DFSS Case Study Development Of A Robust Keypad
Project Summary
Client:
Large Telecommunication Equipment Manufacturer
Challenge: Design of a Robust Key Pad Goal: •Reduce generic contact problem: Target < 0,5% •Reduce total cost of the system: Target > 50% •Increase the click ratio (feeling): Target > 25%
Results: Following Results achieved due to DFSS implementation: •Project Savings > € 3,000,000 •Generic contact problem: Actual ~ 0 •Total cost reduction of the system: Actual > 75% •Increase the click ratio (feeling): Actual > 40% •Reduction in design Cycle Time: 1 months
IDOV Page 1
Identify •Project goals and objectives identified •All Critical To Satisfaction elements collected through Voice Of the Customer (VOC) process •Project team assembled and project plan created
IDOV Page 2
Identify •Detailed analysis of the existing Design and its failure modes •Analysis of Functional Requirements and Design Parameters
1. The key will be pushed down to the rocking point
2. Before clicking down the polydomefoil, the silicone key will be compressed
IDOV Page 3
Identify •Detailed analysis of the existing Design and its failure modes •Analysis of Functional Requirements and Design Parameters
3. After the abrupt click down of the polydome foil, the carbon coated inside of the polydome is contacting the carbon meander on the printed circuit board, genearating a short circuit
4. After this click feeling, usually customer will not push the key anymore, and if at this moment the carbon meander on the printed circuit board is not short circuited, then a contact failure will happen. If some additional force is applied, the carbon meander on the printed circuit will be short circuited, which is the unwanted situation Page 4
Identify Quantifying and measuring of the CTS after completed MSA (Measurement System Analysis):
force diagram
resistance diagram
X
shift “X” = bad function
F2
IDOV Page 5
Design Flow Down And Uncouple/Decouple Constraints Customer Domain
Deliverytime, Deliverytime,Cost Cost
Voice Of the Customer (VOC) (Customer Requirements (Y) Input
Constraints Physical Domain
Deliverytime, Deliverytime,Cost Cost
Generate Concepts
Generate Concepts
Evaluate Concepts
Evaluate Concepts
Input
Input
Disturbance
Disturbance
(Noise, Coupling)
Functional Requirements {FR}
Design Transfer Matrix
{FR} = [ D ]{DV }
Process Domain
Design Variables {DV}
(Noise, Coupling)
Process Transfer Matrix
Process Variables {PV}
{DV } = [ P]{PV }
IDOV Page 6
Design Flow Down And Uncouple/Decouple
{Y} = [I ]{FR}
{FR} = [D]{DV} {Y } = [ F ]{ X } {DV} = [P]{PV}
{Y} = [I] [P][D]{PV} 1424 3 [F ]
[I] = Unity Matrix [P] = Process Transfer Matrix [D] = Design Transfer Matrix [F] = System Transfer Matrix [FR] = Functional Requirements {DV } = Design Variables (Inputs, Control Parameters) {PV } = Process Variables (Inputs, Control Parameters) {Y} = Customer Requirement (CR)
IDOV Page 7
Design Flow Down And Uncouple/Decouple •Transferring Customer Attributes to Keypad Design Functions Y1 : To be able to dial and send SMS to different persons (Spoken, Performance) Y2 : To be able to notice the dialing (unspoken, must-be) Y3 : Quick access to telephone functions (unspoken, excitement) Y4 : Nice actuation of keys (unspoken, must-be) Y5 : To be able to transportation in trouser pocket/bag etc. (unspoken, must-be) Y6 : T be able to dial despite rainy days (unspoken, must-be) Y7 : Cool printing on keys (Spoken, Performance)
FD1 : Provide Dialing/Writing FD2 : Provide Dialing Signal FD3 : Provide Navigation FD4 : Prevent Key Stucking (%) FD5 : Provide Dust Protection (Norm) FD6 : Provide Spray Water Protection (Norm)
⎧Y 1 ⎫ ⎡1 ⎪Y 2 ⎪ ⎢0 ⎪ ⎪ ⎢ ⎪Y 3 ⎪ ⎢0 ⎪ ⎪ ⎢ ⎨Y 4 ⎬ = ⎢0 ⎪Y 5 ⎪ ⎢0 ⎪ ⎪ ⎢ ⎪Y 6 ⎪ ⎢0 ⎪Y 7 ⎪ ⎢ ⎩ ⎭ ⎣0
0 0 1 0 0 1 0 0 0 0 0 0 0 0
0 0 0 0 ⎤ ⎧ F D1 ⎫ 0 0 0 0 ⎥⎥ ⎪⎪ F D 2 ⎪⎪ 0 0 0 0 ⎥ ⎪ FD 3 ⎪ ⎪ ⎥⎪ 1 0 0 0 ⎥ ⎨ FD 4 ⎬ 0 1 0 0 ⎥ ⎪ FD 5 ⎪ ⎪ ⎥⎪ 0 0 1 0 ⎥ ⎪ FD 6 ⎪ ⎪ ⎪ 0 0 0 1 ⎥⎦ ⎩ F D 7 ⎭
FD7 : Provide 3D Surface Printing
IDOV Page 8
Design
Flow Down And Uncouple/Decouple •Transferring Keypad Design Functions to Design Variables FD1 : Provide Dialing
FD11 Transformation of a mechanical linear movement to displacement
FD111 F max
FD112 @ Displacement x max
FD12 Snapping (Click Feeling) (F min at x min)
X D1 : Actuating Keypad System
FD13 Contact on PCB (< 300 Ohm, Contact > 99 %)
FD121 F min
XD111-1 Silicon Hardness
FD131 < 300 Ohm
FD122 @ Displacement x min
X D11 Actuator
FD132 > 99 %
XD111-2 Web Design
XD112 Plunger Height
X D12 Polydome
XD121-1 Thickness
XD122 Height
XD121-2 Diameter
X D13 Carbon PCB / Polydome
XD131-1 Type
XD131-2 Pitch
XD132 Plunger Diameter
IDOV Page 9
Design
Flow Down And Uncouple/Decouple •Transferring Keypad Design Functions to Design Variables Hierarchy 1 FD2 : Dialing Signal
X D2 : Audio Signal (Original Design : Speaker)
FD3 : Navigation
X D3 : Navigation Key
FD4 : Key Stucking (%)
X D4 : Key Gap
FD5 : Dust Protection (Norm)
X D5 : Dust Sealing
FD6 : Spray Water Protection (Norm)
X D5 : Water Sealing
FD7 : 3D Surface Printing
X D7 : Tampon Print
⎧ F D1 ⎫ ⎡ A11 ⎪ FD 2 ⎪ ⎢ 0 ⎪ ⎢ ⎪ ⎪ FD 3 ⎪ ⎢ 0 ⎪ ⎢ ⎪ ⎨ FD 4 ⎬ = ⎢ 0 ⎪ FD 5 ⎪ ⎢ 0 ⎪ ⎢ ⎪ ⎪ FD 6 ⎪ ⎢ 0 ⎪ FD 7 ⎪ ⎢ ⎭ ⎣ 0 ⎩
0
0
0
0
0
A 22 0
0 A33
0 0
0 0
0 0
0
0
A 44
0
0
0 0
0 0
0 0
A55 0
0 A66
0
0
0
0
0
0 ⎤ ⎧ X D1 ⎫ 0 ⎥⎥ ⎪⎪ X D 2 ⎪⎪ 0 ⎥ ⎪ XD 3 ⎪ ⎪ ⎥⎪ 0 ⎥ ⎨ XD 4 ⎬ 0 ⎥ ⎪ XD 5 ⎪ ⎪ ⎥⎪ 0 ⎥ ⎪ XD 6 ⎪ ⎪ ⎪ A77 ⎥⎦ ⎩ X D 7 ⎭
Design Variables XD1, XD2 are integrated in the same physical area (New Design :Actuator) to reduce the complexity
Design Variables XD5, XD6 are integrated in the same physical area (New Design :Polydome foil sticked on PCB) to reduce the complexity
IDOV Page 10
Design
Flow Down And Uncouple/Decouple •Transferring Keypad Design Functions to Design Variables Hierarchy 2
⎧ F D11 ⎫ ⎡ A11 ⎪ ⎢ ⎪ ⎨ F D12 ⎬ = ⎢ 0 ⎪ F D13 ⎪ ⎢ 0 ⎭ ⎣ ⎩
0 A 22 0
0 ⎤ ⎧ X D11 ⎫ ⎪ ⎪ 0 ⎥⎥ ⎨ X D12 ⎬ A33 ⎥⎦ ⎪⎩ X D13 ⎪⎭
Independency of Design Functions Uncoupled Design Transfer Matrix
IDOV Page 11
Design
Flow Down And Uncouple/Decouple •Transferring Keypad Design Functions to Design Variables Hierarchy 3 ⎧ F D111 ⎫ ⎡ A11 ⎪ F D112 ⎪ ⎢ 0 ⎪ ⎢ ⎪ ⎪⎪ F D121 ⎪⎪ ⎢ 0 ⎬=⎢ ⎨ D122 F ⎪ ⎢ 0 ⎪ ⎪ F D131 ⎪ ⎢ 0 ⎪ ⎢ ⎪ ⎪⎩ F D132 ⎪⎭ ⎢⎣ 0
A12 0 0 0 0 0
0 A 23 0 0 0 0
0 0 A34 0 0 0
0 0 A35 0 0 0
0 0 0 A 46 0 0
0 0 0 0 A57 0
0 0 0 0 A58 0
⎧ X D111 − 1 ⎫ ⎪ X D111 − 2 ⎪ ⎪ 0 ⎤⎪ ⎪ X D112 ⎪ 0 ⎥⎥ ⎪ ⎪ X D121 − 1 ⎪ ⎪ 0 ⎥⎪ ⎪ ⎥ ⎨ X D121 − 2 ⎬ 0 ⎥⎪ X D122 ⎪ ⎪ 0 ⎥⎪ ⎥ ⎪ X D131 − 1 ⎪ A69 ⎥⎦ ⎪ ⎪ D131 − 2 X ⎪ ⎪ ⎪⎩ X D132 ⎪⎭
Redundant Design Variables Coupled Design Transfer Matrix but the Sensitivities will be checked experimentely
IDOV Page 12
Design
Concept Generation & Selection Concept Generation, Selection, and Feasibility using structured methodologies:
IDOV Page 13
Design
Preliminary Design Not critical key height (less compression)
Old System
Expensive gold meander on the old PCB system Easier to contact because of the min. pitch and low contact resistance of gold
Critical key height (high compression)
New System
Inexpensive graphite meander on the new PCB system Difficult to contact because of the min. pitch (technology limitation) and high contact resistance
IDOV Page 14
Design
Detailed Design
Hardcap Silicone Polydomfoil Spacerfoil
IDOV Page 15
Design
Functional Block Diagramming •Function Block & P-Diagramming Noise Factors Compression Input Signal : Linear Mechanical Movement
Snap-In Output Signal (Functional Requirement) : F max at x max
F F max
Output Signal (Functional Requirement) : F min at x min
F
1
F min
β x max
x
x min
x
Control Factors (Design Parameters) Silicon Hardness
Plunger Height
Polydome Hight
Web Design
Polydome Thickness Polydome Diameter
Project Boundry (Projekt Focus)
Page 16
Design
Functional Block Diagramming •Function Block & P-Diagramming Noise Factors Force to Signal Converter (Key Contact on PCB Meander) Input Signal : F min at x min
F F max
Output Signal Short on Meander (< 300 Ohm, 99%)
Time Delay
Electric to Acoustic Converter Output Signal (Functional Requirement) : Audio Signal
1
x max
x
Control Factors (Design Parameters) Carbon Type
Plunger Diameter Carbon Meander Pitch
Project Boundry (Projekt Focus)
Page 17
Design
Robustness Noise Factors
Make the Design robust to noise factors.
Control Factors
IDOV Page 18
Noise Factor Effect (Function Dialkeys) Main Effects Plot for Means
Operator
Device
Act.Directio
149
Mean
147
145
143
141
S
er g aa
t n le 1 e u B
2
rt. e V
iz. r o H
CNF2 (Compoung Noise Factor 2) = Op1 + Device 1 + Actuator Direc. Vertical CNF1 (Compound Noise Factor 1) = Op 1 + Device 2 + Actuator Direc. Horizontal
IDOV Page 19
Control Factor Effect for S/N Ratio Function Dialkeys Main Effects Plot for S/N Ratios
PCB Pitch
Plunger Dia
Spacer
PET
Dome Dia
Dome Shape
48,0
S/N Ratio
45,5
43,0
40,5
38,0
6 0,
8 0, D
E 75u
0u 00u 5 1 1
5u mm 2 1 5
l m rma 6m no
t f la
= Control Factor Levels already implemented in design
IDOV Page 20
Control Factor Effect for Means Function Dialkeys
Main Effects Plot for Means
PCB Pitch
Plunger Dia
Spacer
PET
Dome Dia
Dome Shape
280
Mean
240
200
160
120 6 0,
8 0,
D
E
u 75
0u 00u 1 15
5u mm 12 5
al m 6m norm
t fla
IDOV Page 21
Design
Transfer Functions Utilize Design of Experiments to create transfer functions A
B
C
D
E
F
0,6
E
75u
100u
5mm
normal
0,6
E
75u
125u
6mm
flat
0,6
D
150u
100u
5mm
flat
0,6
D
150u
125u
6mm
normal
0,8
E
150u
100u
6mm
normal
0,8
E
150u
125u
5mm
flat
0,8
D
75u
100u
6mm
flat
0,8
D
75u
125u
5mm
normal
First rough samples with holes in the plungers to increase the contact area and forces
Main Effects Plot for Means
A
B
C
D
E
F
220
Mean
190
160
130
100
0,
6
8 0,
D
E
75
u
15
0u
10
0u
12
5u
5m
m
6m
m no
rm
al
fl a
IDOV
t
Page 22
Design
Correction of Transfer Matrix •Transferring Keypad Design Functions to Design Variables FD1 : Dialing
FD11 Transformation of a mechanical linear movement to displacement
FD111 F max
FD12 Snapping (Click Feeling) (F min at x min)
X D1 : Actuating Keypad System
FD13 Contact on PCB (< 300 Ohm, Contact > 99 %)
FD121 F min
X D11 Actuator
XD111-1 Silicon Hardness
FD131 < 300 Ohm
XD111 Web Design
X D12 Polydome
XD121-1 Thickness
XD121-2 Diameter
X D13 Carbon PCB / Polydome
XD131 Type
XD131-2 Pitch
XD121 Plunger Hole Depth FD112 @ Displacement x max
FD122 @ Displacement x min
FD132 > 99 %
XD112 Plunger Height
XD122 Height
Design Variables fixed to a choosed value Patent Pending
XD132 Plunger Diameter XD132 Plunger Hole Dia.
IDOV Page 23
Design
Correction of Transfer Matrix •Correction of Design Transfer Matrix Hierarchy 3 ⎧ F D111 ⎫ ⎡ A11 ⎪ F D112 ⎪ ⎢ 0 ⎪ ⎪ ⎢ ⎪⎪ F D121 ⎪⎪ ⎢ 0 ⎨ ⎬=⎢ F D122 ⎪ ⎪ ⎢ 0 ⎪ F D131 ⎪ ⎢ 0 ⎪ ⎪ ⎢ ⎪⎩ F D132 ⎪⎭ ⎢⎣ 0
0 A 22
0 0
0 0
0 0
0
A303
0
0
0 0
0 0
A 44 0
0 A55
0
0
0
0
0 ⎤ ⎧ X D111 ⎫ 0 ⎥⎥ ⎪⎪ X D112 ⎪⎪ 0 ⎥ ⎪⎪ X D121 ⎪⎪ ⎬ ⎥⎨ 0 ⎥ ⎪ X D122 ⎪ 0 ⎥ ⎪ X D131 ⎪ ⎥⎪ ⎪ A66 ⎥⎦ ⎪⎩ X D132 ⎪⎭
Design Variables introduced in same physical area
Uncoupled Design Transfer Matrix
IDOV Page 24
Design
Flow Down And Uncouple/Decouple •Transferring Keypad Design Variables to Process Variables Hierarchy 3 XD111 : Web Design
X P112 : Insert 1
XD112 : Plunger Height
X P112 : Insert 2
XD121 : Plunger Hole Depth
X P112 : Insert 3
XD122 : Polydome Height
X P1221 : Temperature X P1222 : Cycle-Time X P1223 : Pressure
XD131 : Carbon Type
X P131 : Temperature
XD132 : Plunger Hole Diameter
X P132 : Insert 4
IDOV Page 25
Design
Transfer Functions •Checking design parameter to process variable relation and fixing redundant process variables (cycle time, pressure) to obtain robust parameter XD122 Polydome Height
Polydome Height Temperature
74.0
Cycletime
Pressure
Temperature
Cycletime
Pressure
37.20
71.5 S/N Ratio
Mean
36.95
69.0
36.70 36.45
66.5
36.20
64.0 0 12
0 14
0 15
5
7
9
6
7
11
0 12
0 14
0 15
5
7
9
6
7
11
IDOV Page 26
Design Correction of Transfer Matrix •Transferring Keypad Design Variables to Process Variables Hierarchy 3 XD111 : Web Design
X P112 : Insert 1
XD112 : Plunger Height
X P112 : Insert 2
XD121 : Plunger Hole Depth
X P112 : Insert 3
XD122 : Polydome Height
X P1221 : Temperature X P1222 : Cycle-Time X P1223 : Pressure
XD131 : Carbon Type
X P131 : Temperature
XD132 : Plunger Hole Diameter
X P132 : Insert 4
Process Variables fixed to a choosed value
IDOV Page 27
Optimization
Surface Plot of C NS
quantifying of the optimum functionality of the dialkey
8 7 6
C NS
5 4
Surface Plot of Q D4
3 2, 2,5
X
3,0
1,0
1,1 1 1
1,2
1,3
1,4
1,,5
1,6
Z
3,5
350
300
Q D4
250
200 2, 2,5
X
3,0
1,0
1,1
1,2
1,3
1,4
1,,5
Z
1,6
quantifying of the optimum volume of noise of the Navikey
3,5
IDOV Page 28
Optimization
IDOV Page 29
Control Factors for Taguchi DOE
•
We will focus to change the keypad plunger design to enlarge the contact area
•
The following factors will be defined with 3 levels as controllable factors
Hol e dept h
Hol e di amet er Pl unger di amet er Pl unger Di st ance
IDOV Page 30
Taguchi DOE Analysis for Noise Factors
Main Effects Plot for Means
customer
case
Contact poin
294
Mean
293
292
291
290 r ee Pi
ou Zh
A
B
ak Pe
de Si
IDOV Page 31
Taguchi DOE Analysis for Control Factors •
Analysis result: Main Effects Plot for Means of left key inside
Main Effects Plot for S/NRatios of left key inside Actuator
Offset Plung
Web Thicknes
Plunger Slan
Actuator
Offset Plung
Web Thicknes
Plunger Slan
300 49.2 280
Mean
S/N Ratio
48.4
47.6
240
46.8
220
46.0 2 1. 6* 2. 4*
•
260
.2 *1 .9 *1 3 3.
1* 1. 5* . 2
1 0. A-
A
.1 +0 A
20 0.
27 0.
35 0.
0
6
12
2 1. 6* 2. 4*
.2 *1 .9 *1 3 3.
1* 1. 5* . 2
1 0. A-
A
.1 +0 A
20 0.
27 0.
35 0.
0
6
12
After cross-functional discussion with Suppliers, the following parameter will be used in verify phase for the confirmation run:
IDOV Page 32
Validate
Force
Resistance
Quantifying and measurement of CTS after project completion:
no shift = good function
IDOV Resistance
Displacement Page 33
Validate Validate the Design using Life Cycle Tests Version:
New Keypad System
Number of samples:
2x2 pcs/cavity
Number of hits:
200.000
testing cycles:
Force-Stroke-Measurement 50.000 times --> F/s diagram OK 100.000 times --> F/s diagram OK 150.000 times --> F/s diagram OK
Testing result: No. of cycles 0 50.000 100.000 150.000
4 7 8 9 9 9 ? ? ? ? ? ? 4 cracks 9 9 at cavity 4 only the navi-key was damaged
15 9 ? ? 9
IDOV Page 34
Project Results
•Time frame: Target 4 months, achieved 3 months •Generic contact problem: Target < 0,5%, achieved almost 0 •Reduce total cost of the system: Target > 50%, achieved > 75% •Increase the click ratio (feeling): Target > 25%, achieved > 40% •Savings due to achieved results > 3 Million Euro •Development Cycle Time reduction for the next product generation because of the cancelled second iteration with soft tools •Same keypad system implemented in all the families of the existing and the next product generation •Patent Application for: “Polydomfoil Keypad with Optimized actuators”
Page 35
SSI - Six Sigma & Innovation Josef-Fehler-Str. 43 46397 Bocholt Fon: +49 2871 2745565 Fax: +49 2871 2745564 Mobil: +49 173 2856431 E-Mail:
[email protected] http://www.ssi-team.com
The following are trademarks and service marks of SSI, Six Sigma & Innovation: New Generation Design For Six Sigma ®, Quick Improvement Projects ®, Six Sigma Light ® Six Sigma is a federally registered trademark of Motorola, Inc. MINITAB is a federally registered trademark of Minitab, Inc.
Page 36