The invention relates to sensors for sensing touch and pressure. The invention relates to tactile sensors. The invention relates to pressure sensors. The invention relates to capacitive sensors for sensing touch and pressure. The invention relates to methods for capacitively measuring touch and pressure. The invention relates to computer programs used in connection with measuring touch and pressure.
Many user interfaces require sensing touch without a pressure and pressure separately. For example a user may move his/her finger on a surface, whereby a touch with the surface indicates a location of the finger. This kind of touch is made substantially without pressing the surface. Moreover, to make a selection, the user may press the surface e.g. by his/her finger. This implies a pressure applied onto the surface. Moreover, from a measured signal, the occurrences of only touch and pressing should be separable. More preferably, the pressure applied should be measurable at least with some accuracy, i.e. better accuracy than a binary application of pressure or not. A capacitive sensor for measuring pressure is known e.g. from the applicant's international publication WO 2018/011464.
Often a sensor for measuring both a touch and a pressure is complex, whereby such a sensor is also expensive. In practice, complexity of such sensors is a hindrance for their wide applicability.
For these reasons, a purpose of the present application is to present a simple sensor arrangement for sensing both touch and pressure. More specifically, to present a simple sensor arrangement for sensing, at a first instance of time, a touch substantially without application of pressure, and at a second instance of time, an application of a pressure. An operating principle of the sensor arrangement is capacitive.
As gist of the invention is the use of a capacitive layered sensor structure comprising an electrode layer comprising a first electrode and an electrically permeable and conductive layer. Touch and pressure can be determined from capacitances of the first electrode relative to the surroundings. A first purpose of the electrically permeable and conductive layer, related in particular to the permeability of the layer, is to pass an electric field through the electrically permeable and conductive layer so as to capacitively sense, by the first electrode that is located on a first side of the electrically permeable and conductive layer, an object that is located on a second, opposite, side of the electrically permeable and conductive layer. A second purpose of the electrically permeable and conductive layer, related in particular to the conductivity of the layer, is to form a capacitance in between the electrode and the electrically permeable and conductive layer itself. The capacitive layered sensor structure can be used for such measurements. E.g. an electronic arrangement can be configured to measure a capacitance of the first electrode indicative of a touch (or presence in a vicinity) and another capacitance of the first electrode indicative of a pressure.
A less complex sensor arrangement for the purpose is presented in independent claim 1 of the present application as filed. That sensor arrangement comprises a capacitive layered sensor structure as detailed in the claim 1. Moreover, such a capacitive layered sensor structure can be used to measure both a touch and a pressure as detailed in an independent method claim 13 or the application as filed. Moreover, the method can be run on a computer by running a computer program on a computer. Such a computer program is detailed in claim 16 of application as filed.
1 and 4a2 show, in side views, how touch is sensed by a capacitive layered sensor structure,
2 and 4b2 show, in side views, how pressure is sensed by a capacitive layered sensor structure,
1 and 4c2 show, in side views, how absence of touch and pressure is sensed by a capacitive layered sensor structure,
1 and 4d2 show, in side views, how a suitable first threshold may be selected,
In the figures, the direction Sz indicates a direction of a thickness of the capacitive layered sensor structure. The directions Sx and Sy are perpendicular to each other and to Sz.
The capacitive layered sensor structure 110 comprises an electrode layer 300 comprising a first electrode 301 and a first wire 351 attached to the first electrode 301. To measure touch or pressure, a capacitance of the first electrode 301 is measured. The capacitance of the first electrode 301 is measured relative to surroundings, e.g. relative to at least the electrically permeable and conductive layer 410. The first wire 351 connects the first electrode to measurement electronics, e.g. the electronic arrangement 120. The electrode layer 300 may comprise a substrate 390 (see
As indicated in
Referring to
As for the direction Sz of thickness of the capacitive layered sensor structure 110, Sz may refer to the direction of the thickness of a planar capacitive layered sensor structure 110. The structure 110 is preferably deformable. Therefore, in use, the structure 110 needs not to be planar. However, a non-planar sensor structure 110 may be deformable to a planar shape. In non-planar structures, the direction Sz of the thickness of the structure depends on the point of observation. Moreover, the term thickness of a planar structure refers to the smallest of three orthogonal dimensions of the planar structure.
A first purpose of the of the first insulating layer 210 is to electrically insulate the electrode(s) 301, 302, from the electrically permeable and conductive layer 410 in order to form a capacitance in between the first electrode 301 and the electrically permeable and conductive layer 410. The capacitive layered sensor structure 110 may comprise a second insulating layer 220 arranged such that the electrode layer 300 is arranged in between the first 210 and second 220 insulating layers in the direction of thickness of the thickness of the capacitive layered sensor structure 110. A second purpose of the first insulating layer 210 and the second insulating layer 220, in combination, is to act as a compressible layer (or as compressible layers), i.e. a layer that is deformed by application of pressure. As known e.g. from the prior art publication mentioned above, a change in a distance between the first electrode 301 and another conductor (e.g. layer 400, 410) changes the capacitance therein between. For this functionality, it suffices that only one of the insulating layers 210, 220 is compressible, however, both of them may be compressible. Therefore, at least one of the first insulating layer 210 and the second insulating layer 220 is compressible. In other words, at least one of the first insulating layer 210 and the second insulating layer 220 is configured to be compressed and deform by such a pressure that is to be detected by the capacitive layered sensor structure 110.
Thus, in an embodiment, the first insulating layer 210 is compressible. In another or the same embodiment, the capacitive layered sensor structure 110 comprises a second insulating layer 220 that is compressible. In an embodiment, the first insulating layer 210 is not compressible and the capacitive layered sensor structure 110 comprises a second insulating layer 220 that is compressible. In an embodiment, the first insulating layer 210 is compressible and the capacitive layered sensor structure 110 comprises a second insulating layer 220 that is compressible. In an embodiment, the first insulating layer 210 is compressible and the capacitive layered sensor structure 110 comprises a second insulating layer 220 that is not compressible. In an embodiment, sensor structure 110 does not comprise the second insulating layer 220. The insulating layer that is compressible (i.e. deformable) is arranged in between the electrode layer 300 and an electrically conductive layer. The first insulating layer is arranged in between the layers 300 and 410; while the second insulating layer 220, if present, is arranged in between the layers 300 and 400. In this way, the capacitive layered sensor structure comprises an insulating layer (210, 220) that is compressible and that is arranged in between the electrode layer 300 and an electrically conductive layer (410, 400). The electrically conductive layer (410, 400) may be electrically permeable.
As indicated in
If the electrode layer 300 comprises the substrate 390, as in
The capacitive layered sensor structure 110 comprises the electrically permeable and conductive layer 410. A first purpose of the electrically permeable and conductive layer 410, related in particular to the permeability of the layer 410, is to pass an electric field through the electrically permeable and conductive layer 410 so as to capacitively sense, by the first electrode 301, which is located on a first side of the electrically permeable and conductive layer 410, an object (e.g. 600, see
The electrically permeable and conductive layer 410 increases the capacitance of the first electrode when compared to situation without layer 410, but reduces the amount of capacitance that a touch event is able to generate. Thus, a purpose of the electrically permeable and conductive layer 410 is to reduce the capacitance that a touch event generates. It has been noticed that without the electrically permeable and conductive layer 410 a touch could generate so large signals that it could be even greater than a small pressure. It is noted that a touch or a pressure may be made e.g. by a finger or by a whole hand, which affects the magnitude of the observed signal. Thus, a pressure with a finger could, without the electrically permeable and conductive layer 410, imply a similar signal as a touch with a whole hand. Therefore, for purposes of reliably distinguishing touch and pressure, the electrically permeable and conductive layer 410 is applied. In particular, the electrically conductive property of the layer 410 reduces the signal level generated by a touch.
Moreover, it has been noticed that without the electrical permeability of the layer 410, a layer that is only conducting would reduce the effect of touch to such a degree that touch could not be measured at all. The electrical permeability of the layer 410 has the effect that the touch affects the capacitance of the first electrode 301. Thus, even if a purpose of the electrically permeable and conductive layer is to reduce the capacitance of the first electrode when touch is to be measured, a purpose is to reduce the capacitance only to a measurable level; i.e. not completely remove an effect of the touch on the capacitance.
With reference to
Referring to
Referring to
A purpose of the of the second insulating layer 220 is to electrically insulate the electrode(s) 301, 302 from environment. Electrical contacts to the electrodes in use might cause malfunction of the sensor arrangement 100. Moreover, when the sensor structure comprises a first electrically conductive layer 400, a purpose of the second insulating layer 220 is to insulate the electrode(s) 301, 302 from the first electrically conductive layer 400 and in this way form a capacitance in between the first electrode 301 and the first electrically conductive layer 400. As indicated above, a purpose of the second insulating layer 220 may be to act as the compressible layer or as one of the compressible layers, i.e. as a layer that is deformed by application of pressure.
As indicated by these functions, even if not shown in the Figures, the capacitive layered sensor structure 110 may comprise the second insulating layer 220 even if it does not comprise the first electrically conductive layer 400.
With reference to
The different layers may be attached to each other with adhesive as known per se. However, for clarity, adhesive is not shown in the figures.
In an embodiment, the electronic arrangement 120 is electrically coupled to the electrically permeable and conductive layer 410 in order to measure the capacitance of the first electrode 301 relative to the electrically permeable and conductive layer 410. A common potential, e.g. a ground potential, may be conducted to the electrically permeable and conductive layer 410 at least when measuring the capacitance of the first electrode 301 relative to the layer 410. However, the electronic arrangement 120 need not be electrically coupled to the electrically permeable and conductive layer 410. A capacitance that depends on the degree of deformation can be formed in between the first electrode 301 and a first electrically conductive layer 400 (see
Referring to
However, since the touch of an object 600, or such a small pressure that is classifiable as being indicative of touch (the pressure being generated by the object 600), or the presence of the object 600 in a vicinity of the first electrode 301 is being measured, the signal is somewhat higher than in case nothing is measurable. In particular, in such a case, the value v1 measurable from the first electrode 301 is at least equal to a second threshold tht. The second threshold can be called a touch limit. Such a touch limit (i.e. second threshold) can be used to differentiate a touch (as in
Determining that the first value v1 is at most equal to the first threshold thf and at least equal to the second threshold tht may take place at the first time t1, or the determining may take place later, e.g. in a computer. However, preferably, determining that the first value v1 is at most equal to the first threshold thf and at least equal to the second threshold tht takes place when the object 600 is touching (or mildly pressing) the capacitive layered sensor structure 110, as indicated in
Referring to
Determining that the second value v2 is more than the first threshold thf may take place at the second time t2, or the determining may take place later, e.g. in a computer. However, preferably, determining that the second value v2 is more than the first threshold thf takes place when the object 600 is pressing the capacitive layered sensor structure 110, as indicated in
As detailed above, in this way, the first value v1 is indicative of a touch (or being in a vicinity or a mild pressure) and the second value v2 is indicative of application of a pressure (e.g. a higher pressure).
Referring to
The first electrically conductive layer 400 need not be electrically permeable, since, in general, touch is not sensed from the other side of the capacitive layered sensor structure 110. However, at least in case touch should be sensed from both sides, also the first electrically conductive layer 400 could be made also electrically permeable. For manufacturing reasons, the first electrically conductive layer 400 may be electrically permeable even if touch is not sensed from the other side.
When the capacitive layered sensor structure 110 comprises the first electrically conductive layer 400, the electronic arrangement 120 may be electrically coupled also to the first electrically conductive layer 400 in order to measure the capacitance of the first electrode 301 relative to the first electrically conductive layer 400. As indicated above, measuring in particular the capacitance of the first electrode 301 relative to the first electrically conductive layer 400 improves accuracy. A common potential, e.g. a ground potential, may be conducted to both the first electrically conductive layer 400 and the electrically permeable and conductive layer 410 at least when measuring the capacitance of the first electrode 301 relative to the layers 400, 410.
When the capacitive layered sensor structure 110 comprises the first electrically conductive layer 400, the electronic arrangement 120 may be, but need not be, coupled to the first electrically conductive layer 400. If coupled, a capacitance in between the first electrode 301 and the first electrically conductive layer 400 may be formed. If not coupled, the first electrically conductive layer 400 acts mainly as a shield to reduce disturbances.
The electronic arrangement 120 may be configured to send an output signal Sout. In an embodiment, the output signal Sout is indicative of
In many applications it may be feasible to have information also about the value of the pressure applied on the first electrode 301, not just the information that a pressure is applied. Therefore, in an embodiment, the output signal Sout is indicative of the second value v2.
As indicated above, both touch and pressure can be measured using the capacitive layered sensor structure 110. When using the capacitive layered sensor structure 110 as indicated above, a method for detecting a touch and a pressure is performed. Such a method for detecting a touch and a pressure comprises arranging available a capacitive layered sensor structure 110, of which details have been given above.
Referring to
The size of the object 600, as detailed above, corresponds a reasonable small finger of a typical user. The dielectric constant of the object 600, as detailed above, corresponds to a dielectric constant of a part of a finger.
An embodiment of the method comprises, at least at a second time t2, when the object 600 is compressing at least one of the first and the second insulating layers 210, 220 at such a location that overlaps with the first electrode 301 such that a part of the first insulating layer 210 and a part of the electrically permeable and conductive layer 410 are arranged in between the object 600 and the first electrode 301, measuring a second value v2 indicative of a second capacitance of the first electrode 301 using the capacitive layered sensor structure 110. This may take place at the second time t2, as discussed above and indicated in
This method may be reflected in the arrangement 100. In particular, in an embodiment of the arrangement 100, the capacitive layered sensor structure 110 and the electronic arrangement 120 are, in combination, configured in the following way:
[A] when the object 600 having a volume of at least 1 (cm)3 and a dielectric constant of at least 10 is arranged in the vicinity of the first electrode 301 or touches a surface of the capacitive layered structure at such a location L that overlaps with the first electrode 301 such that a part of the first insulating layer 210 and a part of the electrically permeable and conductive layer 410 are arranged in between the object 600 and the first electrode 301, and when neither the first insulating layer 210 nor the second insulating layer 220 is compressed, a value v1 indicative of the capacitance of the first electrode 301 as measurable from the capacitive layered sensor structure 110 by the electronic arrangement 120 is at most equal to the first threshold thf and at least equal to the second threshold the. Moreover,
[B] at least one of the first insulating layer 210 and the second insulating layer 220 is compressible by the object 600 at such a location L that overlaps with the first electrode such that a part of the first insulating layer 210 and a part of the electrically permeable and conductive layer 410 are arranged in between the object 600 and the first electrode 301, in such a way that a value indicative of the capacitance of the first electrode 301 as measurable from the capacitive layered sensor structure 110 by the electronic arrangement 120 is more than the first threshold thf. It is also noted that all compressions need not imply a capacitance above the first threshold thf, but at least a sufficiently high compression will result in the exceeding the first threshold.
As for the features [A] and [B] hereinabove, naturally the arrangement 100 does not necessarily measure these values at all times. But if, in the situations indicated in
Examples of properties of the object 600 will be discussed below in connection with the method.
As indicated above and in
Referring to
The meaning of the term “vicinity” may dependent on the user needs. In some measurements, a lack of touching may be determined, e.g. when the object 600 is not closer than 15 cm to the electrode. However, in some other measurements, a lack of touching may be determined, e.g. when the object 600 is not closer than 1 cm to the electrode. The value of the second threshold tht can be used to define the meaning vicinity. It may be useful to define the second threshold tht in such a way that when a distance between the object 600 and the first electrode 301 is less than a one dimensional size (e.g. length or width) of the first electrode 301, the resulting signal is at least equal to the second threshold tht. As an example, an object 600 having a volume of at least 1 (cm)3 and a dielectric constant of at least 10 is not arranged in the vicinity of the first electrode 301, when no object 600 having a volume of at least 1 (cm)3 and a dielectric constant of at least 10 is arranged closer than 3 cm (or another distance discussed above, depending on needs) to such a part of a surface of the capacitive layered sensor structure 110 that overlaps with the first electrode 301. Herein the term overlaps means that the overlapping parts are on top of each other in the direction Sz of thickness of the capacitive layered sensor structure 110. Correspondingly, for purposes of definition, an object 600 having a volume of at least 1 (cm)3 and a dielectric constant of at least 10 is arranged in the vicinity of the first electrode 301, when the object 600 having a volume of at least 1 (cm)3 and a dielectric constant of at least 10 is arranged at most 3 cm (or another distance discussed above, depending on needs) apart from such a part of a surface of the capacitive layered sensor structure 110 that overlaps with the first electrode 301 (i.e. the location L).
In an embodiment of the arrangement 100, the capacitive layered sensor structure 110 and the electronic arrangement 120 are, in combination, configured to function in a corresponding manner. This manner is discussed above for the object 600, and the aforementioned properties of the object 600 are also applicable for the capacitive sensor structure discussed above.
Correspondingly, in an embodiment of the arrangement 100, the electronic arrangement 120 is configured, at a third instance of time t3, measure a third value v3 indicative of a third capacitance of the first electrode 301 and determine that the third value t3 is below the second threshold tht. In this way, the third value v3 is indicative of absence of both a touch and an application of a pressure. This has been indicated in
To summarize the situations indicated in
To further clarify the situation (i) the following is noted:
In this way, the actual value of the first threshold thf may be used to fine-tune the pressure required for determining a pressure. Referring to the point (i,b), depending on the application, it may be feasible that very small (but greater than zero) pressures are indicative of a touch rather than a pressure. However, referring to the point (i,a) above, at least one of the first and second insulating layers 210, 220 is compressible by the object 600 such that a value indicative of the capacitance of the first electrode 301 as measurable from the capacitive layered sensor structure 110 by the electronic arrangement 120 is more than the first threshold thf, when the structure 110 is compressed as indicated in the point (i,a). A reasonable large force may be needed for the value indicative of the capacitance of the first electrode 301 to exceed the first threshold thf, depending on the value of the first threshold thf.
To further clarify the situation (ii) the following is noted:
In this way, the actual value of the second threshold tht may be used to fine-tune the distance required for determining a touch. Depending on the application, it may be feasible that only very small distances are considered to indicate a touch; however, in some applications presence in a vicinity may be indicative of a touch.
With reference to
Thus, to further clarify the situation (ii,b) as indicated above, the following is noted:
Thus, in an embodiment, the electronic arrangement 120 is configured to
A corresponding embodiment of the method comprises
Moreover, in an embodiment, the electronic arrangement 120 is configured to
A corresponding embodiment of the method comprises
As indicated above, the touch can be differentiated from the pressing by using a proper electrically permeable and conductive layer 410 and a proper first threshold thf, in combination. If the first threshold thf, is too large, information related to the value of the pressure is lost. If the first threshold thf, is too small, some touch, even without pressing, could be incorrectly determined as being indicative of application of pressure. These situations, however, depend on the use of the sensor arrangement 100. A typical use is a user interface, wherein a user operates the sensor arrangement 100 by a finger or fingers. Pressure in such an application is typically some hundreds of grams per square centimetre (i.e. some tens of kPa). Therefore, in an embodiment, the sensor arrangement 100 is configured such that when at least one of the first insulating layer 210 and the second insulating layer 220 is compressed by the object 600 having a volume of at least 1 (cm)3 and a dielectric constant of at least 10 with a pressure of 10 kPa such that a part of the first insulating layer 210 and a part of the electrically permeable and conductive layer 410 are arranged in between the object 600 and the first electrode 301, a value indicative of the capacitance of the first electrode 301 as measurable from the capacitive layered sensor structure 110 by the electronic arrangement 120 equals a use value vuse. Such a use value vuse is shown in
As indicated above, the touch can be differentiated from absence of touching and pressing by using a proper electrically permeable and conductive layer 410 and a proper second threshold tht, in combination. If the second threshold tht, is too large, some forms of touch are not necessarily identified. If the first threshold thf, is too small, incorrect indication of a touch may be measured.
Moreover, typically measurements include noise. Therefore, even if there is not object 600 in a vicinity of the first sensor 301, a signal measured therefrom is not constant. Thus, the signal measured in such case has a mean (<v>) and a deviation (stdv). In order not to make meaningless measurement, a detection limit thdet may be set to a value equalling the mean added by the deviation (i.e. thdet=<v>+stdv); or to a value equalling the mean added by two deviations (i.e. thdet=<v>+2×stdv). Thus, all signal values that are less than the detection limit thdet may be considered meaningless.
For completeness it is noted that the detection limit thdet is at most equal to the second threshold tht. Typically, the detection limit thdet is less than the second threshold tht. Moreover, the second threshold tht is less than the first threshold thf. Preferably, the second threshold tht is less than the first threshold thf by at least a noise level stdv (discussed in detail above), i.e. preferably, tht<thf−stdv. The first threshold thf may correspond to a situation wherein the compressible layer (210 and/or 220) is compressed to some extent. The first threshold thf may correspond to a situation wherein the compressible layer (210 and/or 220) is fully compressed. In such an extreme case, the presence of pressure would be determined only when the compressible layer (210 and/or 220) is fully compressed, and other compressions would be classified as a touch.
As for the second threshold (i.e. touch limit) tht, what has been said above about the object 600 and the vicinity, applies.
Referring to
As indicated in
However, comparison of these values with the first threshold thf need not be done in microchip 510, but may be done in the computer 520. In addition, comparison of these signal value with the second threshold tht needs not be done in microchip 510, but may be done in the computer 520. However, preferably, comparison of signal value with the second threshold tht is done in the microchip 510. Thus, the signal Sin needs not be indicative of absence of the object 600. Even if the microchip 510 may be configured to determine also a third value v3 that is below the second threshold v2 (see
As indicated in
When the second part of the electronic arrangement 120 (e.g. computer 520) is used, a computer program may run on the computer 520. Such a computer program, when run on the computer 520, is configured to cause the computer 520 to (a) receive information indicative of the first threshold thf, (b) receive information indicative of the second threshold tht, and (c) receive a signal Sin indicative of a first capacitance and a second capacitance measured by a capacitive layered sensor structure 110. The computer program, when run on the computer 520, is further configured: to cause the computer 520 to determine, from the signal Sin a first value v1 indicative of the first capacitance and a second value v2 indicative of the second capacitance; to determine that the first value v1 is at most equal to the first threshold thf; to determine that the second value v2 is more than the first threshold thf; and to generate an output signal Sout. The output signal Sout is indicative of the first value v1 being at most equal to the first threshold thf and the second value v2 being more than the first threshold thf. Unlike the arrangement 100, the computer 520 needs not to determine that the first value v1 is at least equal to the second threshold tht, since such information may have been taken account in the signal Sin as detailed above. Naturally, the computer program, when run on the computer 520, may be further configured: to cause the computer 520 to determine that the first value v1 is at least equal to the second threshold tht. Moreover, the output signal Sout may be indicative of the first value v1 being at least equal to the second threshold tht.
As indicated above, the electronic arrangement 120 may be configured to send an output signal Sout, wherein the output signal Sout is also indicative of the second value v2. Correspondingly, the computer program, when run on the computer 520, may be configured to cause the computer 520 to generate such an output signal Sout, that the output signal Sout is also indicative of the second value v2.
As indicated above, the electronic arrangement 120, e.g. the microchip 510 and the computer 520 in combination, may be configured to determine also a third value v3 that is below the second threshold tht (see
Thus, an embodiment of a computer program, when run on the computer 520, is configured to cause the computer 520 to receive such a signal Sin that is indicative of also a third capacitance measured by a capacitive layered sensor structure 110. Moreover, the embodiment of the computer program, when run on the computer 520, is configured to cause the computer 520 to determine from the signal Sin a third value v3 indicative of the third capacitance; to determine that the first value v1 is at least equal to the second threshold tht; and to determine that the third value v3 is less than the second threshold tht.
An embodiment of a computer program, when run on the computer 520, is configured to cause the computer 520 to receive information indicative of the detection limit thdet and to receive such a signal Sin that is indicative of also a fourth capacitance measured by a capacitive layered sensor structure 110. Moreover, the embodiment of the computer program, when run on the computer 520, is configured to cause the computer 520 to determine from the signal Sin a fourth value v4 indicative of the fourth capacitance; and to determine that the fourth value v4 is less than the second threshold tht and more than the detection limit thdet.
An embodiment of a computer program, when run on the computer 520, is configured to cause the computer 520 to receive information indicative of the detection limit thdet and to receive such a signal Sin that is indicative of also a fifth capacitance measured by a capacitive layered sensor structure 110. Moreover, the embodiment of the computer program, when run on the computer 520, is configured to cause the computer 520 to determine from the signal Sin a fifth value v5 indicative of the fifth capacitance; and to determine that the fifth value v5 is less than the detection limit thdet.
Referring to
Referring to
Preferably the non-conductive area 414 (or at least one such area 414) is arranged at a location of the first electrode 301. More specifically, preferably, the non-conductive area 414 or at least one of the non-conductive areas 414 overlaps with the first electrode 301. As above, also herein the term overlaps means that the overlapping parts are on top of each other in the direction Sz of thickness of the capacitive layered sensor structure 110.
Referring to
Referring to
The non-conductive area 414 may be observable on a small scale. For example, a poorly conducting polymer-based layer may serve as the electrically permeable and conductive layer 410. Such poorly conductive polymer-based material typically comprises a reasonably small amount of conductive particles. Such conductive particles may be particles of some metal (e.g. copper, aluminium, silver, gold) or carbon (including, but not limited to graphene and carbon nanotubes). In addition, some (reasonably) conductive polymers, such as polyaniline, a polyvinyl, and PEDOT:PSS may comprise non-conductive areas in the microscale, depending on their conductivity. The amount of the particles correlates with conductivity, and the conductivity is low, yet the material is conductive, when the amount of particles is just above a percolation threshold of the particles. Thus, on a large scale, the material seems conductive, since the particles percolate, while, on a smaller scale, the material comprises areas that are not electrically connected to each other, i.e. they are non-conductive. In the non-conductive areas, the particles do not percolate. However, the conductive particles that are not connected to other conductive particles may hinder the electrical permeability of electrically permeable and conductive layer 410. Therefore, in a preferable embodiment, the non-conductive area 414 does not comprise particles of electrically conductive material.
Preferably, a ratio of the area of the non-conductive areas 414 and the area of the conductive area 412 is at a proper range. Thus, in an embodiment, the electrically permeable and conductive layer 410 comprises a conductive area 412 and at least one non-conductive area 414. The conductive area 412 may limit the non-conductive area(s) 414 as indicated above. This proper range applies preferably at least nearby the first electrode 301.
To this end, in an embodiment, at least a part of the conductive area 412 overlaps with the first electrode 301. The cross sectional area of the part of the conductive area 412 that overlaps with the first electrode 301 is denoted by A412,301 in
Preferably, the cross sectional area A412,301 of the part of the conductive area 412 that overlaps with the first electrode 301 is at least 5% of a cross sectional area A301 of the first electrode. As for a maximum of the cross sectional area A412,301 of the part of the conductive area 412 that overlaps with the first electrode 301, the conductive area 412 may comprise only one aperture having a size within the aforementioned limits, while the first electrode 301 may be reasonably large. Thus, the upper limit may be at least approximately 100%. Correspondingly, preferably, the cross sectional area A414,301 of the part of the non-conductive area(s) 414 that overlap(s) with the first electrode 301 may be at most 95% of the cross sectional area A301 of the first electrode 301.
Moreover, preferably, the cross sectional area A414,301 of the part(s) of the non-conductive area(s) 414 that overlap(s) with the first electrode 301 is from 0.01 (mm)2 to 100 (mm)2. This size range has been found suitable in particular for application where a touch and pressure of a finger of a user is determined.
If a second electrode 302 is present, preferably the conductive area(s) 412 and the non-conductive area(s) overlap with the second electrode 302 in a similar manner mutatis mutandis.
In an embodiment, the electrically permeable and conductive layer 410 comprises a conductive area 412 that limits (e.g. surrounds) a non-conductive area 414 or non-conductive areas 414, such that a cross sectional area A412 of the conductive area is from 5% to 95% of a total cross sectional area A412+A414 of the conductive area 412 and the non-conductive area 414 or non-conductive areas 414 limited by (e.g. surrounded by) by the conductive area 412
Preferably, the electrode layer 300 comprises a second electrode 302 and a second wire 352 attached to the second electrode 302. This has the effect that the spatial accuracy of the capacitive measurements is improved. Preferably, the first wire 351 connects only the first electrode 301 to the electronic arrangement 120 and the second wire 352 connects only the second electrode 302 to the electronic arrangement 120. This has the effect that the capacitances of the first and second electrode 301, 302 can be measured without multiplexing, which improves the temporal accuracy of the measurements. Thus, in an embodiment, the electronic arrangement 120 is configured to measure a capacitance of the whole area of the first electrode 301 at one instance of time. Correspondingly, in an embodiment, the electronic arrangement 120 is not configured to measure capacitances of parts of the first electrode 301 at subsequent instances of time. However, the measurement principle can be applied also with such a layered sensor structure 110 wherein the first wire 351 connects both the first electrode 301 and the second electrode 302 to the electronic arrangement 120, and the electrode (301, 302), of which capacitance is measured, is determined by multiplexing. For using multiplexing, at least one of the electrically permeable and conductive layer 410 and the first electrically conductive layer 400 may be divided to areas that at least partly overlap with at least one of the first electrode 301 and the second electrode 302, wherein the areas of the layer (410, 400) are not electrically connected to each other.
As indicated above and in
Referring to
As for the materials and thicknesses of the layers of the layered capacitive sensor structure 110, the materials and thicknesses are preferably selected such that the layered capacitive sensor structure 110 is flexible. More preferably, the materials and thicknesses are preferably selected such that the layered capacitive sensor structure 110 is flexible and stretchable, i.e. conformable.
As for the term flexible, a planar flexible material can be bent to a radius of curvature of 10 mm (or less) without breaking the material at a temperature of 20° C. Moreover, the flexible material can be thereafter turned back to the planar form at a temperature of 20° C. without breaking the material. As for the term stretchable, a stretchable material can be stretched by at least 10% in a reversible manner. In particular, a layer of stretchable material can be stretched by at least 10% in a reversible manner a direction that is perpendicular to the direction of thickness of the layer. The reversibility of the stretching is spontaneous, i.e. elastic. Thus, a planar conformable material is flexible as indicated above and stretchable in a direction of the plane of the planar conformable material. A planar conformable material can be arranged to conform a surface of a hemisphere having a radius of 10 cm (or less) at a temperature of 20° C. without introducing significant plastic (i.e. irreversible) deformations to the material.
As for the term compressible, a compressible material can be compressed by at least 10% in a reversible manner. In particular, a layer of compressible material can be compressed by at least 10% in a reversible manner in the direction of the thickness of the layer. The reversibility of the compression is spontaneous, i.e. elastic. Moreover, a Young's modulus of a compressible layer may be less than 1 GPa.
As for suitable materials for the insulating layers 210, 220, 230, a purpose of the insulating layers is to electrically insulate. Therefore, a resistivity of a material of the first insulating layer 210 and a material of the second insulating layer 220 (if present) may be at least 10 Ωm at a temperature of 23° C. This applies also for the third insulating layer 230, if present. Typically, a resistivity of a material of the first insulating layer 210 and a material of the second insulating layer 220 (if present) is at least 100 Ωm at a temperature of 23° C.
As indicated above, at least one of the first and second insulating layers 210, 220 is compressible; however the second insulating layer 220 needs not be present in the solution. The meaning of the term compressible has been discussed above. Suitable materials for a compressible layer include materials from a material group A, wherein the material group A consists of polyurethane (such as thermoplastic polyurethane), polyethylene, poly(ethylene-vinyl acetate), polyvinyl chloride, polyborodimethylsiloxane, polystyrene, acrylonitrile-butadiene-styrene, styrene-butadienestyrene, styrene-ethylene-butylene-styrene ethylene propylene rubber, neoprene, cork, latex, natural rubber, siloxane polymer (such as silicone), and thermoplastic elastomeric gel. Moreover, in order to have reasonable deformations, in an embodiment, a thickness of the compressible layer (210 and/or 220) is at least 0.05 mm, preferably at least 0.3 mm such as at least 0.5 mm. A thickness of the compressible layer (210 and/or 220) is preferably from 0.05 mm to 5 mm, such as from 0.3 mm to 4 mm, such as from 0.5 mm to 2 mm. A Young's modulus in compression of the compressible layer is preferably from 0.01 MPa to 15 MPa, such as from 0.1 MPa to 5 MPa. A Young's modulus in tension may differ from the Young's modulus in compression. Moreover, a material of the compressible layer has preferably a yield strain of at least 10 percent. This ensures that the material can be sufficiently compressed in use.
Preferably, the first insulating layer 210 and the second insulating layer 220 are flexible in the aforementioned sense. Moreover, preferably, a Young's modulus of the first insulating layer 210 is at most 10 GPa, such as at most 5.0 GPa. Moreover, preferably, a Young's modulus of the second insulating layer 220 is at most 10 GPa, such as at most 5.0 GPa.
The first insulating layer 210 or the second insulating layer 220 may act only as a flexible insulator. Suitable materials for a flexible layer include materials from a material group B, wherein the material group B consists of textile, polyimide, polyethylene naphthalate, polyethylene terephthalate, and polyetheretherketone. Suitable materials for a flexible layer also include materials from the material group A as defined above. However, the first insulating layer 210 or the second insulating layer 220 need not be flexible. In such a case, suitable materials further include epoxy and phenolic resin. Examples include FR-4 glass epoxy and cotton paper impregnated with phenolic resin. In particular, the second insulating layer 220 may be hard and/or stiff, if the sensor structure 110 needs not be flexible, and the first insulating layer 210 is compressible.
A thickness of an insulating layer that does not act as a compressible layer may be e.g. up to 5 mm as indicated above. However a thickness of an insulating layer that does not act as a compressible layer may be e.g. less 1 mm, such as less than 0.5 mm, e.g. from 50 μm to 1 mm or from 50 μm to 0.5 mm.
What has been said about the thicknesses and materials of the first and second insulating layers 210, 220 applies to the third insulating layer 230. The third insulating layer 230 needs not to be compressible.
As for the electrode layer 300, preferably also the electrode layer is flexible. More preferably flexible and stretchable.
In an embodiment, the first electrode 301 is made of such material that is stretchable by at least 5% without breaking. Preferably the second electrode 302 is made of such material that is stretchable by at least 5% without breaking. Such material may be e.g. ink or paste. In an embodiment, the first electrode 301 comprises electrically conductive particles, such as flakes or nanoparticles, attached to each other in an electrically conductive manner. In an embodiment, the first electrode 301 comprises electrically conductive particles comprising at least one of carbon (including, but not limited to graphene and carbon nanotubes), copper, silver, and gold. In an embodiment, the first electrode 301 comprises electrically conductive particles comprising carbon. In an embodiment, the first electrode 301 comprises electrically conductive polymer-based material, such as at least one of polyaniline, a polyvinyl (e.g. polyvinyl alcohol or polyvinyl chloride), and PEDOT:PSS (i.e. poly(3,4-ethylenedioxythiophene) polystyrene sulfonate). What has been said about the material of the first electrode 301 applies, in an embodiment, to all electrodes including the second electrode 302. What has been said about the material of the first electrode 301 applies, in an embodiment, to the first wire 351. What has been said about the material of the first electrode 301 applies, in an embodiment, to the second wire 352.
The first electrode 301 may be arranged (e.g. by printing) onto the first insulating layer 210 or the second insulating layer 220 (see
As for the first electrically conductive layer 400, the first electrically conductive layer 400 may be uniformly conductive, e.g. made using conductive ink or paste a uniform amount on a uniform surface. In the alternative, the first electrically conductive layer 400 may be a mesh of conductive yarns, e.g. made using conductive ink or paste or filaments. It may also suffice that the first electrically conductive layer 400 consists of a meandering electrically conductive line. It may also suffice that the first electrically conductive layer 400 comprises multiple separate electrically conductive lines. In an embodiment, at least a part of the first electrically conductive layer 400 is made from a conductive ink. In an embodiment the first electrically conductive layer 400 comprises electrically conductive fabric. In an embodiment, the first electrically conductive layer 400 comprises electrically conductive polymer. Preferably, the first electrically conductive layer 400 is uniformly conductive. As for the term conductive, a conductive material has a resistivity of at most 10 Ωm, measured at a temperature of 23° C. and at an internal elastic strain of 0%; i.e. without compression or tension, i.e. at rest.
What has been said about the material of the first electrically conductive layer 400 applies to the material of the electrically permeable and conductive layer 410, in particular to the conductive area(s) 412 of the electrically permeable and conductive layer 410. However, at least one non-conductive area 414 may be arranged to the electrically permeable and conductive layer 410 at least if its conductivity would be otherwise uniform and high.
Moreover, preferably [A] a conductivity of the first electrically conductive layer 400 at a temperature of 23° C. (at rest) is higher than a conductivity of the electrically permeable and conductive layer 410 at a temperature of 23° C. (at rest) and/or [B] a greater portion of the first electrically conductive layer 400 than of the electrically permeable and conductive layer 410 is covered by conductive material.
Number | Date | Country | Kind |
---|---|---|---|
19397514.1 | May 2019 | EP | regional |
Filing Document | Filing Date | Country | Kind |
---|---|---|---|
PCT/FI2020/050299 | 5/5/2020 | WO |