Controlled oscillators, in particular voltage-controlled oscillators, are suitable for providing an oscillation signal at an output having a frequency which can be controlled as a function of a control signal. In the case of voltage-controlled oscillators, this control signal is a control voltage. In order to set a stable operating point, it is necessary to supply the oscillator with a suitable bias. The noise response of an oscillator is dependent on the set operating point. A poorly set operating point leads to additional noise being generated during operation.

The operating point of conventional oscillators is generally set manually, for example when producing a circuit with an oscillator. In this case, a resistance network for generating the bias of the oscillator is aligned or set either mechanically or by means of laser-trimmable resistors. The generated bias is applied to amplifiers which are connected with positive feedback and produce the desired oscillation signal in conjunction with a resonator. Transistors are typically used as the amplifiers. The feedback circuit and the amplifiers are also referred to as the oscillator core.

In order to control the amplitude of the output signal from the oscillator, it is known from the prior art to tap off the voltage across the resonator, to rectify it and to use the signal obtained after low-pass filtering to control the operating point of the oscillator core. One example of an oscillator having a design in accordance with the prior art is shown in the block diagram in Figure 1. The output A of an amplifier 1 is connected to a resonator 2. A signal is fed back from the resonator 2 to the amplifier 1 with positive feedback. A reference signal Vref is fed to the amplifier 1 , by means of which it is possible to set the gain of the amplifier 1. In the case of this oscillator, the amplitude is set manually.

Figure 2 shows an oscillator with automatic amplitude regulation in accordance with the prior art. In a similar manner to that in Figure 1 , the oscillator has an amplifier 1 and a resonator 2. A signal is fed back from the resonator 2 to the amplifier 1 with positive feedback. A rectifier 3 rectifies the oscillation signal from the resonator 2. The rectified signal is passed from the rectifier 3 to a

low-pass filter 4. The rectified, low-pass-filtered oscillation signal from the resonator 2 is fed to the amplifier 1 instead of the reference signal Vref as the control variable, and the control loop of the amplitude regulation is thus closed.

Figure 3 shows a further example of an oscillator known from the prior art.

The amplifier 1 in this example is in the form of a differential amplifier 6, the gain of which can be set via a controllable current source 7. A resonator 2 is connected with positive feedback to the amplifier in a known manner. The oscillation signal from the resonator 2 is applied to a differential amplifier 8 via a rectifier 3 and a low-pass filter 4, and a signal Vref is fed to said differential amplifier 8 as a further input signal. Using the signal Vref, external control of the gain of the amplifier 1 is possible in addition to the automatic gain regulation.

Figure 4 shows one example of a known controllable current source 7 in conjunction with a differential amplifier 8, as is used in the oscillator described in

Figure 3. The current source 7 is formed by the transistor 11 operated in a common collector circuit. A voltage is tapped off across the emitter resistor 12, and the value of this voltage is essentially proportional to the collector current of the transistor 11. The voltage is fed to the differential amplifier 8, the output of which is connected to the base connection of the transistor 12. As a result, the control loop of the current source 7 is closed. The rectified oscillation signal, coming from the low-pass filter 4, of the resonator 2 is not illustrated in the figure.

Operation of transistors in the linear region, as is used in conventional circuits for control purposes, often brings about low-frequency noise. This noise impairs, inter alia, the phase stability of the oscillation signal in a controlled oscillator.

It is therefore generally desirable to provide a controlled oscillator which has an improved response in terms of phase and/or amplitude noise.

Furthermore, it is often desirable for the oscillator circuit, i.e. the so-called oscillator core, to be used for different frequencies or frequency bands. In addition, the quality factor of the oscillator is often subject to fluctuations, owing to, for

example, component tolerances or technological variations in the production of integrated oscillator circuits. It is therefore likewise desirable to obtain an oscillator which can be very versatile in use and which makes it possible to compensate for tolerances.

An oscillator having the desired properties is specified in independent Patent Claim 1. Further developments and refinements of the oscillator according to the invention are specified in the dependent patent claims.

The oscillator according to the invention has a resonator which is also referred to as a resonance tank. The resonator comprises an inductive element and a capacitive element connected in parallel or series. The inductive element and the capacitive element can be realized by discrete inductive or capacitive components, but it is also possible for the inductive or capacitive properties to be produced by means of a circuit for impedance transformation comprising components having only either the capacitive or the inductive property. At least one of the capacitive or inductive elements of the resonator can be altered in terms of its value, as a result of which the oscillator frequency can be set. At least one amplifier is connected at its output to the resonator, in addition the oscillation signal from the oscillator being fed back with positive feedback to this output. The operating point of the amplifier can be set via a control circuit. Setting of the operating point is to be understood in the sense that, for example, gain, emitter current, transconductance or a bias are altered individually or in combination. In one preferred embodiment, the amplifier is formed by at least one transistor. The transistor may be a bipolar transistor or a field-effect transistor. The oscillation signal is fed to the control circuit, sampled and converted into numerical values. In one advantageous refinement of the invention, the oscillation signal is fed to a peak value detector or average value detector prior to sampling. The sampled signal represents the amplitude or the rms value of the oscillator oscillation. The numerical value of the sampled signal is passed to the control circuit where a comparison of the desired value and the actual value is carried out. The control circuit is associated with a table of parameters. The table of parameters contains control parameters which are associated with various operating points of the amplifier. The control parameters stored in the table of parameters are selected

such that nonlinearities of the amplifier and of the components of the resonator, such as a varactor, i.e. a controllable capacitance, can be compensated for. The parameter(s) selected from the table of parameters is/are fed to a digital-to-analogue converter, the output signal of which is fed to the amplifier for control purposes.

In one embodiment of the oscillator according to the invention, the amplifier is in the form of a differential amplifier which is connected to the resonator. The transistors of the amplifier can be controlled individually or jointly by the control circuit.

In one embodiment of the differential amplifier, the transistors of the amplifier are connected to one another by their emitter connections and are connected to a controllable, current-limiting means, for example a controllable current source. The current through the controllable, current-limiting means is also referred to using the term "tail current". The controllable, current-limiting means is, for example, a transistor in a corresponding circuit. The resonator is connected between the collector connections of the transistors of the amplifier. The oscillation signal, in each case with positive feedback, is fed to the base connections of the transistors of the amplifiers. Since the transconductance of the transistors is dependent on the collector current, the operating point of the amplifiers can be altered by setting the collector current. In the case of feedback with no DC voltage component, the base quiescent potential of the transistors can also be set independently of the feedback. The control circuit in this embodiment makes available two control signals which set the collector current or the base quiescent potential of the transistors.

The quality factor of the oscillator can be set or regulated by regulating the amplitude across the resonator.

As has been described further above, the quality factor of the oscillator is often subject to fluctuations or the oscillator is required to be versatile in use. The fact that the tail currents can be set makes this versatile use possible. In particular, the associated table of parameters and the digital control advantageously make it

possible to match a circuit comprising the oscillator core to the intended use depending on the actual use and to compensate for variations in tolerances.

Furthermore, it is advantageously possible to calibrate the circuit during production. For calibration, the values of the table of parameters are determined during manufacture or in a subsequent test step and stored in the table. The values stored in the table can then be called up during subsequent operation.

The regulation according to the invention of the base quiescent potential is a special case for the regulation of a bias or bias voltage. In the case of the oscillator according to the invention, it is possible to improve the start-up behaviour of the oscillator by regulating the bias voltage. For this purpose, for example in the case of an oscillator with a push-pull circuit, the bias voltage is offset. For this purpose, an offset voltage is fed to the differential input of the differential amplifier, for example to the base connections of the transistors 34 in Figure 7, for the purpose of starting the oscillator. This takes place, for example, by one base connection being set at a first potential and the other base connection being set at a second potential. In one embodiment, the different potentials are set by the base connections being connected to two different voltage dividers by means of two switches. If a corresponding output oscillation is detected, the switches are switched such that an identical base potential is set for both transistors. The time constant of the regulating operation can be set by a corresponding analogue circuit, for example by means of capacitances, or is digitally controlled. The switches are, for example, controlled by a power-on-reset circuit. Alternatively, the switches are controlled by means of a digital control circuit. In the case of the digital control circuit, the base potential set by the digital logic is set after starting and once the switches have opened. The advantage with the digital control circuit is the fact that the oscillator starts to oscillate in any event. In addition, the active switching at the base connections during operation can be dimensioned such that the noise of the circuit is minimized without making it necessary to take into account analogue components of an oscillation excitation circuit. Once the oscillator has started at a first operating point which is optimized for reliable oscillation excitation, there is a changeover to a second operating point during operation, and this operating point is optimized for the noise to be as low as

possible. In an embodiment of the oscillator with improved start-up behaviour, the supply voltage is also detected. The amplitude of the oscillation signal is in this case the regulating variable.

In another exemplary embodiment, the table of parameters contains values which also make possible temperature compensation of components of the oscillator circuit. For this purpose, the control circuit is also fed a temperature signal.

In order to generate an oscillation, an oscillator needs to fulfil the so-called amplitude condition and the so-called phase condition. The phase condition specifies that an oscillation can only come about when the output voltage is in phase with the input voltage. The amplitude condition specifies that an oscillator can only oscillate when the amplifier cancels the attenuation of the feedback. With less feedback, the amplitude of the output signal is reduced, and with greater feedback, the amplitude of the output signal is increased. In order that an oscillator circuit begins to oscillate when the operating voltage is connected, initially the feedback needs to be greater. In this operating state, the amplitude increases exponentially until the amplifier is overdriven. Owing to the overdriving, the feedback signal is reduced to such an extent that no further rise in the output signal takes place. The output voltage from the amplifier, however, is no longer sinusoidal. Therefore, gain regulation must ensure that the gain is reduced before the amplifier is overdriven. Owing to control of the gain according to the invention using values from the table of parameters, it is possible to realize reliable oscillation excitation of the oscillator under all conditions.

In one preferred embodiment of the oscillator circuit, the digital-to-analogue converter operates in accordance with the so-called parallel method. In this case, switches are connected to a voltage source via current-limiting means. The switches are also connected to a common circuit point. Depending on the voltage of the voltage source, the number and the position of the connected switches, an output voltage or a specific output current is set. The switches are in the form of semiconductor switches and comprise, for example, bipolar transistors or field-effect transistors. This type of digital-to-analogue converter has particularly

low noise since the switches are operated in the saturated range. As has been mentioned further above, operation of transistors in the linear range, as is used in conventional circuits for control purposes, often brings about low-frequency noise. This noise impairs, inter alia, the phase stability of the oscillation signal. The operational principle of the digital-to-analogue converter is not essential to the invention. In addition to the parallel method, other operational principles may also be used, for example the weighing method or the counting method.

In one embodiment of the invention, two separate digital-to-analogue converters are provided for one push-pull oscillator for the purpose of setting the operating points of the transistors. In order to start the oscillation, an offset is set by means of the digital-to-analogue converters. During operation, the digital-to-analogue converters are driven by identical values such that there is no offset.

A circuit for digital signal processing may also be used in place of the table of parameters. In this case, for example, a suitable filter characteristic is realized by means of a digital signal processor DSP. The filter properties can thus be matched to the respective operating conditions, such as operating voltage, frequency or temperature, during operation.

The invention will be described in more detail below with reference to the attached drawing, in which:

Figure 1 shows a block diagram of an oscillator having manual amplitude control in accordance with the prior art; Figure 2 shows a block diagram of a known oscillator with automatic amplitude control;

Figure 3 shows a block diagram of a known push-pull oscillator with automatic amplitude control;

Figure 4 shows a controllable current source for the purpose of controlling the amplitude of an oscillator in accordance with the prior art; Figure 5 shows a first block diagram of an oscillator according to the invention; Figure 6 shows a second block diagram of an oscillator according to the

invention; Figure 7 shows a schematic circuit diagram of a push-pull oscillator with operating-point setting according to the invention;

Figure 8 shows a first embodiment of a controllable current source in accordance with the invention; and

Figure 9 shows a second embodiment of a controllable current source in accordance with the invention.

Identical or similar components or elements are provided with the same references in the figures.

Figures 1 to 4 have already been described further above in the prior art and will not be explained in any more detail below.

Figure 5 illustrates a block diagram of a first embodiment of an oscillator according to the invention. The output of an amplifier 1 is connected to a resonator 2. A signal tapped off from the resonator 2 is fed back to the amplifier 1 with positive feedback. The output signal from the resonator 2 is also applied to an analogue-to-digital converter circuit 21. A numerical value is passed from the analogue-to-digital converter circuit 21 to a control circuit 22. A control variable is applied to a digital-to-analogue converter 23 from the control circuit 22, and the output signal from said digital-to-analogue converter 23 is fed to the amplifier 1 as a control variable. The amplitude of the oscillator's oscillation can be controlled by accordingly controlling the amplifier 1 using the digital-to-analogue converter 23.

Figure 6 illustrates a block diagram of a further embodiment of an oscillator according to the invention. A resonator 2 is connected to an amplifier 1. The amplifier 1 comprises, for example, a differential amplifier 6 with capacitive positive feedback, which differential amplifier 6 is associated with a controllable current source 7. The gain can be set by means of the controllable current source. In addition, a circuit 26 for setting the operating point is associated with the differential amplifier 6. The oscillation signal is tapped off from the resonator 2 and fed to an analogue-to-digital converter 21. The numerical value of the analogue-to-digital converter 21 is fed to a control circuit 22 which is also

connected to a table of values 24. An output signal from the control circuit 22 is fed to a digital-to-analogue converter 23, whose output signals are applied to the circuit 26 for setting the operating point of the differential amplifier 6 and to the controllable current source 7.

Figure 7 illustrates a schematic circuit diagram of an oscillator according to the invention. The figure illustrates the resonator 2, the amplifier 1 and the controllable current source 7. A circuit 26 for setting the operating point and the gain of the amplifier is illustrated in the figure as a function block. The amplifier 1 is formed by two transistors 34, whose emitter connections are connected to a transistor 11 of a controllable current source 7. The emitter connection of the transistor 11 of the controllable current source 7 is connected to earth via an emitter resistor 12. The base connection of the transistor 11 of the controllable current source 7 is connected to the circuit 26. The transistor 11 and the emitter resistor 12 form the controllable current source 7 indicated by a dashed frame. The collector connections of the transistors 34 of the amplifier 1 are connected to the base connections of the respective other transistor 34 via capacitors 36. Capacitors 37 are connected to earth from the base connections of the transistors 34. Together with the capacitors 36, they form a capacitive voltage divider which sets the degree of feedback. The resonator 2 is connected to the collector connections of the transistors 34 via resistors 31. However, the resistors 31 are not essential to the circuit and may also be omitted. The resonator 2 in each case comprises a series circuit comprising two inductances 32 or capacitors 33. The series circuits of the components are connected in parallel with one another and are connected to the resistors 31. A supply voltage VDD is supplied to a centre tap of the series circuit comprising the two inductances 32. If the capacitances are varactors, i.e. components having a capacitance which is dependent on a control voltage, this control voltage can be applied to a centre tap (not illustrated in the figure) of the two capacitances. The capacitances can alternatively also be designed such that they can be switched in order to provide a setting option. The above-described circuit forms a push-pull oscillator with capacitive positive feedback. The circuit 26 controls the potential or the current across the base connections of the transistors 34 such that the operating points of the transistors 34 are set optimally as a function of the operating conditions. In this case, it is also

possible for the potentials at the base connections to be set differently for both transistors 34, for example in order to achieve an improved oscillation excitation behaviour. Furthermore, the circuit 26 controls the gain of the amplifier 1 via the controllable current source 7.

Figure 8 illustrates an embodiment of part of the control circuit 26 shown in Figure 7. In the case of the embodiment illustrated in the figure, the digital-to-analogue converter is realized by means of switchable current sources QO, Q1 , Q2, Qn. The switchable current sources QO, Q1 , Q2, Qn can be switched to a common line by means of switches SO, S1 , S2, Sn. The digital-to-analogue conversion takes place by means of correspondingly switching the switches SO, S1 , S2, Sn. The currents I0, 11 , I2, In of the current sources QO, Q1 , Q2, Qn may be identical or weighted such that currents having specific ratios can be connected. The current sources are, for example, realized by means of field-effect transistors, wherein it is possible to set weighting of the currents by means of correspondingly scaling the channel dimensions of the field-effect transistors. However, it is also possible to realize current sources by means of an R-2R resistance network. The common line to which the current sources QO, Q1 , Q2, Qn can be connected is connected to the collector connection of a transistor 11'. The transistor 11' forms, with a transistor 11 and resistors 12 and 12', respectively, a current-mirror circuit. The current flowing through the collect/emitter path of transistor 11 is as high as the current through the transistor 11', which is set by means of the switched current sources or is scaled according to a known ratio. The current through transistor 11 sets the gain of an amplifier 1 of an oscillator, as is shown, for example, in Figure 7. The current mirror in this case corresponds to the controllable current source 7.

Figure 9 illustrates the setting of the current through the controllable current source 7 by means of a resistor network and switches. A series circuit comprising resistors RO, R1 , R2, R3, Rn is connected between a supply voltage Vsuppiy and earth. Taps are provided between the resistors and can be connected to the base connection of a transistor 11 of a controllable current source 7 by means of switches SO, S1 , S2, Sn. The switches SO, S1 , S2, Sn can be driven by means of drive logic of a digital-to-analogue converter. A current through the

collector/emitter path of the transistor can be set by correspondingly selecting a base potential for the transistor 11. The current across the collector/emitter path of the transistor 11 produces a voltage drop across an emitter resistor 12, which sets the potential across the emitter of the transistor 11 and thus also the base potential. The resistors RO, R1 , R2, R3, Rn may have identical values or have specific ratios with respect to one another.

The switches SO to Sn of the circuits shown in Figures 8 and 9 may be realized, for example, by means of field-effect transistors. However, the switches may also be in the form of micromechanical switches. The type of switch is not essential to the invention.

The control circuit 26, illustrated using Figures 8 and 9 with respect to the driving of the controllable current source 7, has corresponding circuits for the purpose of driving the transistors 34 shown in Figure 7. Current-mirrors or switchable resistor networks may likewise be used for the purpose of setting the operating point of the transistors 34 of the amplifier 1 shown in Figure 7.

The oscillator according to the invention can advantageously be used in receiver circuits for modulated RF signals. The application of the oscillator circuit according to the invention is not restricted to this field of use, however. The oscillator according to the invention can be used in any circuit which requires low-noise, stable oscillation.