this <BR> <BR> <BR> <BR> embodiment, like in alternative a) implies >1 reslstivxy as well as a capacitive load on the transducer, which means that the load varies depending on temperature and frequency.

The embodiment according to c) has essentially the same drawbacks as b), i. e. poor separation, since the resistor is equally conductive in both directions, as well as a resistive load from the resistor and a capaci- tive load from the switch in the series pulse generator, which makes the separating circuit temperature-and frequency-dependent.

Previously, a piezoelectric crystal has been com- monly used as transducer. These crystals are not very sensitive to loads in the receiving mode; a certain load

may even be advantageous in many cases in order to attenuate self-oscillations and suppress radial oscil- lations.

More recently, so-called micromechanical transducers have come into use, in which the active oscillating sur- face is composed of a large number of tiny oscillating elements. Using transducers of this type in an ultrasonic device affords several advantages. For example, the os- cillating mass is very small, which means that the oscil- lation is attenuated very quickly following an ultrasonic pulse, which in turn results in a more distinct ultra- sonic pulse with less interfering after-oscillations.

One problem, however, is that they require a high driving voltage during transmission, normally in the range of 60- 200 V, whereas the voltages generated during reception are small, usually only a few millivolts (mV). Thus, the need for separating the transducer from the transmitting stage in the receiving mode is even greater when using micromechanical transducers of this type. It has been found, however, that separating circuits according to prior art do not provide sufficient separating effect to allow the micromechanical transducers to function in a satisfactory manner or to function at all as transducers in ultrasonic devices requiring a high degree of accuracvy of measurement.

Brief Description of the Invention The invention aims at obviating problems and draw- backs associated with prior art and providing an ultra- sonic device which, by using a simple and inexpensive shielding or separating circuit, is capable of effec- tively separating the transmitting and receiving stages from each other in at least the receiving mode. At least these objects. are achieved by means of an ultrasonic device according to claim 1.

The invention is thus based on the understanding that separation, and thus sensitivity and accuracy of measurement, can be significantly improved for all types

of transducers but particularly for micromechanical transducers by the receiving stage and the transmitting stage being interconnected via a diode, the forward di- rection of which is oriented toward the transmitting stage. It will be appreciated, however, that instead of using only one diode it would indeed be feasible to use two or more diodes connected in series or in parallel, the respective forward directions of the diodes being oriented from the transducer towards the transmitting stage in such manner that, in the receiving mode, they block current in the direction from the transmitting stage towards the receiving stage whereas, in the transmitting mode, they conduct current in the direction from the receiving stage towards the transmitting stage.

In an alternative embodiment, the diode or diodes com- prise a so-called PIN diode, which is characterised by its capability to conduct current in the backward direc- tion for a short period of time following forward con- duction. According to the inventive idea, the ultrasonic device is not considered to be in a receiving mode during this time, when the PIN diode conducts current in the reverse direction; instead this is considered to be a transition mode between the transmitting mode and the rece-iv-i-ng mode-which-f-acLl-itates-rap-id charging of the transducer following a transmission pulse. Such an em- bodiment of the ultrasonic device is more expensive than using a regular diode, but is advantageous in cases where a high degree of accuracy of measurement is required.

Preferably, the separating circuit further comprises two resistors, one connecting the receiving side of the diode, i. e. the anode, and the other its transmitting side, i. e. the cathode, to the positive side of a direct- current source. In a preferred embodiment, the resistor connecting the receiving side of the diode is larger than the one connecting the transmitting side of the diode, suitably at least 1.5 times as large, preferably at least 3 times as large and, most preferred, at least 6 times as

large. This means that, in the receiving mode, the volt- age on the receiving side of the diode will be lower than on the transmitting side and the diode will thus be reverse-biased, i. e. no current flows through it. In the preferred embodiment, the transducer is connected between the receiving side of the diode and earth or the negative side of a direct-current source.

In the preferred embodiment, the ultrasonic pulses are transmitted by means of negative electric transmis- sion pulses, i. e. in its initial mode or receiving mode the transducer has a positive voltage which, in the transmitting mode, is reduced to a lower positive voltage or even a negative voltage.

In a preferred embodiment of the invention described below, a zener diode is connected in the opposite direc- tion relative to the diode of the separating circuit.

However, this zener diode cannot be considered to form part of the separating circuit as such but is used only to limit the voltage drop in the ultrasonic device when exciting a transmission pulse.

Brief Description of the Drawings In the drawings Eig. 1 is a schematic circuit diagram of a preferred embodiment of an ultrasonic device according to the invention, in which the parts that are ac- tive during transmission are indicated by con- tinuous lines; Fig. 2 shows the circuit diagram of Fig. 1, the parts that are active during reception being indi- cated by continuous lines; Fig. 3 is a chart illustrating the voltage variation across a transducer as a function of time during a transmission pulse when using a con- ventional silicon diode in the circuit diagrams according to Figs 1 and 2 but without a zener diode;

Fig. 4 is a chart according to Fig. 3 when using a PIN diode in the circuit diagrams according to Figs 1 and 2 but without a zener diode; and Fig. 5 is a diagram according to Fig. 4, with a zener diode according to the circuit diagrams of Figs 1 and 2.

Detailed Description of a Preferred Embodiment of the Invention Figs 1 and 2 show a schematic circuit diagram for an ultrasonic device according to a preferred embodiment of the invention. In the circuit diagram, reference numeral 1 designates a transducer, which may be of piezoelectric type but which, in a preferred embodiment, is of micro- mechanical type and which is adapted to convert electric signals to sound pulses and vice versa. The transducer is connected via a first resistor 2 to a positive direct voltage. The other pole of the transducer is connected to, for example, earth or the negative pole of the bat- tery.

In a connecting point 3 between the transducer 1 and the resistor 2, a diode 4 is connected via its anode side, i. e. the forward direction of the diode is oriented a-aay--fr-om-the-tr-ansducer. In-addit-iora-,-. n.-amplifier--5- is- connected in the same point 3 and adapted to amplify ultrasonic signals received by the transducer and con- verted into electrical signals. The amplified signals can then be used for any desired purpose, such as controlling or measuring, or presented in the form of a graph on a display or on paper.

The cathode side of the diode 4 is connected via a connecting point 6 to the positive direct voltage by means of a second resistor 7 and to earth by means of a switch 8, for example a transistor. The switch 8 is con- trolled by a pulse generator 9. In the preferred embodi- ment as shown, a zener diode 10 is arranged between the connecting point 6 and the switch 8 and adapted to main-

tain the voltage across the transducer 1 at a certain minimum level, thereby, inter alia, reducing the time required to charge the circuit following a transmission pulse. This zener diode may be left out, however, if de- sired.

The components indicated by continuous lines in Fig. 1 are involved in the transmission of an ultrasonic pulse from the transducer 1, and the following process takes place: In an initial mode, a constant direct volt- age is applied across the transducer 1, the direct volt- age being determined by the voltage between the direct- current source and earth as well as the relationship between the resistance of the resistor 2 and the inner resistance of the transducer. In the initial mode, the switch 8 is in a non-conductive blocking state, which means that the connecting point 6 has essentially the same voltage as the direct-current source, since any leakage current through the switch will be small. Thus, the connecting point 6 has a higher voltage than the connecting point 3 and, accordingly, the diode 4 is reverse-biased, i. e. no current flows through it. When an ultrasonic pulse is to be transmitted the pulse generator 9 generates a control pulse of an appropriate length. The control pulse operates the switch 8, causing it to change from its blocking state to a conductive state. The switch will thus start to conduct current and since its inner resistance is small the voltage in the connecting point 6 will be reduced to a value determined by the zener diode 10, the diode 4 starts to conduct current in the forward direction and the voltage in the connecting point 3 will also be reduced to about the same voltage as in the connecting point 6. This results in a significant voltage drop across the transducer 1, i. e. a negative transmission pulse is generated, and an ultra- sonic pulse is transmitted from the transducer.

When the control pulse generated by the pulse gene- rator 9 stops, the switch 8 is turned off, i. e. it re-

turns to its blocking state and the voltages in the connecting points 6 and 3 return to their initial level, primarily by means of charging via the resistors 2 and 7.

Reference is now made to Fig. 2, in which the com- ponents that are primarily involved during reception of ultrasonic pulses are indicated by continuous lines.

The device shown is in its initial mode with the switch 8 in its blocking state and the diode 4 reverse-biased, which results in the voltage in the connecting point 3 being determined by the voltage of the direct-current source relative to earth and the relationship between the resistance of the resistor 2 and the resistance of the transducer 1. Thus, in the receiving state the volt- age across the transducer 1 is essentially constant and ultrasonic pulses reaching the transducer can be con- verted into electrical signals without significant atten- uation. This allows the signal to be detected, amplified in the amplifier 5 and then processed appropriately.

Figs 3 and 4 show two graphs illustrating the volt- age variation as a function of time across the trans- ducer 1 in the ultrasonic device shown in Figs 1 and 2, but without the zener diode 10. The graphs illustrate the development when transmitting an ultrasonic pulse, and the respective graphs relate to the used of two different types of diodes 4. Fig. 3 illustrates the development when using a conventional diode, for example a silicon diode. The horizontal part 11 of the graph represents the voltage across the transducer in the initial mode or the receiving mode. When transmitting a transmission pulse from the pulse generator 9 and, thus, adjusting the switch 8 to its conductive state, the voltage across the transducer will rapidly fall to a value close to zero as illustrated by the vertical section 12 of the graph. The voltage will remain at this low level for the duration of the transmission pulse interval, i. e. when the switch 8 is in its conductive state, as illustrated by graph sec- tion 13. When the transmission pulse stops and the switch

8 returns to its blocking state, the voltage across the transducer will again start to rise as illustrated by the curved section 14 of the graph. As is evident from the curved section 14, the voltage increase across the transducer is relatively slow, which is due to the fact that current is supplied to the transducer only via the resistor 2 and since, in a preferred embodiment, the re- sistance of the latter is high the voltage increase will be slow.

The graph in Fig. 4 represents the voltage across the transducer when using a so-called PIN diode 4. As shown, the first part of the graph is similar to the graph shown in Fig. 3, i. e. the voltage reduction across the transducer occurs in an analogue manner as the switch 8 is changed to its conductive state by means of the control pulse from the pulse generator 9. On the other hand the voltage across the transducer increases con- siderably faster when the switch 8 returns to its blocking state, as illustrated by the graph section 14, which in this embodiment rises considerably more steeply.

This is very advantageous as regards the functioning of the ultrasonic device, since rapid charging of the cir- cuit means that ultrasonic pulses can be transmitted at shorter-time intervals, whlch means, whem using_the ultrasonic device as a flow meter, that a higher degree of accuracy of measurement can be obtained. The reason why a PIN diode will produce a voltage graph of the kind shown in Fig. 4 is that a PIN diode is capable of con- ducting current for a brief moment also in the backward direction immediately after having conducted current in the forward direction. During this time current to the transducer will thus be supplied also via the diode 4 and the resistor 7, the latter having, in a preferred embodiment, a significantly lower resistance than the resistor 2, and the charging time will thereby be re- duced.

Fig. 5 shows a time/voltage chart corresponding to the one in Fig. 4, i. e. the voltage variation across the transducer 1 as a function of time during transmission of an ultrasonic pulse when using a PIN diode 4, the differ- ence being that in this embodiment the zener diode 10 is arranged between the connecting point 6 and the switch 8.

As shown, this graph is similar to the graph in Fig. 4, i. e. rapid recharging as illustrated by graph section 14, the difference being that during excitation of the trans- mission pulse the discharge value will not fall to a value close to zero but will reach a value that is deter- mined by the zener diode 10, more particularly about 27- 28 V in the embodiment shown. Thus, the zener diode con- tributes to a further reduction of the charging time.

Conceivable Modifications of the Invention It will be appreciated that the invention can be modified in various ways within the scope of the appended claims. The components described and illustrated in con- nection with the preferred embodiment may, for instance, be supplemented with further components for various pur- poses. It would also be possible to connect the compo- nents in parallel or in series to additional components of similar type, for example two or more diodes 4, two or mare resistors 2 and 7 respectlvelyr and so on