Single ended voltage to time converter
The single-ended to differential voltage to time converter addresses linearity issues in time-domain ADCs by ensuring minimum pulse durations, enhancing the conversion of short time differences and improving ADC accuracy.
Patent Information
- Application Number
- PCT/EP2023/062131
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2023-05-08
- Publication Date
- 2024-11-14
AI Technical Summary
Time-domain analog-to-digital converters (ADCs) face linearity issues when converting short time durations, particularly in differential ADCs where input voltages are close to each other, leading to poor performance in existing VTC implementations.
A single-ended to differential voltage to time converter is introduced, comprising two VTCs with matched current sources and capacitors, and a pulse-detector circuit to ensure a minimum pulse duration, alleviating linearity problems by guaranteeing non-zero pulse durations for TDCs.
This solution enhances the linearity of time-domain ADCs by ensuring that all pulse durations are above a minimum threshold, improving the accuracy of converting short time differences and reducing non-linearity issues in TDCs.
Smart Images

Figure EP2023062131_14112024_PF_FP_ABST
Abstract
Description
[0001] SINGLE ENDED VOLTAGE TO TIME CONVERTER
[0002] Technical field
[0003] The present invention relates to time-domain analog-to-digital converters, and more specifically to a single ended voltage to time converter for time-domain analog-to- digital converters.
[0004] Background
[0005] A time-domain analog-to-digital converter (ADC) utilizes an intermediate conversion of an analog input sample, e.g. in the form of a voltage, into the timedomain for converting the input sample to the digital domain. One flavor of a timedomain ADC is implemented by a cascade of a voltage-to-time converter (VTC) and a time-to-digital converter TDC. The VTC is a circuit that takes in an analog voltage signal and produces a series of pulses where the delay on each pulse, with respect to an input clock signal, or the delay between pulses on two different output signal lines, is proportional to the input voltage at the time the pulse was generated. The VTC output is fed into a TDC, which measures the delay between pulse edges on two signals and converts the value to a digital representation.
[0006] The VTC may be implemented in various ways depending on requirements on linearity. The more linear VTCs tend to be those where the input signal is sampled on a capacitor and a constant current source discharges the capacitor, thereby resulting in an ideally linear decrease of the voltage on the capacitor. As the voltage reaches a threshold of a comparator, the comparator output toggles from zero to one (or from one to zero, depending on implementation). In the case of a differential ADC, there are two input voltages, the difference between which forms a differential input voltage. Such a differential ADC utilizes two VTCs as described above, one for each of the two input voltages. The differential input voltage is thereby converted into a time difference between the rising edges (or falling edges, depending on implementation) of the two comparator outputs. Summary
[0007] The inventors have realized that many TDC implementations suffer from poor linearity when the time differences, or time durations, to be converted are short. In a differential time-domain ADC as described above, this would e.g. be the case when the differential input voltage is close to 0 (or in other words, when the two input voltages are close to each other). The inventors have realized that this problem can be alleviated by means of a pulse-detector circuit that is connected to the outputs of the two VTCs. The durations of output pulses generated at two outputs of the pulse-detector circuit are converted by two separate TDCs (one for each of the two outputs). Due to an inherent delay in the components of the pulse-detector circuit, these output pulses are guaranteed to have a certain non-zero minimum duration. This, in turn, alleviates the linearity problem of TDCs when converting short time durations, as the durations to be converted are not shorter than said minimum duration. The components in the pulsedetector circuit can be deliberately designed such that the minimum duration is suitable for the TDCs used. The inventors further realized that not all circuits are differential and will therefore not inherently benefit from the increased linearity of two VTCs. It is in view of these considerations and others that the various embodiments of this disclosure have been made. The present disclosure therefor recognizes the fact that there is a need for alternatives to (e.g. improvement of) the existing art described above and it is an object of some embodiments to solve, mitigate, alleviate, or eliminate at least some of the above or other disadvantages. The inventors realized that these problems may be alleviated by means of a single ended to differential voltage to time converter as defined in the claims.
[0008] According to a first aspect, there is provided a single ended to differential voltage to time converter (VTC) for converting an input voltage. The single ended to differential VTC comprises a first VTC comprising an input terminal configured to receive the input voltage, a capacitor, a first current source configured to charge the capacitor of the first VTC, a first comparator circuit configured to compare a voltage at a first terminal of the capacitor of the first VTC with a first reference voltage and to generate a first pulse at an output of the first comparator circuit, and a first sampling switch connected between the input terminal and the capacitor of the first VTC. The single ended to differential VTC further comprises a second VTC comprising an input terminal configured to receive the input voltage, a capacitor, a second current source configured to discharge the capacitor of the second VTC, a second comparator circuit configured to compare a voltage at a first terminal of the capacitor of the second VTC with a second reference voltage and to generate a second pulse at an output of the second comparator circuit, and a second sampling switch connected between the input terminal and the capacitor of the second VTC.
[0009] In one variant, a charge current of the first current source is matched to a discharge current of the second current source.
[0010] In one variant, the capacitor of the first VTC is substantially identical to the capacitor of the second VTC.
[0011] In one variant, during a first phase of operation, the first sampling switch is at a conducting state and the capacitor of the first VTC is configured to, for each sample of the input voltage, be charged to a voltage value of that sample of the input voltage. Further, during a subsequent second phase of operation, the first sampling switch is at a non-conducting state. At the second phase of operation, the first current source is configured to charge the capacitor of the first VTC, and the first comparator circuit is configured to compare the voltage at a first terminal of the capacitor of the first VTC with the first reference voltage and to generate the first pulse at the output of the first comparator circuit.
[0012] In one variant, during a first phase of operation, the second sampling switch is at a conducting state and the capacitor of the second VTC is configured to, for each sample of the input voltage, be charged to the voltage value of that sample of the input voltage. Further, during a subsequent second phase of operation, the second sampling switch is at a non-conducting state. At the second phase of operation, the second current source is configured to discharge the capacitor of the second VTC, and the second comparator circuit is configured to compare the voltage at a first terminal of the capacitor of the second VTC with the second reference voltage and to generate the second pulse at the output of the second comparator circuit. In one variant, the first VTC is further configured to generate, at a first output of the first VTC, a first edge of a first pulse delayed an amount of time corresponding to a magnitude of the sample of the input voltage in relation to the first reference voltage.
[0013] In one variant, the second VTC is further configured to generate, at a first output of the second VTC, a first edge of a second pulse delayed an amount of time corresponding to a magnitude of the sample of the second input voltage in relation to the second reference voltage.
[0014] In one variant, the first VTC is further configured to receive a first auxiliary reference voltage. The first VTC is further configured to selectively generate, depending on a relation between a magnitude of the sample of the input voltage and the first auxiliary reference voltage, at an auxiliary output of the first VTC, a first edge of a first auxiliary pulse delayed an amount of time corresponding to the magnitude of the sample of the input voltage in relation to the first auxiliary reference voltage
[0015] In one variant, the second VTC is further configured to receive a second auxiliary reference voltage. The second VTC is further configured to selectively generate, depending on a relation between a magnitude of the sample of the input voltage and the second auxiliary reference voltage, at an auxiliary output of the second VTC, a first edge of a second auxiliary pulse delayed an amount of time corresponding to the magnitude of the sample of the input voltage in relation to the second auxiliary reference voltage.
[0016] According to a second aspect, there is provided an analog-to-digital converter (ADC) for converting samples of an input voltage. The ADC comprises the single ended to differential VTC according to the first aspect and a differential TDC. The differential TDC comprising a first input terminal coupled to the output of the first comparator circuit of the single ended to differential VTC and configured to receive the first pulse, a second input terminal coupled to the output of the second comparator circuit of the single ended to differential VTC and configured to receive the second pulse, a first output terminal, and a second output terminal. The differential time-to- digital circuit is configured to generate a first digital value at the first output terminal and a second digital value at the second output terminal. A difference between the first digital value and the second digital value is proportional to a temporal difference between a first edge of the first pulse and a first edge of the second pulse.
[0017] In one variant, the differential time-to-digital circuit is configured to generate the first digital value proportional to an amount of time the first edge of the first pulse precedes the first edge of the second pulse.
[0018] In one variant, the differential time-to-digital circuit is configured to generate the first digital value as zero responsive to the first edge of the first pulse succeeding the first edge of the second pulse.
[0019] In one variant, the differential time-to-digital circuit is configured to generate the second digital value proportional to an amount of time the first edge of the second pulse precedes the first edge of the first pulse.
[0020] In one variant, the differential time-to-digital circuit is configured to generate the second digital value as zero responsive to the first edge of the second pulse succeeding the first edge of the first pulse.
[0021] In one variant, the first digital value and / or the second digital value is offset by an offset value determined by the internal time delay of the differential time-to-digital circuit.
[0022] In one variant, the first digital value and the second digital value form an output sample of the ADC.
[0023] In one variant, the ADC further comprises a circuit coupled to the outputs of the differential time-to-digital circuit configured to generate an output sample of the ADC in response to the first digital value and the second digital value.
[0024] In one variant, the differential time-to-digital circuit further comprises a pulsedetector circuit coupled to the outputs of the first VTC and the second VTC and having a first output and a second output.
[0025] In one variant, the pulse-detector circuit is configured to make a transition from a first logic state to a second logic state at a first output of the pulse-detector circuit in response to the first edge of the first pulse whereby a first edge of a third pulse is generated at the first output of the pulse-detector circuit. In this variant, the pulsedetector circuit is further configured to make a transition from the first logic state to the second logic state at a second output of the pulse-detector circuit in response to the first edge of the second pulse whereby a first edge of a fourth pulse is generated at the second output of the pulse-detector circuit.
[0026] In one variant, the pulse-detector circuit is further configured to, in response to both the first output and the second output of the pulse-detector circuit being set in the second logic state, reset both the first and the second output to the first logic state. In response thereto, a second edge of the third pulse is generated at the first output of the pulse-detector circuit and a second edge of the fourth pulse is generated at the second output of the pulse-detector circuit.
[0027] In one variant, the differential time-to-digital circuit comprises a first TDC coupled to the first output of the pulse-detector circuit and a second TDC coupled to the second output of the pulse-detector circuit.
[0028] In one variant, each of the first TDC and the second TDC is a pulse-shrinking TDC.
[0029] In one variant, the pulse-detector circuit comprises a first flip-flop and a second flip-flop, each having a data input, a clock input, a reset input, and an output. The data input of each of the first flip-flop and the second flip-flop is configured to receive a constant signal corresponding to the second logic state. The clock input of the first flipflop is configured to receive the first pulse. The clock input of the second flip-flop is configured to receive the second pulse. The output of the first flip-flop is connected to the first output of the pulse-detector circuit. The output of the second flip-flop is connected to the second output of the pulse-detector circuit. The pulse-detector circuit comprises a logic circuit having a first input connected to the output of the first flipflop, a second input connected to the output of the second flip-flop, and an output connected to the reset inputs of the first flip-flop and the second flip-flop for resetting the first flip-flop and the second flip-flop in response to the outputs of both the first flipflop and the second flip-flop being set in the second logic state.
[0030] In one variant, the logic circuit is further configured to reset the first flip-flop and the second flip-flop in response to a reset pulse.
[0031] In one variant, the ADC comprises circuitry configured to provide said reset pulse in preparation of converting each sample. According to a third aspect, there is provided a receiver circuit comprising the ADC of the second aspect.
[0032] According to a fourth aspect, there is provided an integrated circuit comprising the single ended to differential VTC of the first aspect
[0033] According to a second aspect, there is provided an integrated circuit comprising the ADC of the second aspect.
[0034] According to a second aspect, there is provided an electronic apparatus comprising the ADC of the second aspect.
[0035] In one variant, the electronic apparatus is a communication apparatus.
[0036] In one variant, the communication apparatus is a wireless communication device for a cellular communications system.
[0037] In one variant, the communication apparatus is a base station for a cellular communications system.
[0038] Brief description of the drawings
[0039] Fig. 1 illustrates a communication environment;
[0040] Fig. 2 illustrates a transceiver circuit;
[0041] Figs. 3a, 3b and 3c are block diagrams of an ADC;
[0042] Figs. 4 and 5 are circuit diagrams of VTCs;
[0043] Figs. 6 and 7 show pulse waveforms;
[0044] Figs. 8 and 9 are circuit diagrams of pulse-detector circuits;
[0045] Fig. 10 is a circuit diagram of a pulse-shrinking TDC;
[0046] Fig. 11 is a block diagram of an ADC;
[0047] Figs. 12a and 12b are circuit diagrams of VTCs;
[0048] Fig. 13 show pulse waveforms;
[0049] Fig. 14 is a block diagram of a pulse-detector circuit according to some embodiments of the present disclosure;
[0050] Figs. 15a and 15b are circuit diagrams of partial pulse-detector circuits;
[0051] Fig. 16 is a block diagram of a pulse forwarding circuit according to some embodiments of the present disclosure;
[0052] Fig. 17 illustrates an integrated circuit comprising an ADC; and Fig. 18 illustrates an integrated circuit comprising a single ended to differential VTC.
[0053] Detailed description
[0054] Fig. 1 illustrates a communication environment wherein embodiments of the present invention may be employed. A wireless communication device 1, or wireless device 1 for short, of a cellular communications system is in wireless communication with a radio base station 2 of the cellular communications system. The wireless device 1 may be what is generally referred to as a user equipment (UE). The wireless device 1 is depicted in Fig. 1 as a mobile phone, but may be any kind of device with cellular communication capabilities, such as a tablet or laptop computer, machine-type communication (MTC) device, or similar. Furthermore, a cellular communications system is used as an example throughout this disclosure. However, embodiments of the present invention may be applicable in other types of systems as well, such as, but not limited to, WiFi systems.
[0055] The radio base station 2 and the wireless device 1 are examples of what, in this disclosure, are generically referred to as communication apparatuses. Embodiments are described below in the context of a communication apparatus in the form of the radio base station 2 or the wireless device 1. However, other types of communication apparatuses may be considered as well, such as a WiFi access point or WiFi enabled device.
[0056] Fig. 2 is a block diagram of an embodiment of a transceiver circuit 10, which may be comprised in a communication apparatus, such as the radio base station 2 or the wireless device 1. In the embodiment illustrated in Fig. 2, the transceiver circuit 10 comprises a digital signal processing (DSP) circuit 15. The DSP circuit 15 may e.g. be what is commonly referred to as a baseband processor. The DSP circuit 15 may e.g. be configured to perform various digital signal processing tasks, such as one or more of coding, decoding, modulation, demodulation, fast Fourier transform (FFT), inverse FFT (IFFT), mapping, demapping, etc.
[0057] Furthermore, in the embodiment illustrated in Fig. 2, the transceiver circuit 10 comprises a transmitter circuit 20. The transmitter circuit 20 generally comprises a digital -to-analog converter (DAC) 25. The DAC 25 is connected to the DSP circuit 15 and configured to receive, as an input signal of the DAC 25, a digital representation of a signal to be transmitted from the DSP circuit 15. The DAC 25 is further configured to convert the signal to be transmitted to an analog representation, which is an output signal of the DAC 25. The transmitter circuit 20 further comprises a transmitter (Tx) frontend (FE) circuit 30 connected between the DAC 25 and an antenna 35 connected to the transceiver circuit 10. The Tx FE circuit 30 is configured to transform the output signal from the DAC 25 to a format suitable for transmission via the antenna 35. This may include operations such as frequency up-conversion, filtering, and / or amplification. The Tx FE circuit 30 may comprise one or more mixers, filters, and / or amplifiers, such as power amplifiers (PAs), to perform such operations. The design of such Tx FE circuits is, per se, well known to a person skilled in the field of radio transceiver design, and is not discussed herein in any further detail.
[0058] Moreover, in the embodiment illustrated in Fig. 2, the transceiver circuit 10 comprises a receiver circuit 40. The receiver circuit 40 comprises a receiver (Rx) FE circuit 45 connected to the antenna 35. Furthermore, the receiver circuit 40 comprises an ADC 50. The ADC 50 is connected between the Rx FE circuit 45 and the DSP circuit 15. The Rx FE circuit 45 is configured to transform a signal received via the antenna 35 to a format suitable to be input to the ADC 50. This may include operations such as frequency down-conversion, filtering, and / or amplification. The Rx FE circuit 45 may comprise one or more mixers, filters, and / or amplifiers, such as low-noise amplifiers (LNAs), to perform such operations. The design of such Rx FE circuits is, per se, well known to a person skilled in the field of radio transceiver design, and is not discussed herein in any further detail. The ADC 50 is configured to receive its (analog) input signal from the Rx FE circuit, and convert it to a digital representation to generate the digital output signal of the ADC 50. This digital output signal of the ADC 50 is input to the DSP circuit 15 for further digital signal processing. In some embodiments, the ADC 50 may be configured to receive a radio-frequency (RF) input signal, without any preceding frequency down-conversion.
[0059] In embodiments of this disclosure, the ADC 50 is implemented as a time-domain ADC further described below. Furthermore, embodiments of the ADC 50 disclosed herein are configured to convert samples of an input voltage V, see Fig. 3a. Said samples of the input voltage V are converted to a first digital value xa[n] and a second digital value xb[n]. Samples V[n] are denoted with a sequence index n. A sampling time for sequence index n is below referred to as nT, where T is the sampling period of the ADC 50. However, this is a mere example used for simplicity. For instance, nonuniform sampling may be used, in which case the ADC has no fixed sampling period.
[0060] Fig. 3a illustrates an embodiment of an ADC 50. It is provided with an input terminal 101 configured to receive a series of samples of an input voltage V[n], where n is, as previously mentioned, a sequence index. The ADC 50 comprises a first voltage-to-time converter (VTC) 110a configured to receive the input voltage V[n], The first VTC 110a has an input terminal 112a connected to the input terminal 101 of the ADC 50 for this purpose. The first VTC 110a is further provided with an output 114a. Furthermore, the ADC 50 comprises a second VTC 110b also configured to receive the same input voltage V[n], The second VTC 110b has an input terminal 112b connected to the input terminal 101 of the ADC 50 for this purpose. The second VTC 110b further has an output 114b. The functionality of the VTCs 110a, 110b is further described below.
[0061] The ADC 50 may comprise a pulse-detector circuit 120 coupled to the outputs 114a, 114b of the first VTC 110a and the second VTC 110b. In Fig. 3a, the pulse-detector circuit 120 has a first input 122a connected to the output 114a of the first VTC 110a and a second input 122b connected to the output 114b of the second VTC 110b for this purpose. Furthermore, the pulse-detector circuit 120 has a first output 124a and a second output 124b. The functionality of the pulse-detector circuit 120 is further described below.
[0062] Furthermore, the ADC 50 comprises a first time-to-digital converter (TDC) 130a coupled to, in Fig. 3a, the first output 124a of the pulse-detector circuit 120. In Fig. 3a, the first TDC 130a has an input 132a connected to the first output 124a of the pulsedetector circuit 120 for this purpose. Moreover, the first TDC 130a has an output 134a. The first TDC 130a is configured to provide a first digital value xa[n] at the output 134a.
[0063] The ADC 50 further comprises a second TDC 130b coupled to, in Fig. 3a, the second output 124b of the pulse-detector circuit 120. In Fig. 3a, the second TDC 130b has an input 132b connected to the second output 124b of the pulse-detector circuit 120 for this purpose. Furthermore, the second TDC 130b has an output 134b. The second TDC 130b is configured to provide a second digital value xb[n] at the output 134b.
[0064] A single ended to differential VTC 400 may be formed by the first VTC 110a and the second VTC 110b such that the input terminals 112a, 112b of the first VTC 110a and the second VTC 110b are connected to form an input terminal 101 of the single ended to differential VTC 400. The outputs 114a, 114b of the first VTC 110a and the second VTC 110b form respective outputs 114a, 114b of the single ended to differential VTC 400.
[0065] In Fig. 3a, the pulse detector circuit 120, the first TDC 130a and the second TDC 130b form a differential time-to-digital circuit 300. The first input 122a and the second input 122b of the pulse detector circuit 120 form a respective first input 122a and second input 122b of the differential time-to-digital circuit 300. The output 134a of the first TDC 130a and the output 134b of the second TDC 130b respectively form outputs 134a, 134b (output terminals) of the differential time-to-digital circuit 300.
[0066] In Fig. 3b, an ADC 50 is shown that is similar to the ADC 50 of Fig. 3a but wherein the differential time-to-digital circuit 300 does not comprise the pulse detector circuit 120. The respective inputs 132a, 132b of the first TDC 130a and the second TDC 130b are directly connected to the outputs 114a, 114b of the single ended to differential VTC 400. As in the differential time-to-digital circuit 300 of Fig. 3a, the output 134a of the first TDC 130a and the output 134b of the second TDC 130b respectively form outputs 134a, 134b (output terminals) of the differential time-to-digital circuit 300 in Fig. 3b.
[0067] In Fig. 3c, a variant of the ADCs 50 of Figs. 3a and 3b is shown. In this embodiment, the ADC 50 further comprises a circuit 150 configured to generate an output sample x[n] of the ADC 50 in response to the first digital value xa[n] and the second digital value xb[n]. In Fig. 3c, the output sample x[n] is the difference between the first digital value xa[n] and the second digital value xb[n]. Hence, the circuit 150 may be a subtraction circuit, or adder circuit with one positive and one negative input. However, other mathematical operations may be involved as well, including for instance gain and offset adjustments. The circuit 150 may be provided with one or more additional logical processing circuits (not shown) configured to logically analyze the first digital value xa[n] and the second digital value xb[n] and provide the output sample x[n] based on the logical analysis.
[0068] In Fig. 4, the first VTC 110a of the ADC 50 is shown. The first VTC 110a comprises a capacitor 250a configured to, for each sample of the input voltage V[n], be charged to the voltage value of that sample of the first input voltage V[n] during a first phase of operation (for that sample). Furthermore, according to some embodiments, the first VTC 110a comprises a first current source 260a configured to charge the capacitor 250a of the first VTC 110a during a subsequent second phase of operation (for that sample). Moreover, according to some embodiments, the first VTC 110a comprises a first comparator circuit 270a configured to compare the voltage Vca at a first terminal of the capacitor 250a of the first VTC 110a with a first reference voltage Vrefa and to generate a first pulse at an output of the first comparator circuit 270a. In Fig. 4, the first reference voltage Vrefa is provided to a negative input of the first comparator circuit 270a. In Fig. 4, a second terminal of the capacitor 250a of the first VTC 110a is connected to a signal ground node. Furthermore, in Fig. 4, the other terminal of the capacitor 250a of the first VTC 110a is connected to a positive input of the first comparator circuit 270a. Moreover, in Fig. 4, a sampling switch 280a is connected between the input terminal 112a of the first VTC 110a and the positive input terminal of the first comparator circuit 270a. A sampling control signal smp _p controls the sampling switch 280a such that, for each sample, the sampling switch 280a is closed during said first phase of operation and open during said second phase of operation. The sampling control signal smp _p may further be used as a reset pulse as will be explained in coming sections. For the embodiment illustrated in Fig. 4, the input voltage V provided at the input terminal 112a of the first VTC 110a may be either time continuous or time discrete (e.g. piecewise constant). In the case of a time-continuous input voltage V, said sample V[n] is the value of V at the end of the first phase of operation.
[0069] In Fig. 4, the first current source 260a is connected in series with a switch 290a between the positive input of the first comparator circuit 270a and the signal ground node. A control signal smp m controls the switch 290a such that, for each sample, the switch 290a is open (not conducting) during said first phase of operation, and closed (conducting) during said second phase of operation. Thereby, during the second phase of operation, the capacitor 250a of the first VTC 110a is gradually charged and the voltage Vca at the first terminal of the capacitor 250a of the first VTC 110a is thus gradually increased. When it reaches above the first reference voltage Vrefa, a transition is made at the output of the first comparator circuit 270a, which is the start of the first pulse, i.e. a first edge of the first pulse.
[0070] With reference to Fig. 5, an exemplary embodiment of the second VTC 110b will be presented. The second VTC 110b may form part, e.g. together with the first VTC 110a, of the single ended to differential VTC 400. The second VTC 110b comprises a capacitor 250b configured to, for each sample of the input voltage V[n], be charged to the voltage value of that sample of the input voltage V[n] during the first phase of operation. Furthermore, according to some embodiments, the second VTC 110b comprises a second current source 260b configured to discharge the capacitor 250b of the second VTC 110b during the second phase of operation. Moreover, according to some embodiments, the second VTC 110b comprises a second comparator circuit 270b configured to compare the voltage Vcb at a first terminal of the capacitor 250b of the second VTC 110b with a second reference voltage Vrefb and to generate the second pulse P2 (see e.g. Fig. 6) at an output 114b of the second comparator circuit 270b. Advantageously, the second comparator circuit 270b is configured to compare the voltage across the first terminal of the capacitor 250b of the second VTC 110b with a second reference voltage Vrefb being different from the first reference voltage Vrefa. However, in some embodiments, an offset voltage may be introduced at the comparator circuits 270a, 270b (advantageously with opposite sign due to their reversed input connections), and both comparators circuits 270a, 270b may be provided with the same reference voltage Vrefa, Vrefb. In such an embodiment, also the second comparator circuit 270b is configured to compare the voltage across the first terminal of the capacitor 250b of the second VTC 110b with the first reference voltage Vrefa. In Fig. 5, the first reference voltage Vrefa or the second reference voltage Vrefb is provided to the positive input of the second comparator circuit 270b. In Fig. 5, a second terminal of the capacitor 250b of the second VTC 110b is connected to a signal ground node. Furthermore, in Fig. 5, the other terminal of the capacitor 250b of the second VTC 110b is connected to a negative input of the second comparator circuit 270b. Moreover, in Fig. 5, a sampling switch 280b is connected between the input terminal 112b of the second VTC 110b and the negative input terminal of the second comparator circuit 270b. The sampling control signal smp_p controls the sampling switch 280b such that, for each sample, the sampling switch 280b is closed during said first phase of operation and open during said second phase of operation. As above, for the embodiment illustrated in Fig. 5, the input voltage V provided at the input terminal 112b of the second VTC 110b may be either time continuous or time discrete (e.g. piecewise constant). In the case of a time-continuous voltage V, said sample V[n] is the value of V at the end of the first phase of operation.
[0071] In Fig. 5, the second current source 260b is connected in series with a switch 290b between the negative input of the second comparator circuit 270b and the signal ground node. The control signal smp m controls the switch 290b such that, for each sample, the switch 290b is open during said first phase of operation and closed during said second phase of operation. Thereby, during the second phase of operation, the capacitor 250b of the second VTC 110b is gradually discharged and the voltage Vcb at the first terminal of the capacitor 250b of the second VTC 110b is thus gradually reduced. When it reaches below the first reference voltage Vrefa or the second reference voltage Vrefb (i.e. the one of the first reference voltage Vrefa and the second reference voltage Vrefb that is actually used in the embodiment), a transition is made at the output of the second comparator circuit 270b, which is the start of the second pulse P2, i.e. a first edge P2’ (see e.g. Fig. 6) of the second pulse P2. It should be mentioned that, in embodiments wherein the first reference voltage Vrefa is utilized in both VTCs, an offset is applied to such that the transition at the output of the second comparator circuit 270b is made responsive to the voltage Vcb at the first terminal of the capacitor 250b of the second VTC 110b being below the first reference voltage Vrefa offset by the above mentioned offset.
[0072] According to some embodiments, the VTCs 110a and 110b, the pulse-detector circuit 120, and the TDCs 130a and 130b are configured to operate as follows. The operation is described for a specific sample V[n], i.e. for a given value of n, of the input voltage V. The operation is the same for each sample V[n], i.e. for each value of n. The operation is illustrated with pulse waveforms in Fig. 6. Reference is made below to a first logic state and a second logic state. In the examples presented, the first logic state is considered to be ‘O’, generally represented with a “low” voltage level (such as 0V), and the second logic state is considered to be ‘ 1’, generally represented with a “high” voltage level (i.e. a voltage level higher than the “low” voltage level, e.g. equal to a supply voltage of the circuitry). However, in some embodiments, the first logic state and the second logic state could equally well be ‘ 1’ and ‘O’, respectively, with suitable modifications of the circuitry as would be readily understood by a skilled person. Furthermore, in Fig. 6, the pulses start with a rising edge Pl’, P2’, P3’, P4’ (sometimes called a first edge) and end with a falling edge Pl”, P2”, P3”, P4” (sometimes called a second edge). However, in some embodiments, pulses that are inverted with respect to those shown in Fig. 6 may be used, i.e. the first edge Pl’, P2’, P3’, P4’ is a falling edge and the second edge is Pl”, P2”, P3”, P4” is a rising edge.
[0073] In some embodiments, the first VTC 110a is configured to perform the sampling of the input voltage V, e.g. as described with reference to Fig. 4. That is, in these embodiments, the voltage V at the input 112a of the first VTC 110a is a time- continuous voltage. In other embodiments, the sampling may be performed by a sampling circuit (not shown in the drawings) preceding the first VTC 110a. That is, in these embodiments, the voltage V at the input 112a of the first VTC 110a is a time discrete (e.g. piecewise constant) voltage.
[0074] Similarly, in some embodiments, the second VTC 110b is configured to perform the sampling of the input voltage V, e.g. as described with reference to Fig. 5. That is, in these embodiments, the voltage V at the input 112b of the second VTC 110b is a time- continuous voltage. In other embodiments, the sampling may be performed by a sampling circuit (not shown in the drawings) preceding the second VTC 110b. That is, in these embodiments, the voltage V at the input 112b of the second VTC 110b is a time discrete (e.g. piecewise constant) voltage.
[0075] The first VTC 110a is configured to generate a first pulse Pl (see Figs. 6 and 7) delayed an amount of time tdl that corresponds to the magnitude of the sample of the input voltage V[n] in relation to the first reference voltage Vrefa (best seen in Fig. 7 as Va[n]). In Fig. 6, the starting time for the pulse Pl is labeled tl. Similarly, the second VTC 110b is configured to generate a second pulse P2 delayed an amount of time td2 corresponding to the magnitude of the sample of the input voltage V[n] in relation to the second reference voltage Vrefb (best seen in Fig. 7 as Vb[n]). In Fig. 6, the starting time for the pulse P2 is labeled t2. Furthermore, in Fig. 6, t2 occurs after tl. Of course, which of tl and t2 is first depends on the values of the input voltage V in relation to the reference voltages Vrefa, Vrefb. As illustrated in Fig. 6, the first and second pulses Pl and P2 are ended before the next sample time instant (n+l)T. The pulses Pl and P2 are generally ended by closing the sampling switches 280a, 280b such that the input terminal 101 is connected to the capacitors 250a, 250b and the inputs of the comparator circuits 270a, 270b. That is to say, by operating the sampling control signal smp p. In Figs 6 and 7, this happens at a time instant labeled t6.
[0076] In Fig. 6, the delays tdl and td2 are measured from the sampling time instant nT. However, since a differential representation of the input voltage V is provided by the single ended to differential VTC 400, the signal is represented by the difference between tdl and td2, so any time instant other than nT may be used as reference as well.
[0077] In a conventional differential time-domain ADC as outlined in the background section, a TDC would use one of Pl and P2, whichever comes first, as a start signal, and the other one of Pl and P2 as a stop signal, and thereby measure the absolute difference between tdl and td2 (and use some additional logic circuitry that keeps track of the sign of the difference). If this difference is small, many known TDC implementations have difficulties to accurately measure the difference, resulting in an overall poor linearity of such a conventional differential time-domain ADC. According to embodiments of the ADC 50 disclosed herein, this problem is alleviated by means of the pulse-detector circuit 120, which guarantees a certain (nonzero) minimum duration of time for the TDCs 130a and 130b to measure, as further described below.
[0078] The pulse-detector circuit 120 is configured to make a transition from the first logic state (‘0’) to the second logic state (‘ 1’) at the first output 124a of the pulse-detector circuit 120 in response to the start of the first pulse Pl, as illustrated in Fig. 6. In Fig. 6, this happens at a time instant labeled t3. Furthermore, pulse-detector circuit 120 is configured to make a transition from the first logic state (‘0’) to the second logic state (‘ 1’) at the second output 124b of the pulse-detector circuit 120 in response to the start of the second pulse P2. In Fig. 6, this happens at a time instant labeled t4. Moreover, the pulse-detector circuit is configured to reset both its first and its second output 124a, 124b to the first logic state (‘0’) in response to both the first output 124a and the second output 124b of the pulse-detector circuit 120 being set in the second logic state (‘ 1’). In Fig. 6, this happens at a time instant labeled t5.
[0079] Thereby, as illustrated in Fig. 6, a third pulse P3 is generated at the first output 124a of the pulse-detector circuit and a fourth pulse P4 is generated at the second output 124b of the pulse-detector circuit 120. The difference between the durations of the third pulse P3 and the fourth pulse P4 is a time-domain representation of the input voltage V[n], Advantageously, said difference is proportional to the input voltage V[n], or specifically to V[n]-V0 where V0 is a non-zero middle voltage that will be further explained in later sections.
[0080] The first TDC 130a is configured to receive the third pulse P3 and generate a first digital value xa[n] corresponding to the duration of the third pulse P3. The second TDC 130b configured to receive the fourth pulse P4 and generate a second digital value xb[n] corresponding to the duration of the fourth pulse P4.
[0081] Due to inherent delays in the electronic components used to implement the pulsedetector circuit, the reset of the outputs 124a and 124b of the pulse-detector circuit 120 does not occur instantaneously when both outputs 124a and 124b have been set to the second logic state (‘ 1’). This means that the pulses P3 and P4 have a guaranteed minimum duration. In Fig. 6, it is the fourth pulse P4 that has this minimum duration (which is t5-t4), but depending on the value of the input voltage V[n], it may instead be the third pulse P3 that has the minimum duration. Thus, the durations to be converted by the TDCs 130a and 130b are not shorter than said guaranteed minimum duration. This helps alleviate linearity problems that many TDCs experience when measuring and converting short time durations, e.g. since the TDCs 130a and 130b are operated in a region where the difference between their respective outputs is a relatively linear function of the difference in pulse lengths at their respective inputs, and thus also a relatively linear function of the input voltage V[n], In this region, any non-linearity in the TDCs 130a and 130b, which may cause problems for shorter pulse durations, influences both TDCs 130a and 130b to the same extent, and thus cancels in the differential domain. It should be noted that the components in the pulse-detector circuit may be deliberately designed or dimensioned to provide a suitable value of said guaranteed minimum delay.
[0082] In some embodiments, the first digital value xa[n] and the second digital value xb[n] may form an output sample of the ADC 50. That is, in these embodiments, the output of the ADC 50 is in the form of a vector with two elements, xa[n] and xb[n]. In such embodiments, a signal content of the output sample is represented by the difference between xa[n] and xb[n].
[0083] In order to more fully provide background for the workings of the single ended to differential VTC 400 and the ADC 50, reference is made to the exemplary waveforms of Fig. 7 corresponding to the waveforms of Fig. 6. The waveforms of Fig. 7 will be explained with reference to the exemplary embodiment of the VTCs 110a, 110b presented in reference to Fig. 4 and Fig. 5 but corresponding waveforms may very well be provided with other variants of the VTCs 110a, 110b. At the sampling time nT in Fig. 7, the VTCs 110a, 110b enter the second phase of operation which means that their respective sampling switches 280a, 280b are open.
[0084] In Fig. 7, five waveforms are shown where the top waveform shows a voltage Vca (solid line) at the first terminal of the capacitor 250a of the first VTC 110a overlaid with a voltage Vcb (dash-dotted line) at the first terminal of the capacitor 250b of the second VTC 110b.
[0085] At the sampling time nT, the first current source 260a of the first VTC 110a will charge the capacitor 250a of the first VTC 110a causing the voltage Vca at the first terminal of the capacitor 250a to increase. At the time labeled tl, the voltage Vca at the first terminal of the capacitor 250a reaches the first reference voltage Vrefa. Responsive to this, the first comparator circuit 270a is triggered and the first edge Pl’ of the first pulse Pl is provided at the output 114a of the first comparator circuit 270a. This is shown in the second waveform from the top in Fig. 7.
[0086] Further, at the sampling time nT, the second current source 260b of the second VTC 110b will discharge the capacitor 250b of the second VTC 110b causing the voltage Vcb at the first terminal of the capacitor 250b to decrease. At the time labeled t2, the voltage Vcb at the second terminal of the capacitor 250b reaches the second reference voltage Vrefb. Responsive to this, the second comparator circuit 270b is triggered and the first edge P2’ of the second pulse P2 is provided at the output 114b of the second comparator circuit 270b. This is shown in the third waveform from the top in Fig. 7.
[0087] Responsive to the first edge Pl’ of the first pulse Pl, the pulse-detector circuit 120 is configured to generate a first edge P3’ of the third pulse P3 at the first output 124a of the pulse-detector circuit 120. In Fig. 7, this occurs at the time instant labeled t3 and is shown in the second waveform from the bottom.
[0088] In addition to this, responsive to the first edge P2’ of the second pulse P2, the pulsedetector circuit 120 is configured to generate a first edge P4’ of the fourth pulse P4 at the second output 124b of the pulse-detector circuit 120. In Fig. 7, this occurs at the time instant labeled t4 and is shown in the bottom waveform.
[0089] The pulse-detector circuit 120 may further be configured to, responsive to having generated both the first edge P3’ of the third pulse P3 and the first edge P4’ of the fourth pulse P4 (i.e. both comparators circuits 270a, 270b are triggered), generate a second edge P3” of the third pulse P3 and a second edge P4” of the fourth pulse P4. In Fig. 7, this happens at a time instant labeled t5.
[0090] The VTCs 110a, 110b are further configured to generate a respective second edge Pl”, P2” of the first and second pulses Pl, P2. The second edges Pl”, P2” are generally generated by closing the sampling switches 280a, 280b such that the input terminal 101 is connected to the capacitors 250a, 250b and the inputs of the comparator circuits 270a, 270b. That is to say, by operating the sampling control signal smp_p. This occurs at a time instant labeled t6 which occurs before the next sample time instant (n+l)T.
[0091] In Fig. 7, a middle voltage V0 is indicated at a voltage that is voltage-wise in a middle between the first reference voltage Vrefa and the second reference voltage Vrefb. If it is assumed that a magnitude of a slope of the voltage Vca across the capacitor 250a of the first VTC 110a during charging of that capacitor 250a is equal to a magnitude of the slope of the voltage Vcb across the capacitor 250b of the second VTC 110b during discharging of that capacitor 250b, then the first digital value xa[n] will be assume its minimum value if the input voltage V[n] is below the middle voltage V0 and the second digital value xb[n] will assume its minimum value if the input voltage V[n] is above the middle voltage VO.
[0092] The single ended input voltage sample V[n] is represented by a first voltage difference Va[n] of the first VTC 110a and a second voltage difference Vb[n] of the second VTC 110b. The first voltage difference Va[n] and the second voltage difference Vb[n] form a pair where Va[n] = Vrefa — 7[n] and Vb[n] = 7[n] — Vrefb. This pair represents the input voltage V[n] in relation to the middle voltage VO as the difference between the signals, i.e. Vb [n] — Va[n] = 2 * 7[n] — (Vrefa + Vrefb) = 2 * (7[n] — 70). As can be seen the difference between Vb[n] and Va[n] has a linear relation to V[n], with an offset so that the difference is zero when 7[n] = (Vrefa + Vrefb) / 2 = 70. As explained with reference to Fig. 6, the difference between the durations of the third pulse P3 and the fourth pulse P4 is a time-domain representation of the input voltage Vb[n]- Va[n] = 7[n] — Vrefb — Vrefa + 7[n],
[0093] Generally, the delay tdl indicated by the first pulse Pl may be described as tdl = k(Vrefa — 7[n]) + ma, the delay td2 indicated by the second pulse P2 may be described as td2 = fc(7[n] — Vrefb) + mb where k, ma and mb are constants and ma=mb. An amplitude of the input voltage V[n] is consequently transferred to timedomain as td2-tdl. However, such an exact ideal relation may be challenging to accomplish in practice due to manufacturing tolerance, non-linearity, noise, etc.
[0094] In the following, the workings of the pulse-detector 120 will be described in further detail. Fig. 8 illustrates an embodiment of the pulse-detector circuit 120. In Fig. 8, the pulse-detector circuit 120 comprises a first flip-flop 200a and a second flip-flop 200b, each having a data input “D”, a clock input “elk”, a reset input “reset”, and an output “Q”. This type of flip-flop is commonly referred to as a “D flip-flop”. The data input “D” of each of the first flip-flop 200a and the second flip-flop 200b is configured to receive a constant signal corresponding to the second logic state (‘ 1’).
[0095] The clock input “elk” of the first flip-flop 200a is configured to receive the first pulse Pl. In Fig. 8, the clock input “elk” of the first flip-flop 200a is connected to the first input 122a of the pulse-detector circuit 120 for this purpose. The clock input “elk” of the second flip-flop 200b is configured to receive the second pulse P2. In Fig. 8, the clock input “elk” of the second flip-flop 200b is connected to the second input 122b of the pulse-detector circuit 120 for this purpose.
[0096] Hence, the clock inputs of the flip-flops 200a and 200b are not driven by what would normally be considered as clock signals, but by the pulses Pl and P2, respectively. The term clock input is, however, used since this is common terminology for such inputs in flip-flops.
[0097] Furthermore, in Fig. 8, the output “Q” of the first flip-flop 200a is connected to the first output 124a of the pulse-detector circuit 120. Moreover, in Fig. 8, the output “Q” of the second flip-flop 200b is connected to the second output 124b of the pulse-detector circuit 120.
[0098] In the following discussion, the flip-flops 200a and 200b are assumed to be positive edge triggered, i.e. when a rising edge (or positive edge) is presented at their respective clock inputs “elk”, they transfer the logic value present at their respective data inputs “D” to their respective outputs “Q”. Let us further assume that the data outputs “Q” of both flip-flops 200a and 200b are initially in the first logic state (‘0’) and consider the pulse wave forms in Fig. 6 or Fig. 7 as an elucidating example. At time instant tl, a rising edge (i.e the first edge PL of the first pulse Pl) is presented at the clock input “elk” of the first flip flop 200a. Since the data input “D” of the first flip-flop 200a is set to the second logic state (‘ 1’), the output “Q” of the first flip-flop 200a makes a transition to the second logic state (‘ 1’) in response to this rising edge, which results in the start of pulse P3. In Fig. 6 and Fig. 7, this happens at time instant t3, slightly after tl due to a slight delay in the first flip-flop 200a. Similarly, at time instant t2, a rising edge (i.e the first edge P2’ of the second pulse P2) is presented at the clock input “elk” of the second flip flop 200b. Since the data input “D” of the second flip-flop 200b is also set to the second logic state (‘ 1’), the output “Q” of the second flip-flop 200b makes a transition to the second logic state (‘ 1’) in response to this rising edge, which results in the start of pulse P4. In Fig. 6 and Fig. 7, this happens at time instant t4, slightly after t2 due to a slight delay in the second flip-flop 200b.
[0099] In order to reset the first and second outputs 124a, 124b of the pulse-detector circuit 120 to the first logic state (‘0’) in response to both the first output 124a and the second output 124b of the pulse-detector circuit 120 being set in the second logic state (‘ 1’), embodiment of the pulse-detector circuit 120 illustrated in Fig. 8 comprises a logic circuit 210. The logic circuit 210 has a first input 212a connected to the output “Q” of the first flip-flop 200a and a second input 212b connected to the output “Q” of the second flip-flop 200b. Furthermore, the logic circuit 210 has an output 214 connected to the reset inputs “reset” of the first flip-flop 200a and the second flip-flop 200b for resetting the first flip-flop 200a and the second flip-flop 200b in response to the outputs Q of both the first flip-flop 200a and the second flip-flop 200b being set in the second logic state (‘ 1’). In Fig. 8, it is assumed that the reset inputs “reset” of the flip-flops 200a and 200b are “active high”, i.e. that the flip-flops 200a and 200b are reset when a ‘ 1’ is presented at their respective reset inputs “reset”. Then, the functionality of the logic circuit 210 as described above may be implemented with an AND gate, as illustrated in Fig. 8. If, on the other hand, the reset inputs “reset” of the flip-flops 200a and 200b are “active low”, i.e. the flip-flops 200a and 200b are reset when a ‘0’ is presented at their respective reset inputs “reset”, the functionality of the logic circuit 210 as described above may be implemented with a NAND gate. Other types of implementations may be used in other scenarios, e.g. if the first logic state would be ‘ 1’ and the second logic state would be ‘O’, etc.
[0100] An inherent delay in the logic circuit 210 and the flip-flops 200a and 200b provides the above-mentioned guaranteed minimum duration of the pulses P3 and P4. In order to provide a desired length of said guaranteed minimum duration, different design techniques may be used when designing the logic circuit 210, as would be appreciated by a person skilled in the art of electronic design. For instance, with reference to Fig. 8, the AND gate may be intentionally designed to be “slow”, e.g. using a relatively small width-over-length (W7L) ratios in the transistors in the AND gate, using currentstarving techniques, and / or using a relatively high capacitive load at the output of the AND gate, etc. Alternatively or additionally, delay elements may be connected to the inputs and / or the output of the AND gate.
[0101] In some embodiments, the logic circuit 210 is further configured to reset the first flip-flop 200a and the second flip-flop 200b in response to a reset pulse. This may be useful if, for instance, the first or second flip-flop 200a, 200b is stuck in an erroneous state, e.g. due to noise or an input signal outside the intended input range of the ADC 50, such that the “standard” reset mechanism based on both outputs 124a and 124b of the pulse-detector circuit 120 being in the second logic state (‘ 1’) is not triggered for one or more samples. Then, by including circuitry 210 (see Fig. 9) in the ADC 50 configured to provide said reset pulse in preparation of converting each sample, it may be guaranteed that both outputs 124a and 124b of the pulse-detector circuit 120 are initially in the first logic state (‘0’) for each sample. An example of such an embodiment of the logic circuit 210 is illustrated in Fig. 9. In Fig. 9, the logic circuit 210 comprises the AND gate shown in Fig. 8 and an OR gate. Instead of having the output of the AND gate connected to the reset inputs “reset” of the flip-flops 200a and 200b, the output of the AND gate is instead connected to a first input of the OR gate, and the output of the OR gate is connected to the reset inputs “reset” of the flip-flops 200a and 200b. A second input of the OR gate is connected to the sampling control signal smp _p. The sampling control signal smp _p will provide a reset pulse. With this configuration, the flip-flops 200a and 200b may be reset if said reset pulse is provided at the third input 212c or a ‘ 1’ is provided at the output of the AND gate in the logic circuit 210.
[0102] The TDC 130 may be implemented in a number of different ways and for completeness, an example of a pulse-shrinking TDC 130 is shown in Fig. 10. A pulseshrinking TDC 130 is sometimes referred to as a tapped delay line. The reference number 130 is used in Fig. 10 as a generic reference number representing both the first TDC 130a and the second TDC 130b (i.e. they can both be implemented in the same way). In the same way, the reference number 132 is used in Fig. 10 as a generic reference number representing both the input 132a of the first TDC 130a and the input 132b of the second TDC 130b. Similarly, the reference number 134 is used in Fig. 10 as a generic reference number representing both the output 134a of the first TDC 130a and the output 134b of the second TDC 130b.
[0103] In Fig. 10, the output 134 has N bits 134 1 - 134_N. Furthermore, the pulseshrinking TDC 130 in Fig. 10 comprises N flip-flops of the type described in the context of the embodiments of the pulse-detector circuit 120 illustrated in Figs. 6-7. Moreover, in Fig. 10, the pulse-shrinking TDC 130 comprises a chain of N pulse- shrinking elements 350 1 - 350_N. The output “Q” of each flip-flop is connected to a unique one of the N bits 134 1 - 134 N of the output 134 of the pulse-shrinking TDC 130. Furthermore, the clock input “elk” of each flip-flop is connected to the output of a unique one of pulse-shrinking elements 350 1 - 350_N. Moreover, the reset input “reset” of all flip-flops is driven by the sampling-control signal smp _p, whereby, for each sample, the outputs “Q” of all flip-flops are initially reset to the first logic state (‘0’). The data inputs “D” of all flip-flops are set to the second logic state (‘ 1’). As the pulse provided at the input 132 (from the pulse-detector circuit 120) propagates along the chain of pulse-shrinking elements 350_N - 350 1, the outputs “Q” of a number of flip-flops, namely those whose clock inputs are connected to outputs of the pulseshrinking elements that the pulse propagates through before vanishing, will be set to the second logic state (‘ 1’). The output bits 134 1 - 134 N thus provide a thermometer- coded digital representation of the duration of the pulse provided at the input 132. Pulse-shrinking TDCs are known in the art of time-to-digital conversion, and the functionality is not described in any further detail herein.
[0104] An advantage of pulse-shrinking TDCs is that they have a relatively low energy consumption. For most pulse-durations, only a subset of the flip-flops will be reached by a pulse at its clock input “elk” and contribute to the dynamic energy consumption of the TDC. However, the inventors have identified the pulse-shrinking TDC as one of the types of TDCs that suffer from relatively poor linearity when the duration to be converted is relatively short, as mentioned above. This problem is alleviated with the use of the pulse-detector circuit 120, which, as mentioned above, provides a guaranteed minimum duration of the pulses that are input to the TDCs 130a and 130b.
[0105] From the above, it may be concluded that the difference between the first digital value xa[n] and the second digital value xb[n] is proportional to a difference in timing between the first edge Pl’ of the first pulse Pl and the first edge P2’ of the second pulse P2. Specifically, the first digital value xa[n] will be proportional to how much earlier the first edge Pl’ of the first pulse Pl is than the first edge P2’ of the second pulse P2. That is to say, the first digital value xa[n] is proportional to an amount of time the first edge Pl’ of the first pulse Pl precedes the first edge P2’ of the second pulse P2. In other words, the first digital value xa[n] is proportional to how much closer the input voltage V is to the first reference voltage Vrefa than to the second reference voltage Vrefb. Correspondingly, the second digital value xb[n] will be proportional to how much earlier the first edge P2’ of the second pulse P2 is than the first edge Pl’ of the first pulse Pl. That is to say, the second digital value xb[n] is proportional to an amount of time the first edge P2’ of the second pulse P2 precedes the first edge Pl’ of the first pulse Pl. In other words, the second digital value xb[n] is proportional to how much closer the input voltage V is to the second reference voltage Vrefb than to the first reference voltage Vrefa.
[0106] Generally, the first digital value xa[n] will assume a minimum value if the input voltage V is closer to the second reference voltage Vrefb than to the first reference voltage Vrefa and the second digital value xb[n] will assume its minimum value if the input voltage V is closer to the first reference voltage Vrefa than to the second reference voltage Vrefb.
[0107] In Fig. 7, the slope of the voltage Vca across the capacitor 250a of the first VTC 110a during charging of the capacitor 250a of the first VTC 110a is equal to (but with opposite sign) the slope of the voltage Vcb across the capacitor 250b of the second VTC 110b during discharging of the capacitor 250b of the second VTC 110b. This is an advantage of designing the first current source 260a analogously to the second current source 260b such that the charge current provided by the first current source 260a is equal to the discharge current provided by the second current source 260b. This is true under the assumption that the capacitor 250a of the first VTC 110a is equal to the capacitor 250b of the second VTC 110b. It should be mentioned that the capacitor 250a of the first VTC 110a and the capacitor 250b of the second VTC 110b may, for various reasons, be designed differently. The same is true for the first current source 260a and the second current source 260b. A difference between the capacitors 250a, 250b (or the current sources 260a, 260b) may be compensated by introducing a proportional difference if the current sources 260a, 260b. This will keep the slope of the voltage Vca across the capacitor 250a of the first VTC 110a during charging of the capacitor 250a of the first VTC 110a equal to the slope of the voltage Vcb across the capacitor 250b of the second VTC 110b during discharging of the capacitor 250b of the second VTC 110b. However, in some embodiments, it may be advantageous to design the current sources 260a, 260b and / or the capacitors 250a, 250b such that the slope of the voltage Vca across the capacitor 250a of the first VTC 110a during charging of the capacitor 250a of the first VTC 110a is different from to the slope of the voltage Vcb across the capacitor 250b of the second VTC 110b during discharging of the capacitor 250b of the second VTC 110b.
[0108] As mentioned, preferably the slope of the voltage Vca across the capacitor 250a of the first VTC 110a during charging of the capacitor 250a of the first VTC 110a is equal to the slope of the voltage Vcb across the capacitor 250b of the second VTC 110b during discharging of the capacitor 250b of the second VTC 110b. However, the skilled person will understand, after absorbing all teachings of the present disclosure, that differences between the slopes may be compensated in e.g. the interpretation of the first digital value xa[n] and the second digital value xb[n].
[0109] As mentioned above, the second edges P3”, P4” of the third and fourth pulses P3, P4 are generated responsive to both the first edges P3’, P4’ of the third and fourth pulses P3, P4 having been generated. Due to the internal delay of the pulse detector circuit 120 the shortest pulses P3, P4 output from the pulse-detector circuit 120 last the minimum duration. This means that the digital values xa[n], xb[n] provided by the TDCs 130 will be offset by a digital value corresponding to the minimum duration. In some embodiments, e.g. the circuit 150 may be configured to compensate for this offset.
[0110] The time-domain ADC 50 presented in previous sections alleviates problem of the prior art with reduced linearity when pulse lengths are comparably short. However, a conversion time, a latency, an area and a power consumption of the time-domain ADC 50 will depend (linearly) on a measurement time of the input voltage V, see e.g. tl and t2 in Fig. 7 together with a reciprocal resolution (1 / TLSB) of the time-domain ADC 50. This strong dependency on the resolution may cause a trade-off between a maximum measurement time and a resolution in order to meet a specific conversion rate. To increase the maximum converter sample rate, without compromising the resolution, an alternative, optional, converter architecture will be presented in the following.
[0111] The inventors of the present disclosure have further realized that by adding auxiliary reference voltages to the VTCs 110a, 110b, a duration, i.e. a pulse width, of the signals P3, P4 provided to the TDCs 130a, 130b may be reduced. This auxiliary reference voltage may be utilized to scale a required full scale range of the TDCs 130a, 130b and thereby reduce the conversion time. From this, power consumption, area and conversion time of the differential time-domain ADC 50 may reduced to one third. The timedomain ADC 50 of the present disclosure allows for an increase of a maximum sample rate, and a class-C like efficiency gain.
[0112] In Fig. 11, a time-domain ADC 50 that is a variant of the time-domain ADCs 50 of Figs. 3a-c is shown. The time-domain ADC 50 of Fig. 11 is similar to the time-domain ADCs 50 of Figs. 3a-c and only the differences will be explained in the following. In Fig. 11, the first VTC 110a is provided with a first output 114a and an auxiliary output 115a. Correspondingly, the second VTC 110b is also provided with a first output 114b and an auxiliary output 115b. The pulse detector circuit 120 is coupled to the outputs 114a, 114b, 115a, 115b of the VTCs 110a, 110b. In Fig. 11, the control circuit 120 has a first input 122a connected to the output 114a of the first VTC 110a, a second input 122b connected to the output 114b of the second VTC 110b, a first auxiliary input 123a connected to the auxiliary output 115a of the first VTC 110a and a second auxiliary input 123b connected to the auxiliary output 115b of the second VTC 110b. The pulse detector circuit 120 in Fig. 11 is provided with a first output 124a, a second output 124b and a third output 124c.
[0113] In Fig. 11 the ADC 50 comprises a first TDC 130a provided with an input 132a connected to the first output 124a of the pulse detector circuit 120. The ADC 50 further comprises a second TDC 130b provided with an input 132b connected to the second output 124b of the pulse detector circuit 120. The functionality of a TDC 130a, 130b is analogues to the description provided with reference to Fig. 10. The first TDC 130a is provided with an output 134a and configured to provide the first digital value xa[n] at the output 134a. The second TDC 130b is provided with an output 134b and configured to provide the second digital value xb[n] at the output 134b. A third digital value xc[n] is provided by the pulse detector circuit 120 at the third output 124c of the control circuit 120. In Fig. 11, first digital value xa[n], the second digital value xb[n] and the third digital value xc[n] form an output sample of the ADC 50. That is, in these embodiments, the output of the ADC 50 is in the form of a vector with three elements, xa[n], xb[n], xc[n]. An example of a configuration of a first VTC 110a usable with the ADC in Fig. 11 is illustrated in Fig. 12a, where the first reference voltage Vrefa is provided to the negative input of the first comparator circuit 270a. In Fig. 12a, one terminal of the capacitor 250a is connected to a signal ground node. Furthermore, in Fig. 12a, the other terminal of the capacitor 250a is connected to a positive input of the first comparator circuit 270a and a positive port of an auxiliary comparator circuit 270aa. Moreover, in Fig. 12a, a sampling switch 280a is connected between the input terminal 112a of the first VTC 110a and the positive input terminals of the comparator circuits 270a, 270aa. A sampling control signal smp _p controls the sampling switch 280a such that, for each sample, the sampling switch 280a is closed during said first phase of operation and open during said second phase of operation. The sampling control signal smp _p may also be used as a reset pulse of the pulse detector circuit 210. For the embodiment illustrated in Fig. 12a, the voltage V provided at the input terminal 112a of the first VTC 110a may be either time continuous or time discrete (e.g. piecewise constant). In the case of a time-continuous input voltage V, said sample V[n] is the value of V at the end of the first phase of operation.
[0114] In Fig. 12a, the first current source 260a is connected in series with a switch 290a between the positive inputs of the comparator circuits 270a, 270aa and the signal ground node. A control signal smp m controls the switch 290a such that, for each sample, the switch 290a is open during said first phase of operation and closed during said second phase of operation. Thereby, during the second phase of operation, the capacitor 250a is gradually charged and the voltage Vca at the first terminal of the capacitor 250a is thus gradually increased. When the voltage Vca at the first terminal of the capacitor 250a reaches above the first reference voltage Vrefa, a transition is made at the output of the first comparator circuit 270a, which is the start of the first pulse Pl, i.e. the first edge Pl’ of the first pulse Pl. Correspondingly, when the voltage Vca at the first terminal of the capacitor 250a reaches above the first auxiliary reference voltage Vrefaa, a transition is made at the output of the auxiliary comparator circuit 270aa, which is the start of a first auxiliary pulse Pla, i.e. a first edge Pla’ of the first auxiliary pulse Pla (see Fig. 13). An example of a configuration of a second VTC 110b usable with the ADC 50 of Fig. 11 is illustrated in Fig. 12b, where the second reference voltage Vrefb is provided to a positive input of the first comparator circuit 270b. The second auxiliary reference voltage Vrefba is provided to a positive input of the auxiliary comparator circuit 270ba. In Fig. 12b, one terminal of the capacitor 250b is connected to the signal ground node. Furthermore, in Fig. 12b, the other terminal of the capacitor 250b is connected to negative inputs of the comparator circuits 270b, 270ba. Moreover, in Fig. 12b, a sampling switch 280b is connected between the input terminal 112b of the second VTC 110b and the negative input terminals of the comparator circuits 270b, 270ba. The sampling control signal smp _p controls the sampling switch 280b such that, for each sample, the sampling switch 280b is closed during said first phase of operation and open during said second phase of operation. As above, for the embodiment illustrated in Fig. 12a, the voltage V provided at the input terminal 112b of the second VTC 110b may be either time continuous or time discrete (e.g. piecewise constant). In the case of a time- continuous voltage V, said sample V[n] is the value of V at the end of the first phase of operation.
[0115] In Fig. 12b, the second current source 260b is connected in series with a switch 290b between the negative inputs of the comparator circuits 270b, 270ba and the signal ground node. A control signal smp m controls the switch 290b such that, for each sample, the switch 290b is open during said first phase of operation and closed during said second phase of operation. Thereby, during the second phase of operation, the capacitor 250b is gradually discharged and the voltage Vcb at the first terminal of the capacitor 250b is thus gradually decreased. When the voltage Vcb at the first terminal of the capacitor 250b reaches below the second reference voltage Vrefb, a transition is made at the output of the first comparator circuit 270b, which is the start of the second pulse P2, i.e. the first edge P2’ of the second pulse P2. Correspondingly, when the voltage Vcb at the first terminal of the capacitor 250b reaches below the second auxiliary reference voltage Vrefba, a transition is made at the output of the auxiliary comparator circuit 270ba, which is the start of a second auxiliary pulse P2a, i.e. a first edge P2a’ of the second auxiliary pulse P2a (see Fig. 13). The operation of the time-domain ADC 50 of Fig. 11 will be explained by the pulse waveforms in Fig. 13. Fig. 13 corresponds to Fig. 7 in the naming of the signals and times. Further, in Fig. 13, there is substantially no delay in the pulse detector circuit 120 when it comes to generating the first edges at its outputs (tl and t3 occur at the same time), this is to make the waveforms clearer. In Fig. 13, the first auxiliary reference voltage Vrefaa is negatively offset from a middle voltage V0 and the second auxiliary reference voltage Vrefba is offset correspondingly positively from the middle voltage V0. The skilled person will understand how to modify the plots and the effects on the results if these voltages Vrefa, Vrefaa, Vrefb, Vrefba are different. In some embodiments of the present disclosure, the first auxiliary reference voltage Vrefaa and the second auxiliary reference voltage Vrefba are equal.
[0116] Fig. 13, the first and second auxiliary reference voltages Vrefaa, Vrefba are offset from the middle voltage V0. In Fig. 13, at the sampling time instant nT, the capacitors 250a, 250b of the VTCs 110a, 110b are at the input voltage V[n], As the sampling switches 280a, 280b opens, the switches 290a, 290b may close and charging of the capacitor 250a of the first VTC 110a and discharge of the capacitors 250b of the second VTC 110b may be started. In Fig. 13, the input voltage V[n] is above the second auxiliary reference voltage Vrefba, and thereby also above the second reference voltage Vrefb and the first auxiliary reference voltage Vrefaa. This means that the auxiliary comparator circuit 270ba of the second VTC 110b will trigger, but not the auxiliary comparator circuit 270aa of the first VTC 110a. At the fourth time t4, the value Vcb at the first terminal of the capacitor 250b of the second VTC 110b crosses the second auxiliary reference voltage Vrefba and the auxiliary comparator 270ba of the second VTC 110b triggers and generates the first edge P2a’ of the second auxiliary pulse P2a at the auxiliary output 115b of the second VTC 110b. This is the temporally first of the first edge P2’ of the second pulse P2 and the first edge P2a’ of the second auxiliary pulse P2a, the pulse detector circuit 120 generates the first edge P4’ of the fourth pulse P4 at the second output 124b of the control circuit 120. At the first time tl, the value Vca at the first terminal of the capacitor 250a of the first VTC 110a crosses the first reference voltage Vrefa and the first comparator 270a of the first VTC 110a generates the first edge Pl’ of the first pulse Pl at the first output 114a of the first VTC 110a. This is the temporally first of the first edge Pl’ of the first pulse Pl and the first edge Pl a’ of the first auxiliary pulse Pla, the pulse detector circuit 120 generates the first edge P3’ of the third pulse P3 at the first output 124a of the control circuit 120 in response thereto. Responsive to having generated first edges P3’, P4’ of the third and fourth pulses P3, P4, the outputs 124a, 124b of the control circuit 120 are reset by generating the second edge P3” of the third pulse P3 at the first output 124a of the control circuit 120 and a second edge P4” of the fourth pulse P4 at the second output 124b of the control circuit 120. This occurs at the fifth time t5. Discharging of the capacitor 250b of the second VTC 110b continues, and at the second time t2, the value Vcb at the negative input of the first comparator 270b of the second VTC 110b reaches the second reference voltage Vrefb and the first comparator 270b of the second VTC 110b triggers and generates the first edge P2’ of the second pulse P2 at the first output 114b of the second VTC 110b. As the pulse detector circuit 120 already has generated the first edge P4’ of the fourth pulse P4 responsive to the first edge P2a’ of the second auxiliary pulse P2a, the first edge P2’ of the second pulse P2 may be disregarded by the pulse detector circuit 120. At the sixth time t6, the first, the second and the second auxiliary pulses Pl, P2, P2a are ended by generation of the second edge Pl” of the first pulse Pl at the first output 114a of the first VTC 110a, the second edge P2” of the second pulse P2 at the first output 114b of the second VTC 110b and a second edge P2a” of the second auxiliary pulse P2a at the auxiliary output 115b of the second VTC 110b. This occurs before the next sample time instant (n + 1)T and may be triggered, as previously described, by the sample control signal smp_p.
[0117] From Fig. 13 it is clearly shown that by adding the auxiliary comparators 270aa, 270ba to the VTCs 110a, 110b a duration of a pulse to be converted by the TDCs 130a, 130b is reduced. The waveforms in Fig. 7, are the result of the ADCs 50 of Figs. 3a-c being provided with substantially the same input voltage V[n] as that of Fig. 13. In both Figs. 7 and 13, a conversion time, i.e. a conversion time of the TDCs 130a, 130b, is determined by the difference in duration between the third pulse P3 and the fourth pulse P4. This difference will be reduced for substantially all values of the input voltage V[n] (except when the auxiliary voltages Vrefaa, Vrefba and the input voltage V[n] are all equal). As a result, a conversion range (e.g. a depth) of the TDC 130a, 130b may be reduced, saving power and area of the ADC 50. Further to this, a sampling rate of the ADC 50 may be increased.
[0118] As the skilled person will appreciate, the pulse detector circuit 120 for the ADC of Fig. 11 is advantageously slightly modified compared to the pulse detector circuit 120 of Figs. 8 and 9. To this end, Figs. 15a and 15b each show partial pulse detector circuits 120a, 120b of a pulse detector circuit 120 suitable for use with the ADC 50 of Fig. 11. A first partial pulse detector circuit 120a of the pulse detector circuit 120, Fig. 15a, is configured to provide the third pulse P3 at the first output 124a of the pulse detector circuit 120 and a part xca[n] of the third digital value xc[n], A second partial pulse detector circuit 120b of the pulse detector circuit 120, Fig. 15b, is configured to provide the fourth pulse P4 at the second output 124b of the pulse detector circuit 120 and a part xcb[n] of the third digital value xc[n].
[0119] Fig. 14 shows a block diagram of a control circuit 120 according to some embodiments of the present disclosure. The control circuit 120 comprises a first partial control circuit 120a and a second partial control circuit 120b.
[0120] The first partial control circuit 120a is configured to receive the first pulse Pl at the first input 122a of the control circuit 120. The first partial control circuit 120a is configured to receive the first auxiliary pulse Pla at the first auxiliary input 123a of the control circuit 120. The first partial control circuit 120a is configured to provide the third pulse P3 at an output 125a of the first partial control circuit 120a. The first partial control circuit 120a is further configured to provide a part xca[n] of the third digital value xc[n] at an auxiliary output 125aa of the first partial control circuit 120a.
[0121] The second partial control circuit 120b is configured to receive the second pulse P2 at the second input 122b of the control circuit 120. The second partial control circuit 120b is configured to receive the second auxiliary pulse P2a at the second auxiliary input 123b of the control circuit 120. The second partial control circuit 120b is configured to provide the fourth pulse P4 at an output 125b of the second partial control circuit 120b. The second partial control circuit 120b is further configured to provide a part xcb[n] of the third digital value xc[n] at an auxiliary output 125ba of the second partial control circuit 120b. Figs. 15a and 15b, each show partial control circuits 120a, 120b that may be comprised in the control circuit 120 described above in reference to Fig. 14. In Fig. 15a, the first partial control circuit 120a comprises a first flip-flop 201a and a second flipflop 202a, each having a data input “D”, a clock input “elk”, a reset input “reset”, and an output “Q”. These flip-flops 201a, 202a are advantageously corresponding to flipflops 200a, 200b that was presented with reference to Figs. 8 and 9. As will be clear from the following, the clock inputs “elk” of the flip-flops 201a, 202a are not driven by what would normally be considered as clock signals, but by the pulses Pl, Pla respectively. The term clock input is, however, used since this is common terminology for such inputs in flip-flops.
[0122] The clock input “elk” of the first flip-flop 201a is configured to receive the first pulse Pl. In Fig. 15a, the clock input “elk” of the first flip-flop 201a is connected to the first output 114a of the first VTC 110a for this purpose.
[0123] The clock input “elk” of the second flip-flop 202a is configured to receive the first auxiliary pulse Pla. In Fig. 15a, the clock input “elk” of the second flip-flop 202a is connected to the auxiliary output 115a of the first VTC 110a for this purpose.
[0124] Furthermore, in Fig. 15a, the outputs “Q” of the first flip-flop 201a and the second flip-flop 202a are connected to inputs of an OR-gate 204a of the first partial control circuit 120a. The output 125a of the first partial control circuit 120a is provided by an output of the OR-gate 204a and a part of the third digital value xca[n] is provided by the output “Q” of the second flip-flip 202a.
[0125] In Fig. 15b, the second partial control circuit 120b comprises a first flip-flop 201b and a second flip-flop 202b, each having a data input “D”, a clock input “elk”, a reset input “reset”, and an output “Q”. These flip-flops 201b, 202b are advantageously corresponding to the flip-flops presented with reference to Fig. 15a.
[0126] The clock input “elk” of the first flip-flop 201b is configured to receive the second pulse P2. In Fig. 15b, the clock input “elk” of the first flip-flop 201b is connected to the first output 114b of the second VTC 110b for this purpose.
[0127] The clock input “elk” of the second flip-flop 202b is configured to receive the second auxiliary pulse P2a. In Fig. 15b, the clock input “elk” of the second flip-flop 202b is connected to the auxiliary output 115b of the second VTC 110b for this purpose. Furthermore, in Fig. 15b, the outputs “Q” of the first flip-flop 201b and the second flip-flop 202b are connected to inputs of an OR-gate 204b of the second partial control circuit 120b. The output 125b of the second partial control circuit 120b is provided by an output of the OR-gate 204b and a part of the third digital value xcb[n] is provided by the output “Q” of the second flip-flip 202b.
[0128] In addition, in Figs. 15a and 15b, the first flip-flops 201a, 201b and the second flipflops 202a, 202b of the partial control circuits 120a, 120b are configured to be reset by the sampling control signal smp_p. To this end, the reset inputs “reset” of the flip-flops 201a, 201b, 202a, 202b are connected to the sampling control signal smp_p.
[0129] The partial control circuits 120a, 120b presented with reference to Figs. 15a and 15b are advantageous when the reference voltages Vrefa, Vrefb, Vrefaa, Vrefba are configured such that a maximum pulse length provided by the control circuit 120 is one third of the maximum pulse length provided by the an ADC comprising the VTCs 110a, 110b of Figs. 4 and 5, i.e. VTCs not utilizing auxiliary reference voltages Vrefaa, Vrefba. This is provided when the auxiliary voltages Vrefaa, Vrefba are offset from the reference voltages Vrefa, Vrefb such that Vrefaa = Vrefa — 2 / 3 ■ FSR and Vrefba = Vrefb + 2 / 3 ■ FSR, where FSR is the full-scale range FSR of the ADC 50.
[0130] In Fig. 14, the control circuit 120 was shown comprising a pulse forwarding circuit 127 arranged to receive the output 125a of the first partial control circuit 120a and the output 125b from the second partial control circuit 120b, i.e. the outputs from the OR- gates 204a, 204b in Figs. 15a-b. If no pulse forwarding circuit 127 is present in the control circuit 120, the output 125a of the first partial control circuit 120a and the output 125b from the second partial control circuit 120b form the first output 124a and the second output 124b of the control circuit 120. In embodiments of the control circuit 120 wherein the pulse forwarding circuit 127 is present, the first output 124a and the second output 124b of the control circuit 120 may be provided by the pulse forwarding circuit 127 as will be explained in the following.
[0131] Fig. 16 illustrates an embodiment of the pulse forwarding circuit 127. The function of the pulse forwarding circuit 127 of Fig. 16 is analogues to the pulse-detector circuit 120 of Fig. 8 and will therefore only be explained in short. In Fig 16, the pulse forwarding circuit 127 comprises a first flip-flop 200a and a second flip-flop 200b. In Fig. 16, the clock input “elk” of the first flip-flop 200a is connected to the output 125a of the first partial control circuit 120a and the clock input “elk” of the second flip-flop 200b is connected to the output 125b of the second partial control circuit 120b. Furthermore, in Fig. 16, the output “Q” of the first flip-flop 200a is connected to the first output 124a of the control circuit 120 and the output “Q” of the second flip-flop 200b is connected to the second output 124b of the control circuit 120.
[0132] As with the pulse-detector circuit 120 of Fig. 8, an inherent delay in the logic circuit 210 and the flip-flops 200a, 200b, of the pulse forwarding circuit 127 provides a guaranteed minimum duration of the pulses P3 and P4.
[0133] In the following discussion, the flip-flops 201a, 202a, 201b, 202b are assumed to be positive edge triggered, i.e. when a rising edge (or positive edge) is presented at their respective clock inputs “elk”, they transfer the logic value present at their respective data inputs “D” to their respective outputs “Q”. Let us further assume that the data outputs “Q” of the flip-flops 201a, 202a, 201b, 202b are initially in the first logic state (‘0’) and consider, for simplicity, the pulse wave forms in Fig. 13 as an elucidating example.
[0134] At the fourth time t4, the first edge P2a’ of second auxiliary pulse P2a is provided at the auxiliary output 115b of the second VTC 110b. This is received by the clock input “elk” of the second flip-flop 202b of the second partial pulse detector circuit 120b. Since the data input “D” of the second flip-flop 202b of the second partial pulse detector circuit 120b is set to the second logic state (‘ 1’), the output “Q” of the second flip-flop 202b of the second partial pulse detector circuit 120b makes a transition to the second logic state (‘ 1’) in response to the first edge P2a’ of second auxiliary pulse P2a which is provided to the OR-gate 204b and as part of the third digital value xcb[n]. As a result, the OR-gate 204b of the second partial pulse detector circuit 120b makes a transition to the second logic state (‘ 1’) providing an edge to the clock input “elk” of the second flipflop 200b of the pulse forwarding circuit 127. The second flip-flop 200b of the pulse forwarding circuit 127 transitions to the second logic state (‘ 1’) providing the first edge P4’ of the fourth pulse P4 at the second output 124b of the pulse detector circuit 120.
[0135] At the first time tl, the first edge PL of the first pulse Pl provided at the output 114a of the first VTC 110a. This is received by the clock input “elk” of the first flip-flop 201a of the first partial pulse detector circuit 120a. Since the data input “D” of the first flip-flop 202a of the first partial pulse detector circuit 120a is set to the second logic state (‘ 1’), the output “Q” of the first flip-flop 202a of the first partial pulse detector circuit 120a makes a transition to the second logic state (‘ 1’) in response to the first edge Pl’ of first pulse Pl which is provided to the OR-gate 204a. As a result, the OR- gate 204a of the first partial pulse detector circuit 120a makes a transition to the second logic state (‘ 1’) providing an edge to the clock input “elk” of the first flip-flop 200a of the pulse forwarding circuit 127. The first flip-flop 200a of the pulse forwarding circuit 127 transitions to the second logic state (‘ 1’) providing the first edge P3’ of the third pulse P3 at the first output 124a of the pulse detector circuit 120.
[0136] At the third time instant t3, the first edge P3’ of the third pulse P3 and the first edge P4’ of the fourth pulse have been generated. The part of the third data value xca[n] provided by the first partial pulse detector circuit 120a is at the first logic state (‘0’) and the part of the third data value xcb[n] provided by the second partial pulse detector circuit 120b is at the second logic state (‘ 1’). Based on this, the third logical value xc[n] may be interpreted as “+1” if the output of the part of the third data value xcb[n] provided by the second partial pulse detector circuit 120b is at the second logic state (‘ 1’). Further, the third logical value xc[n] may be interpreted as “-1” if the output of the part of the third data value xca[n] provided by the first partial pulse detector circuit 120a is at the second logic state (‘ 1’). If none of the parts of the third data value xca[n], xcb[n] are at the second logic state (‘ 1’), the third logical value xc[n] may be interpreted as “0”.
[0137] The partial pulse detector circuits 120a, 120b may advantageously be reset by a corresponding logic circuit 210 as presented in reference to Figs. 8 or 9.
[0138] The embodiments of the ADC 50 described herein are suitable for integration on an integrated circuit (IC). Thus, according to some embodiments, there is provided an IC 500 comprising the ADC 50, as schematically illustrated in Fig. 17.
[0139] The embodiments of the single ended to differential VTC 400 described herein are suitable for integration on an integrated circuit (IC). Thus, according to some embodiments, there is provided an IC 600 comprising the single ended to differential VTC 400, as schematically illustrated in Fig. 18. The disclosure above refers to specific embodiments. However, other embodiments than the above described are possible within the scope of the invention. For example, the circuitry used in embodiments of the pulse-detector circuit 120 shown in Fig. 8 and Fig. 9 may also be used as phase detectors in phase-locked loop (PLL). Other types of circuitry that are used in phase-detectors in PLLs may thus be used to implement the pulse-detector circuit 120 in embodiments of the ADC 50. Furthermore, the ADC 50 may be used in other types of electronic apparatuses than communication apparatuses. The different features of the embodiments may be combined in other combinations than those described.
Claims
CLAIMS1. A single ended to differential voltage to time converter, VTC, (400) for converting an input voltage (V), the single ended to differential VTC (400) comprising: a first VTC (110a) comprising: an input terminal (112a) configured to receive the input voltage (V), a capacitor (250a), a first current source (260a) configured to charge the capacitor (250a) of the first VTC (110a) , a first comparator circuit (270a) configured to compare a voltage (Vca) at a first terminal of the capacitor (250a) of the first VTC (110a) with a first reference voltage (Vrefa) and to generate a first pulse (Pl) at an output (114a) of the first comparator circuit (270a), and a first sampling switch (280a) connected between the input terminal (112a) and the capacitor (250a) of the first VTC (110a); a second VTC (110b) comprising: an input terminal (112b) configured to receive the input voltage (V), a capacitor (250b), a second current source (260b) configured to discharge the capacitor (250b) of the second VTC (110b), a second comparator circuit (270b) configured to compare a voltage (Vcb) at a first terminal of the capacitor (250b) of the second VTC (110b) with a second reference voltage (Vrefb) and to generate a second pulse (P2) at an output (114b) of the second comparator circuit (270b), and a second sampling switch (280b) connected between the input terminal (112b) and the capacitor (250b) of the second VTC (110b).
2. The single ended to differential VTC (400) of claim 1, wherein a charge current of the first current source (260a) is matched to a discharge current of the second current source (260b).
3. The single ended to differential VTC (400) of claim 1 or 2, wherein the capacitor (250a) of the first VTC (110a) is substantially identical to the capacitor (250b) of the second VTC (110b).
4. The single ended to differential VTC (400) of any one of the preceding claims, wherein: during a first phase of operation, the first sampling switch (280a) is at a conducting state and the capacitor (250a) of the first VTC (110a) is configured to, for each sample of the input voltage (V), be charged to a voltage value of that sample of the input voltage (V); during a subsequent second phase of operation, the first sampling switch (280a) is at a non-conducting state, the first current source (260a) is configured to charge the capacitor (250a) of the first VTC (110a) , and the first comparator circuit (270a) is configured to compare the voltage (Vca) at the first terminal of the capacitor (250a) of the first VTC (110a) with the first reference voltage (Vrefa) and to generate the first pulse (Pl) at the output of the first comparator circuit (270a).
5. The single ended to differential VTC (400) of any one of the preceding claims, wherein: during a first phase of operation, the second sampling switch (280b) is at a conducting state and the capacitor (250b) of the second VTC (110b) is configured to, for each sample of the input voltage (V), be charged to the voltage value of that sample of the input voltage (V); during a subsequent second phase of operation, the second sampling switch (280b) is at a non-conducting state, the second current source (260b) is configured to discharge the capacitor (250b) of the second VTC (110b), and the second comparator circuit (270b) is configured to compare the voltage (Vcb) at the first terminal of thecapacitor (250b) of the second VTC (110b) with the second reference voltage (Vrefb) and to generate the second pulse (P2) at the output of the second comparator circuit (270b).
6. The single ended to differential VTC (400) of any one of the preceding claims, wherein: the first VTC (110a) is further configured to generate, at a first output (114a) of the first VTC (110a), a first edge (PF) of a first pulse (Pl) delayed an amount of time (tdl) corresponding to a magnitude of the sample of the input voltage (V[n]) in relation to the first reference voltage (Vrefa), and the second VTC (110b) is further configured to generate, at a first output (114b) of the second VTC (110b), a first edge (P2’) of a second pulse (P2) delayed an amount of time (td2) corresponding to a magnitude of the sample of the second input voltage (V[n]) in relation to the second reference voltage (Vrefb).
7. The single ended to differential VTC (400) of claim 6, wherein: the first VTC (110a) is further configured to receive a first auxiliary reference voltage (Vrefaa), and to selectively generate, depending on a relation between a magnitude of the sample of the input voltage (V[n]) and the first auxiliary reference voltage (Vrefaa), at an auxiliary output (115a) of the first VTC (110a), a first edge (Pla’) of a first auxiliary pulse (Pla) delayed an amount of time (tdal) corresponding to the magnitude of the sample of the input voltage (V[n]) in relation to the first auxiliary reference voltage (Vrefaa); and the second VTC (110b) is further configured to receive a second auxiliary reference voltage (Vrefba), and to selectively generate, depending on a relation between a magnitude of the sample of the input voltage (V[n]) and the second auxiliary reference voltage (Vrefba), at an auxiliary output (115b) of the second VTC (110b), a first edge (P2a’) of a second auxiliary pulse (P2a) delayed an amount of time (tda2) corresponding to the magnitude of the sample of the input voltage (V[n]) in relation to the second auxiliary reference voltage (Vrefba).
8. An analog-to-digital converter, ADC, (50) for converting samples of an input voltage (V), the ADC (50) comprising: the single ended to differential VTC (400) according to any one of the preceding claims; and a differential time-to-digital circuit (300) comprising: a first input terminal (122a) coupled to the output (114a) of the first comparator circuit (270a) of the single ended to differential VTC (400) and configured to receive the first pulse (Pl), a second input terminal (122b) coupled to the output (114b) of the second comparator circuit (270b) of the single ended to differential VTC (400) and configured to receive the second pulse (P2), a first output terminal (134a), and a second output terminal (134b), wherein the differential time-to-digital circuit (300) is configured to generate a first digital value (xa[n]) at the first output terminal (134a) and a second digital value (xb[n]) at the second output terminal (134b), wherein a difference between the first digital value (xa[n]) and the second digital value (xb[n]) is proportional to a temporal difference between a first edge (PF) of the first pulse (Pl) and a first edge (P2’) of the second pulse (P2).
9. The ADC (50) of claim 8, wherein the differential time-to-digital circuit (300) is configured to generate the first digital value (xa[n]) proportional to an amount of time the first edge (PF) of the first pulse (Pl) precedes the first edge (P2’) of the second pulse (P2); preferably, the differential time-to-digital circuit (300) is configured to generate the first digital value (xa[n]) as zero responsive to the first edge (PF) of the first pulse (Pl) succeeding the first edge (P2’) of the second pulse (P2).
10. The ADC (50) of claim 8 or 9, wherein the differential time-to-digital circuit (300) is configured to generate the second digital value (xb[n]) proportional to anamount of time the first edge (P2’) of the second pulse (P2) precedes the first edge (PT) of the first pulse (Pl).
11. The ADC (50) of claim 10, wherein the differential time-to-digital circuit (300) is configured to generate the second digital value (xb[n]) as zero responsive to the first edge (P2’) of the second pulse (P2) succeeding the first edge (Pl’) of the first pulse (Pl).
12. The ADC (50) of any one of claims 8 to 11, wherein the first digital value (xa[n]) and / or the second digital value (xb[n]) is offset by an offset value determined by the internal time delay of the differential time-to-digital circuit (300).
13. The ADC (50) of any one of claims 8 to 12, wherein the first digital value (xa[n]) and the second digital value (xb[n]) form an output sample of the ADC (50).
14. The ADC (50) of any one of claims 8 to 12, further comprising a circuit (150) coupled to the outputs (134a, 134b) of the differential time-to-digital circuit (300) configured to generate an output sample (x[n]) of the ADC (50) in response to the first digital value (xa[n]) and the second digital value (xb[n]).
15. The ADC (50) of any one of claims claim 8 to 14, wherein the differential time-to-digital circuit (300) further comprises a pulse-detector circuit (120) coupled to the outputs (114a, 114b) of the first VTC (110a) and the second VTC (110b) and having a first output (124a) and a second output (124b).
16. The ADC (50) of claim 15, wherein the pulse-detector circuit (120) is configured to:- make a transition from a first logic state to a second logic state at a first output (124a) of the pulse-detector circuit (120) in response to the first edge (PF) of the first pulse (Pl) whereby a first edge (P3’) of a third pulse (P3) is generated at the first output (124a) of the pulse-detector circuit (120) ; and- make a transition from the first logic state to the second logic state at a second output (124b) of the pulse-detector circuit (120) in response to the first edge (P2’) of the second pulse (P2) whereby a first edge (P4’) of a fourth pulse (P4) is generated at the second output (124b) of the pulse-detector circuit (120).
17. The ADC (50) of claim 16, wherein the pulse-detector circuit (120) is further configured to:- in response to both the first output (124a) and the second output (124b) of the pulse-detector circuit (120) being set in the second logic state, reset both the first and the second output (124a, 124b) to the first logic state; whereby a second edge (P3”) of the third pulse (P3) is generated at the first output (124a) of the pulse-detector circuit (120) and a second edge (P4”) of the fourth pulse (P4) is generated at the second output (124b) of the pulse-detector circuit (120).
18. The ADC (50) of any one of claims 15 to 17, wherein the differential time- to-digital circuit (300) comprises: a first time-to-digital converter, TDC, (130a) coupled to the first output (124a) of the pulse-detector circuit (120); and a second TDC (130b) coupled to the second output (124b) of the pulse-detector circuit (120).
19. The ADC (50) of claim 18, wherein each of the first TDC (130a) and the second TDC (130b) is a pulse-shrinking TDC.
20. The ADC (50) of any one of claims 15 to 19, wherein the pulse-detector circuit (120) comprises: a first flip-flop (200a) and a second flip-flop (200b), each having a data input (D), a clock input (elk), a reset input (reset), and an output (Q); wherein the data input (D) of each of the first flip-flop (200a) and the second flip-flop (200b) is configured to receive a constant signal corresponding to the second logic state;the clock input (elk) of the first flip-flop (200a) is configured to receive the first pulse (Pl); the clock input (elk) of the second flip-flop (200b) is configured to receive the second pulse (P2); the output (Q) of the first flip-flop (200a) is connected to the first output (124a) of the pulse-detector circuit (120); the output (Q) of the second flip-flop (200b) is connected to the second output (124b) of the pulse-detector circuit (120); and the pulse-detector circuit (120) comprises a logic circuit (210) having a first input (212a) connected to the output (Q) of the first flip-flop (200a), a second input (212b) connected to the output (Q) of the second flip-flop (200b), and an output (214) connected to the reset inputs (reset) of the first flip-flop (200a) and the second flip-flop (200b) for resetting the first flip-flop (200a) and the second flip-flop (200b) in response to the outputs (Q) of both the first flip-flop (200a) and the second flip-flop (200b) being set in the second logic state.
21. The ADC (50) of claim 20, wherein the logic circuit (210) is further configured to reset the first flip-flop (200a) and the second flip-flop (200b) in response to a reset pulse.
22. The ADC (50) of claim 21, comprising circuitry configured to provide said reset pulse in preparation of converting each sample.
23. A receiver circuit (40) comprising the ADC (50) of any one of claims 6 to 19.
24. An integrated circuit (600) comprising the single ended to differential VTC (400) of any one of claims 1 to 7.
25. An integrated circuit (500) comprising the ADC (50) of any one of claims 8 to 22.
26. An electronic apparatus comprising the ADC (50) of any one of the claims 8 to 22.
27. The electronic apparatus of claim 26, wherein the electronic apparatus is a communication apparatus.
28. The electronic apparatus of claim 27, wherein the communication apparatus is a wireless communication device for a cellular communications system.
29. The electronic apparatus of claim 27, wherein the communication apparatus is a base station for a cellular communications system.
Citation Information
Patent Citations
High-precision voltage conversion ratio search circuit of reconfigurable charge pump and charge pump system
CN114337258A
High-speed analog-to-digital conversion circuit based on time domain
CN115102548A
Sub-ranging voltage-to-time-to-digital converter
US9323226B1
Cited By
Time-domain analog-to-digital converter
US12785714B2
