Patent Application
20170277348
The present disclosure generally relates to capacitive touch sensing systems, and more particularly relates to methods of eliminating unwanted capacitances being sensed using a scheme called guarding.
Capacitive sensing systems usually consist of a capacitive sensor matrix and capacitive sensing devices implementing a specific capacitive sensing technique. One general type of capacitive sensor technology is projected capacitance (PCAP) technology, wherein electric field lines project beyond touchable surface. The projected field lines consist of two kinds of field lines: ones that terminate at adjacent sensors and ones that terminate at far away conducting surfaces in the environment, which are often grounded. The field lines that terminate at adjacent sensors are due to capacitance across sensor pairs, often called mutual capacitance and voltage applied across the sensors. The field lines that terminate at far away surfaces are due to inherent capacitance, often called self-capacitance, of a particular sensor and voltage applied to it with respect to the environment. Based on which of the two capacitance is sensed, there are two capacitance sensing techniques; one is called self-capacitance sensing and other called mutual-capacitance sensing.
A PCAP sensor with mutual capacitance sensing generally includes sensors arranged in rows and columns such that capacitance at each cross-point can be sensed. When the projected fields across two sensors are interrupted by an object such as a finger, there is change in capacitance across the sensors and it can be sensed as touch. A PCAP sensor with self-capacitance sensing generally includes sensors arranged in an arbitrary pattern covering an entire touchable area called a touch panel.
The mutual capacitance sensing technique is generally implemented such that various sensor capacitances distributed across the touch panel are sensed a row at a time. In such technique, by measuring charge transfer across two terminals of each sensor capacitor, it is possible to sense mutual capacitance explicitly. On the other hand, the self-capacitance sensing technique can be implemented to sense all the sensor capacitors at once in parallel or individually in a sequence. Sensing in a sequence reduces the cost of circuitry, and hence a device, implementing such sensing technique and is often desirable in touch panels with less sensors.
While it is possible to sense self-capacitance explicitly using parallel sensing, it is inherently impossible to sense only self-capacitance using sequential sensing. This inherent inability to isolate self-capacitance from total capacitance makes such sensing technique vulnerable to changes in mutual capacitance due to the presence of water and other conductive fluids on the touch panel. To overcome such shortcoming of sequential self-capacitance sensing technique, often additional schemes are employed to drive sensors adjacent to the sensor being sensed to emulate parallel sensing. Such additional schemes, which are an integral part of the sequential self-capacitance sensing technique is called guarding.
As detailed in U.S. Pat. Nos. 8,866,793 and 9,001,083, both of which are fully incorporated herein by reference to the extent permitted by law, the presence of water changes mutual capacitance and can greatly affect the reliability of overall touch sensing system. U.S. Pat. Nos. 8,866,793 and 9,001,083 address this problem by providing conductive structures proximate capacitive touch pads and a scheme for altering the electrical potential of the conductive structures to compensate for the effect of mutual capacitance, based on external conditions such as water or an intervening separator, e.g., a glove. The compensation for mutual capacitance improves the water immunity and therefore the reliability of the overall touch sensing system.
In the context of a touch sensing system with water immunity using sequential self-capacitance sensing, it is possible to reduce the resources needed such as analog to digital converters (ADCs) and signal conditioning circuitry called analog front end (AFE) by cycling through the sensors one at a time. However as stated before in doing so it becomes necessary to include a guarding scheme in order to drive identical waveforms or signals on adjacent sensors to avoid inclusion of mutual capacitance.
In the context of charge-transfer based sensing schemes such as drive voltage measured charge, it is necessary for the AFE to include a charge measuring circuit, which converts incoming signal in the form of charge into a voltage signal. One of the techniques for converting incoming charge into voltage uses a charge integrator circuit. In such an AFE, the input common mode of the integrator is set somewhere close to half the supply voltage to allow maximum voltage swing and to make design of such an circuit feasible. The voltage at which the touch sensors settle to during charge transfer phase is generally set to virtual ground which is generally half the supply voltage.
In
As can be seen, the sensing waveform has four phases: a charge phase φ1, a positive integration phase φ2, a discharge phase φ3, and a negative integration phase φ4, in that order.
In this example, the final voltage at the output VINT of the integrator after one sensing cycle can be expressed as follows:
In this relationship, CM is the mutual capacitance due to an adjacent sensor or conductor, CS is the sensor self-capacitance, and CW, (as used throughout this specification) is capacitance introduced by the presence of a conductive liquid or gel, e.g., water. Similarly, QS is the charge due to self-capacitance of the sensor, while QM is the charge due to the mutual capacitance (which includes that introduced by the conductive liquid or gel). CINT is a capacitance of the integrator.
In the relationship above, an internal reference factor of ½ or 50% is used as a convenient voltage for ease of implementing the integrator circuit design.
In the absence of guarding, the sensor output signal VSENSE includes charge from mutual capacitance QM, which can change in the presence of the conductive liquid or gel, e.g., water. Again, the additional capacitance CW represents this change or influence. This can lead to detecting the liquid or gel, e.g., water, as false touches.
For the purposes of this illustration, the sensor has a 1 pF capacitance, the integrating capacitor has a capacitance of 1 pF, the mutual capacitance is 0.1 pF and Vdd is 1V. The waveforms include 5 complete cycles, each cycle including a positive and a negative charge phase for the sensor. The effect of the mutual capacitance is assumed to be a charge of 0.5V and is reflected in the 0.5V bias of the sensor signal in
In
In can be appreciated that in the absence of the charge QM from the mutual capacitance, the integrator output voltage VINT should be 1V after one complete cycle and 5V after 5 cycles. However, since the charge QM from the mutual capacitance also gets integrated, it can be seen that the final integrator output voltage calculates to 5.5V instead of 5V.
However, although using a guarding signal identical to the sensor of interest driving signal should completely eliminate the charge QM, it introduces other problems. Since such a waveform needs to be driven to a middle voltage other than Vdd or ground, it is not possible to do so using digital drivers. To remedy this problem, some systems have used a replica circuit inside the sensor driving chip to generate the identical guarding waveform, which is then driven on the adjacent sensors. An example of such replica circuit can be found in Cypress Semiconductor, Inc.'s programmable system on a chip (PSoC®) CapSense™ based sensing system, where there is a copy of a sigma-delta converter just to derive the identical waveform. However such a technique is wasteful in terms of requiring additional analog circuitry.
On the other hand, if adjacent sensors or conductors are driven with a digital signal swinging from Vdd to ground, it will result in over-guarding. In this case, given all of the same parameters above, the integrator output voltage calculates to 4.5V, which is due to twice as much signal being subtracted as compared to that inherently present due to presence of the mutual capacitance CM. In the over-guarding case, the relationship for the integrator voltage becomes:
It can be appreciated that this outcome still depends on the charge QM which includes the capacitance CW and hence the presence of a conductive liquid or gel, e.g., water.
In
As can be seen in
In
Another way to look at it, both transitions of the guard signal occur when the sensor signal VSENSE is at mid voltage.
The present disclosure provides one or more inventions in which a digital guarding scheme is used to eliminate integration of undesired mutual capacitance.
In an embodiment, there is provided a digital guard waveform that is used to drive a conductive element adjacent a capacitive sensor of interest in a manner such that a time averaged value of additional charge due to guarding of the adjacent conductive element is equal and opposite to that added due to the presence of mutual capacitance due to that conductive element. This allows for the use of a simple digital guarding scheme while achieving near perfect, if not perfect, guarding at the same time.
In an embodiment, there is provided a method of driving a capacitive touch sensing system, comprising: generating a guard waveform; during a sensing cycle, guarding a capacitive sensor of interest by applying the guard waveform to at least one conductive element adjacent to the capacitive sensor of interest; and integrating a voltage of the capacitive sensor of interest during the sensing cycle, wherein, a time averaged value of additional charge due to the guarding of the adjacent capacitive sensor is equal and opposite to that charge added due to the presence of mutual capacitance between capacitive sensor of interest and the adjacent capacitive sensor.
In an embodiment the adjacent conductive element is another capacitive sensor.
In an embodiment, the sensor of interest is only guarded once per sensing cycle. The guard signal transitions are occur in such a way that only one edge of the guard signal, be it a rising edge or a falling edge, lies within the time when sensor is at mid voltage. Hence only one transition of the guard signal is effective.
In an embodiment, the integrator output VINT adheres to the following relationship:
where,
the internal reference is B (which is a percentage of Vdd), and A is 1-B,
Q
S
=C
S
V
dd, and
Q
M=(CM+CW)Vdd.
Again, CM is the mutual capacitance due to an adjacent sensor or conductor, CS is the sensor capacitance, and CW, (as used throughout this specification) is capacitance introduced by the presence of a conductive liquid or gel, e.g., water. Similarly, QS is the charge from the sensor, while QM is the charge from the mutual capacitance (which includes that introduced by the conductive liquid or ge). CINT is a capacitance of the integrator.
In an embodiment, the integrator output VINT adheres to the following relationship, given an internal reference factor of ½:
In an embodiment, a processor device includes logic that effects driving of a capacitive sensor device such that a sensor of interest is only guarded during a positive charge phase and a rising edge of a guard waveform is positioned during charge time of the sensor of interest, when the sensor of interest is not connected to an integrator.
In an embodiment, the processor device logic effects driving of the capacitive sensor device such that an integrator output VINT adheres to the following relationship, given an internal reference factor of ½:
In an embodiment, the processor device logic effects driving of the capacitive sensor device such that an integrator output VINT adheres to the following relationship:
where the internal reference is B (which is a percentage of Vdd), and A is 1-B.
In an embodiment, a non-transitory storage device includes machine readable or implementable logical instructions that when executed by the machine cause the machine to effect of a capacitive sensor such that a sensor of interest is only guarded during a positive charge phase and a rising edge of a guard waveform is positioned during charge time of the sensor of interest, when the sensor of interest is not connected to an integrator.
In an embodiment, the logical instructions effect driving of the capacitive sensor such that an integrator output VINT adheres to the following relationship, given an internal reference factor of ½:
In an embodiment, the logical instructions effect driving of the capacitive sensor such that an integrator output VINT adheres to the following relationship:
where the internal reference factor is B (which is a percentage of Vdd), and A is 1-B.
These and other aspects and features of the disclosed embodiments are described in greater detail below with reference to the accompanying drawings.
The accompanying drawings constitute a part of this specification and illustrate an embodiment of the invention and together with the specification, explain the invention.
The present disclosure is herein described in detail with reference to embodiments illustrated in the drawings, which form a part here. Other embodiments may be used and/or other changes may be made without departing from the spirit or scope of the present disclosure. The illustrative embodiments described in the detailed description are not meant to be limiting of the subject matter presented here. One skilled in the art recognizes that numerous alternative components and embodiments may be substituted for the particular examples described herein and still fall within the scope of the invention.
Embodiments of the present disclosure provides a new or improved guarding method and devices using same for capacitive touch sensor systems in which the capacitive influence of a conductive liquid or gel, e.g., water, can be eliminated or minimized.
In this improved guarding method one or more conductive elements adjacent, i.e., nearest neighbor to, a sensor element of interest are driven with a guard signal. The conductive elements may themselves be another capacitive sensor element.
In this improved guarding scheme, it is preferable to utilize the correct polarity and timing of the guard signal with respect to the sensing waveform. This is done by, essentially, guarding only during a portion of the driving time of a capacitive sensor device, preferably half of the time, such that a sensor of interest is only guarded during a positive charge phase. To that end, the guard signal transitions are positioned in such a way that only one of the edges of the guard signal waveform lies within the time when the sensor is at mid voltage. Hence only one of the guard signal transitions is effective for guarding.
Preferably, a rising edge of a guard waveform is positioned during charge time of the sensor of interest, when the sensor of interest is not connected to an integrator.
The effective expression for the final voltage output at the integrator, given an internal reference factor of 50% or ½ of Vdd, is then:
The relationships noted above still hold, namely QS=CS Vdd and QM=(CM+CW)Vdd.
Further, CM is the mutual capacitance due to an adjacent sensor or conductor, CS is the sensor capacitance, and CW, (as used throughout this specification) is capacitance introduced by the presence of a conductive liquid or gel, e.g., water. Similarly, QS is the charge from the sensor, while QM is the charge from the mutual capacitance (which includes that introduced by the conductive liquid or ge). CINT is a capacitance of the integrator
Hence although there may be over-guarding during the positive charge phase, by the end of first cycle, all of the charge coming from mutual capacitance is subtracted and the integrator output signal VINT will not depend on the presence of a conductive liquid or gel, e.g., water. As one can see in the waveform of
Comparing
Additionally, the scheme disclosed herein can be used where the internal reference voltage is not exactly 50% of the supply voltage. In such case, the voltage expression becomes:
where the internal reference factor is B (which is a percentage of Vdd, and A is 1-B.
But by the end of one complete cycle, the overall effect is similar to that when the internal reference was at 50% of Vdd, i.e., the result is still VINT=QS/CINT. Hence this scheme is independent of internal voltage reference.
For example, if the internal reference voltage is 60% of Vdd, then A is 40% and B is 60% and the relationship resolves as:
There are various ways to ensure that the guarding waveform is such that only one edge of the guarding signal lies within an integration phase. One way is to use a cyclo-stationary method. Another is to use a non-cyclo-stationary method.
In a cyclo-stationary method, timings of the various transitions of the guarding waveform are fixed with respect to the sensing waveform. Thus the guarding waveform can be derived from the sensing waveform, or the states of the various phases of the sensing waveform (i.e., the charge phase φ1, the positive integration phase φ2, the discharge phase φ3, and the negative integration phase φ4). To that end a counter or clock is used to determine the elapse of each phase of the sensing cycle. In essence, When the counter or clock reaches a desired value for a given phase, the sensing cycle transitions to the next phase. This process is repeated until the system is shut down or is put in a standby or sleep mode.
In a specific implementation, the phases φ1, φ2, φ3, and φ4 can have assigned to them a number of clock cycles M, O, P, and Q, respectively. The sensing signal generator can be a state machine, with each phase being a state of the machine. A signal or variable representative of the phase/state of the machine can then be generated. Then the guarding waveform can be generated by delaying generation of the guarding waveform by one clock cycle such that a desired guard waveform transition occurs within the desired phase.
For example, in the case off a positive guarding scheme, the guard waveform can be derived by decoding when the phase/state signal is in phase 2, the positive integration phase, and delaying the generation of the guard waveform by one clock cycle from phase 2. The width of the guard signal is such that the next transition will occur in phase 3 or within a delay from the beginning of phase 3.
In a touch sensing system employing a Sigma-Delta method or multi-cycle integration approach, the final sensing result is only obtained after a predetermined number of cycles, e.g., N cycles, where N is a programmable number. Such sensing systems are also called over-sampling systems, where N specifies a factor by which the signal is over-sampled. This need to use N cycles provides for flexibility in the guarding scheme.
In that regard, a sensor being sensed is either connected to Vdd or Ground and then connected to the charge integrator, which sets the voltage of the sensor to the reference voltage Vref. Then, the per cycle contribution from mutual capacitance is QM=Vref*CM. In this case, Vref is less than Vdd and, hence, even if one employs one guard waveform transition per cycle, it will result in over-guarding. Under these conditions, the above described implementation will be effective.
When the sensing system is operating such that the sensor is discharged and connected to a charge integrator with reference voltage Vref, the sensing waveform will transition from Ground to Vref and back to ground on every cycle. This scheme uses unipolar signaling and hence will not have a charge phase and a positive integration phase. Instead it will only have the discharge phase and the negative integration phase.
In such sensing system, it is possible to completely cancel the mutual capacitance contribution as long as: QM,total=NQM−MQG, where QM=Vref*CM, QG−Vdd*CM, and M=N*Vref/Vdd. Thus the sensing system needs to make sure it guards only M out of N sensing cycles. In this implementation, there will be some cycles which are not guarded, and hence all of the sensing cycles are not the same. As a result, this is a non-cyclo-stationary implementation.
In
The touch substrate 12 is intended to be representative of any suitable combination of layers making up a capacitive touch sensing device, including any protective layers that might be desired. Such other layers are well known.
In
In
In
In
In
In
Alternatively, the chip 66 could comprise a state machine made of hardwired logic devices that implement the desired control scheme. In that case, the core 62 and 64 would be replaced by equivalent hardware logic circuitry, which equivalent circuitry is known or easily determined.
In
In
In
The preceding description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the following claims and the principles and novel features disclosed herein.
While various aspects and embodiments have been disclosed, other aspects and embodiments are contemplated. The various aspects and embodiments disclosed are for purposes of illustration and are not intended to be limiting, with the true scope and spirit being indicated by the following claims.
