Systems and methods for self-capacitance sensing, display systems

Abstract

A system and method for self-capacitance sensing, a display system is described. Each channel of the array has a self-capacitance Ci that changes in response to the presence or absence of a local touch event. Ci is discharged for a discrete discharge time using a discharge current that is local noise suppressed to obtain a discharge voltage level that differs with the presence or absence of a local touch event, and each channel is read by an analog front end (AFE) by outputting a voltage output of the channel based on the discharge voltage level. The voltage output of the output channel is achieved by at least passively mixing the discharge voltage level to generate an upconverted channel signal pair, sampling the upconverted channel signal pair to obtain a differential voltage sample, and amplifying the differential voltage sample to generate Vout, where Vout is indicative of the absence or presence of a touch event local to the channel.

Description

[0001] This application claims priority to U.S. Patent Application No. 18 / 177,745, filed March 2, 2023, entitled “LOW-NOISE SELF-CAPACITOR SENSING FOR CAPACITIVE TOUCH PANELS”, the entire contents of which are incorporated herein by application. Technical Field

[0002] This invention relates to sensing circuits, and more particularly, to a system and method for self-capacitance sensing, and a display system, such as for use in large capacitive touch panels. Background Technology

[0003] Many modern electronic applications include integrated touch panels, such as touchscreen displays. Typically, the touch sensing layer of a touchscreen display uses capacitive sensing to determine when and where a user touches the display. Display noise can couple into the touch sensing layer, manifesting as noise in the readout of the capacitive touch sensing information. This display noise coupling tends to increase over time, making it increasingly challenging to provide sufficiently low-noise readout circuitry for such applications.

[0004] Typically, the touch sensing layer of a display comprises an array of "mutual capacitors" and "self capacitors." For example, there are self capacitors for each row and each column of the array, and mutual capacitors at each row-column intersection. Mutual capacitors in a touch panel tend to be the primary sensing elements because they often provide more accurate information about the location of a touch (e.g., a finger). Moreover, self-capacitance sensing can provide a useful alternative (or supplementary) source of touch sensing information, especially in situations where mutual capacitance sensing is often inaccurate (e.g., when the user's finger is wet).

[0005] However, self-capacitance sensing can be more challenging because its signal level is much smaller compared to that obtained through mutual capacitance sensing. The capacitance change sensed in self-capacitance during a touch event may often be only a fraction of its underlying capacitance value. To reliably sense this small change in capacitance, the sensing circuit can be designed to effectively cancel out the underlying capacitance value with sufficiently low readout noise. While several conventional methods exist, these methods tend to have limitations. Summary of the Invention

[0006] The embodiments disclosed herein include systems and methods for low-noise self-capacitance sensing in capacitive touch panels, such as displays integrated into touchscreen electronic devices. For example, a touch panel array is integrated with a display panel and has multiple touch sensing channels. Each channel has a corresponding channel self-capacitance Ci and a corresponding touch capacitance Ctouch, the corresponding channel self-capacitance Ci including a corresponding underlying self-capacitance Cs corresponding to display noise coupled to the channel by capacitance from the display panel, and the corresponding touch capacitance Ctouch changing in response to the presence of a touch event localized to the channel. Each channel is coupled to an analog front-end (AFE) via a voltage input Vin node, the voltage of which is related to the voltage across Ci of the channel. The AFE includes a discharge stage and a sensing stage. The discharge stage is configured to couple the Vin node with a locally noise-suppressed discharge current Iout for discrete discharge times to discharge the Vin node to a discharge voltage level that varies depending on the presence or absence of a touch event localized to the channel. The sensing stage outputs the voltage of the output channel by passively mixing at least the discharge voltage level to generate an upconverted channel signal pair; sampling the upconverted channel signal pair to obtain a differential voltage sample; and amplifying the differential voltage sample to generate Vout, where Vout indicates the absence or presence of a touch event local to the channel.

[0007] According to a first set of embodiments, a method is provided for self-capacitance sensing in a touch panel array integrated with a display panel, the touch panel array having multiple channels, each channel having a corresponding channel self-capacitance and a corresponding touch capacitance, wherein the corresponding channel self-capacitance includes a corresponding basic self-capacitance corresponding to display noise, the display noise capacitance being coupled to a channel from the panel, and the corresponding touch capacitance changing in response to the presence of a touch event in a local channel. The method includes: in a first phase of a readout cycle, charging a voltage input Vin node to a charging voltage level, the Vin node being coupled to one of the plurality of channels such that the voltage at the Vin node is correlated with the voltage across the corresponding self-capacitance Ci of the channel coupled thereto; in a second phase of the readout cycle following the first phase, discharging the Vin node with a locally noise-suppressed discharge current Iout for discrete discharge times to discharge the Vin node to a discharge voltage level, wherein the discharge voltage level is a first voltage level in the absence of a touch event localized to the channel and a second voltage level in the presence of the touch event localized to the channel; and in a third phase of the readout cycle following the second phase, outputting a voltage output Vout of the channel by: passively mixing the discharge voltage level at least to generate an up-converted channel signal pair; sampling the up-converted channel signal pair to obtain a differential voltage sample; and amplifying the differential voltage sample to generate Vout, wherein Vout indicates the absence or presence of the touch event localized to the channel.

[0008] In some such embodiments, the readout cycle comprises two half-cycles. The method may further include: in the first half-cycle of the readout cycle: configuring a passive mixer in a pass-through configuration; and using the passive mixer in the pass-through configuration to perform the charging, discharging, and output. The method may further include: in the second half-cycle of the readout cycle: configuring the passive mixer in a switched configuration; and using the passive mixer in the switched configuration to perform the charging, discharging, and output. In such embodiments, the passive mixer receives the discharge voltage level of the Vin node at a first input, receives another discharge voltage level associated with an adjacent channel or receives a common-mode reference voltage level at a second input, and generates the upconversion channel signal pair at the first and second outputs by: in the pass-through configuration, coupling the first input to the first output and coupling the second input to the second output; and in the switched configuration, coupling the first input to the second output and coupling the second input to the first output.

[0009] According to another set of embodiments, a system is provided for self-capacitance sensing in a touch panel array integrated with a display panel, the touch panel array having multiple channels, each channel having a corresponding channel self-capacitance and a corresponding touch capacitance, wherein the corresponding channel self-capacitance includes a corresponding base self-capacitance corresponding to display noise, the display noise capacitance being coupled to the channel from the panel, and the corresponding touch capacitance changing in response to the presence of a touch event in a local channel. The system includes: a voltage input Vin node coupled to one of the plurality of channels, such that Vin represents the voltage across the corresponding self-capacitance Ci of the channel; a discharge stage configured to couple the Vin node to a locally noise-suppressed discharge current Iout for discrete discharge times during a discharge phase to discharge the Vin node to a discharge voltage level such that the discharge voltage level is a first voltage level in the absence of a touch event localized to the channel and a second voltage level in the presence of the touch event localized to the channel; and a sensing stage coupled to the discharge stage to output a voltage output (Vout) of the channel by: passively mixing the discharge voltage level at least to generate an up-converted channel signal pair; sampling the up-converted channel signal pair to obtain a differential voltage sample; and amplifying the differential voltage sample to generate the Vout, wherein the Vout indicates the absence or presence of the touch event localized to the channel.

[0010] In some such embodiments, the system further includes: a plurality of instances of the Vin node, each instance being coupled to a corresponding channel of the plurality of channels, such that the i-th instance Vin_i of the Vin node represents the voltage across a corresponding i-th self-capacitance Ci_i of the corresponding i-th channel coupled thereto; a plurality of instances of the discharge stage, wherein the i-th instance of the discharge stage is configured to: couple the Vin_i to an i-th locally noise-suppressed discharge current Iout_i during a discharge phase to discharge the Vin_i to a corresponding i-th discharge voltage level, each corresponding i-th discharge voltage level being the first voltage level in the absence of a touch event localized to the corresponding i-th channel. The second voltage level is provided in the presence of a touch event localized to the corresponding i-th channel; and multiple instances of the sensing level are coupled to multiple instances of the discharge level, wherein the i-th instance of the sensing level is configured to output a corresponding i-th voltage output Vout_i for the corresponding i-th channel by: passively mixing at least the i-th discharge voltage level to generate an i-th upconversion channel signal pair, sampling the i-th upconversion channel signal pair to obtain an i-th differential voltage sample, and amplifying the i-th differential voltage sample to generate an i-th voltage output instance, wherein the i-th voltage output instance indicates the absence or presence of the touch event localized to the i-th corresponding channel.

[0011] Some such embodiments of a system for self-capacitance sensing are provided as part of a display system. The display system includes: a display panel; a touch panel array integrated with the display and having a plurality of channels, each channel having a corresponding channel self-capacitance and a corresponding touch capacitance, wherein the corresponding channel self-capacitance includes a corresponding base self-capacitance corresponding to display noise, the display noise capacitance being coupled to a channel from the panel, the corresponding touch capacitance changing in response to the presence of a touch event in a local channel; and a plurality of instances of the system as claimed in claim 1, each instance of the Vin node being coupled to a corresponding channel of the plurality of channels.

[0012] The following figures, description and claims provide a more detailed description of the foregoing, their implementations and the features of the disclosed technology. Attached Figure Description

[0013] The accompanying drawings, which are mentioned and form part of this document, illustrate embodiments of the present disclosure. The drawings, together with the description, serve to explain the principles of the invention.

[0014] Figure 1 A simplified portion of an illustrative conventional touch panel array with rows and columns is shown.

[0015] Figure 2 A simplified portion of the touch sensing environment is shown to further illustrate self-capacitance sensing.

[0016] Figure 3 A high-level block diagram of a self-capacitor sensing environment for discrete-time sensing is shown.

[0017] Figure 4 The simplified illustrative waveforms and timings for the three stages of the operation are shown.

[0018] Figure 5 A high-level block diagram is shown for a self-capacitor sensing environment for discrete-time sensing using a conventional pre-charged capacitor (PCC) implementation with a discharge stage having frequency-domain up-conversion (FUC) for noise reduction.

[0019] Figure 6 It shows in Figure 5 Simplified illustrative waveforms and timings of several stages of circuit operation.

[0020] Figure 7 A high-level block diagram is shown for a self-capacitor sensing environment for discrete-time sensing using a conventional resistance-to-time conversion (RTC) implementation of a discharge stage with frequency-domain up-conversion (FUC) for noise reduction.

[0021] Figure 8 It shows in Figure 5 Simplified illustrative waveforms and timings of several stages of circuit operation.

[0022] Figure 9 A simplified block diagram of a low-noise discrete-time self-capacitance sensing system according to an embodiment described herein is shown.

[0023] Figure 10 A circuit block diagram of an illustrative example of a noise-suppressing discharge current generator according to several embodiments described herein is shown.

[0024] Figure 11 A circuit block diagram is shown for another illustrative example of a noise-suppressed discharge current generator coupled to a shared ramp generator, according to several embodiments described herein.

[0025] Figure 12 A circuit block diagram illustrating an illustrative implementation of a differential sensing stage according to several embodiments described herein is shown.

[0026] Figure 13 It shows the relationship with Figure 12 Illustrative graphs and timing diagrams of several signals related to the operation of the circuit.

[0027] Figure 14 A circuit block diagram illustrating an illustrative implementation of a differential discharge stage according to several embodiments described herein, the differential discharge stage including an example of a noise-suppressed discharge current generator coupled to a bias generator.

[0028] Figure 15 An example configuration of a capacitive touch sensing environment is shown, in which, Figure 14 Differential discharge stage and Figure 12 The differential sensing circuit of the passive mixer is coupled in the implementation.

[0029] Figure 16 An illustrative configuration of a capacitive touch sensing environment according to various embodiments described herein is shown, wherein a single-ended discharge stage is coupled to a single-ended sensing stage.

[0030] Figure 17 A flowchart is shown of an illustrative method 1700 for performing self-capacitance sensing in a touch panel array according to an embodiment described herein.

[0031] In the accompanying drawings, similar parts and / or features may have the same reference numerals. Furthermore, various components of the same type may be distinguished by adding a second numeral after their subsequent reference numerals to differentiate them. If only the first reference numerals are used in the description, then the description also applies to any of the similar parts having the same first reference numerals, regardless of the second reference numerals. Detailed Implementation

[0032] Numerous specific details are set forth in the following description in order to provide a thorough understanding of the invention. However, those skilled in the art will understand that the invention can be practiced without one or more of these details. In other instances, features and techniques known in the art will not be described further for the sake of brevity.

[0033] Many modern electronic applications include integrated touch panels, such as touchscreen displays. Typically, the touch sensing layer of a touchscreen display uses capacitive sensing to determine when and where a user touches the display. Display noise can couple into the touch sensing layer, manifesting as noise in the readout of capacitive touch sensing information. This display noise coupling tends to increase over time, making it increasingly challenging to provide sufficiently low-noise readout circuitry for such applications. Figure 1A simplified portion of an illustrative touch panel array 100 with rows and columns is shown. Typically, the touch panel array 100 is used to sense when and where a user touches the array by sensing changes in capacitance and by correlating the sensed changes with specific row-column intersections. Mutual capacitors (Cm) 110 (illustrated as 110aa to 110nn) are located at each row-column intersection. Furthermore, each row and column of the array 100 is coupled to a corresponding self-capacitor (Cs) 105 (illustrated as Csrx 105ra to 105rn and Cstx 105ta to 105tn). Row-oriented self-capacitors are labeled "Csrx", and column-oriented self-capacitors are labeled "Cstx".

[0034] As used herein, a touch event is considered any touch interaction with the touch panel array 100, which can be detected by any one or more of the touch sensing circuits 120. A touch event is considered "local to" a particular row or column line when it is sufficiently close to that row or column line to manifest as a change in capacitance (mutual capacitance and / or self-capacitance) detectable at least by the touch sensing circuit 120 coupled to that particular row or column line. Correspondingly, a touch event is considered local to a particular self-capacitance 105 when it is sufficiently close to a particular row or column line coupled to a self-capacitance 105 to manifest as a change in self-capacitance detectable at least by the touch sensing circuit 120 coupled to that particular row or column line. Similarly, a touch event is considered local to a particular mutual capacitor 110 when it is sufficiently close to that mutual capacitor 110 to manifest as a change in mutual capacitance detectable at least by the touch sensing circuit 120 receiving a signal driven by the mutual capacitor 110. Similarly, when a touch event is close enough to any part of the touch panel array 100 to manifest as a change in mutual capacitance and / or self capacitance that can be detected at least by a particular touch sensing circuit 120, the touch event is hereby considered to be local to that particular touch sensing circuit 120.

[0035] For example, a touch event (e.g., placing a finger) occurring at the circled row-column intersection 115 can cause a detectable change in capacitance associated with the mutual capacitor 110bc, the row-direction self-capacitor 105rb, and the column-direction self-capacitor 105tc. Therefore, the touch event can be considered to be at least localized to: the third column line, the second row line, the row-direction self-capacitor 105rb, the column-direction self-capacitor 105tc, the mutual capacitor 110bc, the touch sensing circuit 120rb, and the touch sensing circuit 120tc. In some cases, the same touch event can be localized to multiple adjacent row lines, column lines, self-capacitors 105, and / or mutual capacitors 110 (i.e., therefore the touch event is detectable relative to them).

[0036] Although not explicitly shown, the touch panel array 100 can be integrated as part of a display, such as a touchscreen display for an electronic device. The grid of row and column lines effectively provides multiple touch sensing channels. Mutual capacitors 110 in the touch panel array 100 tend to be primary sensing elements because they often provide more accurate information about the location of a touch (e.g., a finger). The mutual capacitance of one of the mutual capacitors 110 is typically measured by driving a signal through the column and row lines coupled to the mutual capacitor 110 and measuring the output. For example, measuring the capacitance of mutual capacitor 110bc may involve coupling a driver (not shown) to a column line corresponding to a self-capacitor 105tc in the column direction. The driver can transmit a signal through the column line, which is coupled via mutual capacitor 110bc to the row line corresponding to the self-capacitor 105rb in the row direction. This signal can then be received at a touch sensing circuit 120rb coupled to the row line and measured to detect any change in capacitance indicating the presence of a touch event at mutual capacitor 110bc.

[0037] In addition to mutual capacitance 110 sensing, self-capacitance 105 sensing can provide a useful alternative (or supplementary) source of touch sensing information, especially where mutual capacitance 110 sensing is often inaccurate (e.g., when the user's fingers are wet). Although self-capacitance 105... Figure 1 While shown as discrete components, each self-capacitor 105 represents the total parasitic capacitance on a corresponding row or column line, which is coupled to the self-capacitor 105 as shown. For example, as described above, the touch panel array 100 can be integrated with a display, and the parasitic capacitance of the self-capacitor 105 can be manifested according to capacitively coupled display noise. Measuring the self-capacitor 105 typically does not involve drive signals via the row or column lines of the touch panel array 100. Instead, as... Figure 1 As shown, each row line and each column line is coupled to a corresponding touch sensing circuit 120 (shown as touch sensing circuits 120ra to 120rn for row lines and touch sensing circuits 120ta to 120tn for column lines). As described herein, a touch event approaching a particular row line or column line can manifest as a change in self-capacitance sensed by the touch sensing circuit 120 coupled to that row line or column line.

[0038] Figure 2 A simplified portion of the touch sensing environment 200 is shown to further illustrate self-capacitance sensing. The illustrated environment shows a touch panel array 215 integrated with the display panel 220 (e.g., by...). Figure 1(Implemented by touch panel array 100). Typically, the common cathode layer of display panel 220 (e.g., which can be considered a local ground plane) faces touch panel array 215. Capacitively coupled display noise 225 from display panel 220 manifests at touch panel array 215 as self-capacitance (i.e., parasitic capacitance) on the row and column lines of touch panel array 215, which can be expressed as (e.g., as...). Figure 1 (As shown) Self-capacitors 105 coupled to those row and column lines. To avoid making the figures too complex, only one self-capacitor 105 is shown.

[0039] For reference Figure 1 Each row and column line discussed can be coupled to a corresponding touch sensing circuit 120 (only one is shown). Embodiments of the touch sensing circuit 120 typically include an analog front-end (AFE) 230 that generates a channel-sensing output voltage Vout 235. It is desirable to include an up-conversion stage in the AFE 230 to up-convert the input direct-current (DC) signal to a higher frequency to help eliminate low-frequency noise in the AFE. The AFE 230 path then includes a subsequent down-converting mixer (DCM) 245 to shift the signal back to DC. A filter 240 following the DCM 245 helps maintain the DC signal while eliminating higher-frequency spurious signals. For example, without up-conversion, the output of the AFE 230 is typically a DC amplitude (e.g., generated by averaging the amplitude over a series of unipolar pulses). The upconversion stage causes the output of AFE 230 to be a series of pulses with alternating polarity, thus representing the detected signal as an amplitude at a specific frequency (i.e., non-DC, with effectively removed low-frequency noise). DCM 245 can then mix the output of AFE 230 with a sine wave of the same frequency to generate a DC output signal corresponding to the DC input signal.

[0040] When no touch event occurs locally to the touch sensing circuit 120, the touch sensing circuit 120 generates Vout 235 based on the channel capacitance corresponding to the base capacitance value of the respective self-capacitor 105. When a touch event occurs, the amount of self-capacitance exhibited by the self-capacitor 105 changes. For example, as shown, a finger 210 touching the touch panel array 215 can manifest as a touch capacitor 205 providing a parallel capacitance path to ground. This can effectively increase the apparent self-capacitance of any self-capacitor 105 locally to the touch event. Therefore, the touch sensing circuit 120 generates Vout 235 based on the increased channel capacitance corresponding to the base capacitance value of the respective self-capacitor 105 plus the additional parallel capacitance provided by the touch event (i.e., channel capacitance = Cs 105 + Ctouch 205).

[0041] This type of self-capacitor 105 sensing can be effective, but it is often more challenging than mutual capacitor 110 sensing, at least because self-capacitor sensing often involves signal levels much smaller than those obtained using mutual capacitor 110 sensing. The capacitance change induced in the self-capacitor 105 during a localized touch event may often be only a fraction of its underlying capacitance value. For example, the capacitance measured between touch and non-touch conditions may typically differ by less than 0.1%. To reliably sense such minute capacitance changes, the sensing circuitry can be designed to effectively cancel out the underlying capacitance value with very low readout noise.

[0042] Figure 3 A high-level block diagram of a self-capacitor sensing environment 300 for discrete-time sensing is shown. Environment 300 typically includes an input stage 301, a discharge stage 302, and a sensing stage 303. Input stage 301 may represent a touch panel array (e.g., as seen at the input of coupled touch sensing circuitry 120). Figure 1 100 touch panel arrays Figure 2 Specific row lines or column lines (channels) of a touch panel array 215, etc. For example... Figure 2 As shown, the input to the touch sensing circuit 120 can be an AFE 230, which can be implemented as a discharge stage 302 and a sensing stage 303 (i.e., as used herein, the AFE 230 is considered to include both the discharge stage 302 and the sensing stage 303). Although specific components are shown in a particular arrangement and / or stage for clarity, alternative embodiments may rearrange and / or regroup the components without departing from the scope of the embodiments described herein.

[0043] Input stage 301 represents a specific channel of the capacitive touch sensing array, as seen in AFE 230. As described above, the base capacitance of the channel's self-capacitor 105 corresponds to the display noise 225 from the capacitive coupling of the integrated display panel. The total self-capacitor Ci of the channel can be simply expressed as self-capacitor 105 in parallel with touch capacitor 205 (i.e., Ci = Cs + Ctouch). In the absence of any touch event localized to self-capacitor 105, the amount of the additional touch capacitor 205 can be zero, or in the presence of a touch event localized to self-capacitor 105, the amount of the additional touch capacitor 205 can be a detectable value (e.g., Ctouch > 0). Input stage 301 is further shown as having impedance, as in Figure 3 The output of input stage 301 can be represented by the input voltage level at Vin 310, the input voltage node of AFE 230, as indicated by the resistor.

[0044] Figure 3 The operation of the intermediate-stage switch typically follows three stages corresponding to the timing of the three switches K1 330, K2 335, and K3 340. The timing of these stages is controlled by the phase-controlled switch controller 360. To ensure... Figure 3 The description is clearer. Figure 4 A simplified illustrative waveform and timing diagram at 402 is shown for the three stages of operation. Switch labels are used to indicate the signals used to control a particular switch (e.g., in...). Figure 4 The signal marked "K1330" is used for control. Figure 3 (The signal of switch K1 330 in the middle).

[0045] In the first stage 402a, K1 330 closes for a charging period. For example... Figure 3 As shown, K1 330 is closed, coupling the capacitor input 301 to the source voltage Vcc via resistor Rp, thereby charging the channel self-capacitance Ci. Closing K1 330 also effectively couples Vin 310 to Vcc (e.g., via the input impedance as shown), such that in the first stage 402a, the voltage drop between Vin 310 and ground across the capacitor input 301 connected in series with Rp increases to Vcc (or increases to a predetermined level below Vcc based on the input impedance). After Ci is charged, K1 330 is opened. Figure 4 An illustrative graph showing the voltage levels of the Vin 310 as it charges to full charge is provided. The thicker solid line represents the behavior of the Vin 310 in the absence of a touch event (Ci = Cs), while the thinner dashed line represents the behavior of the Vin 310 in the presence of a touch event (Ci = Cs + Ctouch).

[0046] In the second stage 402b, K2 335 is closed for a predetermined discharge time T 405 (both K1 330 and K3 340 are open). The predetermined discharge time T 405 is also referred to herein as the "discrete discharge time," and therefore, the self-capacitance sensing method described herein can be considered a type of discrete-time sensing method. Figure 3 As shown, K2 335 is closed, and Vin 310 is coupled to discharge block 325. The voltage across Ci and Rp (i.e., related to Vin 310) is discharged for a predetermined amount of time by the discharge current Iout 320 provided by discharge block 325. The discharge rate through discharge block 325 is inversely proportional to Ci. Therefore, the difference in Ci between the presence and absence of a local touch event results in a difference in the discharge rate through discharge block 325. Figure 4 The illustrative graph of Vin 310 illustrates this effect. The discharge at Vin 310 is slower when a partial touch event is present (dashed line) than when no partial touch event is present (solid line), due to the increased capacitance of Ci from the additional Ctouch 205 when a partial touch event is present. Components (at least those of discharge stage 302) are used to draw a certain amount of charge from Vin 310 such that at the end of T 405 (i.e., after the discrete discharge time): Vin has decreased to the discharge reference level Vcm 315 in the absence of a partial touch event (i.e., when Ci = Cs); or Vin has decreased to a level detectably different from Vcm 315 (e.g., higher than Vcm 315) in the presence of a partial touch event. Thus, after T 405, when K2 335 is disconnected, the remaining charge on Ci (and the corresponding level of Vin 310) is detectably different between touch and non-touch event conditions.

[0047] In stage 402c, K3 340 is open (K1 330 and K2 335 are closed). For example... Figure 3 As shown, K3 340 is closed, and Vin 310 is coupled to amplifier block 350. Amplifier block 350 is used to amplify the difference between Vin 310 under touch event and non-touch event conditions, so that the difference can be read out as a reliably detectable difference in the channel sense output voltage Vout 235 output by AFE 230. Figure 4 An illustrative graph of Vout 235 is shown, and the illustrative difference of Vout 235 between touch event and non-touch event conditions is labeled Vsense 410. For a typical amplifier block 350 with gain α, Vout 235 can be expressed as:

[0048]

[0049] In some implementations, amplifier block 350 compares Vin 310 with a discharge reference level Vcm 315. For example, as described above, parameters (e.g., T 405, Iout 320, etc.) are set such that in the absence of a local touch event, Vin 235 decays to a level substantially equal to Vcm 315 in the second stage 402b, or in the presence of a local touch event, Vin 310 decays to a detectable level different from (e.g., greater than) Vcm 315. For capacitors, it is known that the change of capacitor current Ic over time is related to its capacitance and voltage: Ic = C*(dV / dt). In the context of this exemplary implementation, this relationship can be restated as: Id*T = (Vcc – Vcm)*Cs. Amplifier block 350 can amplify the difference between Vin 310 and Vcm 315 in the third stage 402c, so that the generated Vout 235 is essentially zero in the absence of a touch event (where Vin≈Vcm), or the generated Vout 235 exhibits a non-zero Vsense 410 level in the presence of a touch event (where Vin>Vcm).

[0050] As shown in the figure, the embodiment may include, or communicate with, a phase-controlled switch controller 360. The phase-controlled switch controller 360 may output control signals to set the states of switches, such as K1 330, K2 335, and K3 340. For example, the switch may be a transistor, and the control signals may be used to turn the transistor on or off. The phase-controlled switch controller 360 may include its own timing control (e.g., a clock, counter, etc.), or the phase-controlled switch controller 360 may communicate with an additional component that controls the timing of the signals output by the phase-controlled switch controller 360.

[0051] As described above, the signal level can be very low when self-capacitor 105 senses a touch event. For example, at the end of the second stage 402b, the difference in level between touch and non-touch conditions for Vin 310 can be very small. Detection in the third stage 402c depends on the distinction between touch and non-touch levels, which can depend on reliably eliminating the underlying capacitance value of Cs 105. For example, the presence of additional noise on Vin 310 or Vcm 315 reduces the margin available for reliably distinguishing touch and non-touch conditions. The embodiments described herein include various novel techniques for reducing several conventional sources of detection suppression noise in the context of self-capacitor-based touch event sensing. The term "detection suppression noise" is used herein to refer to a type of noise that tends to reduce the effectiveness of self-capacitor-based sensing of touch events. For example, Cs 105 originates from capacitively coupled display noise 225, but this noise is common to all channels (at least for adjacent channels) and will eventually cancel out by correlation, therefore it is not considered detection suppression noise. Some of the described techniques aim to reduce detection suppression noise in discharge stage 302. Other techniques described are designed to reduce detection suppression noise in sensing stage 303.

[0052] One way to reduce detection suppression noise is through frequency-domain up-conversion (FUC), which involves changing the detection polarity in different cycles, for example, by switching the polarity in each cycle. The use of FUC effectively upconverts the signal received from the touch sensing channel, thereby shifting Vout 235 (i.e., the signal of interest) from the DC domain to a higher frequency domain, while leaving a component in the DC domain that suppresses detection noise and facilitates low-frequency noise removal. Before discussing the novel approach to applying such techniques in this paper, two examples of applying FUC to conventional capacitive touch sensing methods are provided: the pre-charged capacitor (PCC) method and the resistance-to-time conversion (RTC) method.

[0053] Figure 5 A high-level block diagram of a self-capacitor sensing environment 500 for discrete-time sensing is shown, which uses a conventional pre-charged capacitor (PCC) implementation of a discharge stage 302, and the discharge stage 302 has frequency domain up-conversion (FUC) for noise reduction. Figure 5 Parallel description, Figure 6 It shows in Figure 5 The simplified illustrative waveforms and timing of several stages of circuit operation 610. For example... Figure 5As shown, environment 500 includes an AFE coupled to a capacitive channel of the touch panel array. The channel has a self-capacitance Ci, represented by a self-capacitance Cs 105 (capacitively coupled display noise 225), connected in parallel with a touch capacitor 205, which varies from zero capacitance in the absence of a local touch event to a non-zero capacitance in the presence of a local touch event. The AFE is shown as being implemented by a PCC discharge block 510 and a differential amplifier block 520. The PCC discharge block 510 includes a charging capacitor Cc 505 configured to pre-charge during the first phase 610a for effectively discharging the channel self-capacitance Ci.

[0054] Typically, FUC is achieved by setting switches K4 535 (shown as K4a535a to K4d 535d) to position '1' (corresponding to the first polarity) during the first half-cycle of operating cycle 605, and setting switches K4 535 to position '2' (corresponding to the second opposite polarity) during the second half-cycle of operating cycle 605. Additionally, according to reference... Figure 3 and Figure 4 The three phases 610 described in phase 402 (shown as phases 610a to 610c) can operate in the same manner for each half-cycle portion of period 605. In the first phase 610a, switch K1 330 is closed, thereby effectively charging the channel self-capacitance Ci (and node Vin 310). Simultaneously, in the PCC discharge block 510, switches K2a 335a and K2b 335b are in position '1', such that switch K2b 335b precharges Cc 505 (opposite to the charging of Ci), while switch K2a 335a keeps Cc 505 isolated from Vin 310. At the end of the first phase 610a, switch K1 330 is opened, and switches K2a 335a and K2b 335b switch to position '2'. This effectively floats node Vin 310 at the charging level and couples it in parallel with both Ci and Cc 505 (i.e., Cs105, Ctouch 205, and Cc 505 are coupled in parallel between Vin 310 and ground). Subsequently, in the third stage 610c, switch K3340 closes, and Vin 310 is coupled to the input of differential amplifier 520. As shown, amplifier 520 can generate a differential Vout 235 based on the difference between Vin 310 and the discharge reference level Vcm 315. As shown, stages 610a to 610c can be effectively repeated as stages 610d to 610f, where switch K4 535 is set to position "2" to reverse the polarity of the charging and discharging operations. Figure 6A simplified illustrative representation of Vout 235 is shown. It can be seen that applying FUC to a PCC-based sensing method effectively shifts Vout 235 from DC to a higher frequency domain (e.g., this frequency could be approximately 0.5 divided by the period 605).

[0055] Although Figure 5 and 6 The methods described may be effective, but they often have several limitations. One limitation is that adding FUC requires the circuit to be able to switch the polarity of charging and discharging operations, which can increase the circuit's considerable complexity. Another limitation is that PCC-based methods often occupy unwanted space. For example, the capacitance of Cc 505 is typically significantly smaller than that of Cs 105, such that the coupling of Cc 505 to Vin 310 is sufficient to draw charge, allowing the amount of charge Qd drawn from Cs 105 to essentially stabilize Vin 310 to Vcm315 within the desired amount of time (in the absence of touch events). Typically, Qd = Id * T, where Id is the discharge current. For it to function properly, the capacitance of Cc 505 is typically chosen to be approximately one-third the capacitance of Cs 105 (e.g., if the capacitance of Cs 105 is 1 nF, then Cc 505 could be approximately 330 pF). Therefore, implementing this PCC-based approach involves providing a Cc 505 for each channel (e.g., each instance of a Cs 105 can have a corresponding instance of a Cc 505). In particular, when there are dozens or more channels in a touch panel, and further, with higher self-capacitance values, the Cc 505 instances occupy a relatively large silicon area, which may be undesirable for many applications. Another limitation of the PCC-based approach is its reliance on the complete stabilization of the Cc 505 in each discharge cycle. While such complete stabilization can be beneficial (e.g., it can result in low sensitivity to clock jitter), it adds a considerable amount of time to readout per Vout 235 (e.g., for each channel). As the number of channels increases, this approach may fail to maintain the desired update rate. Adding FUC exacerbates this limitation, as it may be necessary to include multiple discharge cycles of the Cc 505 for each precharge so that a single Cc 505 can be reused for each read, and this multiple discharge cycle may consume even more time per readout cycle.

[0056] To avoid the large space loss and other limitations associated with PCC-based methods, some conventional implementations use a resistive method to discharge Ci over discrete time quantities. Figure 7A high-level block diagram of a self-capacitor sensing environment 700 for discrete-time sensing is shown, which uses a conventional resistance-to-time (RTC) implementation of a discharge stage 302, and the discharge stage 302 has frequency-domain up-conversion (FUC) for noise reduction. Figure 7 Parallel description, Figure 8 It shows in Figure 7 The simplified illustrative waveforms and timing of several stages of circuit operation 810 are shown. Figure 3 and Figure 5 In this embodiment, environment 700 includes an AFE coupled to a capacitive channel of the touch panel array. The channel has a self-capacitance Ci, represented by a self-capacitance Cs105 (capacitively coupled display noise 225), connected in parallel with a touch capacitor 205, which varies from zero capacitance in the absence of a local touch event to a non-zero capacitance in the presence of a local touch event. The AFE is shown as being implemented by an RTC discharge block 710 and a differential amplifier block 720. The RTC discharge block 710 includes two discharge paths, each corresponding to a different operating polarity. Each path has a corresponding discharge resistor Rd 705 (shown as Rda705a and Rdb705b) for effectively discharging Ci.

[0057] Typically, FUC is achieved by setting switches K4 735 (shown as K4a 735a and K4b 735b) to position '1' (corresponding to the first polarity) in the first half-cycle of operating cycle 805, and setting switches K4 735 to position '2' (corresponding to the second opposite polarity) in the second half-cycle of operating cycle 805. Additionally, according to reference... Figure 3 and Figure 4 The three phases 810 described in phase 402 (shown as phases 810a to 810c) can operate in the same manner for each half-cycle portion of period 805. See also Figure 3 and 4 (and similar to) Figure 5 and 6As described in the operation, K1 330 is closed during the first stage 810a, and Vin 310 is coupled to Vcc or a source voltage of opposite polarity Vss, thereby charging Ci (Cs 105 and Ctouch 205 in parallel) to the charging level. Simultaneously, the corresponding switches K2 335 in the two discharge branches (i.e., switches K2a 335a and K2b 335b) are opened, making the RTC discharge block 710 effectively inactive. In the second stage 810b, switch K2 335 is closed (after switch K1 330 is opened), and one of Rda 705a or Rdb 705b is coupled to Vin 310 according to the polarity setting implemented by switch K4 735. In this case, Ci discharges to a certain discharge level (i.e., corresponding to the capacitance of Ci, which corresponds to the presence or absence of a touch event) through the coupled Rd 705. In stage 810c, switch K3 340 is closed, and Vin 310 is coupled to the input of differential amplifier block 720. As shown, amplifier 720 can generate differential Vout 235 based on the difference between Vin 310 and the discharge reference level Vcm 315. As shown, stages 810a to 810c can be effectively repeated as stages 810d to 810f, where switch K4 735 is set to position "2" to reverse the polarity of the charging and discharging operations. Figure 8 A simplified illustrative representation of Vout 235 is shown. It can be seen that, similar to the PCC-based approach, applying FUC to the RTC-based sensing method effectively shifts Vout 235 from DC to a higher frequency domain.

[0058] Therefore, the RTC-based method can be designed to generate similar outputs as the PCC-based method, except that the RTC discharge block 710 does not rely on multiple instances of large capacitors (as in the case of the PCC-based implementation), and can correspondingly save significant space. However, because current and voltage are inversely proportional in a resistor, the amount of charge discharged through Rd 705 varies with Vin 210 in the second stage 810b. Therefore, the discharge provided by the RTC discharge block 710 causes signal changes represented by variations in Ci (e.g., between touch and non-touch conditions) to leak through Rp 705, resulting in very large signal loss. Furthermore, the RTC discharge block 710 can be highly sensitive to clock jitter in the second stage 810b. Clock noise can cause slight variations in the pulse width used to control the on / off timing of K2 335, effectively altering T 805. It is known that the capacitor current Ic is related to the changes in its capacitance and voltage over time: Ic = C*(dV / dt). If the pulse width time is increased due to clock jitter Tj, then for the discharge current Id, the voltage error Vin_e caused by jitter at Vin 310 can be described as: Vin_e=Tj*Id / (Cs+Ctouch).

[0059] The embodiments described herein provide several novel features to reduce detection suppression noise (i.e., to generate a high signal-to-noise ratio (SNR) for detection). To illustrate these features, Figure 9 A simplified block diagram of a low-noise discrete-time self-capacitance sensing system 900 according to an embodiment described herein is shown. (See reference...) Figure 3 The described sensing environment 300 for a specific touch sensing channel (i.e., a specific row or column line of a touch panel array) is shown as including an input stage 301, a discharge stage 302, and a sensing stage 303. Furthermore, as referenced... Figure 3 The timing of the stages is controlled by the phase-controlled switch controller 360. The sensing system 900 is shown as including the same stages, but with an additional background of multiple touch sensing channels (e.g., dozens of row and column lines) of a touch panel array, each touch sensing channel being read out via an AFE.

[0060] Specifically, Figure 9 Assume the touch panel array has N touch sensing channels (N is an integer greater than 1). By convention, touch sensing channels can correspond to labels 'a'–'n' appended to the end of a reference numeral. For example, each of the N touch sensing channels can be associated with a corresponding input stage 301, which herein may be referred to as input stages 301a to 301n according to prescribed conventions. To avoid Figure 9Due to its complexity, only a limited number of instances of each element in the diagram are explicitly shown. For example, only input stage 301a (corresponding to input stage 301 of the first touch sensing channel according to the prescribed convention) is explicitly shown, but the diagram is intended to implicitly represent N instances of the same name, input stages 301a to 301n.

[0061] Each of the N instances of input stage 301 includes a corresponding total channel self-capacitance Ci 905, which represents the parasitic capacitance on the channel from capacitively coupled display noise (i.e., the self-capacitance Cs of the base quantity) plus the additional touch capacitance (i.e., Ctouch) in the presence of a touch event. (See above, for example, reference...) Figure 3 and Figure 4 The readout of the corresponding Ci 905 for each touch sensing channel follows a series of stages. For example, during the first stage (e.g., based at least on the control timing of switch K1 330), the corresponding Ci 905 is charged to set the corresponding Vin 310 to a charging voltage level. In the second stage (e.g., based at least on the control timing of switch K2 335), Ci is discharged for a period of time, thereby bringing the corresponding Vin 310 to a discharge voltage level. As described above, the sensing system 900 is designed such that changes in Ci 905 between touch and non-touch conditions cause changes in the discharge rate, which manifests as a detectable difference in the discharge voltage level on Vin 310 at the end of the second stage. Therefore, during some discrete discharge times (e.g., Figure 4 After T405), when K2 335 is disconnected, the residual charge on Ci (and the corresponding level of Vin 310) is detectably different between touch event and non-touch event conditions.

[0062] As shown in the figure, the discharge of Ci can be performed by a corresponding instance of one of N instances of discharge stage 302 associated with the channel, which can be coupled to a corresponding instance of one of N instances of noise-suppressed discharge current generator 910 associated with the channel. In some embodiments, a group of instances of noise-suppressed discharge current generator 910 shares bias generator 950. In some embodiments, a single instance of bias generator 950 is shared by all N instances of noise-suppressed discharge current generator 910 (i.e., by all touch sensing channels). In other implementations, each of the plurality of instances of bias generator 950 is shared by a corresponding subset of the N instances of noise-suppressed discharge current generator 910. Embodiments of noise-suppressed discharge current generator 910 and bias generator 950 are discussed in more detail below (e.g., in Figure 10 , Figure 11 , Figure 14 and Figure 15 middle).

[0063] In the third stage (e.g., at least based on the control timing of switch K3 340), the discharge voltage level represented by the corresponding Vin 310 at the end of the second stage is converted to the corresponding Vout 235 by the corresponding instance of sensing stage 303. As shown, each instance of sensing stage 303 may include a corresponding instance of passive mixer 920, sample and hold (S / H) block 930, and amplifier (Amp) block 940. Embodiments of sensing stage 303 and its various components are discussed in more detail below (e.g., in...). Figure 12 , Figure 13 , Figure 15 and Figure 16 middle).

[0064] Sensing system 900 is illustrated in a configuration for differential sensing. Therefore, embodiments may include N-1 instances of sensing stage 303 to support N touch sensing channels (only three sensing stages 303a to 303c of the N-1 sensing stages 303a to 303(n-1) are explicitly shown). For example, Vin 310a and Vin 310b (for the first and second touch sensing channels) serve as differential inputs to the first sensing stage 303a, Vin 310b and Vin 310c (for the second and third touch sensing channels) serve as differential inputs to the second sensing stage 303b, and so on. Any suitable alternative arrangement may be used to provide differential inputs to sensing stage 303. For example, by assigning Vin 310a and Vin 310b to the differential inputs of the first sensing stage 303a, and Vin 310c and Vin 310d to the differential inputs of the second sensing stage 303b, only N / 2 sensing stages can be used. In other embodiments, some or all of the instances of sensing stages 303 can be used as single-ended (i.e., non-differential). For example, N instances of sensing stages 303 can be used, each instance having one input coupled to a corresponding touch sensing channel among N touch sensing channels, and another input coupled to a reference voltage (e.g., Vcm). Figure 16 An example of this single-ended implementation is described in the document.

[0065] Figure 10A circuit block diagram of an illustrative example of a noise-suppressed discharge current generator 910 according to several embodiments described herein is shown. As shown, the noise-suppressed discharge current generator 910 may include a rotator 1020 and a bias generator 950. The bias generator 950 is used to generate a bias voltage Vb for a plurality of current sources of the noise-suppressed discharge current generator 910 based on a feedback Vfb from one or more current sources in the current sources 1025 of the noise-suppressed discharge current generator 910. In some embodiments, the bias generator 950 is entirely limited to the noise-suppressed discharge current generator 910. For example, each instance of the noise-suppressed discharge current generator 910 includes an instance of the bias generator 950. In other implementations, some or all of the bias generators 950 are implemented separately from (and electrically coupled to) the noise-suppressed discharge current generator 910. For example, a single instance of bias generator 950 is coupled to and receives feedback from multiple instances of noise-suppressed discharge current generator 910. Some embodiments of bias generator 950 generate Vb as a constant bias voltage. Other embodiments of bias generator 950 generate Vb according to a specific bias profile, such as a substantially constant slope ramp. For example, each instance of bias generator 950 may include a ramp generator, or multiple instances of bias generator 950 may share a ramp generator to generate a constant slope ramp used by bias generator 950 to generate its Vb.

[0066] As shown in the figure, the noise-suppressed discharge current generator 910 may include M current sources 1025 (shown as current sources 1025a to 1025m), all of which are biased by the same Vb. The current sources 1025 can be implemented in any suitable manner. Each current source 1025 is represented as a transistor connected in series with a resistor. Each transistor has a gate coupled to Vb, a drain coupled to a corresponding source port of the rotator 1020, and a source grounded (Gnd) via a resistor. The rotator 1020 is configured to have M source ports (labeled '1'–'M'), a feedback port (labeled 'xi'), and a drain port (labeled 'x(M-1)'). M can be any positive integer greater than 1.

[0067] The rotator 1020 can be driven by a clock Clk_r. Each rotation has M Clk_r cycles. In the i-th cycle of the M Clk_r cycles, the i-th source port of the M source ports is coupled to the feedback port, and the remaining M-1 source ports are coupled to the drain ports. For example, in the first Clk_r cycle, the first source port '1' is coupled to the feedback port, such that current source 1025a is feedback coupled to bias generator 950. In the illustrated configuration, the current at the feedback port corresponds to the current at the first source port, and this current is converted into a feedback voltage by a resistor coupled to the local source voltage Vdd. Meanwhile, the other source ports '2'–'M' are coupled to the drain ports, such that current is provided by the rotator 1020 from the M-1 current sources 1025 (i.e., from current sources 1025b to 1025m) other than current source 1025a. In this configuration, the feedback loop with current source 1025a attempts to cancel the noise contribution from current source 1025a. Therefore, the current generated by current source 1025a is effectively based on the noise-cancelled Vb, which corresponds to the base bias voltage minus a certain amount of noise-cancelled Vn_ch1 on the first current source 1025a. Because Vb is shared by all current sources 1025, all current sources 1025 will be biased to generate current in the first Clk_r cycle, which is reduced by the noise-cancelled Vn_ch1. However, each of the remaining M-1 current sources 1025 will also contribute its own noise during the first Clk_r cycle. Therefore, the total current-generating noise in the first Clk_r cycle will include the sum of the noise contributions from current sources 1025b-1025m minus M-1 times the noise contribution from current source 1025a. For example, the total current-generating noise in the first Clk_r cycle can be expressed as:

[0068] Total_noise_cycle1=sum(Vn_ch2,…,Vn_chM)–(Vn_ch1*(M-1)).

[0069] Similarly, the total current-generated noise for any i-th Clk_r cycle (after the first Clk_r cycle) can be expressed as:

[0070] Total_noise_cyclei=sum(Vn_ch1,…,Vn_ch(i-1),Vn_ch(i+1),…,Vn_chM)–(Vn_chi*(M-1)).

[0071] After M Clk_r cycles, it can be seen that the noise from each current source 1025 is effectively canceled, making the total current-generated noise zero throughout the entire Clk_r cycle, as follows:

[0072] Total_noise=[sum(Vn_ch1,…,Vn_chM)*(M-1)]–[sum(Vn_ch1,…,Vn_ch20)*(M-1)]=0.

[0073] Therefore, as the rotator 1020 rotates through all current sources 1025, any noise from the current sources 1025 is canceled out, and the resulting current is actually M-1 times the unit current Iunit of each individual current source 1025 biased by Vb. The rotation period is much shorter than each readout period, allowing the rotator 1020 to operate in a single discharge phase (i.e., Figure 4 During stage 402b), the circuit rotates multiple times through all M current sources 1025. Thus, the i-th discharge current Iout_i generated by any i-th instance of the noise-suppressed discharge current generator 910 (and therefore by any i-th instance of the discharge stage 302 including the noise-suppressed discharge current generator 910) is (M-1)*Iunit. For example, refer to... Figure 9 Node Vin 310a is coupled to a first instance of discharge stage 302a, which includes a first instance of a noise-suppressed discharge current generator 910a. When discharge stage 302a is activated during the discharge phase (e.g., by closing switch K2 335), node Vin 310a is coupled to node Iout_a (i.e., the Iout_i node shown corresponding to the i-th instance, but referred to as the Iout_a node corresponding to the first instance) to generate a discharge current of (M-1)*Iunit.

[0074] In some embodiments, as shown, the noise-suppressed discharge current generator 910 further includes a fine-tuning current source 1030 in a current path connected in parallel with the rotator 1020. The fine-tuning current source 1030 can be biased by the same Vb, but can be used to draw an adjustable current amount, denoted as α*Iunit. Typically, α represents a fractional value, such that the fine-tuning current source 1030 facilitates fine-tuning of the discharge current. M can be selected to coarsely determine the discharge current as (M-1)*Iunit, and α can be adjusted to fine-tune the discharge current to (M-1+α)*Iunit. For example, if M is 20 and α is 0.4, the noise-suppressed discharge current generator 910 can be used to generate a discharge current of 19.4*Iunit.

[0075] Figure 11 A circuit block diagram of another illustrative example of a noise-suppressed discharge current generator 910 coupled to a shared ramp generator 1110, according to several embodiments described herein, is shown. The noise-suppressed discharge current generator 910 includes a rotator 1020, a current source 1025, a fine-tuning current source 1030, and related circuitry to reference... Figure 10It operates in the same manner as described. As shown, the bias generator 950 includes a shared ramp generator 1110 and a local bias generator 1120. (As generally referenced...) Figure 10 The bias generator 950 is described. Figure 11 The local bias generator 1120 generates Vb for biasing the current source 1025 (e.g., and the fine-tuning current source 1030).

[0076] An embodiment of the local bias generator 1120 is implemented as an operational amplifier. For example, the operational amplifier output Vb is a function of comparing the bias reference voltage generated by the bias generator 950 with the feedback voltage Vfb from the rotator 1020. In some embodiments, the operational amplifier is implemented by chopping to help reduce noise from the operational amplifier itself (e.g., the operational amplifier's so-called "flicker" noise). As used herein, "chopping" refers to the switching of a differential input or output, such that the signal is shifted to a higher frequency (based on the switching frequency) and can effectively cancel out lower frequency noise. As shown, the operational amplifier can be chopping-implemented at its input and output. While this particular implementation of the local bias generator 1120 as an operational amplifier with input / output chopping is only... Figure 11 This is shown in the context of (e.g., in the specific environment of the illustrated rotator 1020 configuration, the shared ramp generator 1110 configuration, etc.), but this implementation of the local bias generator 1120 can be incorporated into any other embodiment described herein.

[0077] In some embodiments, bias generator 950 generates a bias reference voltage as a constant reference voltage, such that local bias generator 1120 generates Vb as a substantially constant bias voltage, and noise-suppressed discharge current generator 910 generates a substantially constant discharge current. In other embodiments, bias generator 950 generates a bias reference voltage with a non-constant curve (e.g., according to a linear function, nonlinear function, etc.), such that local bias generator 1120 generates Vb as a non-constant bias voltage, and noise-suppressed discharge current generator 910 generates a non-constant discharge current. For example, the illustrated shared ramp generator 1110 is used to generate a ramp reference voltage Vref_ramp that follows a substantially constant negative slope. Therefore, local bias generator 1120 generates Vb as a ramp-down bias voltage, and noise-suppressed discharge current generator 910 generates a ramp-down discharge current (i.e., a discharge current generated by ramping down). Specifically, the shared ramp generator 1110 is configured such that the discharge current generated by the noise-suppressed discharge current generator 910 from the resulting ramp is generated in each discharge phase (i.e., Figure 4The second stage (402b) begins very large and ends very small at the end of each discharge stage.

[0078] For example, refer to Figure 3 In the context of this embodiment, the shared ramp generator 1110 can be configured such that, in the absence of a touch event, the total discharge current on each discharge stage causes Vin 310 to stabilize at Vcm 315 at the end of the discharge phase. In some embodiments, the discharge current generated by the ramp towards the beginning of each discharge phase is sufficiently high to reduce any noise contribution from the discharge current itself generated by the ramp. Furthermore, since Vin 310 is relatively high at the beginning of each discharge phase, a higher voltage drop can be provided across Rp, which supports the use of a higher discharge current at the beginning of the discharge phase. As mentioned above, the voltage error at Vin 310 due to clock jitter during the discharge phase is related to the discharge current (i.e., a higher discharge current over the same jitter time results in a larger error at Vin 310). The relatively low discharge current generated by the ramp towards the end of each discharge phase reduces the impact of clock jitter on detection.

[0079] Return to Figure 11 The illustrated shared ramp generator 1110 includes a capacitor C1, four resistors R1 to R4, and a network of two ramp switches KR1 and KR2. KR1 is closed during the first ramp generation phase (e.g., it may correspond in time to...). Figure 4 The first stage 402a is used to charge Ci. This couples C1 between the ramp source voltage Vdd and Vref_ramp. The voltage at Vref_ramp is a portion of Vdd controlled by a voltage divider of resistors R1 and R2 (coupled between Vdd and ground). After charging C1, the voltage across C1 is the difference between Vdd and Vref_ramp. In the second ramp generation stage, KR1 is open and KR2 is closed. This couples C1 in parallel with R3, causing C1 to discharge through R3, resulting in Vref_ramp rising to Vdd in a substantially linear manner. The timing of the second ramp generation stage can correspond to the discharge stage used to discharge Ci (e.g., ...). Figure 4 The timing of the second stage (402b). For example, the discharge of C1 may begin simultaneously with or slightly earlier than the discharge stage (e.g., during the closing phase). Figure 4 or Figure 9 (before switch K2 335 in the middle), and can end at the same time as the discharge phase or slightly later than the discharge phase.

[0080] Resistor R4 is coupled to C1 such that the fundamentally linear sloping current profile of C1 is reflected by the current profile through R4. R4 is also coupled to the feedback path of the local bias generator 1120, such that the output voltage (i.e., Vb) of the local bias generator 1120 has a correspondingly fundamentally linear sloping voltage profile. Therefore, the noise-suppressed discharge current generator 910 can generate a ramp-generated discharge current that is proportional to the current discharged from C1 (and follows the current's sloping profile). As described above, the same instance of the shared ramp generator 1110 can be coupled to multiple instances of the local bias generator 1120 (each instance limited to a corresponding instance of the noise-suppressed discharge current generator 910), such that the same sloping bias current and Vref_ramp can be used by multiple local bias generators 1120 to generate their respective bias voltages. U.S. Patent Application No. 18 / 164,605, filed on February 5, 2013, entitled “Self-Capacitance Sensing for Capacitive Touch Panels,” provides further descriptions and embodiments of ramp bias current generation, which are incorporated herein by reference in their entirety.

[0081] Figure 12 A circuit block diagram illustrating an illustrative implementation of the differential sensing stage 1200 according to the various embodiments described herein is shown. For greater clarity, Figure 13 It shows the relationship with Figure 12 Illustrative graphs and timing diagrams of several signals related to the operation of the circuit. The differential sensing stage 1200 can be... Figure 9 and / or Figure 3 The implementation method of the sensing level 303 instance. (And...) Figure 9 Consistent with the description, the differential sensing stage 1200 is shown as including a passive mixer 920, a sample-and-hold block 930, and an amplifier block 940. For example, although not explicitly labeled, the differential sensing stage 1200 may represent the i-th instance of sensing stage 303i, which has an associated i-th instance of the passive mixer 920i, sample-and-hold block 930i, and amplifier block 940i. The differential sensing stage 1200 may receive differential input signals corresponding to the respective Vin 310 nodes of the two channels, which are generally referred to as Vin 310i of the i-th channel and Vin310(i+1) of the (i+1)-th channel.

[0082] As described herein, the fundamental value of the self-capacitance of any touch sensing channel is based on the display noise capacitively coupled to that channel from the integrated display panel (e.g., to that particular row or column line of the touch panel array). It is generally assumed that while the display noise varies across the display panel, it tends to have very small local variance. For example, it can be assumed that the display noise coupled to two directly adjacent channels of the touch panel array is similar enough to be considered common-mode noise by the sensing circuitry described herein. Similarly, adjacent but not directly adjacent channels may experience capacitively coupled display noise that is sufficiently similar to be considered common-mode noise by the sensing circuitry described herein. Figure 12 The intermediate differential receiving channel generally refers to any i-th channel and (i+1)-th channel that are considered "adjacent". In this context, the term "adjacent" generally means channels that are directly adjacent to each other on the touch panel (e.g., and any associated metal row or column lines, mutual capacitors, etc.), within a few rows of each other on the touch panel, or otherwise close enough to each other that the display noise coupled to these channels can be considered as common-mode noise for the cancellation purposes described herein. Thus, in some embodiments, the i-th channel and the (i+1)-th channel are directly adjacent channels of the touch panel array (e.g., directly adjacent row lines or directly adjacent column lines). In other embodiments, the i-th channel and the (i+1)-th channel are not directly adjacent, but are close enough to experience sufficient common-mode display noise for cancellation by the sensing circuitry described herein.

[0083] An embodiment of the passive mixer 920 is implemented as a chopper that switches its input-output paths according to the mixer clock Ck_mix 1210. Each cycle of Ck_mix 1210 comprises two cycles: in the first cycle, the passive mixer 920 is in a pass-through configuration, whereby it couples the first input to the first output and the second input to the second output; in the second cycle, the passive mixer 920 is in a switched configuration, whereby it couples the first input to the second output and the second input to the first output. In some embodiments, each readout cycle of the i-th touch sensing channel may be executed as two readout cycles, each readout cycle switching the passive mixer 920 between the pass-through and switched configurations. For example, Figure 13 The complete Ck_mix 1210 cycle 1305 representing a dual readout cycle is shown. The first half of cycle 1305 is the first readout cycle, where K1 330, K2 335, and K3 340 follow at least the reference... Figure 4The timing is described, and Ck_mix1210 is HIGH (corresponding to the first, e.g., pass-through setting). The latter half of cycle 1305 is the second read cycle, which can be the same as the first read cycle (i.e., K1 330, K2 335, and K3 340 follow at least reference...). Figure 4 (Description of timing), except that Ck_mix 1210 is LOW (corresponding to the second, for example, swap setting).

[0084] The differential output of the passive mixer 920 can be passed to a sample-and-hold (S / H) block 930. As shown, the sample-and-hold block 930 may include two branches, each branch coupled between a corresponding output of the passive mixer 920 and a sampling reference voltage Vsh 1205. Each branch may be nominally identical, including a corresponding S / H resistor Rsh 1215 and a corresponding S / H switch Ksh 1220 (i.e., the first branch includes a first S / H resistor Rsh 1215a and a first switch Ksh 1220a, and the second branch includes a second S / H resistor Rshb 1215a and a second switch Kshb 1220b). An embodiment also includes a sample-and-hold capacitor Csh 1225 coupled between the outputs of the passive mixer 920.

[0085] S / H switch 1220 switches according to the same timing sequence as the K3 340 instance, such as Figure 4 and Figure 13 As shown. For example, in the first readout cycle, Ck_mix 1210 sets the passive mixer 920 to a pass-through configuration. See reference... Figure 13In the third stage 402c of the first readout cycle, switch K3 340 is closed. This allows the discharge voltage level of the i-th channel (i.e., the voltage at node Vin310i) to pass through the upper input and output of passive mixer 920, and allows the discharge voltage level of the (i+1)-th channel (i.e., the voltage at Vin 310(i+1)) to pass through the lower input and output of passive mixer 920. Simultaneously, both S / H switches Ksh 1220 are closed. This causes the voltage at node Vin 310i to be sampled in the first branch of sample-and-hold block 930 as the voltage across Rsha 1215a (i.e., the difference between Vsh 1205 and the voltage at node Vin 310i), and the voltage at node Vin310(i+1) to be sampled in the second branch of sample-and-hold block 930 as the voltage across Rshb 1215b (i.e., the difference between Vsh 1205 and the voltage at node Vin 310(i+1)). Csh 1225 charges to the voltage difference between the branches of sample-and-hold block 930 (i.e., the differential voltage), and after switches K3 340 and Ksh 1220 are turned off, Csh 1225 holds this differential voltage (e.g., until the next third stage 402c). The second readout cycle can be performed in essentially the same way, except that Ck_mix 1210 sets the passive mixer 920 to a switched configuration (e.g., as Figure 13 (The latter half of cycle 1305 is shown). Again, in the third stage 402c of the second readout cycle, switch K3 340 and both S / H switches Ksh 1220 are closed. This causes the discharge voltage levels of the i-th channel and the (i+1)-th channel to be swapped through the passive mixer 920, such that the voltage at node Vin 310i is sampled in the second branch of the sample-and-hold block 930, and the voltage at node Vin 310(i+1) is sampled in the first branch of the sample-and-hold block 930. Therefore, in the first readout cycle, Csh 1225 effectively holds the sampled value Vin 310i–Vin 310(i+1); and in the second readout cycle, Csh 1225 effectively holds the sampled value Vin310(i+1)–Vin 310i.

[0086] Thus, the passive mixer 920 and the sample-and-hold block 930 operate together to effectively upconvert the signal via frequency domain upconversion (FUC). (See above reference.) Figures 5 to 8 As described, applying FUC to conventional PCC-based and / or RTC-based methods often involves at least doubling the discharge circuitry and exchanging copies between each half-cycle. For example, each instance of the sensing circuitry has at least two copies of the discharge circuitry, one configured for one polarity and one configured for the opposite polarity. Figure 12The passive mixer 920 and sample-and-hold block 930 in the implementation provide up-conversion of the channel signal without relying on the doubling of the discharge circuit and the exchange between copies of different polarities.

[0087] The differential output of sample-and-hold block 930 is actually an up-converted hybrid version of the voltage levels of two (e.g., adjacent) touch sensing channels corresponding to its self-capacitance. The differential output from sample-and-hold block 930 can be passed to the differential input of amplifier block 940. Amplifier block 940 can be implemented using one or more amplifiers of any suitable type. In some implementations (not shown), each differential voltage is passed to a corresponding differential amplifier in a first stage, which compares the differential voltage to a reference voltage level; and in a second stage, the output of the differential amplifier is passed to a subtractor to determine the difference between the outputs and eliminate common-mode noise. The implementation shown uses a differential difference amplifier (DDA) 1230 to effectively provide front-end subtraction of common-mode noise and amplification of the signal in a single stage, such that the desired signal is amplified without noise.

[0088] The DDA1230 generates differential output voltages Von 235n and Vop 235p based on a first differential input pair 1235p and 1235n, a second differential input pair 1237p and 1237n, and a feedback network 1240. The differential output pairs Von 235n and Vop 235p essentially correspond to amplified versions of the difference between the self-capacitance response of the i-th channel and the self-capacitance response of the (i+1)-th channel, where the fundamental self-capacitances of the two channels (i.e., the display noise from common-mode capacitive coupling) are canceled out. Each differential output from the sample-and-hold block 930 is coupled to a corresponding differential input from the first differential input pair 1235p and 1235n. The feedback network 1240 includes two nominally identical feedback branches to effectively set the feedback gain of the DDA 1230. For example, in the illustrated implementation, the first feedback branch includes a first feedback resistor Rfba and a first feedback capacitor Cfba, and the second feedback branch includes a second feedback resistor Rfbb and a second feedback capacitor Cfbb. The first feedback branch is coupled between the positive differential output voltage Vop 235p and the second negative differential input (1237n), the second feedback branch is coupled between the negative differential output voltage Von 235n and the second positive differential input 1237p, and a third feedback resistor Rfbc is coupled between the branches. The feedback gain can be a function of the ratio between the third feedback resistor Rfbc and the first feedback resistor Rfba and the second feedback resistor Rfbb.

[0089] The capacitively coupled display noise (i.e., the underlying self-capacitance) is coupled to both the first differential input pair 1235p and 1235n in essentially the same way (i.e., both because it is assumed to be common-mode noise and further because it has been mixed by the passive mixer 920). Therefore, when the differential output voltages Von235n, Vop235p are generated, this common-mode portion of the received signal is immediately suppressed by the DDA 1230. The feedback signal then essentially only amplifies (i.e., amplifies) the desired signal portion of the channel signal received at the first differential input pair 1235p and 1235n. Figure 13 An illustrative representation of the Vout 235 waveform output by the DDA 1230 is shown (i.e., corresponding to the differential output voltages Von 235p and Vop 235n). The waveform shown illustrates at least a complete cycle 1305, including a first readout cycle corresponding to a first passive mixer 920 configuration (e.g., pass-through) and a second readout cycle corresponding to a second passive mixer 920 configuration (e.g., switch-through).

[0090] Figure 14 A circuit block diagram illustrating an illustrative implementation of a differential discharge stage 1400 according to several embodiments described herein is shown. The differential discharge stage 1400 includes an example of a noise-suppressed discharge current generator 910 coupled to a bias generator 950. The differential discharge stage 1400 may be... Figure 3 or Figure 9 The implementation method of discharge stage 302. Figure 10 and Figure 11 Each diagram in the figure illustrates a single Iout_i node, which represents a single generated discharge current for the i-th touch sensing channel (i.e., for the i-th instance of the discharge stage 302i, which includes the i-th instance of the noise-suppressed discharge current generator 910i). Figure 14 In the diagram, the noise-suppressed discharge current generator 910 is shown to simultaneously generate two discharge currents for two (e.g., adjacent) touch sensing channels, denoted as Iout_i and Iout_(i+1). For example, the discharge currents for the first two touch sensing channels are Iout_1 and Iout_2, corresponding to the first and second instances of discharge stage 302. Therefore, the noise-suppressed discharge current generator 910 shown is configured for differential operation, for example, for use with... Figure 12 The implemented sensing level coupling. For example, Figure 14 The noise-suppressed discharge current generator 910 can separately... Figure 14 The nodes labeled Iout_i and Iout_(i+1) are... Figure 12 The nodes labeled Vin 310i and Vin 310(i+1) are coupled to... Figure 12The differential sensing stage 1200 is coupled. In some such differential implementations, for N touch sensing channels, there may be fewer than N instances of noise-suppressed discharge current generators 910 (e.g., there may be N-1 instances).

[0091] The noise-suppressed discharge current generator 910 receives a bias voltage Vb from the bias generator 950. In the illustrated implementation, the noise-suppressed discharge current generator 910 can operate without the local bias generator 1120 because noise suppression does not depend on feedback from the current source 1025 to the bias generator 950 (e.g., as shown in the figure). Figure 10 and Figure 11 (As shown). Embodiments of the noise-suppressed discharge current generator 910 can be operated with any suitable Vb to generate a corresponding discharge current according to any suitable curve. For example, Vb can have a constant voltage curve, a ramp voltage curve, or any other suitable voltage curve. The bias generator 950 shown can be shared by multiple instances of the noise-suppressed discharge current generator 910.

[0092] The bias generator 950 shown includes a bias voltage generator 1420 and a main current source 1425. The bias voltage generator 1420 is represented as an operational amplifier. The differential input of the operational amplifier is coupled to a bias reference voltage Vbref. Vbref can be a constant reference voltage, a ramp reference voltage, or any suitable reference voltage used to generate Vb with the desired voltage profile. The second differential input of the operational amplifier (bias voltage generator 1420) is feedback coupled to the main current source 1425. Specifically, the drain of the main current source 1425 can be coupled to the feedback voltage Vfb node; and the Vfb node is also coupled to the local source voltage Vdd via a resistor and to the second differential input of the operational amplifier. The output of the operational amplifier (bias voltage generator 1420) drives the main current source 1425 such that the current through the main current source 1425 is controlled by Vbref and Vfb. The same Vb used to drive the main current source 1425 is used to drive the current source 1025 (e.g., and the fine-tuning current source 1030) such that the current regulated by the feedback of the main current source 1425 is mirrored to all current sources of the noise-suppressed discharge current generator 910.

[0093] As shown in the figure, an embodiment of the noise-suppressed discharge current generator 910 includes a pair of current sources 1025 (shown as 1025a and 1025b) and a chopper 1450 controlled by a chopper clock signal Ck_chop 1455. Each cycle of Ck_chop 1455 comprises two cycles. In the first cycle (e.g., the first half-cycle of each cycle), the chopper 1450 is in a pass-through configuration, wherein it couples a first current path corresponding to Iout_i to the first current source 1025a and a second current path corresponding to Iout_(i+1) to the second current source 1025b. In the second cycle (e.g., the second half-cycle of each cycle), the chopper 1450 is in a switch configuration, wherein it couples the first current path corresponding to Iout_i to the second current source 1025b and the second current path corresponding to Iout_(i+1) to the first current source 1025a. Switching between configurations can cause the noise contribution from each current source 1025 to be present in each channel only half the time. Furthermore, chopping can effectively convert any noise from the current source 1025 into common-mode noise on adjacent channels, and this common-mode noise can be transmitted in a similar differential manner (e.g., as in...). Figure 9 (In the middle) the sensing level 303 is suppressed.

[0094] In some embodiments, the noise-suppressed discharge current generator 910 may further include a fine-tuning current source 1030 for each branch (i.e., a first fine-tuning current source 1030a associated with a first current source 1025a and a second fine-tuning current source 1030b associated with a second current source 1025b). The fine-tuning current source 1030 can be used to fine-tune the amount of current generated by each branch. For example, when using the chopper 1450, it is desirable to ensure that each branch of the noise-suppressed discharge current generator 910 is substantially identical, such that the discharge current generated by the channel is substantially the same throughout the entire cycle of Ck_chop 1455 (i.e., for two cycles). However, even if these branches are designed to be nominally identical (i.e., these components are intended to be identical in design), differences can naturally exist due to process variations (e.g., the difference between 1025a and 1025b). Furthermore, the channels to which the branches are coupled may be different. For example, the self-capacitance associated with the i-th channel and the (i+1)-th channel will also be different. The fine-tuning current source 1030 can be adjusted to ensure that each branch produces essentially the same amount of discharge current by compensating for any physical or other differences between branches.

[0095] As mentioned above, Figure 14 The differential discharge stage 1400 can be coupled to a differential sensing circuit, such as... Figure 12 The differential sensing stage 1200 shown is illustrated. Figure 15An example configuration of a capacitive touch sensing environment 1500 is shown, in which, Figure 14 The differential discharge stage 1400 and its implementation Figure 12 The passive mixer 920 is coupled to the differential sensing circuitry. Unless otherwise described, the discharge stage 302i shown is in conjunction with the reference... Figure 14 It operates in the same manner as described, and the sensor level 303i shown is consistent with the reference. Figure 12 and Figure 13 The same manner of operation is described. Environment 1500 also includes N instances of input stage 301 (denoted as input stages "301a...n", representing 301a to 301n), each instance having a corresponding channel self-capacitance Ci 905 (denoted as "Ci_a...n 905a...n", representing Ci_a 905a to Ci_n 905n). As described herein, the output of the i-th input stage 301i can be its corresponding Vin310i.

[0096] In the configuration shown, the corresponding outputs of the N input stages 301 (corresponding Vin 310 nodes) can be switchably coupled to the corresponding AFE among the N AFEs 230 (shown as AFE 230a to AFE 230n). For example, at a first moment, the switch at the input of each AFE 230 is in a first state, wherein: the first AFE 230a is coupled to Vin 310a and Vin 310(a+1); the i-th AFE 230i is coupled to Vin 310i and Vin 310(i+1); and the last AFE 230n is coupled to Vin 310n and Vcm 315. At the second time point, the switches at the inputs of each AFE 230 change to the second state, in which: the first AFE 230a is now coupled to Vcm 315 and Vin 310a; the i-th AFE 230i is now coupled to Vin310(i-1) and Vin 310i; and the last AFE 230n is now coupled to Vin 310(n-1) and Vin310n.

[0097] As shown in the figure, a node pair received by any i-th AFE 230i (e.g., Vin 310i and Vin310(i+1) nodes) is coupled to both the input of passive mixer 920 and the input of chopper 1450. Passive mixer 920 switches between pass-through and switched configurations at a switching rate controlled by Ck_mix 1210, and chopper 1450 switches between pass-through and switched configurations at a switching rate controlled by Ck_chop 1455. An embodiment of this configuration can be implemented such that the switching rate of passive mixer 920 is R times the switching rate of chopper 1450, where R is an integer greater than 1. For example, the frequency at which passive mixer 920 switches its configuration can be twice the frequency at which chopper 1450 switches its configuration.

[0098] While the above embodiments illustrate different implementations, other embodiments may include a single-end implementation. Figure 16 An illustrative configuration of a capacitive touch sensing environment 1600 according to various embodiments described herein is shown, wherein a single-ended discharge stage is coupled to a single-ended sensing stage. N instances are shown to support N touch sensing channels of a touch panel array. Although not explicitly shown, each of the N input stages 301 (shown as input stages 301a to 301n) includes a corresponding self-capacitance Ci, which includes a base capacitance representing display noise from the capacitive coupling of the integrated display and an additional touch capacitance in the presence of a touch event. N instances of a single-ended discharge stage 302 are shown as discharge stages 302a to 302n. Each instance of discharge stage 302 may include Figure 10 or Figure 11 A single-ended embodiment of a noise-suppressed discharge current generator 910 (e.g., including a rotator 1020).

[0099] As described above, the output of each discharge stage 302 can be considered as a corresponding Vin 310 node (shown as nodes 310a to 310n). For example, Vin corresponds to the voltage across Ci (e.g., considering one or more impedance sources in input stage 301), and this Vin in the first stage (e.g., Figure 4 During phase 402a (based on the timing of switch K1 330), it is charged to the charging voltage level, and in the second phase (e.g., Figure 4 During phase 402b (based on the timing of switch K2 335), it is discharged to the discharge voltage level. In the third phase (e.g., Figure 4 In stage 402c), each Vin 310 can be coupled via switch K3 340 to a corresponding instance of N instances of sensing stages 303 (shown as sensing stages 303a to 303n). As previously described (e.g., refer to...) Figure 9Each instance of sensing stage 303 may include a passive mixer 920, a sample-and-hold block 930, and an amplifier block 940 (shown as passive mixers 920a to 920n, sample-and-hold blocks 930a to 930n, and amplifier blocks 940a to 940n).

[0100] As described above, each passive mixer 920 includes two inputs. In the illustrated single-ended configuration, the first input of each passive mixer 920 (via switch K3 340) is coupled to an associated Vin 310 (e.g., the first input of passive mixer 920a is coupled to Vin 310a). The second input of each passive mixer 920 is coupled to a common-mode reference voltage level Vcm 315. In some embodiments, all instances of the passive mixer 920 receive the same Vcm 315. In other embodiments, different instances of the passive mixer 920 may receive different reference levels corresponding to different values ​​of Vcm 315. In some such embodiments, each of multiple regions of the display panel may tend to capacitively couple the corresponding regional level of display noise to a channel of the touch panel array that is physically adjacent to these regions. The value of Vcm 315 for the sensing stage 303 associated with the channel physically adjacent to a particular region may reflect the corresponding regional level of display noise.

[0101] As described above, each passive mixer 920 can switch between a pass-through configuration and a switched configuration in response to the Ck_mix 1210 signal (e.g., which may be the same for all passive mixers 920 or different for different passive mixers 920). Therefore, the output of the i-th passive mixer 920 switches between differential output pairs at Vin 310i and Vcm 315, and differential output pairs at Vcm 315 and Vin 310i, respectively. Correspondingly, the sampled voltage at the i-th sample-and-hold block 930i switches between a differential voltage represented as Vin 310i to Vcm 315 and a voltage represented as Vcm 315 to Vin 310i. Thus, the differential sample and hold voltage can be obtained by the i-th instance of the amplifier block 940 in accordance with the above-referenced differential embodiment (e.g., in...). Figure 12 or Figure 15 It is used in the same way as described in the text.

[0102] Figure 17A flowchart illustrating an illustrative method 1700 for performing self-capacitance sensing in a touch panel array according to embodiments described herein is shown. As described herein, the embodiments operate in the context of a touch panel array integrated with a display panel having multiple channels (e.g., touch sensing channels, such as row lines and / or column lines). Each channel has a corresponding channel self-capacitance and a corresponding touch capacitance, the corresponding channel self-capacitance including a corresponding base self-capacitance corresponding to display noise coupled to the channel by capacitance from the display panel, the corresponding touch capacitance changing in response to the presence of a touch event localized to the channel. Method 1700 may represent a readout cycle of the channel. For example, method 1700 is shown starting at stage 1704, labeled “Start of Readout Cycle”. Stages 1708, 1712, and 1716 may represent a first, second, and third stage of the readout cycle.

[0103] At stage 1708 (e.g., at the first stage of the readout cycle), an embodiment of method 1700 can charge the voltage input Vin node to a charging voltage level. The Vin node is coupled to one of a plurality of channels such that the voltage at the Vin node is related to the voltage across the corresponding self-capacitor Ci of the channel coupled thereto. Once charged, the charging voltage level of the Vin node can be a source voltage (e.g., Vcc), a portion of a source voltage based on one or more impedances associated with the channel, etc.

[0104] At stage 1712 (e.g., at the second stage of the readout cycle following the first stage), embodiments of method 1700 can discharge the Vin node for a discrete discharge time using a locally noise-suppressed discharge current Iout. Iout and the discrete discharge time can be configured such that the Vin node is discharged to a discharge voltage level that is a first voltage level in the absence of a channel-specific touch event and a second voltage level in the presence of a channel-specific touch event. For example, after the discrete discharge time has elapsed, the remaining charge on Ci (and the corresponding level of Vin 310) is detectably different between touch event and non-touch event conditions.

[0105] As described herein, a locally noise-suppressed discharge current Iout can be provided by an embodiment of the discharge stage 302 described herein, which includes a noise-suppressed discharge current generator. Embodiments of the noise-suppressed discharge current generator include a rotating current source. In some embodiments, this rotating current source is, for example... Figure 10 or Figure 11The rotator is implemented iteratively through M configurations (M being an integer greater than 1). For example, the rotator has M source ports, each coupled to a corresponding current source among the M current sources, each current source being biased by a bias voltage (e.g., constant bias voltage, ramp bias voltage, etc.) generated by a bias generator at least based on bias feedback. The rotator also includes a feedback port coupled to the bias generator feedback, and a drain port coupled to the Vin node via a discharge current path. In the m-th condition of the M conditions (i.e., m is a counter from 1 to M), the iteration includes: generating bias feedback based on the m-th current source among the M current sources at the feedback port; and generating Iout at least based on the sum of the currents through the M-1 current sources other than the m-th current source. In other embodiments, the rotating current source is, for example, Figure 14 or Figure 15 This is achieved through the switching configuration of the chopper in the system.

[0106] At stage 1716 (in the third stage of the readout cycle following the second stage), an embodiment of method 1700 may output a voltage output Vout for the channel. This output may be implemented by an embodiment of sensing stage 303 described herein. This output may include at least passively mixing the discharge voltage level to generate an up-converted channel signal pair; sampling the up-converted channel signal pair to obtain a differential voltage sample; and amplifying the differential voltage sample to generate Vout, wherein Vout indicates the presence of a touch event localized to the channel.

[0107] In some embodiments, each readout cycle is performed as two half-cycles, each half-cycle having a passive mixer switched to a different configuration. As described herein, switching the passive mixer in this manner can produce a frequency-domain up-conversion effect without doubling the circuitry used for charging, discharging, sensing, etc. For example, as... Figure 17As shown, the first half-cycle of each readout cycle can be achieved by setting the passive mixer to a pass-through configuration at stage 1706 and executing stages 1708, 1712, and 1716 while the passive mixer is in a pass-through configuration. The second half-cycle of each readout cycle can be achieved by setting the passive mixer to a switched configuration at stage 1718 and executing stages 1708, 1712, and 1716 while the passive mixer is in a switched configuration. As described herein, the passive mixer receives the discharge voltage level of the Vin node at its first input. At the second input, the passive mixer receives another discharge voltage level associated with an adjacent channel (e.g., the inputs receive Vin_i and Vin_(i+1), respectively), or the passive mixer receives a common-mode reference voltage level (e.g., Vcm). The passive mixer can be implemented in stage 1716 to generate up-converted channel signal pairs at the first and second outputs in such a way that: during the first half-cycle, with the passive mixer in a pass-through configuration, the first input is coupled to the first output and the second input is coupled to the second output; and during the second half-cycle, with the passive mixer in a switched configuration, the first input is coupled to the second output and the second input is coupled to the first output.

[0108] It should be understood that when an element or component is referred to herein as being "connected to" or "coupled to" another element or component, it may be connected to or coupled to that other element or component, or there may be intermediate elements or components. Conversely, when an element or component is referred to herein as being "directly connected to" or "directly coupled to" another element or component, there are no intermediate elements or components between them. It should be understood that although the terms "first," "second," "third," etc., may be used herein to describe various elements, components, or regions, these elements, components, or regions should not be limited by these terms. These terms are only used to distinguish one element or component from another. Therefore, without departing from the teachings of the invention, the first element or first component discussed below may be referred to as the second element or second component. As used herein, the terms "logic low," "low state," "low level," "logic low level," "low," or "0" are used interchangeably. The terms "logic high," "high state," "high level," "logic high level," "high," or "1" are used interchangeably.

[0109] As used herein, the terms “a,” “an,” and “the” can include both singular and plural references. It should also be understood that, when used in this specification, the terms “comprising,” “including,” “having,” and variations thereof specify the presence of the stated features, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, steps, operations, elements, components, and / or combinations thereof. Conversely, when used in this specification, the term “consisting of” specifies the stated features, steps, operations, elements, and / or components, and excludes additional features, steps, operations, elements, and / or components. Furthermore, as used herein, the word “and / or” can refer to and include any possible combination of one or more of the related listed items.

[0110] While the invention has been described herein with reference to illustrative embodiments, this description is not intended to be limiting. Rather, the illustrative embodiments are intended to enable those skilled in the art to better understand the spirit of the invention. Many details of well-known processes and manufacturing techniques have been omitted so as not to obscure the scope of the invention. Various modifications to the illustrative embodiments and other embodiments will be apparent to those skilled in the art following the description. Therefore, the appended claims are intended to cover any such modifications.

[0111] Furthermore, some features of the preferred embodiments of the invention can be used advantageously without the need for corresponding use of other features. Therefore, the foregoing description should be considered merely as illustrative of the principles of the invention and not as a limitation thereof. Those skilled in the art will understand variations of the above embodiments that fall within the scope of the invention. Therefore, the invention is not limited to the specific embodiments and examples discussed above, but is limited by the appended claims and their equivalents.

Claims

1. A system for self-capacitance sensing, said system being used in a touch panel array integrated with a display panel, said touch panel array having multiple channels, each channel having a corresponding channel self-capacitance and a corresponding touch capacitance, wherein, The corresponding channel self-capacitance includes a base self-capacitance corresponding to display noise, the display noise capacitance being coupled to a channel from the panel, and the corresponding touch capacitance changing in response to the presence of a touch event in a local channel; the system includes: A voltage input Vin node is coupled to one of the plurality of channels, such that the voltage at the Vin node is related to the voltage across the corresponding self-capacitance Ci of the channel coupled thereto. A discharge stage is configured to couple the Vin node to a locally noise-suppressed discharge current Iout for a discrete discharge time during a discharge phase to discharge the Vin node to a discharge voltage level, such that the discharge voltage level is a first voltage level in the absence of a touch event localized to the channel and a second voltage level in the presence of the touch event localized to the channel; and A sensing stage, coupled to the discharge stage, outputs a voltage output Vout of the channel by: passively mixing the discharge voltage level to generate an upconverted channel signal pair; sampling the upconverted channel signal pair to obtain a differential voltage sample; and amplifying the differential voltage sample to generate Vout, wherein Vout indicates the absence or presence of the touch event localized to the channel; The discharge stage includes: A bias generator for generating a bias voltage based at least on bias feedback; M current sources, each biased by the bias voltage, where M is an integer greater than 1; and The rotator has M source ports, one feedback port, and one drain port, wherein each source port is coupled to a corresponding current source among the M current sources; the feedback port is fed back to the bias generator; and the drain port is coupled to the Vin node via a discharge current path. The rotator is configured to iterate M configurations sequentially to generate the m-th condition among the M conditions: the bias feedback based on the m-th current source among the M current sources at the feedback port; and the Iout based on the sum of the currents through the M-1 current sources other than the m-th current source.

2. The system according to claim 1, further comprising: A phase-controlled switch controller is used to control the readout of the channel during the readout cycle, such that: In the first stage, the phase-controlled switch controller guides the Vin node to charge to the charging voltage level; In the discharge phase following the first stage, the phase-controlled switch controller guides the discharge of the Vin node to the discharge voltage level through the discharge stage; as well as In the third stage, the phase-controlled switch controller guides the output of the channel's Vout through the sensing stage.

3. The system according to claim 1, wherein, The discharge stage further includes: A fine-tuning current source, which is biased by the bias voltage. In the m-th condition, each of the M-1 current sources (excluding the m-th current source) generates a unit current Iunit, where Iunit is 1 / (M-1) of Iout. The fine-tuning current source generates an adjustable current, which is the adjustable portion α of the unit current. The fine-tuning current source is coupled to the Vin node, which is connected in parallel with the rotator, such that Iout = (M–1+α)*Iunit.

4. The system according to claim 1, wherein, The bias generator includes: A shared bias generator is used to generate a bias reference voltage; and A local bias generator is used to generate the bias voltage based on the bias reference voltage and the bias feedback.

5. The system according to claim 4, wherein, The shared bias generator produces a constant bias reference voltage.

6. The system according to claim 4, wherein, The shared bias generator includes a shared ramp generator that generates the bias reference voltage to have a sloping voltage profile.

7. The system according to claim 4, wherein, The local bias generator includes an operational amplifier that receives the bias reference voltage at a first differential input, receives the bias feedback at a second differential input, and generates the bias voltage at an output based at least on the switching between the first and second differential inputs.

8. The system according to claim 1, wherein, The sensing level includes: Passive mixers have the following features: A first input terminal is coupled to the Vin node to receive the discharge voltage level; and The second input terminal is used to receive the common-mode reference voltage level Vcm. The passive mixer is used to passively mix the discharge voltage level with the Vcm, thereby generating the upconversion channel signal pair by iteratively switching between the pass-through configuration and the switching configuration of the first input terminal and the second input terminal at a switching rate controlled by the mixer clock.

9. The system according to claim 8, wherein, The sensing level also includes: Amplifier block having at least a first differential input pair; and A sample and hold block, coupled to the passive mixer, is used to sample the upconversion channel signal pair to obtain the differential voltage sample and hold the differential voltage sample on a charge storage device coupled to both ends of the first differential input pair.

10. The system according to claim 9, wherein, The amplifier block also includes: Feedback network; and A differential amplifier (DDA) having a differential output pair, a first differential input pair, and a second differential input pair negatively fed back to the differential output pair via the feedback network.

11. The system according to claim 1, further comprising: The Vin node has multiple instances, each instance being coupled to a corresponding channel among the multiple channels, such that the i-th instance Vin_i of the Vin node represents the voltage across the i-th self-capacitor Ci_i of the corresponding i-th channel coupled to it. Multiple instances of the discharge stage, wherein the i-th instance of the discharge stage is configured to: couple Vin_i to the i-th locally noise-suppressed discharge current Iout_i during a discharge phase to discharge Vin_i to a corresponding i-th discharge voltage level, the corresponding i-th discharge voltage level being the first voltage level in the absence of a touch event localized to the corresponding i-th channel and the second voltage level in the presence of the touch event localized to the corresponding i-th channel; and Multiple instances of the sensing level are coupled to multiple instances of the discharge level, wherein the i-th instance of the sensing level is configured to output a corresponding i-th voltage output Vout_i for the corresponding i-th channel by: passively mixing at least the i-th discharge voltage level to generate an i-th upconversion channel signal pair, sampling the i-th upconversion channel signal pair to obtain an i-th differential voltage sample, and amplifying the i-th differential voltage sample to generate an i-th voltage output instance, wherein the i-th voltage output instance indicates the absence or presence of the touch event localized to the corresponding i-th channel.

12. The system according to claim 11, further comprising: A bias generator is used to generate a bias voltage. Among them, a plurality of the plurality of instances of the discharge stage are used to generate a discharge current with local noise suppression based on the bias voltage.

13. The system according to claim 11, wherein, The i-th instance of the discharge level includes: The first current source and the second current source are each biased by a free bias voltage; and The chopper has a first input terminal coupled to Vin_i via a first discharge current path, a second input terminal coupled to the (i+1)th instance of the Vin port via a second discharge current path, a first output terminal coupled to the first current source, and a second output terminal coupled to the second current source. The chopper is used to switch between a pass-through configuration and a switch configuration at a rate controlled by a chopper clock. The pass-through configuration couples the first input terminal to the first output terminal and the second input terminal to the second output terminal. The switch configuration couples the first input terminal to the second output terminal and the second input terminal to the first output terminal.

14. The system according to claim 13, wherein, The i-th instance of the discharge level also includes: A first fine-tuning current source, biased by the bias voltage and coupled in parallel with the first discharge current path; and The second fine-tuning current source is biased by the bias voltage and coupled in parallel with the second discharge current path. The i-th instance of the discharge stage generates Iout_i on the first discharge current path based on the current from both the first fine-tuning current source and one of the first or second current sources. The i-th instance of the discharge stage generates the (i+1)-th locally noise-suppressed discharge current on the second discharge current path based on the currents from both the second fine-tuning current source and the first current source or another current source among the second current sources.

15. The system according to claim 13, wherein, The bias voltage is a ramp bias voltage.

16. A display system, comprising: Display panel; A touch panel array, integrated with the display panel and having multiple channels, each channel having a corresponding channel self-capacitance and a corresponding touch capacitance, wherein the corresponding channel self-capacitance includes a basic self-capacitance corresponding to display noise coupled to the channel by capacitance from the display panel, and the corresponding touch capacitance changes in response to the presence of a touch event local to the channel; as well as In multiple instances of the system as described in claim 1, each instance of the Vin node is coupled to a corresponding channel among the multiple channels.

17. A method for self-capacitance sensing, the method being applied in a touch panel array integrated with a display panel, the touch panel array having multiple channels, each channel having a corresponding channel self-capacitance and a corresponding touch capacitance, wherein, The corresponding channel self-capacitance includes a base self-capacitance corresponding to display noise, the display noise capacitance being coupled to a channel from the panel, and the corresponding touch capacitance changing in response to the presence of a touch event in a local channel, the method comprising: In the first phase of the readout cycle, the voltage input Vin node is charged to the charging voltage level. The Vin node is coupled to one of the plurality of channels such that the voltage at the Vin node is related to the voltage across the corresponding self-capacitance Ci of the channel coupled thereto. In the second stage of the readout cycle following the first stage, the Vin node is discharged with a locally noise-suppressed discharge current Iout for discrete discharge times, thereby discharging the Vin node to a discharge voltage level, wherein the discharge voltage level is a first voltage level in the absence of a touch event localized to the channel and a second voltage level in the presence of the touch event localized to the channel; and In the third stage of the readout cycle following the second stage, the voltage output Vout of the channel is output by: passively mixing the discharge voltage level to generate an upconverted channel signal pair; sampling the upconverted channel signal pair to obtain a differential voltage sample; and amplifying the differential voltage sample to generate Vout, wherein Vout indicates the absence or presence of the touch event localized to the channel; The discharge includes: Iterate through M configurations of the rotator in sequence, where M is an integer greater than 1, and the rotator includes: There are M source ports, each source port is coupled to a corresponding current source among the M current sources, each of the M current sources is biased by a bias voltage generated by a bias generator based at least on bias feedback. Feedback port, which is fed back to the bias generator; and The drain port is coupled to the Vin node via the discharge current path. Among the M conditions, the iteration includes the following: The bias feedback is generated at the feedback port based on the m-th current source among the M current sources; and The Iout is generated based at least on the sum of the currents from M-1 current sources other than the m-th current source.

18. The method of claim 17, further comprising: In the first half-cycle of the readout cycle: Set the passive mixer to pass-through configuration; as well as The passive mixer in the pass-through configuration is used to perform the charging, discharging, and output; and In the second half-cycle of the readout cycle: Configure the passive mixer to a switching configuration; and The passive mixer with the aforementioned switching configuration is used to perform the charging, discharging, and output processes. The passive mixer receives the discharge voltage level of the Vin node at its first input terminal, receives another discharge voltage level associated with an adjacent channel or receives a common-mode reference voltage level at its second input terminal, and generates the up-conversion channel signal pair at its first and second output terminals through the following operations: In the pass-through configuration, the first input terminal is coupled to the first output terminal, and the second input terminal is coupled to the second output terminal; and In the switching configuration, the first input terminal is coupled to the second output terminal, and the second input terminal is coupled to the first output terminal.

Images