Upload
others
View
9
Download
0
Embed Size (px)
Citation preview
Metal mesh projected-capacitive
touch screens simulated with
Fieldscale SENSETM
March, 2017
IntroP-cap touch sensors
Extensively used today in portable devices, such assmartphones and tablets.
Growing usage in industrial and automotiveapplications.
Mutual-capacitive touch sensors (instead of self-capacitive ones) are more often used as they offer:
real-time detection of multiple touches
higher resolution
less vulnerability to electromagnetic interference.Source: Atmel® www.electronicproducts.com/Analog_Mixed_Signal_ICs/Communications_Interface/Touchscreens_large_and_small.aspx
Source: Atmel®www.atmel.com/Microsite/maxtouch-t-series-automotive/default.aspx
IntroITO
Sensor electrodes are typically made ofIndium Tin Oxide (ITO) due to its highoptical transmittance.
ITO disadvantages:
o limited supply
o considerable cost
o inflexibility, not suitable for wearables and flexible touch screens
ohigh resistivity, slow response, not suitable for large touch screens
Model of a 3x3 diamond double layer electrode array.
IntroMetal mesh
“ITO-alternatives”:
carbon nanotubes
conductive polymers
graphene
silver nanowires
metal meshPros:o flexibilityo low resistivityo high optical transmissiono low cost
Source: Bison Optronics Co., Ltd. http://www.bisonoptronics.com
Simulated modelDescription of the simulated p-cap sensor
Stack Up Configuration
Material Thickness Dielectric constant (εr)
Cover glass variable (0.5/1/2 mm) variable (6/8/10)
OCA 100 um 4.0
Transmitters 200 nm -
PET film 50 um 3.2
OCA 50 um 3.5
Receivers 200 nm -
PET film 50 um 3.2
OCA 100 um 4.0
Polarizer 200 um 5.0
Display (or shielding) 600 um -Source: Synaptics (SID, Boston 2012, Session M-3).
Simulated modelMetal mesh geometry
3x3 electrode array
Total array dim. 15mm x 15 mm
Margins around the electrodes: s = 500 um
s gg
Electrode width: w = 5 mm
Gap between electrodes: g = 100 um
Transmitters (Tx) Receivers (Rx)
ww
Simulated modelMetal mesh geometry
Line pitch (Lp): o 250 um
o 500 um
o 1000 um
Line width (Lw): o 1 um
o 3 um
o 5 um
Simulated modelFinger size & position
Cylinder with hemispherical tip:
o Length: 10 mm
o Diameter: 8 mm
Finger placed just above the centralelectrode node, touching the coverglass
Simulation Settings in Fieldscale SENSESimulation inputs & computed quantities
Capacitance Computation
BEM Method
Display and finger wereassumed as grounded, perfectconductors
The mutual capacitance, Cm,between the central Tx & Rxwas obtained, as affected bythe finger presence
Effect of finger on the mutual capacitance between electrodes
Transmitter
Receiver
Display (or shielding)
Cm
CRx-display
Finger
Electric field lines
Finger
Simulation Settings in Fieldscale SENSESimulation inputs & computed quantities
Resistance Computation
BEM Method
The edges of the metal meshelectrodes were defined as “ports”
Sheet resistance, RS, of metal mesh was set equal to 0.1 Ohm/sq
Figure: Cell resistance definition
Source Ports
Sink Ports
Receivers
Transmitters
RRx : unit cell resistance
Simulation Settings in Fieldscale SENSESimulation inputs & computed quantities
RC constantRx electrode - unit cell
𝝉𝑹𝒙 = 𝑹𝑹𝒙(𝑪𝒎𝟎 + 𝑪𝑹𝒙−𝑫𝒊𝒔𝒑𝒍𝒂𝒚)
RRx: resistance of Rx- electrode per unit cell
Cm0: mutual capacitance between transmitter and receiver without finger presence
CRx-Display: capacitance between Rx electrode and display (or shielding)
Capacitances used for RC constant extraction
Transmitter
Receiver
Display (or shielding)
Cm0
CRx-Display
Simulation ResultsEffects of metal mesh geometry; cover glass thickness 1 mm, εr = 8
Line pitch (um) Line width (um) Cm0 (without finger, in pF)Cm (finger touching the cover glass,
in pF)
250 1 1.994 1.933
250 3 2.495 2.431
250 5 2.697 2.634
500 1 0.876 0.827
500 3 1.105 1.050
500 5 1.252 1.195
1000 1 0.307 0.278
1000 3 0.398 0.364
1000 5 0.459 0.421
Simulation ResultsEffects of metal mesh geometry; cover glass thickness 1 mm, εr = 8
Mutual capacitance, Cm0, between central Tx & Rx electrodes For a typical line pitch (250 ~ 500 um)
Cm0 is generally between 1 and 2.5 pF
Cm0 becomes higher as the metal meshdensity increases, that is:
o with decreasing line pitch
o with increasing line width (smaller effect)
Parallel planes:C = εA/d
A increases C increases0.0
0.5
1.0
1.5
2.0
2.5
3.0
0 1 2 3 4 5 6
Cm
0(w
ith
ou
t fi
nge
r, p
F)
Line width (um)
line pitch = 250 um
line pitch = 500 um
line pitch = 1000 um
Simulation ResultsEffects of metal mesh geometry; cover glass thickness 1 mm, εr = 8
Reduction of Cm due to finger presence
ΔCm is greater for a denser metal mesh:
o narrower line pitch
o wider line width (smaller effect)
For line pitch = 250 um the effect of linewidth on ΔCm becomes negligible.
0.00
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0 1 2 3 4 5 6
ΔC
m(p
F)
Line width (um)
line pitch = 250 um
line pitch = 500 um
line pitch = 1000 um
Simulation ResultsEffects of metal mesh geometry; cover glass thickness 1 mm, εr = 8
Percentage reduction of Cm (%) due to finger presence
In contrast with Cm0 and ΔCm, the ratioΔCm/Cm0 increases as the metal meshdensity decreases
o with increasing line pitch
o with decreasing line width (smallereffect)
Higher ΔCm/Cm0 results in highersensitivity of the touch sensor. This isnot always preferable, as it may lead tounintended touch detection in case ofhovering finger.
0
1
2
3
4
5
6
7
8
9
10
0 1 2 3 4 5 6
ΔC
m/C
m0
(%)
Line width (um)
line pitch = 250 um
line pitch = 500 um
line pitch = 1000 um
Simulation ResultsEffects of metal mesh geometry; cover glass thickness 1 mm, εr = 8
Cell Resistance
As expected, a denser metal mesh provides a lower Rcell
For a typical metal mesh geometry (line width=3um & line pitch=500 um):Rcell ≈ 20 Ω
0
20
40
60
80
100
120
140
0 1 2 3 4 5 6
Re
sist
ance
(Ω)
Line width (um)
line pitch = 250 um
line pitch = 500 um
line pitch = 1000 um
Simulation ResultsEffects of metal mesh geometry; cover glass thickness 1 mm, εr = 8
RC constantRx electrode - unit cell
A denser metal mesh results in a smaller RC constant, that is, faster response of the touch sensor.
However, there are optical limitations regarding metal mesh density.
0
50
100
150
200
250
300
350
0 1 2 3 4 5 6
RC
co
nst
ant
Line width (um)
line pitch = 250 um
line pitch = 500 um
line pitch = 1000 um
Simulation ResultsEffects of metal mesh geometry; cover glass thickness 1 mm, εr = 8
Optimum design of p-cap sensors requires:
o Cm0 in accordance with controller specifications: 1-2 pF
o ΔCm higher than controller sensitivity: 0.01 – 0.1 pF
o ΔCm/Cm0: usually 4 – 8 %
o Low RC constant
• Metal mesh line pitch:
o optimum value is 250 ~ 500 um
o for a coarser metal mesh ΔCm/Cm0 is higher, but Cm0, ΔCm are greatly reduced and also RC constant increases
• Metal mesh line width (1 ~ 3 um):
o does not significantly affect Cm
o reducing line width improves optical transmission, but increases RC constant, due to higher R
Simulation ResultsEffects of finger position – along z axis over the central node
0
1
2
3
4
5
6
0 1 2 3 4 5 6 7 8 9 10
ΔC
m/C
m0
(%)
finger distance (mm)
ΔCm/Cm0 is 4.9% when the fingertouches the screen surface, but isreduced to almost zero for fingerdistance ~ 5 mm.
This ensures that no false-positivetouch events by hovering fingers aredetected by the controller.
Simulated touch sensor model
Metal mesh line width 3 um
Metal mesh line pitch 500 um
Cover glass thickness 1 mm
Cover glass εr 8
Simulation ResultsEffects of finger location – along x-y level over the central node
When the finger moves from thecenter to the corner of the node,ΔCm/Cm0 becomes less than half(from 4.9% to 2.1%).
This variation enables the accuratedetection of the finger location duringeach charge cycle.
Conclusions
Simulations with Fieldscale SENSETM show that metal meshdensity (line width and line pitch dimensions) has a greatimpact on the performance of p-cap touch sensors.
As metal mesh becomes more coarse:
• optical transmission increases.
• ΔCm/Cm0 increases: the touch sensor becomes more sensitive.
• but RC constant increases: the touch sensor has slower response.
Contact:
Reply within 24 hours
+30 2310 947484
www.fieldscale.com
Run the Simulations You Want