Precision digital-to-analog conversion in the presence of variable and uncertain fractional-valued bit contributions.
A digital-to-analog converter with a sub-binary sequence of capacitors addresses variable fractional bit contributions, ensuring precise impedance matching and efficient power transfer in plasma processing chambers.
Patent Information
- Application Number
- JP2023072238
- Authority / Receiving Office
- JP · JP
- Patent Type
- Patents
- Current Assignee / Owner
- Priority Date
- 2018-05-10
- Filing Date
- 2023-04-26
- Publication Date
- 2024-12-09
- Estimated Expiration
- 2039-05-10
AI Technical Summary
Existing impedance matching networks in plasma processing chambers face challenges due to variable and uncertain fractional bit contributions from capacitors, leading to gaps in capacitance values that hinder efficient power transfer and impedance matching.
A digital-to-analog converter system using a sub-binary sequence of switched capacitors, where the contribution of each capacitor is carefully selected to minimize gaps in capacitance values, ensuring a monotonic relationship between digital input and analog output.
The solution reduces capacitance gaps, enabling precise and efficient impedance matching, thereby improving power transfer efficiency in plasma processing systems.
Smart Images

Figure 0007596434000008 
Figure 0007596434000009 
Figure 0007596434000010
Abstract
Description
[Technical field]
[0001] (Claim of priority under 35 U.S.C. § 119) This application for patent claims priority to Provisional Application No. 62 / 669,454, entitled "PRECISION DIGITAL TO ANALOG CONVERSION IN THE PRESENCE OF VARIABLE AND UNCERTAIN FRACTIONAL BIT CONTRIBUTIONS," filed May 10, 2018, which is assigned to the assignee hereof and expressly incorporated herein by reference.
[0002] The present invention relates generally to plasma processing, and more particularly, but not exclusively, to a system, method and apparatus for impedance-matched radio frequency power delivery from a radio frequency generator to a plasma load in a semiconductor processing chamber. [Background technology]
[0003] In the semiconductor manufacturing world, manufacturers produce plasma processing chambers that utilize radio frequency (RF) power to generate the plasma. To achieve efficient power transfer between an RF generator ("generator") and a plasma load, an impedance matching network ("matching network") is often used to match the load impedance to a desired input impedance, typically 50 Ω. The plasma load impedance can vary depending on variables such as generator frequency, power, chamber pressure, gas composition, and plasma ignition. The matching network accounts for these variations in the load impedance by varying electrical elements, possibly including solid-state variable capacitors internal to the matching network, to maintain the desired input impedance.
[0004] As a preliminary note, this disclosure will often discuss the "contribution" of a capacitor that is switched in and out of a circuit. Note that a capacitor's contribution to the overall capacitance of a circuit is not the same value as the capacitance of the capacitor. For example, a 5pF capacitor may contribute 5pF when switched into the circuit (on state) and 0pF when switched out of the circuit (off state). Instead, the "contribution" of the capacitor also takes into account the capacitance of the switch used to switch the capacitor in and out of the circuit. Given a switch with a 6pF capacitance in the off state in series with the 5pF capacitor mentioned above, the combination has a 5pF capacitance when the switch is on and approximately 2.73pF when the switch is off. [ka] Therefore, this 5pF capacitor has a capacitance of 2.27pF between its off and on states. [ka] contributes the difference (or contribution) and thus can be said to contribute to or have a value of 2.27pF. This disclosure will often refer to this as the "contribution" of the capacitor, but this is actually the change in capacitance of the capacitor and switch combination in the on and off states, rather than the actual capacitance of the capacitor alone.
[0005] FIG. 1 illustrates a typical generator, matching network, and plasma load system. A generator 102 transmits RF power via a transmission line 108 (e.g., a coaxial cable) to a matching network 104 and then via an electrical connection 110 onto a plasma load 106. The matching network 104 varies its internal electrical elements (e.g., variable capacitors) so that the input impedance of the matching network 104 approaches a desired input impedance. The matching network typically contains reactive elements, meaning elements that store energy in electric and magnetic fields, as opposed to resistive elements that dissipate power. The most common reactive elements are capacitors, inductors, and coupled inductors, although others, such as distributed circuits, are also used.
[0006] Some matching networks include two or more variable reactance elements, such as the variable capacitors 202 and 204 shown in FIG. 2. Each of the variable capacitors can include multiple switched capacitors. The variable capacitors can include multiple fractional and full value capacitors that are switched in and out of the variable capacitors 202 and 204, respectively. Typically, the switched capacitors are switched into the individual variable capacitors 202 or 204 such that when all the switches are on, the switched capacitors are all in parallel. The value of the full value capacitor is dictated by the maximum current that either the switch or the capacitor can handle. For example, if the maximum voltage on the variable capacitor is 1000V, the operating frequency is 13.56MHz, and the switch can handle 4A, then the maximum value that the full value capacitor can have is 47pF. For example, if the switch has an off-state capacitance of 8.2pF, then the switch and capacitor combination (e.g., C in FIG. 3A) can be 1000pF. N and S.W. NA switch (see reference 1) adds 47pF to the total capacitance of the variable capacitor when the switch is on and 7pF when the switch is off. Such a capacitor-switch pair adds a change in capacitance of 7pF and 40pF to the baseline variable capacitor value, which is the "value" or "contribution" associated with the capacitor-switch combination when the switch state is changed. The amount of capacitance achieved as these switched capacitors are switched in and out of the variable capacitor can be mapped to a "capacitor setting", which can range from 0 to a convenient number such as 2047. A typical mapping is from 0 to 2, although other mappings such as 1.1 to 2.3 are also possible. n A mapping down to -1 (n is an integer) is generally convenient for implementation in digital hardware and software. When neither fractional value nor full value capacitors are switched into the variable capacitor, the variable capacitor has a finite but minimum capacitance corresponding to the lowest capacitor setting, which is 0 in this example. When all fractional value and full value capacitors are switched into the variable capacitor, the variable capacitor can have a maximum capacitance corresponding to the highest capacitor setting, which is 2047 in this example. By switching a combination of fractional value and full value capacitors into the variable capacitor, capacitances between this minimum and maximum can be achieved, thereby varying the capacitance of the variable capacitor.
[0007] Although two variable capacitors 202 and 204 are shown in FIG. 2, in other embodiments, one variable capacitor or more than two variable capacitors can be used. One or more inductors may also be arranged before or between the variable capacitors, and the illustrated configuration is only one of many implementations. The variable reactance section 206, including the variable capacitors 202, 204, can be arranged with, for example, fixed reactance sections 208 and 210, for example, comprising a capacitor and an inductor, although other embodiments of the fixed reactance sections 208 and 210 can also be implemented. Together, the fixed and variable reactance sections 208, 210, and 206 can constitute a matching network.
[0008] In other words, solid-state matching to create a variable capacitor using switched capacitors converts a digital capacitor setting to an analog capacitor value. The conversion is a form of digital-to-analog conversion in which fractional value (meaning less than full value) and full value capacitors are added together to set the variable capacitor value. For example, fractional value capacitors adding capacitances of 5pF, 10pF, and 20pF, along with five full value capacitors, each adding 40pF to a baseline value of 50pF, can theoretically create an effective variable capacitor that ranges from 50pF (baseline value) to 285pF (50+5×40+5+10+20=285). Each of these "contribution factors" or "values" represents the change in capacitance of the variable capacitor when a given capacitor or capacitors are switched into the circuit. As described above, the actual capacitors that are switched in and out of the variable capacitor have values that differ from the value they contribute to the total capacitance because, among other reasons, the switches used to switch the capacitors in and out of the circuit contribute to the effective capacitance of the circuit.
[0009] At least one existing solution recognizes that component tolerances can lead to gaps between capacitance values that a variable capacitor can achieve (see US 2018 / 0076788), but known solutions involve rounding larger capacitor values to the next standard value and continuing to use a binary sequence of fractional value capacitors, which mitigates some of the gap during increasing capacitance immediately after a full value capacitor is added, but increases the gap just before another full value capacitor is to be added as the capacitance increases. Summary of the Invention [Means for solving the problem]
[0010] The following presents a simplified summary related to one or more aspects and / or embodiments disclosed herein. As such, the following summary should not be considered an extensive overview related to all discussed aspects and / or embodiments, nor should the following summary be considered to identify key or essential elements related to all discussed aspects and / or embodiments or to delimit the scope associated with any particular aspect and / or embodiment. Thus, the following summary has the sole purpose of presenting certain concepts related to one or more aspects and / or embodiments related to the mechanisms disclosed herein in a simplified form in order to proceed to the detailed description presented below.
[0011] Some embodiments of the present disclosure may be characterized as a digital-to-analog converter (DAC) including a digital input, N contribution factors, an interconnection network, and an analog output. Each of the N contribution factors may have an on state and an off state controlled by the digital input. The interconnection network may be coupled to each of the N contribution factors and the analog output. The interconnection network may be configured to provide a sum of the contribution ratios of the N contribution factors to the analog output. For each of the N contribution factors, the contribution ratio of one of the N contribution factors may be the change in the analog output when a state of one of the N contribution factors is changed from off to on and the states of all remaining N-1 contribution factors remain the same. For each of the N contribution factors, the contribution ratio of at least one of the N contribution factors may vary as the analog output is varied, and a maximum-to-minimum ratio of the contribution ratio of at least one of the N contribution factors is at least 1.1. The N contribution factors may be ordered from smallest to largest average contribution ratio to form an ordering. The gap size D is less than or equal to twice the maximum contribution rate of the first of the N contributors. For k ranging from 2 to N, the maximum contribution rate of the kth contributor is less than or equal to D plus the sum of the minimum contribution rates of contributors 1 through k-1.
[0012] Other embodiments of the present disclosure may also be characterized as a matching network comprising one or more variable capacitors, each of which includes a digital input, N switched capacitors, an interconnect network, and an analog output. Each of the N switched capacitors may have an on state and an off state controlled by the digital input and the output. The interconnect network may be coupled to each of the N switched capacitors and configured to provide a sum of the contributions of the N switched capacitors to the capacitance of the variable capacitor across the terminals of the output. For each one of the N switched capacitors, the contribution of one of the N switched capacitors is the change in the variable capacitor capacitance when the state of one of the switched capacitors is switched from off to on and the states of all the remaining N-1 switched capacitors remain the same. For each one of the N switched capacitors, the contribution of at least one of the switched capacitors varies across a range based on the states of the remaining N-1 switched capacitors. The N switched capacitors may be ordered from smallest to largest through an average of the range of each switched capacitor's contribution to the variable capacitor capacitance. The gap size D is less than or equal to twice the maximum contribution of the first of the N switched capacitors. For k ranging from 2 to N, the maximum contribution of the kth switched capacitor can be less than or equal to D+the sum of the minimum contributions of switched capacitor 1 through switched capacitor k-1.
[0013] Another embodiment of the present disclosure may be characterized as a method of forming a variable capacitor, the method including providing a digital input, providing N switched capacitors coupled together via an interconnect topology, providing an analog output from the interconnect topology as a sum of the capacitance contributions of the N switched capacitors, and selecting the N capacitors to form a sub-binary sequence, where a maximum contribution of the kth switched capacitor is less than or equal to a gap size D plus the sum of the minimum contributions of the 1st through (k-1)th switched capacitors.
[0014] In another embodiment, a method for using a matching network is disclosed, the method comprising providing a digital input to a variable capacitor of a matching network, and setting a switch configuration corresponding to the digital input, the switch configuration controlling an on or off state of N switches in the variable capacitor, each of the N switches making a series connection between one of the N capacitors and an interconnection topology, the interconnection topology providing a sum of the capacitances of the N capacitors to an output, and for each of the N capacitors, the contribution to the sum of the capacitances is reduced when one of the N capacitors is switched from on to off and all remaining N-1 pcs The method may include: a difference in the sum of capacitances when the states of the capacitors remain the same, the contributions to the sum of capacitances varying across a range based on the states of the remaining N-1 capacitors, a gap size D between the maximum contribution of one of the N capacitors and the minimum contribution of successive ones of the N capacitors being less than or equal to twice the maximum contribution of the first of the N capacitors, the N capacitors being ordered according to average contributions across each of their ranges of contributions, for k ranging from 2 to N, the maximum contribution to the sum of capacitances of the kth capacitor being less than or equal to D+the sum of the minimum contributions of the first through (k-1)th capacitors; and passing a voltage or current through a matching network, where the impedance experienced by the voltage or current is affected by the sum of the capacitances of the N capacitors. The present specification also provides, for example, the following items: (Item 1) 1. A digital-to-analog converter, comprising: Digital input and N contribution factors, each having an on state and an off state controlled by the digital input; an interconnection network coupled to each of the N contribution factors and to an analog output, the interconnection network configured to provide a sum of the contributions of the N contribution factors to the analog output; For each of the N contributing factors, the contribution ratio of one of the N contributors is the change in the analog output when the state of one of the N contributors is changed from off to on and the states of all the remaining N-1 contributors remain the same; a contribution ratio of at least one of the N contribution factors varies as the analog output is varied, and a maximum to minimum ratio of the contribution ratio of at least one of the N contribution factors is at least 1.1. Interconnected Networks and Equipped with The N contribution factors are ordered from smallest to largest average contribution rate to form an ordering; the gap size D is less than or equal to twice the maximum contribution ratio of a first one of the N contribution factors; For k ranging from 2 to N, the maximum contribution of the kth contributor is less than or equal to the sum of the minimum contributions of D+contributor 1 through contributor k-1. Digital / Analog Converter. (Item 2) 2. The digital-to-analog converter of item 1, wherein the average contribution rate of at least one of the N contribution factors is the square root of the product of the maximum contribution rate and the minimum contribution rate of at least one of the N contribution factors. (Item 3) There are four or more fractional contribution factors, a first ratio of an average contribution rate of contributor 4 in the ordinality of the N contributors to an average contribution rate of contributor 3 in the ordinality of the N contributors is at least 1.2; a second ratio of an average contribution rate of contributor 4 in the ordinality of the N contributors to an average contribution rate of the first contributor in the ordinality of the N contributors is less than 6; Item 1. A digital-to-analog converter as described in item 1. (Item 4) There are five or more decimal contribution factors, a third ratio of an average contribution rate of contributor 5 in the ordinality of the N contributors to an average contribution rate of contributor 4 in the ordinality of the N contributors is at least 1.2; a fourth ratio of an average contribution rate of a contributor 5 in the ordinality of the N contributors to an average contribution rate of the first contributor in the ordinality of the N contributors is less than 12; Item 1. A digital-to-analog converter as described in item 1. (Item 5) There are six or more fractional contributors, a fifth ratio of an average contribution rate of contributor 6 in the ordinality of the N contributors to an average contribution rate of contributor 5 in the ordinality of the N contributors is at least 1.2; a sixth ratio of an average contribution rate of a sixth contributor in the ordination of the N contributors to an average contribution rate of the first contributor in the ordination of the N contributors is less than 20; Item 1. A digital-to-analog converter as described in item 1. (Item 6) 2. The digital-to-analog converter of claim 1, wherein the analog output has a monotonic relationship to the digital input, and a maximum gap size of the analog output per increment of the digital input is less than D. (Item 7) 2. The digital-to-analog converter of claim 1, wherein the analog output value is the capacitance of a variable capacitor. (Item 8) 8. The digital-to-analog converter of claim 7, wherein the contribution factor is a capacitor in series with a switch. (Item 9) 9. The digital-to-analog converter of claim 8, wherein the switch is a PIN diode. (Item 10) A matching network comprising one or more variable capacitors, each of the one or more variable capacitors having: Digital input and N switched capacitors, each having an on state and an off state controlled by said digital inputs and outputs; an interconnection network coupled to each of the N switched capacitors, the interconnection network configured to provide a sum of the contributions of the N switched capacitors to a capacitance of the variable capacitor across the terminals of the output; For each one of the N switched capacitors: the contribution of one of the N switched capacitors is the change in the variable capacitor capacitance when the state of one of the switched capacitors is switched from off to on and the states of all remaining N-1 switched capacitors remain the same; the contribution of at least one of the switched capacitors varies across a range based on the state of the remaining N-1 switched capacitors. Interconnected Networks and Equipped with the N switched capacitors are ordered from smallest to largest through an average of each switched capacitor's range of contribution to the variable capacitor capacitance; a gap size D is less than or equal to twice the maximum contribution of a first of said N switched capacitors; For k ranging from 2 to N, the maximum contribution of the kth switched capacitor is less than or equal to the sum of the minimum contributions of D+ switched capacitor 1 through switched capacitor k-1. Matching network. (Item 11) Item 11. The matching network of item 10, wherein the range of contributions of at least one of the switched capacitors is such that a ratio of a maximum contribution of at least one of the switched capacitors to a minimum contribution of at least one of the switched capacitors is at least 1.1. (Item 12) a first ratio of an average contribution of a fourth of the N switched capacitors to an average contribution of a third of the N switched capacitors is at least 1.2; a second ratio of an average contribution of a fourth one of the N switched capacitors to an average contribution of a first one of the N switched capacitors is less than 6. Item 11. The matching network according to item 10. (Item 13) a third ratio of an average contribution of a fifth of the N switched capacitors to an average contribution of a fourth of the N switched capacitors is at least 1.2; a fourth ratio of an average contribution of a fifth of the N switched capacitors to an average contribution of a first of the N switched capacitors is less than 12. Item 11. The matching network according to item 10. (Item 14) a fifth ratio of an average contribution of a sixth of the N switched capacitors to an average contribution of a fifth of the N switched capacitors is at least 1.2; a sixth ratio of an average contribution of a sixth of the N switched capacitors to an average contribution of a first of the N switched capacitors is less than 20. Item 11. The matching network according to item 10. (Item 15) Item 11. The matching network of item 10, wherein the variable capacitor capacitance has a monotonic relationship to the digital input, and a maximum gap size of the variable capacitor capacitance per increment of the digital input is less than D. (Item 16) 1. A method of forming a variable capacitor, the method comprising: Providing a digital input; providing N switched capacitors coupled together via an interconnect topology; providing an analog output from said interconnect topology as a sum of the capacitance contributions of said N switched capacitors; selecting the N capacitors to form a sub-binary sequence, wherein a maximum contribution of a kth switched capacitor is less than or equal to a gap size D plus the sum of minimum contributions of a first switched capacitor through a (k-1)th switched capacitor; A method comprising: (Item 17) Item 17. The method of item 16, wherein the gap size D is less than or equal to two times the maximum contribution of a first of the N switched capacitors. (Item 18) 17. The method of claim 16, wherein each of the N switched capacitors has a range of contributions to a sum of capacitance contributions, and for at least one of the N switched capacitors, the range is such that a ratio of maximum to minimum contributions of at least one of the N switched capacitors is at least 1.1. [Brief description of the drawings]
[0015] Various objects and advantages of the present disclosure and a more complete understanding will become apparent and be more readily understood by reference to the following detailed description and appended claims, when considered in connection with the accompanying drawings.
[0016] [Figure 1] FIG. 1 illustrates a typical generator, matching network, and plasma load system.
[0017] [Diagram 2] FIG. 2 illustrates a plasma processing system including a matching network having a variable reactance section.
[0018] [Figure 3A] FIG. 3A illustrates a variable capacitor embodiment that utilizes a switched capacitor.
[0019] [Figure 3B] FIG. 3B illustrates an equivalent circuit for FIG. 3A.
[0020] [Figure 4] FIG. 4 illustrates a plot of analog output as a function of digital input, where binary fractional values are used.
[0021] [Diagram 5] FIG. 5 illustrates the steps in the analog output when the digital input is incremented by one for the relationship illustrated in FIG.
[0022] [Figure 6] FIG. 6 illustrates the contribution to the analog output due to the minimum fractional value contribution factor as a function of the digital input.
[0023] [Figure 7] FIG. 7 illustrates the contribution to the analog output by the second least-fractional value contribution factor as a function of the digital input.
[0024] [Figure 8] FIG. 8 illustrates the contribution to the analog output by the fifth least decimal value contribution factor as a function of the digital input.
[0025] [Figure 9] FIG. 9 illustrates the contribution to the analog output by the third full-value contribution factor (a total of 10th contribution factor, including 7 fractional-value contribution factors) as a function of the digital input.
[0026] [Figure 10] FIG. 10 illustrates the contribution to the analog output by the 13th full-value contribution factor (out of the 20th contribution factor, which includes 7 fractional-value contribution factors) as a function of the digital input.
[0027] [Figure 11] FIG. 11 shows a method for calculating the contribution factor values to ensure the maximum step in the analog output when the digital input is incremented by one.
[0028] [Figure 12] FIG. 12 is a more detailed embodiment of the method described and shown with respect to FIG.
[0029] [Figure 13] FIG. 13 illustrates analog output as a function of switch configuration when values of the individual contribution factors in one embodiment are calculated according to the method of FIG.
[0030] [Figure 14] FIG. 14 illustrates the change in the analog output as the switch configuration is incremented by one and the values of the individual contribution factors are calculated according to the method of FIG.
[0031] [Figure 15] FIG. 15 illustrates the mapping between digital inputs and switch configurations.
[0032] [Figure 16] FIG. 16 illustrates the analog output as a function of the digital input when the mapping of FIG. 15 is applied between the switch configuration and the digital input and the values of the individual contribution factors are calculated according to the method of FIG.
[0033] [Figure 17] FIG. 17 illustrates the change in the analog output when the digital input is incremented by one, the mapping of FIG. 15 is applied between the switch configuration and the digital input, and the values of the individual contribution factors are calculated according to the method of FIG. 11.
[0034] [Figure 18A] FIG. 18 illustrates an embodiment of a digital-to-analog (DAC) and its equivalent circuit. [Figure 18B] FIG. 18 illustrates an embodiment of a digital-to-analog (DAC) and its equivalent circuit.
[0035] [Figure 19] FIG. 19 illustrates a power system including a power supply, a matching network, and a load.
[0036] [Figure 20]FIG. 20 illustrates a block diagram depicting physical components that may be utilized to realize a controller of the interconnect network, DAC, or matching network in FIG. 3A, according to an example embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0037] A problem that arises in the context of variable capacitors utilizing switched capacitors is that each switched capacitor contributes a variable amount to the total capacitance when switched, depending on the total capacitance of the variable capacitor. For example, as illustrated in FIG. 6 in combination with FIG. 13, a 7.5pF fractional value capacitor with a 4.17pF contribution (meaning that switching this 7.5pF capacitor into the circuit should nominally change the total capacitance of the variable capacitor by 4.17pF) may actually contribute a total of 4.4pF when the total capacitance of the variable capacitor (e.g., 202 or 204) is low, such as 1nF, while that same 7.5pF fractional value capacitor may only contribute 3.3pF when the total capacitance of the variable capacitor is high, such as 3.5nF. This variability is in part a result of the parasitic effects of the structures used to connect the capacitors in the variable capacitor to the common input and output nodes. The plots in FIG. 6-10, discussed in detail later, show examples of simulated capacitance values for the capacitors as a function of switch configuration. A switch configuration is used herein to mean some combination of switched capacitors that are switched in and out of a variable capacitor to produce the capacitance for the variable capacitor.
[0038] For example, in FIG. 6, a simulated plot of the capacitance contribution of a 7.5 pF capacitor as a function of switch configuration is seen. The switch configurations are on or off, thus showing different configurations of switches coupling the switched capacitor in and out of the variable capacitor. For each of the switch configurations, there is a resulting capacitance contribution for this 7.5 pF capacitor, as shown on the y-axis. In this non-limiting example, as the number of switched capacitors switched into the variable capacitor increases, the contribution from the 7.5 pF capacitor generally decreases (although the local value does in fact go up and down as the switch configurations increase). To derive these simulated results, the 7.5 pF capacitor is switched on and off for every switch configuration, and the capacitance difference between these two states, i.e., the contribution, is determined for every switch configuration. Component tolerances may also contribute to this variability. Although the simulations shown in FIGS. 6-10 show what appears to be a downward relationship between capacitance contribution and switch configuration, it should be noted that this is merely illustrative and that in other embodiments the minimum and maximum contributions may occur anywhere within the plots.
[0039] Traditionally, variable capacitors utilizing switched capacitors are formed from a binary sequence of "fractional value contribution factors" such as 5pF, 10pF, and 20pF, along with one or more "full value contribution factors" such as 40pF. In this example, if only these four capacitors are implemented into the variable capacitor, the resulting variable capacitor may have a "baseline value" of 50pF (the capacitance contribution with all switched capacitors switched out of the variable capacitor), and thus may contribute a capacitance of 50 to 125pF. The maximum contribution of 125pF results from the baseline value of 50pF plus the contribution of all four capacitors, 75pF (5+10+20+40), when switched "on" or into the variable capacitor, which gives 125pF. Note that these values are not the actual values of the four capacitors, but rather are simply the change in capacitance that occurs when one of these capacitors is switched in. The lowest switch configuration, providing a 50pF contribution, may have a switch configuration or digital capacitor setting of 0 (the lowest value of the variable capacitor), and the highest switch configuration, providing a 125pF contribution, corresponding to a switch configuration or digital capacitor setting of 31 (or another convenient number, e.g., 1). In this example, the contributions may have a 5pF step size. In other words, a digital input of 0 to 31 sets the variable capacitor analog output from 50 to 125pF, with 5pF steps between the total capacitance values.
[0040] If the present variable capacitor were modified to include three additional full-value 40 pF capacitors, the baseline contribution would increase from 50 pF to 66 pF, and the variable capacitor would contribute from 66 pF to 261 pF (66 + 40 x 4 + 20 + 10 + 5 = 261). By switching a combination of these fractional and full-value capacitors into the variable capacitor, a range of capacitance values between a minimum (e.g., 66 pF) and a maximum (e.g., 261 pF) can be achieved. For such a variable capacitor, the expected step ("step size") in the capacitor contribution as the switch configuration or digital input is incremented by one would again be the same as 5 pF, i.e., the minimum fractional value contribution factor. In practice, it has been found that the actual step size can be many times larger than the expected step size. A larger step size can lead to a gap in the capacitor value that can be achieved. Typically, matching networks are designed such that gaps in the capacitor values can match the input impedance of the matching network to the desired input impedance with sufficient tolerance below a certain threshold, e.g., 5 pF. These gaps are therefore problematic because they can, in some circumstances, leave the match non-impedance matchable.
[0041] To help illustrate the solutions disclosed herein to these problems, FIG. 3A illustrates a variable capacitor circuit diagram embodying aspects of the present disclosure. In particular, FIG. 3A illustrates a variable capacitor that utilizes a switched capacitor. In FIG. 3A, capacitor C 1 -C N are switch SW 1 -SW N Each switch has an off-state capacitance C SWwhere the capacitor and switch are connected through an interconnect network. For analysis purposes, the interconnect network should model the effect of the network connecting the capacitor and switch to the terminals of the variable capacitor, which may include parasitic elements such as switch drivers. From this model of the variable capacitor, an equivalent network for the variable capacitor, such as that shown in FIG. 3B, can be derived, in which the parasitic series inductance L S is extracted from the network. This parasitic inductance can be included in the design of the matching network. L S The capacitor C in FIG. 3B that remains after removing parasitics such as 100 nH inductor (L in FIG. 3B) is the capacitor value that will be used in further analysis of the capacitor. While this may seem like a minor detail, not removing the nominal parasitics can lead to difficulties in determining the range of values that a particular switched capacitor will contribute to the total capacitance as the total capacitance is varied. For example, if C in FIG. 3B is a 20 nH inductor (L in FIG. 3B) s ) in series with the MOSFET, which varies from 1 to 4 nF, and the capacitor, which varies from 1.17 nF to 9.54 nF when measured at 13.56 MHz in Figures 3A and 3B, C e The parasitic elements will produce C e can be extracted by measuring C, fitting a circuit model to the measured data, and assuming that C in FIG. 3B is frequency independent.
[0042] Although the prior art has sought to address these gaps in variable capacitor values based on capacitor tolerance issues, parasitics can make these gaps even larger than would be expected when only capacitor tolerances are considered. That is, each fractional value capacitor has a variable contribution depending on the total capacitance provided by the capacitor setting. As more of the capacitors are switched into the variable capacitor, the contribution from any one of the capacitors, especially the smaller one, changes. This can lead to larger gaps than would be expected by the prior art, which simply considers gaps caused by device tolerances, not parasitics.
[0043] 4 shows the total capacitance of the variable capacitor as a function of switch configuration when binary fractional capacitor values are used in a circuit such as that depicted in FIG. 3A, leading to a gap in the total capacitance values. Referring to FIG. 3A, the switch configurations are: [ka] where M is the number of full-value capacitors switched into the variable capacitor, N frac is the number of fractional capacitors, S k is the switch SW k is on, it is 1, and the switch SW k is off, it is zero. For switched variable capacitors employing fractional switches, one can simply consider every possible switch configuration, regardless of whether the switches are associated with fractional-value or full-value capacitors, but this becomes impractical for switched variable capacitors with a large number of full-value capacitors (e.g., greater than 15 or greater than 20). For example, consider the example of a variable capacitor with 5 fractional-value and 45 full-value capacitors. Considering all combinations of fractional-value capacitors, but turning on the full-value capacitors in a defined sequence, the number of possible combinations considered is 46×2 5= 1472. If we do not make the distinction between fractional and full value capacitors and consider all possible combinations, the number of possible combinations is 2 50 , which is an astronomical number (if measurements were made every microsecond, it would take 35 years to measure all the configurations).
[0044] Figure 5 shows one illustration of the gradual increase in capacitance for the variable capacitor as the switch configuration is incremented by one. From Figures 4 and 5, it can be seen that a large gap exists in the total capacitance. The gap is about 8 times the expected 5pF gap (since the smallest decimal contribution factor is 5pF in this example) and is a result of not considering the effect of the interconnect network of Figure 3A and the resulting changing contribution of the switched capacitor to the total capacitance of the variable capacitor as the total capacitance is varied, as depicted in Figures 6-10.
[0045] The proposed solution is to consider the impact of all effects, including but not limited to the impact of the interconnect network of FIG. 3A and component tolerances, and choose a "sub-binary" sequence of fractional values (rather than a "binary" sequence) when designing the variable capacitor.
[0046] For purposes of this disclosure, a "sub-binary" sequence is a sequence of numbers where the ratio between consecutive numbers is less than 2. For example, for capacitor values, a binary sequence may include 5pF, 10pF, 20pF, and 40pF, where each value is two times greater than the previous value, while a sub-binary sequence may include 5pF, 8.4pF, 14.1pF, 23.8pF, and 40pF, or 5pF, 7.6pF, 11.6pF, 17.6pF, 26.7pF, and 40pF, where each value is less than two times greater than the previous value. Also, note that because the sub-binary sequences disclosed herein involve values that are closer together than in prior art binary equivalents, one or more extra capacitors (contributing factors) may be required in the variable capacitor to ensure that the sequence can cover the same value range as a conventional binary sequence. For example, to cover a range of binary sequences including values 5 (fractional value), 10 (fractional value), 20 (fractional value), and 40 (full value), an exemplary sub-binary sequence may include additional fractional values to cover the same range (e.g., 5 (fractional value), 8.4 (fractional value), 14.1 (fractional value), 23.8 (fractional value), and 40 (full value)). Thus, in some sense, the proposed solution actually adds complexity, cost, and components, and thus would not be an obvious design choice to one of ordinary skill in the art.
[0047] FIG. 11 shows a procedure for creating a sub-binary sequence of capacitor values in a variable capacitor utilizing switched capacitors with a total of N switched capacitors and a maximum capacitor value of maxC, ensuring that the maximum gap dC in the variable capacitor value (total capacitance of the variable capacitor) is not exceeded. The method 1100 of FIG. 11 will be described with reference to the example circuit and its capacitance plots shown in FIGS. 6-10. Before starting the procedure (block 1105), an initial design of the variable capacitor as shown in FIG. 3A can be analyzed. The nature of the interconnect network of FIG. 3A can be analyzed using standard techniques (e.g., electromagnetic (e.g., HFSS) and circuit analysis (e.g., SPICE and MATLAB®) simulation software). With the nature of the interconnection network known, and using an initial guess of the switched capacitor values that may be used (e.g., assuming a binary sequence of contributions such as 5, 10, 20, and 40 pF, with an estimate of the switch capacitance of, e.g., 6 pF, results in a sequence of 8.5, 14.2, 24.8, and 45.3 pF), the total capacitance of the variable capacitor can be calculated for all switch configurations. When nominal parasitic elements are removed from the circuit, an equivalent circuit such as that of FIG. 3B can be arrived at. For each of the switch configurations (x-axis in FIGS. 6-10), each switched capacitor is switched in and out of the variable capacitor circuit to find the simulated contribution of the switched capacitor to the total capacitance of the variable capacitor (C in FIG. 3B) as a function of switch configuration. The result of this procedure is a set of data as depicted in FIG. 6 for all the switched capacitors in the variable capacitor, including the simulated minimum contribution and the simulated maximum contribution for each of the switched capacitors. For example, the 7.5 pF capacitor in Figure 6 has a maximum contribution Cmax(1) of approximately 4.4 pF and a minimum contribution Cmin(1) of approximately 3.3 pF. This type of data can be simulated for each of the switched capacitors in the variable capacitor.For example, Figures 6-10 represent data for 4 of the 20 switched capacitors in a variable capacitor having 7 fractional value and 13 full value switched capacitors. Given the Cmax and Cmin from these sets of data for each switched capacitor, the method 1100 can then use these simulated minimum and maximum contributions for each of the switched capacitors and perform an iterative set of steps to determine optimal switched capacitor values that reduce the gap between values found in the prior art (e.g., the gap in Figure 4).
[0048] 11 shows a method 1100 in which the first fractional value is selected to have a maximum contribution Cmax(1) less than the desired gap size dC, and all subsequent fractional values can be selected such that the maximum contribution Cmax(k) of the next fractional value is equal to or less than the sum of the minimum contributions of all previously selected fractional values plus the maximum step size. In equation form, this can be written for the kth contribution factor, where k>1, as follows: [ka]
[0049] This results in a sub-binary ascending set of decimal values that may contain one or more additional decimal values than a typical binary sequence. Furthermore, once the Cmax(k) value is determined, the actual capacitor can be selected with the understanding that the selection is often limited to standard capacitor values (hence the notion that Cmax(k) is equal to or less than a certain value).
[0050] For example, a simulation may show that the first decimal value has a maximum contribution Cmax(1) of 5pF and a minimum contribution Cmin(1) of 3pF (e.g., taking into account parasitic and tolerance effects). Using Equation 1, the maximum contribution Cmax(2) of the second decimal value may be equal to or less than the minimum contribution of the first decimal value plus the desired step size, i.e., 3pF+5pF=8pF. The simulation may further show a minimum contribution Cmin(2) for this second decimal value of 6pF. Equation 1 then gives a Cmax(3) for the third decimal value that is equal to or less than the minimum contribution of the first and second decimal values (3pF+6pF=9pF) plus the step size (e.g., 5pF), i.e., 3pF+6pF+5pF=14pF. The simulation may then show that the minimum contribution Cmin(3) of the third decimal value is 11pF. Equation 1 then gives Cmax(4) for the fourth decimal point, which is equal to or less than the sum of the smallest decimal points for the previous decimal points plus the step size, i.e., 3pF+6pF+11pF+5pF=25pF. Simulations can then show a minimum contribution Cmin(4) for this fourth decimal point of 19pF. Thus, for this example step size dC of 5pF, the maximum contributions of the first four decimal points can be equal to or less than 5pF, 8pF, 14pF, and 25pF.
[0051] Given these maximum contributions Cmax(k) for the fractional value capacitors, the maximum contributions for the full value capacitors can be selected. Full value capacitors also use Equation 1 and therefore have a Cmax(5) less than or equal to 3pF+6pF+11pF+19pF+5pF=44pF. Any number of full value capacitors with this Cmax(5) may be implemented.
[0052] Given these maximum contributions Cmax(k), actual capacitor values C(k) that result in these maximum contributions are then selected, keeping in mind that the capacitors are standard values, and therefore the actual capacitor values may not correspond to the calculated C(k) values determined above. Once the actual capacitor values have been selected, the sub-binary solution can be iterated to ensure that the selected capacitor values still lead to a viable solution, and modifications to the selected capacitors can be made if necessary.
[0053] Details of FIG. 11 will now be described using a specific example as the discussion proceeds. Starting with a first switched capacitor k=1 (block 1110), block 1115 develops a set of data such as FIG. 6 for the first switched capacitor (e.g., capacitance C(1)=7.5 pF) and finds a range of simulated contributions for the first switched capacitor (e.g., 3.27-4.38 pF according to FIG. 6). The switch capacitance C(1) of FIG. 3A is SW (e.g., 6 pF), C 1 and SW 1 The expected contribution of the combination of pF and C in FIG. 3A is the difference between their combined capacitances in the on and off states (e.g., 7.5-1 / (1 / 7.5+1 / 6)pF=4.17pF). k (Note that k is C(k) in FIG. 11). In the example of FIG. 6, the actual capacitance contribution varies from 78% to 105% of the expected 4.17 pF contribution. Block 1115 then targets the maximum gap dC (e.g., 5 pF) and determines the value of the first switched capacitor C(1) that should result in a maximum contribution Cmax(1) of the first capacitor that is less than or equal to the maximum gap dC. In the example, the nominal contribution of the first switched capacitor is 5 / 1.05=4.76 pF, ensuring that the maximum contribution is less than 5 pF (i.e., the gap size dC divided by the maximum ratio of the actual contribution and the expected contribution determined by simulated data (e.g., FIG. 6) and the expected contribution obtained through calculation).
[0054] Returning to the embodiment of FIG. 6, the switched capacitor therefore: [ka] The first switched capacitor should have a capacitance C(1) of 7.73 pF (e.g., 8.23 pF-0.5 pF). Block 1115 can then subtract the capacitor tolerance from this calculated value to arrive at the desired capacitance C(1) of the first switched capacitor. In this example, assuming a capacitor tolerance of 0.5 pF, the maximum capacitance C(1) for the first switched capacitor is 7.73 pF (e.g., 8.23 pF-0.5 pF). Block 1115 can then determine the closest standard capacitance that is smaller than C(1), which in this example is 7.5 pF. Thus, the first switched capacitor capacitance C(1) in this example as determined in step 1115 is 7.5 pF.
[0055] In this example, the capacitors used to simulate the minimum and maximum contributions and the capacitors selected via block 1115 are the same. However, in other cases, the simulated capacitors and those selected by block 1115 may be different. In these cases, subsequent analysis (i.e., after all switched capacitor values are calculated, the circuit is analyzed with the new capacitor values and all calculations in FIG. 11 are repeated) typically converges quickly to a final set of switched capacitor values C(k). With the first capacitor value C(1) at hand, the determination of the second capacitor value C(2) involves a maximum step size dC and the minimum contribution of the first capacitor value C(1). Thus, the minimum contribution of the first switched capacitor is determined (block 1125). The first capacitor C 1The minimum expected contribution Cmin(1) of is calculated as 0.78×[(7.5−0.5)−1 / (1 / (7.5−0.5)+1 / 6))pF=2.94pF, which is found to be less than the simulated minimum seen in FIG. 6, which considers only parasitic effects and not component tolerances. The second fractional value can be selected (k=2, step 1125) such that the maximum contribution Cmax(2) of the second fractional value is as large as possible while remaining equal to or less than the minimum contribution Cmin(1) of the first fractional value plus the maximum step size dC. In this example, the second capacitor C 2 The maximum contribution of Cmax(2) should therefore be less than or equal to 2.94 + 5 pF = 7.94 pF. 2 The simulated maximum and minimum contributions Cmax(2) and Cmin(2), respectively, of the MOSFET are again found via simulation and calculation as described above. In our example, FIG. 7 shows a MOSFET with a value of 11 pF, C 2 The simulation results show that the C 2 The contribution rate of C 2 The expected contribution of the 2 The expected contribution of should therefore be less than or equal to 7.94 / 1.03pF=7.7pF, which in turn means that 2 This means that the capacitance of the capacitor should be less than or equal to 11.66pF. Assuming a 2% capacitor tolerance, the maximum value of C(2) should be less than or equal to 11.66 / 1.02pF=11.43pF. The largest standard capacitor value that is less than 11.43pF is 11pF. C 2The value C(2) is thus selected to be equal to 11 pF, which also results in the same value used in the simulation. This selected capacitor value of C(2) = 11 pF is below maxC, which in this embodiment is set to 180 pF based on the maximum current that the switch can handle (block 1130), and thus method 1100 iterates (block 1135) and, since the total number of switched capacitors (N) has not been reached (block 1120), selects a capacitor value for C(3) (block 1125).
[0056] C 2 The minimum contribution rate Cmin(2) of C is found to continue as such, as method 1100 loops through its calculation.
Chemical formula
[0057] Method 1100 can be looped through step 1125 until all contributing factors are considered (i.e., when nk < N + 1 in decision 1120 or 1150). This results in a set of fractional sub - binary ascending order, which can contain one or more additional fractional values over a typical binary sequence. The process can end at step 1120 when N fractional values are selected and no full - value capacitors are required, or when some full - value capacitors are selected at step 1140. Since many full - value capacitors may be required, when determining the maximum contribution rate for full - value capacitors at step 1125, the maximum value for all full - value capacitors should be used. As an example, referring to FIGS. 9 and 10, capacitor C(10) is found to contribute up to a maximum of 1.07 to the expected contribution rate, and capacitor C(20) is found to contribute up to a maximum of 1.09 to the expected contribution rate. Considering all full - value capacitors, the maximum ratio of the maximum value to the expected contribution rate can be found to be 1.10, and thus a factor of 1.1 should be used to find the maximum value that full - value capacitors should take.
[0058] As described above, the solutions disclosed herein add complexity and cost to previous attempts because one or more extra bits may be required in the sub-binary sequence to cover the range of the binary sequence. Despite the added complexity and cost, the sub-binary sequence unexpectedly allows for smaller gaps and allows a user or administrator to reliably select a maximum step size, thereby allowing for more precise adjustment of the match.
[0059] FIG. 12 illustrates another embodiment of a procedure for creating a sub-binary sequence of capacitor values to ensure that the maximum gap in variable capacitor values dC is not exceeded in a variable capacitor utilizing switched capacitors, with a total of N switched capacitors, each having a maximum possible capacitor value of maxC. First, in block 1210, a simulation of a variable capacitor with N switched capacitors connected by an interconnection network, such as the variable capacitor in FIG. 3A, is created, and initial capacitor values are selected for the switched capacitors. Using this simulation, the properties of the interconnection network are then analyzed in block 1215, and an equivalent circuit without nominal parasitics, such as the circuit in FIG. 3B, is generated using those properties. Next, in block 1220, a simulated range of the contribution of each switched capacitor to the total capacitance of the variable capacitor is found for each of the switch configurations. To find the simulated contribution of the switched capacitor, each switched capacitor is switched in and out of the variable capacitor circuit, and the difference between the capacitance of the variable capacitor in the two switch states is determined to be the simulated contribution of the switched capacitor. This contribution determination process is repeated for all switch configurations, generating a mapping between the switch configurations and the simulated contributions of each switched capacitor in Figures 6-10, etc. Block 1221 calculates the maximum capacitance, maxC, for the full-value switched capacitor based on the maximum current the capacitor and switch can handle.
[0060] Then, starting with a first switched capacitor k=1 (block 1225), the maximum allowable gap dC in the capacitance of the variable capacitor is selected in block 1230, and the capacitance C(1) of the first switched capacitor is selected to be as large as possible while still having its simulated maximum contribution Cmax(1) less than dC. Cmax(1) is found using the simulated contribution data produced in block 1220 and is then used along with the capacitance of the switch portion of the first switched capacitor and the tolerance of the first switched capacitor to calculate the maximum desired capacitance of the first switched capacitor. Block 1230 may then determine the closest standard capacitance that is less than this maximum desired capacitance and assign the closest standard capacitance value to the first switched capacitor as C(1). The method 1200 then proceeds to a second switched capacitor k=2 (block 1235) and checks whether the switched capacitor is included in the total number of switched capacitors N in block 1240. If the next switched capacitor is not included in the total number of switched capacitors, the method 1200 proceeds to end in block 1285 since all switched capacitors in the variable capacitor have been assigned capacitance values. Otherwise, the method 1200 continues in block 1250, where the minimum expected contribution Cmin(k) of all switched capacitors that have been assigned capacitance values is summed with dC to provide an upper limit on Cmax(k) (block 1255). Each Cmin(k) value is calculated using the simulation contribution data produced in block 1220 and the expected contribution of each switched capacitor, taking into account the capacitor tolerances. Block 1255 then determines the closest standard capacitance that results in a maximum contribution that is less than or equal to this maximum desired capacitance contribution Cmax(k) and assigns that closest standard capacitance value to the current-switched capacitor as C(k).
[0061] The assigned capacitor value C(k) is then compared to the maximum capacitance maxC of a full value capacitor in block 1265 (recall block 1221). If the assigned value C(k) of the current switched capacitor exceeds the maximum capacitance maxC for a full value switched capacitor, C(k) is overwritten to be equal to maxC in block 1270. In addition, the remaining switched capacitors are also assigned a capacitor value of maxC by iteratively iterating to the next switched capacitor in block 1275 and assigning it the value of maxC in block 1270 until the iteration number reaches a switched capacitor number that exceeds the total number of switched capacitors N, as determined in block 1280. Otherwise, if the assigned C(k) in block 1255 does not exceed maxC, the method 1200 continues to the next switched capacitor in block 1260 and repeats the process described above, beginning with block 1240.
[0062] Figure 13 shows an example of total capacitance (C in Figure 3B) as a function of switch configuration for a variable capacitor using switched capacitors, with the capacitor value calculated according to the procedure of Figure 11. Figure 14 shows the steps in capacitor value as the switch configuration is incremented by 1 as a function of switch configuration. Figure 14 is a confirmation that the procedure of Figure 11 ensures that the steps in capacitor value do not exceed the desired maximum step size dC (i.e., none of the steps shown are greater than 5 pF).
[0063] Given the calculated contributions for fractional and full value capacitors, it may be desirable to have a linear or monotonic relationship between the capacitor setting, which may be a digital input applied to the variable capacitor, and the resulting capacitance of the variable capacitor. FIG. 13 shows the relationship between the variable capacitor capacitance and the switch configurations for a variable capacitor with switched capacitor values selected using method 1100. Thus, an additional step to methods 1100 and 1200 may include determining a mapping between capacitor settings and switch configurations. This process is performed for physically constructed variable capacitors incorporating capacitors with values selected as described above to ensure that the maximum step is less than the desired step size. Because of the way the capacitor values are calculated, it is guaranteed that if all switch configurations are used, a sequence of switch configurations can be found that are mapped to capacitor settings such that capacitor settings from 0 to 1 (or 0 to 2047 or another preferred mapping) correspond to the actual variable capacitor capacitance from minimum to maximum in a monotonically increasing manner, with no steps greater than the desired step size dC. FIG. 15 is an example of such a mapping between capacitor settings on the horizontal axis and switch configurations on the vertical axis. The vertical axis is the switch configuration that results in a capacitor value that most closely matches the capacitor value that should correspond to the capacitor setting on the horizontal axis. The process involves measuring the capacitance for all switch configurations, for example with a network analyzer, removing nominal parasitics to arrive at a model such as FIG. 3B, and then selecting those switch configurations that are closest to each of the discretized capacitor settings. The discretization of the capacitor settings should be done so that they do not contribute significantly to the step size. For example, if the physical capacitors vary from 50 to 285 pF and the desired maximum step size is 5 pF, one may choose to use 0 to 1023 for the discretization. This results in an additional error of (285-50) / 1024=0.23 pF.If the physical capacitors use a sub-binary sequence consisting of nominal contributions of 4 fractional value bits and 5 full value bits (e.g., 4, 7, 12.5, 22 pF fractional value bits and 5 full value bits of 40 pF), the total number of possible combinations is 2 to the power of 9, or 512. This means that for some capacitor settings, the switch configuration should be repeated if more than 512 switch settings, e.g., 2048, are used (e.g., capacitor settings of 500, 501, and 502 may all use the same switch configuration and therefore correspond to the same capacitor value). Conventional digital-to-analog conversion schemes use (285-50) / 5=47 steps rounded up to the next power of 2, thus typically 64 steps rather than 1023 steps. Of course, one can also use the same remapping technique for capacitors not selected according to the sub-binary scheme, but doing so does not guarantee a maximum step size.
[0064] The switch configuration is a specific and unique combination of on and off states of the contributors, and the switch configuration is mapped to a number of intervals such that one endpoint of the interval corresponds to a minimum analog output from the digital-to-analog converter and the other endpoint of the interval corresponds to a maximum analog output from the digital-to-analog converter, and the analog output of the digital-to-analog converter is monotonic over the interval, allowing steps in the interval to be sufficiently small such that the absolute value of a corresponding step in the analog output from the digital-to-analog converter is less than D.
[0065] FIG. 16 shows the correspondence between capacitor settings and variable capacitor capacitances using the mapping between switch configurations and capacitor settings illustrated in FIG. 15 and described above.
[0066] FIG. 17 illustrates the change in variable capacitor capacitance as the capacitor setting is incremented by one, the mapping of FIG. 15 is applied between the switch configuration and capacitor setting, and the switched capacitor value is calculated according to the method of FIG. 11.
[0067] The following example highlights the effectiveness of the disclosed solution. Consider two consecutive binary numbers 10110111 and 10111000, which define two capacitor switch configurations. The right least significant bit (LSB) of each binary number defines whether the first capacitor contribution is added to the sum (1 means add, 0 means don't add), the second number from the right defines whether the second capacitor contribution is added to the sum, and so on. To increment a binary number, one may proceed from the LSB, find the first 0, change that 0 to a 1, and change everything to the right to zero. All bits to the left of the 0 that was changed to a 1 are unchanged. These bits do not change, but the effective capacitance that they add to the sum may change as shown in Figures 6-10. However, the contributions of these unchanged bits do not change much for small changes in capacitor value, and the change may be bounded as dCu. The change resulting from the fractional value added change does not exceed the maximum contribution of the fourth bit minus the sum of the minimum contributions of the first through third bits. However, the procedure of FIG. 11 ensures that the maximum contribution of the fourth bit is less than the sum of the minimum contributions of the lower bits plus the desired step size, and therefore the total change is less than dCu + the desired step size. The total capacitance can therefore be from the baseline value to the baseline value + the sum of the minimum contributions of all capacitors, with successive capacitor settings increasing by no more than dCu + the desired step size. Thus, the capacitor range from the baseline value to the baseline value + the sum of the minimum contributions of all capacitors can be covered in steps that do not exceed dCu + the desired step size. In most practical applications, dCu is negative as shown in FIG. 6-10, and the individual capacitor contributions decrease as the capacitor settings increase. If dCu is positive but less than the desired step size, it can be taken into account by reducing dC x dCu in the procedure of FIG. 11.
[0068] In the presence of variable contributions from switched capacitors, the techniques described herein for creating a variable capacitor with a monotonic relationship between capacitor setting and variable capacitor capacitance can be applied more generally. For example, in a digital-to-analog converter where individual current sources are summed together to produce a total current output, the same techniques can be applied to digital-to-analog converters where the contributions of the individual current sources do not contribute the same amount to the output current as the total output current is varied. We can more broadly describe the combination of capacitors and switches or individual current sources as contributors that contribute to a combined analog output. In the case of a variable capacitor, the analog output is the capacitance presented to the terminals of the output, and in the case of current sources, the analog output is the total current output from the analog outputs of the current sources.
[0069] FIG. 18A illustrates a digital-to-analog converter (DAC) and an equivalent circuit (FIG. 18B). The DAC 1800 can include a digital input 1802, N contribution factors 1804, an interconnection network 1806, and an analog output 1808. Each of the N contribution factors 1804 can have an on state and an off state controlled by the digital input 1802. The interconnection network 1806 can be coupled to each of the N contribution factors 1804 and the analog output 1808. The interconnection network 1806 can be configured to provide a sum of the contributions of the N contribution factors 1804 to the analog output 1808. For each of the N contribution factors 1804, the contribution of one of the N contribution factors 1804 is the change in the analog output when the state of one of the N contribution factors is changed from off to on and the remaining states of all of the N-1 contribution factors 1804 remain the same. For example, FIG. 6-10 shows plots of data derived from turning the state of a contribution factor 1804 on and off while leaving the state of all other contribution factors 1804 the same. This process provides simulated maximum and minimum contribution rates for each contribution factor 1804. In other words, the contribution rate of each contribution factor varies as the analog output is varied. The ratio of maximum to minimum contribution rates for at least one of the contribution factors 1804 is at least 1:1. The average contribution rate of each of the N contribution factors 1804 can be found, and the N contribution factors 1804 can be ordered from smallest to largest average contribution rate to form an ordering. The average can be determined, in one case, as the square root of the product of the maximum and minimum contribution rates of one of the N contribution factors 1804. The gap size D can be less than or equal to twice the maximum value of the contribution rate of the first one of the N contribution factors 1804. The maximum contribution of each of the N contributors 1804 can then be defined for the kth contributor, where k is from 2 to N. The maximum contribution of the kth contributor is less than or equal to the sum of the minimum contributions of D+contributor 1 (i.e., k=1) through contributor k-1. In equation form, this is: [ka]
[0070] The DAC 1800 can use fractional and full value contribution factors in the same manner as fractional and full value capacitors described above. In one embodiment, four or more fractional value contribution factors are used, and a first ratio of the average contribution rate of contribution rate of contribution rate of contribution rate of contribution rate of the N contributors in the ordinality of the N contributors to the average contribution rate of contribution rate of the first contributor in the ordinality of the N contributors is at least 1.2, and a second ratio of the average contribution rate of contribution rate of contribution rate of contribution rate of the N contributors in the ordinality of the N contributors to the average contribution rate of the first contributor in the ordinality of the N contributors is less than 6. In another embodiment, there are five or more fractional value contribution factors, and a third ratio of the average contribution rate of contribution rate of contribution rate of contribution rate of contribution rate of contribution rate of the N contributors in the ordinality of the N contributors to the average contribution rate of contribution rate of contribution rate of contribution rate of the first contributor in the ordinality of the N contributors is at least 1.2, and a fourth ratio of the average contribution rate of contribution rate of contribution rate of contribution rate of contribution rate of the N contributors in the ordinality of the N contributors to the average contribution rate of the first contributor in the ordinality of the N contributors is less than 12. In another embodiment, there are six or more decimal-value contributors, and a fifth ratio of the average contribution rate of contributor 6 in the ordinality of the N contributors to the average contribution rate of contributor 5 in the ordinality of the N contributors is at least 1.2, and a sixth ratio of the average contribution rate of contributor 6 in the ordinality of the N contributors to the average contribution rate of the first contributor in the ordinality of the N contributors is less than 20.
[0071] For a practical and efficient control scheme, there should be a monotonic relationship between the contribution factor settings (i.e., the digital input 1802 settings) and the analog output 1808. Thus, the analog output 1808 may have a monotonic relationship to the digital input 1802, where the maximum gap size in the analog output 1808 per increment of the digital input 1802 is less than D.
[0072] The DAC 1804 can take a variety of forms. For example, the DAC 1804 can be a variable capacitor, for example, for use in a matching network (e.g., for plasma processing). The contributors 1808 can each be a capacitor in series with a respective switch, thereby switching the capacitor in and out of the variable capacitor. In one embodiment, the switches can be PIN diodes.
[0073] Although the term "gap size" is used for both variables dC and D, these are not necessarily the same gap size. Gap size dC refers to the gap size selected during testing and design of the variable capacitor. Gap size D refers to the gap size based on an existing system, such as a DAC or variable capacitor seeking to determine if it is using sub-binary or binary sequences of fractional value contribution factors.
[0074] FIG. 19 illustrates a power system with a matching network 1904, which may utilize the embodiment of the DAC illustrated and described with respect to FIG. 18. The power system 1900 may include a power supply 1902 coupled to and providing power to a load 1906 through a matching network 1904. The matching networks 1904 may each include one or more variable capacitors 1910, 1912, which may include N switched capacitors, an interconnection topology, an analog output of the interconnection topology, and a digital input (all as shown in FIG. 18). The matching network 1904 may include a digital input 1914 that may be coupled to or control the digital inputs of each of the one or more variable capacitors 1910, 1912. The load 1906 may include a plasma load, such as found in a plasma processing chamber or a remote plasma source, where a remote plasma is generated remotely and then delivered into the processing chamber.
[0075] The methods described in connection with the embodiments disclosed herein may be embodied directly in hardware, in processor executable code encoded in a non-transitory tangible processor-readable storage medium, or in a combination of the two. For example, referring to FIG. 20, shown is a block diagram depicting physical components that may be utilized to realize the interconnection network, DAC 1800, or controller of matching network 1904 in FIG. 3A, according to an exemplary embodiment. As shown, in this embodiment, a display portion 2012 and a non-volatile memory 2020 are coupled to a bus 2022, which is also coupled to a random access memory ("RAM") 2024, a processing portion (including N processing components) 2026, an optional field programmable gate array (FPGA) 2027, and a transceiver component 2028, including N transceivers. Although the components depicted in FIG. 20 represent physical components, FIG. 20 is not intended to be a detailed hardware schematic. Thus, many of the components depicted in Figure 20 may be realized by a common construct or distributed among additional physical components. Additionally, it is contemplated that other existing and yet to be developed physical components and architectures may be utilized to implement the functional components described with reference to Figure 20.
[0076] The display portion 2012 generally operates to provide a user interface for a user, and in some implementations the display is realized by a touch screen display. In general, the non-volatile memory 2020 is a non-transient memory that functions to store (e.g., persistently store) data and processor executable code (including executable code associated with effecting the methods described herein). In some embodiments, for example, the non-volatile memory 2020 includes boot loader code, operating system code, file system code, and non-transient processor executable code to facilitate the execution of the methods described with reference to Figures 11 and 12 further described herein.
[0077] In many implementations, the non-volatile memory 2020 is realized by a flash memory (e.g., NAND or ONENAND memory), although it is contemplated that other memory types may be utilized as well. While it may be possible to execute code from the non-volatile memory 2020, executable code in the non-volatile memory is typically loaded into RAM 2024 and executed by one or more of the N processing components in the processing portion 2026.
[0078] The N processing components in communication with the RAM 2024 generally execute instructions stored in the non-volatile memory 2020 and operate to enable control of the interconnect network, DAC 1800, or matching network 1904 in FIG. 3A, or for selection of a sub-binary sequence of capacitors. For example, non-transient processor executable code for effecting the methods described with reference to FIGS. 11 and 12 may be persistently stored in the non-volatile memory 2020 and executed by the N processing components in communication with the RAM 2024. As one skilled in the art would appreciate, the processing portion 2026 may include a video processor, digital signal processor (DSP), microcontroller, graphics processing unit (GPU), or other hardware processing component, or a combination of hardware and software processing components (e.g., FPGA or FPGAs including digital logic processing portions).
[0079] Additionally or alternatively, the processing portion 2026 may be configured to effect one or more aspects of the methodologies described herein (e.g., the methods described with reference to FIGS. 11 and 12). For example, non-transient processor-readable instructions may be stored in the non-volatile memory 2020 or RAM 2024 and, when executed on the processing portion 2026, cause the processing portion 2026 to perform control of the interconnect network, DAC 1800, or matching network 1904 in FIG. 3A, or selection of a sub-binary sequence of capacitors. Alternatively, non-transient FPGA configuration instructions may be persistently stored in the non-volatile memory 2020 and accessed by the processing portion 2026 (e.g., during boot-up) to configure the hardware configurable portion of the processing portion 2026 to effect the functionality of the control of the interconnect network, DAC 1800, or matching network 1904 in FIG. 3A.
[0080] The input components 2030 operate to receive signals (e.g., in the case of a controller, a user input, or in the case of an interconnection topology, a digital input) that indicate one or more aspects of user control over a DAC or matching network, or for selection of a sub-binary sequence of capacitors. The output components generally operate to provide one or more analog or digital signals to effect operational aspects of the controller. For example, the output portion 2032 may provide an analog output.
[0081] The depicted transceiver component 2028 includes N transceiver chains, which may be used to communicate with external devices via wireless or wired networks. Each of the N transceiver chains may represent a transceiver associated with a particular communication scheme (e.g., WIFI, Ethernet, Profi bus, etc.).
[0082] Some portions are presented in terms of algorithms or symbolic representations of operations on data bits or binary digital signals stored within a computing system memory, such as a computer memory. These algorithmic descriptions or representations are examples of techniques used by those skilled in the data processing arts to convey the substance of their work to others skilled in the art. An algorithm is a self-consistent sequence of operations or similar processes leading to a desired result. In this context, operations or processes involve physical manipulations of physical quantities. Typically, though not necessarily, such quantities may take the form of electrical or magnetic signals capable of being stored, transferred, combined, compared, or otherwise manipulated. It has proven convenient at times, primarily for reasons of common usage, to refer to such signals as bits, data, values, elements, symbols, characters, terms, numbers, numeric values, or the like. It should be understood, however, that all of these and similar terms are to be associated with the appropriate physical quantities, but are merely convenient labels. Unless specifically stated otherwise, throughout this specification, discussions utilizing terms such as "processing," "computing," "calculating," "determining," and "identifying" or the like should be understood to refer to the actions or processes of a computing device, such as one or more computers or similar electronic computing devices or devices, that manipulate or transform data represented as physical electronic or magnetic quantities in memories, registers, or other information storage, transmission, or display devices of a computing platform.
[0083] As will be appreciated by one of ordinary skill in the art, aspects of the present invention may be embodied as a system, method, or computer program product. Thus, aspects of the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment (including firmware, resident software, microcode, etc.), or an embodiment combining software and hardware aspects, which may all generally be referred to herein as a "circuit," "module," or "system." Additionally, aspects of the present invention may take the form of a computer program product embodied in one or more computer-readable medium(s) having computer-readable program code embodied thereon.
[0084] As used herein, the recitation of "at least one of A, B, and C" is intended to mean "any of A, B, C, or any combination of A, B, and C." The foregoing description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present disclosure. Various modifications of these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be applied to other embodiments without departing from the spirit or scope of the present disclosure. Thus, the present disclosure is not intended to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims
1. 1. A digital-to-analog converter comprising: Digital input and Analog output; N contribution factors coupled to said analog output via an interconnection network; Equipped with Each of the N contributors comprises a capacitor; the maximum contribution rate of the kth contributor is less than or equal to the gap size D plus the sum of the minimum contribution rates of the first through (k-1)th contributors, the interconnection network is configured to provide a sum of the contributions of the N contribution factors to the analog output; the analog output includes a capacitance value of a variable capacitor; The gap size D corresponds to a range of capacitance values that the variable capacitor cannot achieve.
2. 2. The digital-to-analog converter of claim 1 , wherein a contribution rate of at least one of the N contribution factors varies as the analog output is varied, and a ratio of a maximum value to a minimum value of the contribution rate of the at least one of the N contribution factors is at least 1.
1.
3. There are four or more fractional contribution factors, The N contribution factors are ordered from smallest to largest average contribution rate to form an ordering; a first ratio of an average contribution rate of a contributor 4 in the ordinal order of the N contributors to an average contribution rate of a contributor 3 in the ordinal order of the N contributors is at least 1.2; 2. The digital-to-analog converter of claim 1 , wherein a second ratio of an average contribution rate of contribution factor 4 in the ordering of the N contribution factors to an average contribution rate of the first contributor in the ordering of the N contribution factors is less than 6.
4. 4. The digital-to-analog converter of claim 3, wherein for an nth contribution factor of the N contribution factors, an average contribution rate of the nth contribution factor is the square root of the product of the maximum and minimum contribution rates of the nth contribution factor.
5. There are five or more fractional contribution factors; The N contribution factors are ordered from smallest to largest average contribution rate to form an ordering; a third ratio of an average contribution rate of contributor 5 in the ordinal order of the N contributors to an average contribution rate of contributor 4 in the ordinal order of the N contributors is at least 1.2; 2. The digital-to-analog converter of claim 1 , wherein a fourth ratio of an average contribution rate of contribution factor 5 in the ordering of the N contribution factors to an average contribution rate of the first contributor in the ordering of the N contribution factors is less than 12.
6. 6. The digital-to-analog converter of claim 5, wherein for an nth contribution factor of the N contribution factors, an average contribution rate of the nth contribution factor is the square root of the product of the maximum and minimum contribution rates of the nth contribution factor.
7. There are six or more fractional contribution factors; The N contribution factors are ordered from smallest to largest average contribution rate to form an ordering; a fifth ratio of an average contribution rate of a contribution factor 6 in the ordinal order of the N contribution factors to an average contribution rate of a contribution factor 5 in the ordinal order of the N contribution factors is at least 1.2; 2. The digital-to-analog converter of claim 1 , wherein a sixth ratio of an average contribution rate of a sixth contributor in the ordering of the N contributors to an average contribution rate of the first contributor in the ordering of the N contributors is less than 20.
8. 8. The digital-to-analog converter of claim 7, wherein for an nth contribution factor of the N contribution factors, an average contribution rate of the nth contribution factor is the square root of the product of the maximum and minimum contribution rates of the nth contribution factor.
9. 2. The digital-to-analog converter of claim 1, wherein the analog output has a monotonic relationship to the digital input, and a maximum gap size in the analog output per increment of the digital input is less than D.
10. 2. The digital-to-analog converter of claim 1, wherein the capacitors of the N contribution factors are in series with a switch.
11. 11. The digital-to-analog converter of claim 10, wherein the switch is a PIN diode.
12. The digital-to-analog converter of claim 1 , wherein the N contribution factors form a subbinary sequence.
13. 2. The digital-to-analog converter of claim 1, wherein the gap size D is less than or equal to twice the maximum contribution rate of the smallest of the N contribution factors.
14. A matching network comprising one or more variable capacitors, each of the one or more variable capacitors having: Digital input and Analog output; N contribution factors coupled to said analog output via interconnect means; Equipped with the N contribution factors form a sub-binary sequence; the maximum contribution rate of the kth contributor is less than or equal to the gap size D plus the sum of the minimum contribution rates of the first through (k-1)th contributors, the analog output is configured to provide a sum of the contributions of the N contribution factors; The gap size D corresponds to the range of capacitance values that the variable capacitor cannot achieve.
15. 15. The matching network of claim 14, wherein a range of contribution rates of at least one of the N contribution factors varies as others of the N contribution factors are switched in and out of the matching network, such that a ratio of a maximum contribution rate of the at least one of the N contribution factors to a minimum contribution rate of the at least one of the N contribution factors is at least 1.
1.
16. a first ratio of an average contribution of a fourth one of the N contributing factors to an average contribution of a third one of the N contributing factors is at least 1.2; 15. The matching network of claim 14, wherein a second ratio of an average contribution of the fourth one of the N contribution factors to an average contribution of the first one of the N contribution factors is less than 6.
17. a third ratio of an average contribution rate of a fifth one of the N contributors to an average contribution rate of a fourth one of the N contributors is at least 1.2; 15. The matching network of claim 14, wherein a fourth ratio of an average contribution of the fifth of the N contribution factors to an average contribution of the first of the N contribution factors is less than 12.
18. a fifth ratio of an average contribution of a sixth one of the N contributors to an average contribution of a fifth one of the N contributors is at least 1.2; 15. The matching network of claim 14, wherein a sixth ratio of an average contribution of the sixth of the N contribution factors to an average contribution of the first of the N contribution factors is less than 20.
19. 1. A power supply system comprising: a source of AC or DC power; a matching network comprising one or more variable capacitors; Equipped with Each of the one or more variable capacitors comprises: Digital input and an analog output configured to couple to a load; N contribution factors coupled to said analog output via interconnect means; Equipped with the maximum contribution rate of the kth contributor is less than or equal to the gap size D plus the sum of the minimum contribution rates of the first through (k-1)th contributors, the interconnection means being configured to provide a sum of the contributions of the N contribution factors; The gap size D corresponds to a range of capacitance values that the variable capacitor cannot achieve.
20. 20. The power supply system of claim 19, further comprising a plasma processing chamber coupled to the analog output.
Citation Information
Patent Citations
Integrated tunable impedance network
JP2016119671A
Impedance matching network using heat pipe inductor
US20180076788A1
Variable capacitance capacitor array, variable capacitance capacitor array device and circuit module
WO2008001914A1
Impedance matching device
WO2018062109A1
