Embodiments of the invention relate to electronic systems, and more particularly, to frequency multipliers.
A rotary traveling wave oscillator (RTWO) is a type of electronic oscillator in which a traveling wave moves around a closed differential loop that includes a crossover for reversing the polarity of the traveling wave each transit of the loop. Additionally, the traveling wave's energy is preserved by maintaining amplifiers distributed around the loop. At any point along the loop, a differential clock signal is available by tapping the loop. The frequency of the differential clock signal is determined by the time taken by the traveling wave to propagate around the loop, and the phase of the differential clock signal is determined by the position along the loop that the differential clock signal is tapped from.
RTWOs can be used in a variety of applications, including, for example, radio frequency systems, optical networks, and/or chip-to-chip communication. For instance, an RTWO can be used in a frequency synthesizer to generate an output clock signal having a controlled phase and frequency relationship to a reference clock signal.
RTWO-based frequency multipliers are provided herein. In certain embodiments, an RTWO-based frequency multiplier includes an RTWO that generates a plurality of clock signal phases of a first frequency, and an edge combiner that processes the clock signal phases to generate an output clock signal having a second frequency that is a multiple of the first frequency. The edge combiner can be implemented as a logic-based combining circuit that combines the clock signal phases (which can be square-wave pulses) from the RTWO. For example, the edge combiner can include parallel stacks of transistors operating on different clock signal phases, with the stacks pulling down (or up) an output node to generate the output clock signal of multiplied frequency.
In one aspect, a frequency multiplier includes an RTWO including a differential transmission line connected as a ring, the differential transmission line configured to carry a traveling wave, wherein the RTWO is configured to generate a plurality of clock signal phases of a first frequency. The frequency multiplier further includes an edge combiner configured to receive the plurality of clock signal phases and to generate an output clock signal having a second frequency that is a multiple of the first frequency.
In another aspect, a method of frequency multiplication is provided. The method includes generating a plurality of clock signal phases of a first frequency using an RTWO that includes a differential transmission line connected as a ring, providing the plurality of clock signal phases from the ring of the RTWO to an edge combiner, and combining the plurality of clock signal phases to generate an output clock signal having a second frequency that is a multiple of the first frequency using the edge combiner.
In another aspect, a frequency multiplier includes an RTWO including a differential transmission line connected as a ring, the differential transmission line configured to carry a traveling wave, wherein the RTWO is configured to generate a plurality of clock signal phases of a first frequency. The frequency multiplier further includes means for edge combining the plurality of clock signal phases to generate an output clock signal having a second frequency that is a multiple of the first frequency.
The following detailed description of embodiments presents various descriptions of specific embodiments of the invention. However, the invention can be embodied in a multitude of different ways. In this description, reference is made to the drawings, where like reference numerals may indicate identical or functionally similar elements. It will be understood that elements illustrated in the figures are not necessarily drawn to scale. Moreover, it will be understood that certain embodiments can include more elements than illustrated in a drawing and/or a subset of the elements illustrated in a drawing. Further, some embodiments can incorporate any suitable combination of features from two or more drawings.
As persons having ordinary skill in the art will appreciate, a rotary traveling wave oscillator (RTWO) includes a differential transmission line connected in a ring with an odd number of one or more crossovers (for instance, a Mobius ring), and a plurality of maintaining amplifiers electrically connected along a path of the differential transmission line. Additionally, each of the crossovers reverses the polarity of a wave propagating along the differential transmission line, and the maintaining amplifiers provide energy to the wave to compensate for the differential transmission line's losses.
In certain implementations, the ring is partitioned into segments evenly distributed around the ring, with each segment including a pair of conductors extending from the differential transmission line and to which a maintaining amplifier and at least one tuning capacitor array are connected between. For example, the maintaining amplifier can be implemented using a pair of back-to-back inverters that compensate for the segment's losses and ensure differential operation, while the tuning capacitor array serves to tune the oscillation frequency of the RTWO over a wide tuning range and/or to provide a fine frequency step size.
RTWOs can be used in a variety of applications, including, for example, radio frequency systems, optical networks, and/or chip-to-chip communication. For instance, an RTWO can be used in a frequency synthesizer to generate an output clock signal having a controlled phase and frequency relationship to a reference clock signal.
An RTWO has an ability to generate multiple clock signal phases at millimeter-wave (mmW) frequencies, while achieving low phase noise (PN). For example, RTWO's can be used as a local oscillator (LO) in mmW radars operating in the 77-81 GHz band.
In certain applications, a clock signal from an RTWO is multiplied using a frequency multiplier. However, such frequency multiplication can be difficult to achieve without degrading performance parameters, particularly when the multiplication factor is greater than two.
For example, cascading conventional frequency doublers is not power-efficient and/or suffers from challenges in designing inter-stage matching circuits for rejecting unwanted spurious harmonics. Additionally, cascading doublers is not area efficient since isolation between doublers should be maintained to avoid subharmonic coupling.
Conventional frequency quadruplers can also be used, but suffer from a number of drawbacks including, but not limited to, poor DC-to-RF efficiency.
RTWO-based frequency multipliers are provided herein. In certain embodiments, an RTWO-based frequency multiplier includes an RTWO that generates a plurality of clock signal phases of a first frequency, and an edge combiner that processes the clock signal phases to generate an output clock signal having a second frequency that is a multiple of the first frequency. The edge combiner can be implemented as a logic-based combining circuit that combines the clock signal phases (which can be square-wave pulses) from the RTWO. For example, the edge combiner can include parallel stacks of transistors operating on different clock signal phases, with the stacks pulling down (or up) an output node to generate the output clock signal of multiplied frequency.
In certain implementations, the output of the edge combiner is coupled to inductor-capacitor (LC) filter. Such an LC filter can have an impedance that is tunable based on an oscillation frequency of the RTWO. For example, in an implementation in which the RTWO-based frequency multiplier is a frequency quadrupler, the impedance of the LC filter can be tuned to about four times the RTWO's oscillation frequency.
Including the edge combiner allows frequency multiplication to be achieved while running the RTWO at a relatively low oscillation frequency. Implementing the RTWO at lower frequency achieves a number of benefits including, but not limited to, low transmission line losses, a larger number of segments, and/or clock signal phases closer to ideal square waveforms.
In certain implementations, controllable components, such as correction capacitors placed in the RTWO's segments, are used to correct for phase error mismatch of the clock signal phases provided to the edge combiner. The values of such controllable components can be set in a variety of ways. In a first example, a harmonic in the output clock signal (for example, a fifth harmonic) is observed and the controllable components are adjusted to reduce or minimize the power level of the harmonic. In a second example, digital data from a time-to-digital converter (TDC) is processed by a finite state machine (FSM) to set the values of the controllable components. The FSM can monitor the digital data from the TDC using any suitable metric for providing phase alignment.
In certain implementations, the edge combiner is placed inside the ring of the RTWO, thereby aiding in providing balanced routes for routing the clock signal phases from the RTWO's ring to the edge combiner. In certain implementations, an output buffer is included for buffering the output clock signal from the edge combiner.
In one example application, an RTWO-based frequency multiplier serves as a frequency quadrupler that combines four square-wave pulses (each of which can be differential) from a co-located 10-GHz RTWO in order to generate an output clock signal at 40 GHz with good harmonic rejection of RTWO harmonics. Such an RTWO-based frequency quadrupler can be followed by a frequency doubler to generate an LO signal suitable for servicing the 77-81 GHz band, for instance, for a radar application. Although one specific application of an RTWO-based frequency multiplier is provided, the RTWO-based frequency multipliers disclosed herein can be used in a wide variety of applications.
In the illustrated embodiment, the RTWO 1 provides various clock signal phases of a frequency f1 to the edge combiner 2. Each of the clock signal phases has a different phase, and are obtained from tapping a ring of the RTWO 1 at different positions. The clock signal phases can be singled-ended or differential signals. In certain implementations, the clock signal phases are provided from the output of buffers distributed around the ring at different positions. For example, an input of such a buffer can be connected to a particular position along the ring, and an output of the buffer can provide a clock signal phase of a desired value.
The frequency f1 of the clock signal phases is set by a period of the traveling wave propagating around the RTWO's ring. The RTWO 1 can include controllable capacitors (for example, tuning capacitor arrays in the RTWO's segments) that can be controlled to set the frequency f1 to a desired value.
As shown in
The output clock signal CLKOUT is a multiple of the RTWO's frequency. For example, the frequency multiplier 10 can serve to provide, frequency doubling, frequency tripling, frequency quadrupling, or any other desired frequency multiplication. Moreover, in comparison to conventional frequency multipliers, the RTWO-based frequency multiplier 10 provides frequency multiplication with high DC-to-RF efficiency.
The frequency multiplier 20 of
In the illustrated embodiment, the RTWO 40 includes a differential transmission line including a first conductor 31 and a second conductor 32. As shown in
In the illustrated embodiment, the RTWO's differential transmission line 31-32 is connected in a closed-loop and is folded at each of four corners. However, the RTWO's differential transmission line can be implemented in other ways, including, for example, different implementations of folding and/or routing of the conductors 31 and 32.
The RTWO 40 further includes a first differential buffer 36a, a second differential buffer 36b, a third differential buffer 36c, and a fourth differential buffer 36d used to provide various clock signal phases to the edge combiner 41. In the example, the buffers 36a-36d each have differential inputs and differential outputs. However, other implementations are possible, for example, single-ended input/differential output, single-ended input/single-ended output, or differential input/single-ended output.
In the embodiment of
The edge combiner 41 processes the clock signal phases to generate a differential clock signal provided on a differential transmission line TL+, TL−. The differential clock signal is filtered by the LC filter 42. As shown in
In the embodiment of
In the illustrated embodiment, the LC filter 42 also receives a supply voltage VDD used to power the edge combiner 41. In particular, the inductor LT is implemented as metal coil winding around the edge combiner 41 and having a center tap that receives the supply voltage VDD. Accordingly, the inductor LT is also used as a radio frequency choke for providing a DC supply voltage to the edge combiner 41, in this example. However, other implementation are possible.
With continuing reference to
The output buffer 43 is biased by a bias voltage VBIAS, in this example. In certain implementations, one or more parameters of the output buffer 43 (such as a bias, resonance, impedance, etc.) are adjusted based on a selected oscillation frequency of the RTWO 41. Accordingly, the output buffer 43 is tuned based on an oscillation frequency of the RTWO 41, in some embodiments.
The segment 70 of
As shown in
In the embodiment of
In the illustrated embodiment, the coarse capacitor array 62 includes an array of three selectable coarse capacitors of capacitance Ccrs, while the fine capacitor array 63 includes an array of thirty-one selectable fine capacitors of capacitance Cfin. The segment decoder 65 controls the capacitance values of the coarse capacitor array 62 and the fine capacitor array 63 based on a coarse control word (crs, 2-bit, in this example) and a fine control word (fin, 5-bit, in this example). The segment decoder 65 is implemented with thermometer decoding, in this embodiment. By controlling a capacitance of the coarse capacitor array 62 and the fine capacitor array 63, an oscillation frequency of the RTWO is controlled.
The segment 70 of
Accordingly, in some embodiments, the correction capacitor array 64 is used to provide phase adjustments to clock signal phases used by an edge combiner of an RTWO-based frequency multiplier.
In the illustrated embodiment, the edge combiner 101 is implemented differentially and includes a first half circuit connected between a non-inverted output terminal Vq+ and ground, and a second half circuit connected between an inverted output terminal Vq− and ground. Although a differential edge combiner is depicted, the teachings herein are also applicable to single-ended configurations.
The first half circuit includes n-type field-effect transistors (NFETs) N0, N1, N2, N3, N4, N5, N6, and N7, while the second half circuit includes NFETs N8, N9, N10, N11, N12, N13, N14, and N15. NFETs N0 and N1 are connected in series in a first stack, NFETs N2 and N3 are connected in series in a second stack, NFETs N4 and N5 are connected in series in a third stack, and NFETs N6 and N7 are connected in series in a fourth stack, with the first through fourth stacks connected in parallel with one another between the non-inverted output terminal Vq+ and ground. Additionally, NFETs N8 and N9 are connected in series in a fifth stack, NFETs N10 and N11 are connected in series in a sixth stack, NFETs N12 and N13 are connected in series in a seventh stack, and NFETs N14 and N15 are connected in series in an eight stack, with the fifth through eighth stacks connected in parallel with one another between the inverted output terminal Vq− and ground.
As shown in
Accordingly, the first half circuit of the edge combiner 101 includes four pairs of NFET transistors that selectively activated based on timing of the clock signal phases to generate a first current Lq+ used to pull down the non-inverted output terminal Vq+. Additionally, the second half circuit of the edge combiner 102 includes another four pairs of NFET transistors that selectively activate based on timing of the clock signal phases to generate a second current Lq− used to pull down the inverted output terminal Vq−. The first current Lq+ and the second current Lq− correspond to differential components of an output current having a frequency that is a multiple (four times, in this example) of the frequency of the clock signal phases.
In the illustrated embodiment, the LC filter 102 includes a tunable capacitor array 103 and an inductor 104. The tunable capacitor array 103 includes a plurality of selectable capacitors of capacitance Cq, while the inductor 104 has an inductance 104. The tunable capacitor array 103 and the inductor 104 are connected in parallel with one another between the non-inverted output terminal Vq+ and the inverted output terminal Vq−. Although an example of an LC filter with a tunable capacitor and a fixed inductor is shown, other implementations are possible, including configurations in which an inductor is tunable (for instance, by way of selectable inductors).
In the example of
In particular, the tunable capacitor array 140 includes slices CS<0>, CS<1>, . . . CS<7>. Additionally, the slices CS<0>, CS<1>, . . . CS<7> receive control bits code<0>, code<1>, . . . code<7>, respectively, and inverted control bits codeb<0>, codeb<1>, . . . codeb<7>, respectively.
Each slice of the tunable capacitor array 140 is implemented using NFET transistors N16, N17, N18, N19, and N20 and using a differential implementation of metal-oxide-metal (MOM) capacitors with capacitance Cq. Each slice is connected between the non-inverted terminal Vq+ and the inverted terminal Vq−.
As shown in
In the illustrated embodiment, each pair of metal clock routes has a balanced or matched length to aid in matching a propagation delay of the clock signal phases to the edge combiner layout 205.
In certain implementations, tunable components are provided for correcting for phase error of one or more of the clock signal phases, thereby aligning the actual clock signal phase with a desired or ideal phase value (as indicated by the subscript). Examples of such tunable components include, but are not limited to, correction capacitors in the RTWO segments (for example, to provide a local capacitance correction near a point at which a given clock signal phase is tapped from the RTWO's ring) and/or components for adjusting a delay of buffers used to provide the clock signal phases from the RTWO ring to the edge combiner.
In certain implementations, grounded shields are included in the pairs of metal clock routes 201-204 to provide shielding for reducing noise. For example, in the illustrated embodiment, grounded conductors are positioned beneath each pair of metal clock routes. The grounded conductors are routed with each pair of metal clock routes and correspond to one example of a grounded shield.
In the illustrated embodiment, the input transformer T1 includes a differential input that receives a differential input signal from a differential transmission line TL+, TL−. The input transformer T1 further includes a differential output that provides a differential output signal across the gates of NFETs N21 and N22. The differential input is connected to a first winding of the input transformer T1, while the differential output is connected to a second winding of the input transformer T1. The capacitor C1 is connected across the differential input of the input transformer T1, while the capacitor C2 is connected across the differential output of the input transformer T1. Additionally, a bias voltage VBIAS is provided at a center tap of the first winding to aid in controlling a common-mode voltage of the differential transmission line TL+, TL−.
With continuing reference to
The output transformer T2 receives the amplified signal from the NFETs N21 and N22 at a differential input, and provides an output signal Vo from one terminal of a differential output (with the other terminal of the differential output grounded, in this example). The capacitor C3 is connected across the differential input of the output transformer T2, while the capacitor C4 is connected across the differential output of the output transformer T2. The differential input is connected to a first winding of the output transformer T2, while the differential output is connected to a second winding of the output transformer T2. Additionally, a power supply voltage VDD is provided at a center tap of the first winding of the output transformer T2 to aid in powering the NFETs N21 and N22.
Although one embodiment of an output buffer is depicted, the teachings herein are applicable to output buffers implemented in a wide variety of ways.
In the illustrated embodiment, an RTWO ring formed from conductors 31 and 32 and a crossover 33 is depicted. Additionally, capacitors associated with a first segment 301, a second segment 302, a third segment 303, and a fourth segment 304 are depicted.
There are many potential sources of phase error between the input clock signal phases to an edge combiner of an RTWO-based frequency multiplier. Examples of such sources of phase error include, but are not limited to, mismatch between the RTWO phases, asymmetric routing of a clock tree layout, local mismatches in the input buffers, local mismatches of the edge combiner's transistors, and/or asymmetric transmission line routing between the edge combiner and an output buffer.
To help alleviate such errors, correction components can be included. For example, in the embodiment of
The values of the correction components (for example, capacitances of correction capacitor arrays in the RTWO's segments) can be provided in a wide variety of ways. In one example, RTWO phase calibration is performed by observing a fifth harmonic of the RTWO frequency at the system output.
In the illustrated embodiment, the phase-correction MOM switched-capacitor array for each segment is set to its middle value Cmid. By fixing the phase-correction capacitance for the chosen side of the RTWO ring (left segment 301, in this example) and sequentially tuning the other sides by ΔC1 (8×3-bit control), ΔC2, and ΔC3, the RTWO clock phase signals can be aligned to desired phases values at the edge combiner's input, thereby significantly lowering the fifth harmonic level of the frequency multiplier.
In the illustrated embodiment, an RTWO ring 341 is depicted as well as circuitry 342 implemented within the RTWO ring 341. The circuitry 342 within the RTWO ring 341 includes clock buffers 343a, 343b, 343c, and 343d, a clock tree 345, tuning capacitors Ca1, Ca2, Cb1, Cb2, Cc1, Cc2, Cd1, and Cd2, a time-to-digital converter (TDC) (implemented as TDC latches 347a, 347b, 347c, and 347d, in this embodiment), a finite-state machine (FSM) 348, and an edge combiner 349.
In the embodiment of
During calibration, the RTWO frequency is locked at a prime fractional multiple of a reference clock CLKREF frequency used for controlling the FSM 348 and the TDC latches 347a, 347b, 347c, and 347d. Thus, an even distribution of TDC output codes is achieved. The outputs of the TDC latches 347a, 347b, 347c, and 347d are accumulated and the FSM 348 generates a histogram used to adjust the clock tree path delay by way of digitally-controlled capacitors, in this embodiment. For example, capacitor adjustment can be performed until the histogram bins are equalized.
The frequency multiplier 410 of
Devices employing the above described schemes can be implemented into various electronic devices. Examples of electronic devices include, but are not limited to, RF communication systems, consumer electronic products, electronic test equipment, communication infrastructure, radar systems, etc.
The foregoing description may refer to elements or features as being “connected” or “coupled” together. As used herein, unless expressly stated otherwise, “connected” means that one element/feature is directly or indirectly connected to another element/feature, and not necessarily mechanically. Likewise, unless expressly stated otherwise, “coupled” means that one element/feature is directly or indirectly coupled to another element/feature, and not necessarily mechanically. Thus, although the various schematics shown in the figures depict example arrangements of elements and components, additional intervening elements, devices, features, or components may be present in an actual embodiment (assuming that the functionality of the depicted circuits is not adversely affected).
Although this invention has been described in terms of certain embodiments, other embodiments that are apparent to those of ordinary skill in the art, including embodiments that do not provide all of the features and advantages set forth herein, are also within the scope of this invention. Moreover, the various embodiments described above can be combined to provide further embodiments. In addition, certain features shown in the context of one embodiment can be incorporated into other embodiments as well.
The present application claims priority to U.S. Provisional Patent Application No. 63/199,912, filed Feb. 2, 2021 and titled “RTWO-BASED FREQUENCY MULTIPLIER,” the entirety of which is hereby incorporated herein by reference.
Number | Date | Country | |
---|---|---|---|
63199912 | Feb 2021 | US |