Method and Apparatus of Capacitively Coupling an Adjustable Capacitive Circuit in a VCO

Abstract

Capacitive adjustment in an RCL resonant circuit is typically performed by adjusting a DC voltage being applied to one side of the capacitor. One side of the capacitor is usually connected to either the output node or the gate of a regenerative circuit in an RCL resonant circuit. The capacitance loading the resonant circuit becomes a function of the DC voltage and the AC sinusoidal signal generated by the resonant circuit. By capacitively coupling both nodes of the capacitor, a DC voltage can control the value of the capacitor over the full swing of the output waveform. In addition, instead of the RCL resonant circuit driving a single differential function loading the outputs, each output drives an independent single ended function; thereby providing two simultaneous operations being determined in place of the one differential function.

Description

BACKGROUND OF THE INVENTION

CMOS (Complementary Metal Oxide Semiconductor) is the primary technology used to construct integrated circuits. N-channel devices and P-channel devices (MOS device) are used in this technology which uses fine line technology to consistently reduce the channel length of the MOS devices. Current channel lengths are 40 nm, the power supply of VDD equals 1.2V and the number of layers of metal levels can be 8 or more.

Oscillators and frequency adjust of the oscillators are fundamental components in electronic systems such as communication systems. Typically, the highest performance circuits in a given technology are usually measured in some form of an on-chip free running oscillator, such as; a ring oscillator using transistors or a resonant oscillator that uses transistors and reactive components in a regenerative connection. The oscillators are powered by a first plurality of DC voltages to generate clock signals that adjust and synchronize the operation of the die or integrated circuit chip. The oscillators typically include a frequency adjust circuit to adjust the frequency of the oscillator. Once these clock signals are generated in the die, a Phase Lock Loop (PLL) can be used to compare a reference clock with the generated clock signal to maintain the frequency of operation of the oscillator at a stable value as is well known in the art.

The resonant circuit can be a differential resonant circuit that generates a clock signal and an inverse (180° phase difference) clock signal, simultaneously, on its two output nodes. These two outputs are applied to a first differential function which can perform a first plurality of related computations where the first differential function can be a differential amplifier, a mixer, a filter, a counter, a divider, etc. If a second differential function is desired, a second differential function is coupled to the differential output nodes and loads the resonant circuit thereby decreasing the operating frequency. An increase in power dissipation may be required to compensate for the frequency reduction.

Varactors, also known as a variable capacitance diode, can be operated in the reversed biased state to provide a capacitance for the resonant oscillator. The reverse bias is provided by biasing the nodes of the varactor with a second plurality of DC bias voltages. These DC bias voltages bias the MOS device to create a variable capacitance. DC paths between the first plurality of DC voltages and second plurality of DC adjust voltages can exist which can cause the resonant oscillator to behave more non-linearly due to the varactor or MOS device being DC coupled or partially DC coupled to the resonant oscillator.

BRIEF SUMMARY OF THE INVENTION

Various embodiments and aspects of the inventions will be described with reference to details discussed below, and the accompanying drawings will illustrate the various embodiments. The following description and drawings are illustrative of the invention and are not to be construed as limiting the invention. Numerous specific details are described to provide a thorough understanding of various embodiments of the present invention. However, in certain instances, well-known or conventional details are not described in order to provide a concise discussion of embodiments of the present inventions.

One of the embodiments of the disclosure introduces a DC adjustable capacitance into a resonant circuit by using a differential AC coupling technique. The resonant circuit is a differential resonant circuit that generates a clock signal and the clock signal's inverse, simultaneously, on one of the two output nodes, respectively. A DC adjustable capacitance is formed from an MOS device by applying a first DC bias voltage to the source and drain nodes and a second DC bias voltage to the gate of the MOS device. Two MOS devices are connected back to back where the source and drain nodes of one MOS device is connected to the source and drain nodes of the second device. These two MOS devices are coupled to the output nodes of the resonant circuit using a first and a second outer capacitor. The inventive technique uses the first and the second outer capacitor coupled in series with the capacitance of the two back to back MOS devices to place effectively four series capacitors between the two output nodes of the differential resonant circuit. The first and second DC bias voltages can continuously adjust the capacitance value of the two DC adjustable capacitances and provides a fine frequency adjust of the resonant circuit by varying the overall capacitance value of the four series capacitors presented to the resonant circuit.

Another embodiment of the disclosure eliminates DC paths between the first plurality of DC voltages and second plurality of DC bias voltages by isolating these paths using a coupling capacitor. This DC isolation insures that the varactor or MOS device presents a substantially constant capacitance to the resonant oscillator for the full duration of an oscillation cycle when the DC bias voltages are set to a given value. The non-linear capacitive behavior of the varactor or MOS device is varied when these DC bias voltages are altered to a second given value. Once again the varactor or MOS device presents a substantially constant capacitance to the resonant oscillator for the full duration of a longer/shorter oscillation cycle when the DC bias voltages are set to this second given value. Thus, the non-linear behavior of the DC adjustable capacitance of the MOS device is shielded from the normal operation of the resonant oscillator.

Another embodiment of the disclosure is that the two outer capacitors coupled in series present an upper maximum capacitance to the differential resonant circuit and can be used to isolate or restrict large parasitic capacitances formed by the MOS devices. The two series capacitors couples in a third and a fourth adjustable series capacitance of the fine adjust circuit to the nodes of the resonant circuit. Two MOS devices are coupled back to back to form the third and the fourth adjustable series capacitance. The capacitance of the two MOS back to back devices is varied electronically using a first DC voltage applied to the source and drain nodes and a second DC voltage applied to the gate of the MOS devices. The third and fourth adjustable capacitors can be varied causing the load on the resonant circuit to vary.

The coarse adjust capacitance circuit uses two additional outer capacitors and a switch. The outer capacitors are coupled in series with the switch to the resonant circuit. Each switch (MOS device) can have a large gate width to decrease their resistance. The low impedance of the MOS device couples the two outer capacitors into the resonant circuit. Each node of the resonant circuit would be presented with a capacitive load comprised of the parallel combination of these two additional outer capacitors. This capacitive load that can be added to the resonant circuit is enabled by the switch and is used to coarsely adjust the capacitance loading the outputs of the resonant circuit.

Another embodiment of the disclosure presents how a second function can be coupled to the output node of a resonant circuit without necessarily decreasing the operating frequency. This is achieved by independently loading the first output node of the differential resonant circuit or differential VCO with a first function (that can perform computations) and has a single input while the second output node is loaded with a second function (that can perform different computations). The first or second functions are single-ended functions and can include computations such as a single input amplifier, a mixer, a filter, a counter, a divider, etc. The capacitive load that the first function and second function present to the output nodes of the differential resonant circuit is equivalent to the capacitive load that the first differential function presented to the first and second output nodes. The first and second functions perform different computations yet load the differential resonant circuit with an equivalent load similar to the first differential function mentioned earlier. This is to insure that the load on the differential resonant circuit remains identical and balanced.

Each output node of the resonant circuit can be loaded by additional computational blocks, as long as the summations of the total capacitive load of all computational blocks coupled to each output node match each another. However, due to the larger capacitive load, the power dissipation may increase as the number of total single-ended computational blocks increases to three or more. The balanced load insures that the resonant circuit generates a first waveform and a second waveform that is substantially 180° out of phase with the first waveform.

BRIEF DESCRIPTION OF THE DRAWINGS

Please note that the drawings shown in this specification may not necessarily be drawn to scale and the relative dimensions of various elements in the diagrams are depicted schematically. The inventions presented here may be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be through and complete, and will fully convey the scope of the invention to those skilled in the art. In other instances, well-known structures and functions have not been shown or described in detail to avoid unnecessarily obscuring the description of the embodiment of the invention. Like numbers refer to like elements in the diagrams.

FIG. 1 a depicts a circuit diagram of a conventional capacitive coupled oscillator.

FIG. 1 b shows a block diagram of the circuit in FIG. 1a.

FIG. 2 a illustrates a circuit diagram of a capacitive coupled oscillator in accordance with the present invention.

FIG. 2 b shows a block diagram of the circuit in FIG. 2a in accordance with the present invention.

FIG. 3 a presents the fine frequency capacitive adjustment circuit in accordance with the present invention.

FIG. 3 b depicts a simplified diagram of FIG. 3a in accordance with the present invention.

FIG. 3 c shows a further simplified diagram of FIG. 3b in accordance with the present invention.

FIG. 3 d illustrates the coarse frequency capacitive adjustment circuit in accordance with the present invention.

FIG. 4 a depicts a cross sectional view if the MOS capacitor in accordance with the present invention.

FIG. 4 b illustrates the Capacitance vs. Voltage transfer curve of the MOS capacitor in FIG. 4a in accordance with the present invention.

FIG. 5 a depicts the simulated Frequency vs. Bias Voltage results of the oscillator of FIG. 2a in accordance with the present invention.

FIG. 5 b shows the simulated Gain vs. Frequency results of the oscillator of FIG. 2a in accordance with the present invention.

FIG. 6 a illustrates the oscillator providing differential output clocks to a conventional single function.

FIG. 6 b shows the oscillator providing differential output clocks to differential amplifier.

FIG. 6 c presents the oscillator providing each of its differential output clocks to one of the two separate functions in accordance with the present invention.

FIG. 6 d depicts the oscillator providing each of its differential output clocks to a divide by 3 circuit and a divide by 4 circuit in accordance with the present invention.

FIG. 7 shows a phase lock loop (PLL) with the differential oscillator driving two different functions in accordance with the present invention.

FIG. 8 a presents a divide by 4 schematic in accordance with the present invention.

FIG. 8 b depicts a divide by 3 schematic in accordance with the present invention.

DETAILED DESCRIPTION OF THE INVENTION

FIG. 1 a illustrates a differential oscillator that generates an output and inverse output and uses a diode adjust to adjust the frequency. The circuit contains a bi-symmetry of components: L₁ and L₂, C₁ and C₂, MN_a and MN_b, etc. where each of the matched pairs are substantially equal to one another. The circuit connects an inductor L₂ to device MN_b with the gate of the device capacitively coupled to the inverse output node. Similarly, a second inductor L₁ connected to device MN_a with the gate of the device capacitively coupled to the output node. The two devices MN_a and MN_b are cross coupled through coupling capacitors C₁ and C₂.

The gate of both devices MN_a and MN_b in FIG. 1a are set to a voltage V_tune using a frequency adjust circuit. The frequency adjust circuit includes the high impedance resisters R₁ and R₂ the diodes D₁ and D₂, and the two DC adjust voltages V_tune and V_tune1. The cathode of varactor D₁ is coupled to the gate 1-3 of MN_a and the cathode of varactor D₂ is coupled to the gate 1-4 of MN_b. Their anodes are tied together coupled to and adjusted by a second DC supply V_tune1. The capacitive load on inverse output 1-1 is C_1.1 while the capacitive load on the output 1-2 is C_1.2. The two outputs are presented to a differential amplifier 1-5 that generates a positive 1-7 and a negative output 1-6. By setting the DC adjust voltage V_tune to a voltage greater than the threshold voltage of the n-channel devices enables the n-channel devices. Then, the DC adjust voltage V_tune1 can be adjusted to vary the voltage across the diode thereby adjusting the capacitive load applied to the gates of MN_a and MN_b. As the DC adjust voltage V_tune1 is varied, the frequency of oscillation of this differential oscillator varies as well.

In FIG. 1b provides a block diagram of FIG. 1a. VDD is coupled to the tapped inductor circuit, the far end of the tapped inductor circuit is coupled to the outputs 1-1 and 1-2. These two output nodes are coupled in turn to the capacitively coupled regenerative circuit which includes and MN_a, MN_b, C₁ and C₂. The gates 1-3 and 1-4 of devices MN_a and MN_b are driven through a high impedance network by a DC voltage V_tune to set the bias point. The DC coupled diode circuit is adjusted by the two DC adjust voltages V_tune and V_tune1.

Note in FIG. 1a that the frequency adjust circuitry is directly coupled to the gates of MN_a and MN_b. Thus, the gates of MN_a and MN_b receive both a DC adjust voltage component from the frequency adjust circuit and an AC component from the opposing output through the capacitors C₁ and C₂. Thus, these gates will be adjusted by this DC adjust voltage with a superimposed waveform capacitively coupled through C₁ and C₂ from the outputs.

An exemplary circuit is described which prevents any gate in the oscillator from receiving any DC adjust voltage component from the frequency adjust circuit as discussed previously. This is achieved by isolating all DC adjust voltages generated by the frequency adjust circuitry by using a capacitive coupling inventive technique. FIG. 2a illustrates such an exemplary circuit where the frequency adjust circuitry which contains the adjustable capacitance is capacitively coupled to the outputs preventing any DC adjust voltage from connecting to the oscillator. FIG. 2a illustrates a circuit that capacitively couples the adjustable capacitors of a frequency adjust circuit to a RLC resonant circuit (oscillator). The RLC resonant circuit contains at least one course adjust capacitance and a fine adjust capacitance. The course capacitive adjustment introduces lumped capacitance in steps. The fine adjust capacitance adjustment uses DC adjusts to continuously vary the capacitance between the coarse steps.

FIG. 2 a uses two regenerative cross coupled MOS devices. The outputs of the circuit are at nodes 2-1 and 2-2 and each output node has a lump capacitance of C_1.3 and C_1.4 for, respectively. The first regenerative circuit consists of two N-channel devices M₁ and M₂ cross coupled between the outputs 2-1 and 2-2. The second regenerative circuit is located at the top and consists of the two P-channel devices M₅ and M₆ that are cross coupled between the outputs 2-1 and 2-2. An inductor circuit formed from a single inductor L₃ connects output 2-1 to 2-2. This inductor circuit can he a combination of inductors in series or in parallel; however, the present invention shows a single inductor. The coarse capacitances are located within block 2-6 and it is introduced into this differential circuit by the series combination of capacitors C_5a-c and C_6a-c. The connection is complete when switch S_1a-c is digitally enabled. The block 2-6 contains three-coarse adjusts: the a version, the b version, and the c version for the coarse capacitors and switches. Although, any number of coarse steps can he included. The total coarse capacitance, for this case, would be ½(C_5a+C_5b+C_5c) assuming C_5a=C_6a, C_5b=C_6b and C_5c=C_6c.

FIG. 2 a provides a fine adjust that is located within block 2-5 and uses a string of four series capacitors; C₃, capacitance of MOS device M₃, capacitance of MOS device M₄ and capacitance C₄. Assume that C₃ is substantially equal to C₄ and that the two devices M₃ and M₄ are substantially equal. The center point of the string is where a DC voltage V_cnt is applied to the source and drain nodes of devices M₃ and M₄. To enable setting the gate voltages of M₃ and M₄, a high impedance resistor string is formed between nodes 2-3 and 2-4 by resisters R₃ and R₄. The center point of these two resisters is applied to a DC adjust voltage called V_bias. In this circuit block 2-5, the two DC adjust voltages V_bias and the V_cnt can be adjusted independently of each other. Furthermore, these two DC adjust voltages are independent of the sinusoidal voltages being formed at the output nodes of 2-1 and 2-2 due to the presence of the coupling capacitors C₃ and C₄. In addition, these two coupling capacitors C₃ and C₄ block any low frequency noise introduced by the two DC adjust voltages V_bias and the V_cnt. Thus, this innovative circuit can introduce capacitance into the RLC resonant circuit in coarse steps and/or can be varied continuously within a step independent of the AC operation occurring at the output nodes of this RLC resonant circuit or oscillator.

A block diagram of FIG. 1b is provided in FIG. 2b. Starting from the top, VDD is connected to the regenerative circuit. The two lower outputs of the regenerative circuit are connected to the output nodes 2-1 and 2-2. All remaining blocks connect to these two nodes. The inductor circuit is shown just below the regenerative circuit connected to the output nodes. The AC coupled fine frequency adjust circuit uses the DC adjusts V_bias and V_cnt to adjust the bias applied to the two MOS devices M₃ and M₄ and is connected to the two output nodes. The AC coupled course frequency adjust circuit uses a digital adjust to enable or disable the switches S_1A-C. Finally a second regenerative circuit is connected to the output nodes 2-1 and 2-2 and connects the RLC resonant circuit to ground.

FIG. 3 a presents the AC coupled fine frequency adjust circuit 2-5 that is connected to nodes 2-1 and 2-2. The center point of this symmetric network is V_bias between the resisters R₃ and R₄ and the DC adjust voltage V_cnt which couples to the sources and drains of the MOS devices M₃ and M₄. This differential circuit is capacitively coupled to the RLC resonant differential circuit to present an adjustable capacitance to both output nodes 2-1 and 2-2.

In FIG. 3b, a simplified representation of the MOS devices is illustrated. The MOS devices M₃ and M₄ have been substituted by adjustable capacitors C_M3 and C_M4. The adjustment to these capacitors is through the two DC adjust voltages V_bias and V_cnt, but because of the differential nature of the resonant RLC network, the two DC adjust voltages V_bias and V_cnt behave as AC grounds. Thus, the output node 2-1 is loaded with the capacitance C₃ in series with the parallel combination of C_M3 and R₃. However, R₃ is very large in comparison to the AC impedance of C_M3 and can be disregarded. The fine capacitance equals CF_2-1=(C₃)(C_M3)/((C₃)+(C_M3)). As C_M3 is adjusted; the capacitive value on node 2-1 varies accordingly. The resistors have been removed in FIG. 3C further simplify the AC coupled fine frequency adjust circuit. The four capacitors being in series is very evident.

A coarse capacitive adjustment technique using a plurality of digitally adjusted switches coupled to two capacitors in series to enable/disable and connect the two series capacitors to the resonant circuit's output nodes. Each digital increment adjusts the capacitive loading presented to the resonant circuit's output in discrete steps. In between these discrete steps, the fine frequency adjust to the resonant circuit can be used to provide a continuous capacitive load.

FIG. 3 d presents the block 2-6 which contains the AC coupled fine frequency adjust circuit between output nodes 2-1 and 2-2. Switches S_1a, S_1b and S_1c, enable or disable a capacitively coupled path between the two output nodes. On each side of this switch is a capacitance, for example, C_5a is to the left of switch S_1a and capacitance C_6b is to the right of switch S_1a. This switch S_1a is ohmically isolated from each of the two output nodes by a capacitance. The capacitance value of C_5a substantially equals to the capacitance value of C_6a. Once the switch S_1a is enabled a virtual ground is established at this point because of the differential nature of the RLC resonant circuit. Each output node sees a coarse capacitance equal to CC_2-1=(C_5a)(C_6a)/((C_5a)+(C_6a)). If C_5a substantially equals C_6a, then the capacitance loading each node is C_5a/2. The coarse capacitor values can be arranged linearly weighted, digitally weighted, or a combination of the two.

The MOS devices M₃ and M₄ behaving as adjustable capacitors have a cross-section as illustrated in FIG. 4a. The gate is separated from the Si p-type substrate by an oxide (SiO₂). FIG. 4b provides the C-V curve 4-1 for the MOS capacitor. When the potential applied to the gate is more positive than the devices threshold voltage, the device is depleted and will be in a state called inversion. But at high frequencies, the inversion layer is not able to respond so the charge depletion layer needs to make up the difference. As this potential applied to the gate is decreased, the MOS device losses the ability to form an inversion layer and the space-charge region starts to decrease in depth. The series combination of the oxide capacitance and depletion capacitance cause the capacitance to increase. Finally, when the potential to the gate is more negative than the substrate, the interface between the oxide and substrate device goes into the accumulation state by attracting holes at the interface. The capacitance continues to increase until only the capacitance of the oxide is presented to the circuit.

FIG. 5 a provides a graph of frequency (GHz) versus the bias voltage applied to the adjustable fine frequency adjust circuit. Superimposed over this graph are the coarse steps of the coarse adjust. The continuous curve is due to the voltage applied at the VCO voltage adjust input (AC coupled fine frequency adjust circuit) by the PLL (Charge pump and loop filter output) to MOS varactor devices. The steps between the continuous curves are coarse steps adjusted by the enablement/disablement of the digital switches. For example waveforms 5-1, 5-2 and 5-3 are individual continuous curves that vary as the bias voltage changes from 0 V to 1.2 V. The waveform 5-1 has a frequency range from about 8.4 GHz to 8.9 GHz. The curve 5-3 starts out at 7.3 GHz and spreads out to 7.6 GHz. Each one of the course adjusts is indicated by the difference between two waveforms at a given bias. For example, displacement 5-4 illustrates a coarse difference between the waveform 5-1 and 5-2 where an additional coarse capacitance has been removed. A second course step is illustrated by 5-5. One can readily see that by using a combination of fine and coarse adjust one can vary the frequency of oscillation of the RLC resonant circuit from 7.4 GHz all the way to 8.9 GHz.

FIG. 5 b illustrates three waveforms 5-6, 5-7 and 5-8 where each curve peaks at a different frequency. The curves are plotted with V_cnt (fine adjust) fixed to the middle value of its entire swing (600 mV is mid value between 0V to 1.2V). The curve with lowest frequency 5-8 has all coarse digitally adjusted capacitors switched on. The curve with highest frequency 5-6 has all coarse digitally adjusted capacitors switched off and the curve in the middle 5-7 corresponds to half of the coarse digitally adjusted capacitors enabled by their corresponding switches. This graph shows how the Quality factor degrades or varies across digital tuning range.

FIG. 6 a illustrates a Voltage Controlled Oscillator (VCO) being adjusted by at least one DC adjust voltage input 6-1 for fine capacitance adjustment and at least one digital adjust 6-10 for coarse capacitance adjustment. For example, the circuit described in FIG. 2a can be used to represent the VCO. The VCO is a resonant RLC circuit and generates an oscillatory output and an inverse oscillatory output, in this case V_out and V_out. The capacitive load of a first interconnect coupling V_out to a first differential input of Function A 6-2 is C_1.3. The capacitive load of a second interconnect coupling V_out to a second differential input of Function A 6-2 is C_1.4. The input capacitance of each differential input of Function A 6-2 is typically equal. The interconnect is formed by patterned metal layers in a die which has a capacitance based on the geometric structure and position of these metal layers within the structure of the die. When a VCO drives a differential input, the patterned metal layers of the differential interconnects are similar to maintain similar capacitance. Thus, the total capacitive load on each of the output nodes should equal each other. Since the capacitive load on the VCO is equalized, the output and inverse output are 180° apart.

FIG. 6 b illustrates the Function A substituted with a differential amplifier 6-3. The input capacitance on the positive and negative input terminals of the differential amplifier is substantially equal. Therefore the interconnect capacitance C_1.3 is made substantially equal to the interconnect C_1.4 in the VCO thereby providing a clock signal to the differential amplifier that is 180° separated in phase.

In FIG. 6c, V_out couples to the input of the Function B 6-4 by a first interconnect that has a capacitive load of C_1.5. While. V_out couples to the input of the Function C 6-5 by a second interconnect that has a capacitive load of C_1.6. The Function B 6-4 can be completely different from the Function C 6-5; therefore, the input capacitance of the Function B and Function C generally will be different since these are different blocks. This input capacitance of Function B is called the B function capacitance while the input capacitance of Function C is called the C function capacitance. Any difference between the B function capacitance and C function capacitance needs to be compensated for in either the capacitive load of C_1.5 or C_1.6. The patterned metal layers forming the interconnect can be independently adjusted to ensure that each of the outputs of the VCO has identical capacitive loads, which insures that Function B and Function C are being driven by waveforms 180° apart.

Another aspect of these functions is that each of these functions become single ended. Referring back to FIG. 6b, the two complimentary outputs of the VCO are applied to one function a differential amplifier which requires that the two inputs are driven by a differential signal. However, the functions in FIG. 6c require that each output drives a function that only requires a single input.

In FIG. 6d, the Function B and Function C are substituted by a “divide by 3” 6-6 and “divide by 4” 6-7 functions. In addition, buffers 6-8 and 6-9 are inserted before each function and the outputs of the VCO. These buffers, being identical, equalize the capacitive load presented to the VCO. Thus, the capacitive load C_1.7 at each of the two outputs of the VCO would be equal if the structural layout of the interconnect are equivalent. Thus, each output of the VCO drives a separate independent function where the capacitive load on the two outputs are equalized to insure that the VCO generates two signals 180° out of phase with each other.

An embodiment illustrating this innovative technique is illustrated in FIG. 7. A phase lock loop (PLL) is formed by the PFD (Phase and Frequency detect) feeding a charge pump 7-1. The output of the charge pump is low-pass filtered 7-2 providing a DC voltage 6-1 that is applied to the VCO 7-3. The output of the VCO is loaded by a buffer 6-8 while the inverse output of the VCO is loaded by a buffer 6-9. Since the two buffers can be sized similarly, the capacitive load at the output of the VCO is equalized. This is a second inventive technique of equalizing and therefore presenting the same capacitive load to the outputs of the VCO. The buffer 6-8 drives the “divide by 3” 6-6 while the second buffer 6-9 drives the “divide by 4” 6-7. The “divide by 4” 6-7 is applied to the “divide by N” 7-6. The second buffer 6-9, the “divide by 4” 6-7 and the “divide by N” 7-6 form the feedback path going back to the PFD from the output of the VCO. The feedback path is compared in the PFD against the input reference frequency f_ref. Furthermore, the output of the “divide by 4” 6-7 is tapped by the buffer 7-5 to provide a clock signal f₂. The output of the “divide by 3” 6-6 is tapped by the buffer 7-4 to generate f₁. The two buffers can he adjusted in size to guarantee that the waveforms f₁ and f₂ maintain a desired phase separation.

FIG. 8 a depicts the schematic diagram for the “divide by 4” 6-7 where the input in₁ is applied to first FF (flip-flop) 8-1. The output of the first FF is divided by 2 and applied to the second FF 8-2 to further divide the signal. The buffer 8-3 outputs the “divided by 4” signal as Out₁. This buffer can be sized to drive a larger load.

FIG. 8 b depicts the schematic diagram for the “divide by 3” 6-6 where the input in₂ is applied to clock the first FF (flip-flop) 8-4 and second FF 8-5 to further divide the signal. The output of the first FF is divided by 2 and applied to the AND gate 8-6. The buffer 8-7 outputs the “divided by 3” signal as Out₂. This buffer can be sized to drive a larger load. Also, although not illustrated, buffers can be inserted between the outputs of the VCO and the inputs in₁ and in₂ to provide gain if required.

Finally, it is understood that the above description are only illustrative of the principle of the current invention. Various alterations, improvements, and modifications will occur and are intended to be suggested hereby, and are within the spirit and scope of the invention. This invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided so that the disclosure will be thorough and complete, and will fully convey the scope of the invention to those skilled in the arts. It is understood that the various embodiments of the invention, although different, are not mutually exclusive. In accordance with these principles, those skilled in the art may devise numerous modifications without departing from the spirit and scope of the invention. Although the circuits were described using CMOS, the same circuit techniques can be applied to depletion mode devices and BJT or biploar circuits, since this technology allows the formation of current sources and source followers. When a device is specified, the device can be a transistor such as an N-MOS or P-MOS. The CMOS or SOI (Silicon on Insulator) technology provides two enhancement mode channel types: N-MOS (n-channel) and P-MOS (p-channel) devices or transistors.

Claims

1. A differential VCO comprising: an output node and an inverse output node;a source and a drain node of a first and a second MOS device connected to a first DC voltage;a gate of said first MOS device coupled to a second DC voltage by a first impedance;a gate of said second MOS device coupled to said second DC voltage by a second impedance;said gate of said first MOS device coupled by a first capacitor to said output node: andsaid gate of said second MOS device coupled by a second capacitor to said inverse output node.

2. The VCO of claim 1, further comprising: an inductor connected between said output node and said inverse output node.

3. The VCO of claim 2, further comprising: at least one regenerative circuit coupled to said output node and said inverse output node.

4. The VCO of claim 1, further comprising: a first end of a switch coupled by a third capacitor to said output node; anda second end of said switch coupled by a fourth capacitor to said inverse output node.

5. The VCO of claim 4, whereby an enabled switch capacitively couples a coarse capacitance substantially equal to said third and said fourth capacitors in series.

6. The VCO of claim 1, whereby said first DC voltage and said second DC voltage can be adjusted to adjust a capacitance of said first and said second MOS device.

7. The VCO of claim 6, whereby said capacitance of said first and said second MOS device is due to a state selected from the group consisting of inversion, depletion and accumulation.

8. The VCO of claim 6, whereby said capacitance of said first and said second MOS device remains substantially constant independent of sinusoidal signals at said outputs.

9. The VCO of claim 1, whereby said first impedance substantially equals said second impedance, andsaid first capacitance substantially equals said second capacitance.

10. A differential VCO comprising: at least one inductor coupled between an output node and an inverse output node;a source and a drain node of a first and a second MOS device connected to a first DC voltage;a gate of said first MOS device coupled to a second DC voltage by a first impedance;a gate of said second MOS device coupled to said second DC voltage by a second impedance;said gate of said first MOS device coupled by a first capacitor to said output node; andsaid gate of said second MOS device coupled by a second capacitor to said inverse output node.

11. The VCO of claim 10, further comprising: at least one regenerative circuit coupled to said output node and said inverse output node.

12. The VCO of claim 10, further comprising: a first end of a switch coupled by a third capacitor to said output node; anda second end of said switch coupled by a fourth capacitor to said inverse output node.

13. The VCO of claim 12, whereby an enabled switch capacitively couples a coarse capacitance substantially equal to said third and said fourth capacitors in series.

14. The VCO of claim 10, whereby said first DC voltage and said second DC voltage can be adjusted to adjust a capacitance of said first and said second MOS device.

15. The VCO of claim 14, whereby said capacitance of said first and said second MOS device is due to a state selected from the group consisting of inversion, depletion and accumulation.

16. The VCO of claim 14, whereby said capacitance of said first and said second MOS device remains substantially constant independent of the sinusoidal signal at said outputs.

17. The VCO of claim 10, whereby said first impedance substantially equals said second impedance, andsaid first capacitance substantially equals said second capacitance.

18. A method of isolating a capacitance of MOS devices from an output node and an inverse output node of a VCO comprising the steps of: connecting a source and a drain node of a first and a second MOS device to a first DC voltage;connecting a gate of said first MOS device to a second DC voltage by a first impedance;connecting a gate of said second MOS device to said second DC voltage by a second impedance;coupling said gate of said first MOS device by a first capacitor to said output node; andcoupling said gate of said second MOS device by a second capacitor to said inverse output node, thereby isolating said capacitance of MOS devices from said output node and said inverse output node of said VCO.

19. The method of claim 18, further comprising the steps of: adjusting said first DC voltage and said second DC voltage to place said first and said second MOS devices in a state selected from the group consisting of inversion, depletion and accumulation.

20. The method of claim 18, further comprising the steps of: coupling at least one inductor between said output node and said inverse output node; andmaintaining said capacitance of said first and said second MOS device substantially constant independent of sinusoidal signals at said outputs.