ADPD106 [ADI]
Photometric Front Ends;型号: | ADPD106 |
厂家: | ADI |
描述: | Photometric Front Ends |
文件: | 总66页 (文件大小:884K) |
中文: | 中文翻译 | 下载: | 下载PDF数据表文档文件 |
Photometric Front Ends
Data Sheet
ADPD105/ADPD106/ADPD107
FEATURES
GENERAL DESCRIPTION
Multifunction photometric front end
The ADPD105/ADPD106/ADPD107 are highly efficient,
photometric front ends, each with an integrated 14-bit analog-
to-digital converter (ADC) and a 20-bit burst accumulator that
works with flexible light emitting diode (LED) drivers. The
accumulator is designed to stimulate an LED and measure
the corresponding optical return signal. The data output and
functional configuration occur over a 1.8 V I2C interface on the
ADPD105 or SPI on the ADPD106 and ADPD107. The control
circuitry includes flexible LED signaling and synchronous
detection.
Fully integrated AFE, ADC, LED drivers, and timing core
Enables best-in-class ambient light rejection capability
without the need for photodiode optical filters
Three 370 mA LED drivers
Flexible, multiple, short LED pulses per optical sample
20-bit burst accumulator enabling 20 bits per sample period
On-board sample to sample accumulator, enabling up to
27 bits per data read
Low power operation
SPI, I2C interface, and 1.8 V analog/digital core
Flexible sampling frequency ranging from 0.122 Hz to 3820 Hz
FIFO data operation
The analog front end (AFE) features best-in-class rejection of signal
offset and corruption due to modulated interference commonly
caused by ambient light.
APPLICATIONS
Couple the ADPD105/ADPD106/ADPD107 with a low
capacitance photodiode of <100 pF for optimal performance.
The ADPD105/ADPD106/ADPD107 can be used with any LED.
The ADPD105 is available in a 2.46 mm × 1.4 mm WLCSP and a
4 mm × 4 mm LFCSP. The SPI only versions, ADPD106 and
ADPD107, are available in a 2.46 mm × 1.4 mm WLCSP.
Wearable health and fitness monitors
Clinical measurements, for example, SpO2
Industrial monitoring
Background light measurements
Rev. A
Document Feedback
Information furnished by Analog Devices is believed to be accurate and reliable. However, no
responsibility is assumed by Analog Devices for its use, nor for any infringements of patents or other
rights of third parties that may result from its use. Specifications subject to change without notice. No
license is granted by implication or otherwise under any patent or patent rights of Analog Devices.
Trademarks and registeredtrademarks arethe property of their respective owners.
One Technology Way, P.O. Box 9106, Norwood, MA 02062-9106, U.S.A.
Tel: 781.329.4700
Technical Support
©2017 Analog Devices, Inc. All rights reserved.
www.analog.com
ADPD105/ADPD106/ADPD107
Data Sheet
TABLE OF CONTENTS
Features .............................................................................................. 1
LED Driver Pins and LED Supply Voltage.............................. 31
LED Driver Operation............................................................... 31
Determining the Average Current ........................................... 31
Determining CVLED ..................................................................... 32
LED Inductance Considerations.............................................. 32
Recommended Start-Up Sequence .......................................... 33
Reading Data............................................................................... 33
Clocks and Timing Calibration................................................ 34
Optional Timing Signals Available on GPIO0 and GPIO1 .. 35
Calculating Current Consumption.......................................... 37
Optimizing SNR per Watt......................................................... 38
Applications....................................................................................... 1
General Description......................................................................... 1
Revision History ............................................................................... 3
Functional Block Diagrams............................................................. 4
Specifications..................................................................................... 6
Temperature and Power Specifications ..................................... 6
Performance Specifications......................................................... 7
Analog Specifications................................................................... 8
Digital Specifications ................................................................... 9
Timing Specifications ................................................................ 10
Absolute Maximum Ratings.......................................................... 12
Thermal Resistance .................................................................... 12
Recommended Soldering Profile ............................................. 12
ESD Caution................................................................................ 12
Pin Configurations and Function Descriptions ......................... 13
Typical Performance Characteristics ........................................... 17
Theory of Operation ...................................................................... 19
Introduction................................................................................ 19
Dual Time Slot Operation......................................................... 19
Time Slot Switch......................................................................... 20
Adjustable Sampling Frequency............................................... 22
State Machine Operation........................................................... 23
Normal Mode Operation and Data Flow................................ 23
AFE Operation............................................................................ 25
AFE Integration Offset Adjustment......................................... 25
I2C Serial Interface...................................................................... 27
SPI Port ........................................................................................ 28
Typical Connection Diagram ................................................... 30
Optimizing Power by Disabling Unused Channels and
Amplifiers.................................................................................... 39
TIA ADC Mode.......................................................................... 40
Digital Integrate Mode............................................................... 42
Pulse Connect Mode.................................................................. 45
Synchronous ECG and PPG Measurement Using TIA ADC
Mode ............................................................................................ 45
Register Listing ............................................................................... 48
LED Control Registers............................................................... 52
AFE Global Configuration Registers....................................... 54
System Registers ......................................................................... 59
ADC Registers ............................................................................ 63
Data Registers ............................................................................. 64
Required Start-Up Load Procedure ......................................... 64
Outline Dimensions....................................................................... 65
Ordering Guide .......................................................................... 66
Rev. A | Page 2 of 66
Data Sheet
ADPD105/ADPD106/ADPD107
REVISION HISTORY
1/2017—Rev. 0 to Rev. A
Added Table 16 and WLCSP Input Configurations Section.....21
Changes to Register 0x14, PD1 to PD4 Input Configurations
Section ..............................................................................................21
Changes to Typical Connection Diagram Section .....................30
Added Figure 37 and Table 22.......................................................31
Added Measuring Voltages Using the Current Input Section,
Figure 52, and Figure 53.................................................................47
Changes to Table 28 ........................................................................48
Updated Outline Dimensions........................................................65
Changes to Ordering Guide...........................................................66
Added ADPD106................................................................. Universal
Changes to Features Section and General Description Section.......1
Changes to Figure 1...........................................................................4
Added Figure 2; Renumbered Sequentially ...................................5
Changes to Table 2 ............................................................................6
Changes to Table 9 ..........................................................................12
Added Figure 6 and Table 11; Renumbered Sequentially..........13
Added Figure 8 and Table 13 .........................................................15
Changes to ADPD105 LFCSP Input Configurations
Section ..............................................................................................20
Added Register 0x14, PD1 to PD8 Input Configurations Section
and Figure 19 to Figure 21 .............................................................20
7/2016—Revision 0: Initial Version
Rev. A | Page 3 of 66
ADPD105/ADPD106/ADPD107
Data Sheet
FUNCTIONAL BLOCK DIAGRAMS
AVDD
DVDD
TIME SLOT
SWITCH
ANALOG BLOCK
PDC
AFE:
PD1-2
SIGNAL CONDITIONING
ADPD105/
ADPD106/
ADPD107
TIA
BPF
±1 INTEGRATOR
V
V
V
V
BIAS
WLCSP VERSIONS
VREF
1µF
AFE:
SIGNAL CONDITIONING
TIME SLOT A
TIA
BPF
DATA
±1 INTEGRATOR
14-BIT
ADC
BIAS
TIME SLOT B
DATA
AFE:
PD3-4
SIGNAL CONDITIONING
AFE
GPIO0
GPIO1
CONFIGURATION
TIA
BPF
±1 INTEGRATOR
A
B
PD3-4 ON
BIAS
ADPD105/
ADPD107
ONLY
SLOT
DIGITAL
DATAPATH
AND
INTERFACE
CONTROL
SELECT
AFE:
SIGNAL CONDITIONING
MOSI
TIA
BPF
ADPD106/
ADPD107
ONLY
MISO
SCLK
CS
±1 INTEGRATOR
BIAS
SDA
SCL
ADPD105
ONLY
LEDX3
LEDX2
LEDX1
LED3
LED2
LED1
LED3 LEVEL AND TIMING CONTROL
LED2 LEVEL AND TIMING CONTROL
LED1 LEVEL AND TIMING CONTROL
LED3 DRIVER
LED2 DRIVER
DGND
AGND
LED1 DRIVER
V
LED
LGND
Figure 1. Block Diagram for ADPD105/ADPD106/ADPD107 WLCSP (Chip Scale Package) Versions
Rev. A | Page 4 of 66
Data Sheet
ADPD105/ADPD106/ADPD107
AVDD
DVDD
TIME SLOT
SWITCH
ANALOG BLOCK
PDC
PD1
AFE:
ADPD105
LFCSP
SIGNAL CONDITIONING
TIA
BPF
±1 INTEGRATOR
PD5
V
V
V
V
BIAS
VREF
AFE:
SIGNAL CONDITIONING
PD2
PD6
1µF
TIME SLOT A
TIA
BPF
DATA
±1 INTEGRATOR
14-BIT
ADC
BIAS
TIME SLOT B
DATA
AFE:
SIGNAL CONDITIONING
AFE
PD3
PD7
GPIO0
GPIO1
CONFIGURATION
TIA
BPF
±1 INTEGRATOR
A
B
BIAS
SLOT
SELECT
DIGITAL
DATAPATH
AND
INTERFACE
CONTROL
AFE:
SIGNAL CONDITIONING
PD4
PD8
TIA
BPF
±1 INTEGRATOR
BIAS
SDA
SCL
LEDX3
LEDX2
LEDX1
LED3
LED2
LED1
LED3 LEVEL AND TIMING CONTROL
LED2 LEVEL AND TIMING CONTROL
LED1 LEVEL AND TIMING CONTROL
LED3 DRIVER
LED2 DRIVER
DGND
AGND
LED1 DRIVER
V
LED
LGND
Figure 2. Block Diagram for ADPD105 LFCSP Version
Rev. A | Page 5 of 66
ADPD105/ADPD106/ADPD107
Data Sheet
SPECIFICATIONS
TEMPERATURE AND POWER SPECIFICATIONS
Table 1. Operating Conditions
Parameter
Test Conditions/Comments
Min
Typ
Max
Unit
TEMPERATURE RANGE
Operating Range
Storage Range
POWER SUPPLY VOLTAGES
VDD
−40
−65
+85
+150
°C
°C
Applied at the AVDD and DVDD pins
1.7
1.8
1.9
V
AVDD = DVDD = 1.8 V, TA = 25°C, unless otherwise noted.
Table 2. Current Consumption
Parameter
Symbol
Test Conditions/Comments
Min Typ Max Unit
POWER SUPPLY (VDD) CURRENT
VDD Supply Current1
SLOTx_LED_OFFSET = 25 μs; LED_PERIOD =19 μs; LED peak
current = 25 mA, single-channel mode
1 Pulse
100 Hz data rate; Time Slot A only
100 Hz data rate; Time Slot B only
100 Hz data rate; both Time Slot A and Time Slot B
100 Hz data rate; Time Slot A only
100 Hz data rate; Time Slot B only
92
75
119
175
155
281
μA
μA
μA
μA
μA
μA
10 Pulses
100 Hz data rate; both Time Slot A and Time Slot B
Peak VDD Supply Current
(1.8 V)
IVDD_PEAK
4-Channel Operation
1-Channel Operation
Standby Mode Current
VLED SUPPLY CURRENT2
Average Supply Current
VLEDA or VLEDB
9.3
4.5
0.3
mA
mA
μA
IVDD_STANDBY
Peak LED current = 25 mA; LED pulse width = 3 μs
50 Hz data rate
100 Hz data rate
200 Hz data rate
50 Hz data rate
1 Pulse
3.75
7.5
15
38
75
μA
μA
μA
μA
μA
μA
10 Pulses
100 Hz data rate
200 Hz data rate
150
1 VDD is the voltage applied at the AVDD and DVDD pins.
2 VLED applies to the external LED supply voltage for any given LED being driven by the ADPD105/ADPD106/ADPD107 LED drivers under the listed conditions.
Rev. A | Page 6 of 66
Data Sheet
ADPD105/ADPD106/ADPD107
PERFORMANCE SPECIFICATIONS
AVDD = DVDD = 1.8 V, TA = full operating temperature range, unless otherwise noted.
Table 3.
Parameter
Test Conditions/Comments
Min
Typ
Max
Unit
DATA ACQUISITION
Resolution
Resolution/Sample
Resolution/Data Read
LED DRIVER
LED Current Slew Rate1
Rise
Single pulse
64 to 255 pulses
64 to 255 pulses and sample average = 128
14
20
27
Bits
Bits
Bits
Slew rate control setting = 0; TA = 25°C; ILED = 70 mA
Slew rate control setting = 7; TA = 25°C; ILED = 70 mA
Slew rate control setting = 0, 1, 2; TA = 25°C; ILED = 70 mA
Slew rate control setting = 6, 7; TA = 25°C; ILED = 70 mA
LED pulse enabled
Voltage above ground required for LED driver operation
AFE width = 4 μs2
AFE width = 3 μs
Time Slot A only; normal mode; 1 pulse; SLOTA_LED_OFFSET = 23 μs; SLOTA_
LED_PERIOD = 19 μs
Time Slot B only; normal mode; 1 pulse; SLOTA_LED_OFFSET = 23 μs; SLOTA_
LED_PERIOD = 19 μs
Both time slots; normal mode; 1 pulse; SLOTA_LED_OFFSET = 23 μs; SLOTA_
LED_PERIOD = 19 μs
Time Slot A only; normal mode; 8 pulses; SLOTA_LED_OFFSET = 23 μs; SLOTA_
LED_PERIOD = 19 μs
240
mA/μs
mA/μs
mA/μs
mA/μs
mA
1400
3200
4500
Fall
LED Peak Current
Driver Compliance Voltage
LED PERIOD
370
0.6
19
17
V
μs
μs
Sampling Frequency3
0.122
0.122
0.122
0.122
0.122
0.122
3230 Hz
3820 Hz
1750 Hz
2257 Hz
2531 Hz
1193 Hz
Time Slot B only; normal mode; 8 pulses; SLOTA_LED_OFFSET = 23 μs; SLOTA_
LED_PERIOD = 19 μs
Both time slots; normal mode; 8 pulses; SLOTA_LED_OFFSET = 23 μs; SLOTA_
LED_PERIOD = 19 μs
CATHODE PIN (PDC) VOLTAGE
During All Sampling Periods
Register 0x54, Bit 7 = 0x0; Register 0x3C, Bit 9 = 14
Register 0x54, Bit 7 = 0x0; Register 0x3C, Bit 9 = 0
1.8
1.3
1.8
1.3
1.55
0
1.8
1.3
1.55
0
1.8
1.3
1.8
1.3
1.55
0
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
V
During Slot A Sampling
During Slot B Sampling
During Sleep Periods
Register 0x54, Bit 7 = 0x1; Register 0x54, Bits[9:8] = 0x04
Register 0x54, Bit 7 = 0x1; Register 0x54, Bits[9:8] = 0x1
Register 0x54, Bit 7 = 0x1; Register 0x54, Bits[9:8] = 0x2
Register 0x54, Bit 7 = 0x1; Register 0x54, Bits[9:8] = 0x35
Register 0x54, Bit 7 = 0x1; Register 0x54, Bits[11:10] = 0x04
Register 0x54, Bit 7 = 0x1; Register 0x54, Bits[11:10] = 0x1
Register 0x54, Bit 7 = 0x1; Register 0x54, Bits[11:10] = 0x2
Register 0x54, Bit 7 = 0x1; Register 0x54, Bits[11:10] = 0x35
Register 0x54, Bit 7 = 0x0; Register 0x3C, Bit 9 = 1
Register 0x54, Bit 7 = 0x0; Register 0x3C, Bit 9 = 0
Register 0x54, Bit 7 = 0x1; Register 0x54, Bits[13:12] = 0x0
Register 0x54, Bit 7 = 0x1; Register 0x54, Bits[13:12] = 0x1
Register 0x54, Bit 7 = 0x1; Register 0x54, Bits[13:12] = 0x2
Register 0x54, Bit 7 = 0x1; Register 0x54, Bits[13:12] = 0x3
PHOTODIODE INPUT PINS/
ANODE VOLTAGE
During All Sampling Periods
During Sleep Periods
1.3
V
V
Cathode voltage
1 LED inductance is negligible for these values. The effective slew rate slows with increased inductance.
2 Minimum LED period = (2 × AFE width) + 5 μs.
3 The maximum values in this specification are the internal ADC sampling rates in normal mode. The I2C read rates in some configurations may limit the output data rate.
4 This mode may induce additional noise and is not recommended unless absolutely necessary. The 1.8 V setting uses VDD, which contains greater amounts of
differential voltage noise with respect to the anode voltage. A differential voltage between the anode and cathode injects a differential current across the capacitance
of the photodiode of the magnitude of C × dV/dt.
5 This setting is not recommended for photodiodes because it causes a 1.3 V forward bias of the photodiode.
Rev. A | Page 7 of 66
ADPD105/ADPD106/ADPD107
Data Sheet
ANALOG SPECIFICATIONS
AVDD = DVDD = 1.8 V, TA = full operating temperature range, unless otherwise noted. Compensation of the AFE offset is explained in
the AFE Operation section.
Table 4.
Parameter
Test Conditions/Comments
Min Typ
Max Unit
INPUT CAPACITANCE
100
pF
PULSED SIGNAL CONVERSIONS, 3 ꢀs WIDE LED PULSE1
4 ꢀs wide AFE integration; normal operation,
Register 0x43/Register 0x45 = 0xADA5
Transimpedance amplifier (TIA) feedback resistor
ADC Resolution2
25 kΩ
50 kΩ
100 kΩ
200 kΩ
3.27
1.64
0.82
0.41
nA/LSB
nA/LSB
nA/LSB
nA/LSB
ADC Saturation Level
TIA feedback resistor
25 kΩ
26.8
13.4
6.7
ꢀA
ꢀA
ꢀA
ꢀA
50 kΩ
100 kΩ
200 kΩ
3.35
Ambient Signal Headroom on Pulsed Signal
TIA feedback resistor
25 kΩ
23.6
11.8
5.9
ꢀA
ꢀA
ꢀA
ꢀA
50 kΩ
100 kΩ
200 kΩ
2.95
PULSED SIGNAL CONVERSIONS, 2 ꢀs WIDE LED PULSE1
ADC Resolution2
3 ꢀs wide AFE integration; normal operation,
Register 0x43/Register 0x45 = 0xADA5
TIA feedback resistor
25 kΩ
4.62
2.31
1.15
0.58
nA/LSB
nA/LSB
nA/LSB
nA/LSB
50 kΩ
100 kΩ
200 kΩ
ADC Saturation Level
TIA feedback resistor
25 kΩ
37.84
18.92
9.46
ꢀA
ꢀA
ꢀA
ꢀA
50 kΩ
100 kΩ
200 kΩ
4.73
Ambient Signal Headroom on Pulsed Signal
FULL SIGNAL CONVERSIONS3
TIA feedback resistor
25 kΩ
50 kΩ
100 kΩ
200 kΩ
12.56
6.28
3.14
1.57
ꢀA
ꢀA
ꢀA
ꢀA
TIA Saturation Level of Pulsed Signal and Ambient Level TIA feedback resistor
25 kΩ
50 kΩ
100 kΩ
200 kΩ
50.4
25.2
12.6
6.3
ꢀA
ꢀA
ꢀA
ꢀA
Rev. A | Page 8 of 66
Data Sheet
ADPD105/ADPD106/ADPD107
Parameter
Test Conditions/Comments
Min Typ
Max Unit
SYSTEM PERFORMANCE
Total Output Noise Floor
Normal mode; per pulse; per channel; no LED; photodiode
capacitance (CPD) = 70 pF
25 kΩ; referred to ADC input
1.0
LSB rms
25 kΩ; referred to peak input signal for 2 μs LED pulse
25 kΩ; referred to peak input signal for 3 μs LED pulse
25 kΩ; saturation signal-to-noise ratio (SNR) per pulse
per channel4
4.6
3.3
78.3
nA rms
nA rms
dB
50 kΩ; referred to ADC input
1.2
2.8
2.0
76.6
1.7
1.9
1.4
73.6
2.75
1.6
LSB rms
nA rms
nA rms
dB
LSB rms
nA rms
nA rms
dB
LSB rms
nA rms
nA rms
dB
50 kΩ; referred to peak input signal for 2 μs LED pulse
50 kΩ; referred to peak input signal for 3 μs LED pulse
50 kΩ; saturation SNR per pulse per channel4
100 kΩ; referred to ADC input
100 kΩ; referred to peak input signal for 2 μs LED pulse
100 kΩ; referred to peak input signal for 3 μs LED pulse
100 kΩ; saturation SNR per pulse per channel4
200 kΩ; referred to ADC input
200 kΩ; referred to peak input signal for 2 μs LED pulse
200 kΩ; referred to peak input signal for 3 μs LED pulse
200 kΩ; saturation SNR per pulse per channel4
1.1
69.5
−37
DC Power Supply Rejection Ratio (DC PSRR)
dB
1 This saturation level applies to the ADC only and, therefore, includes only the pulsed signal. Any nonpulsatile signal is removed prior to the ADC stage.
2 ADC resolution is listed per pulse when the AFE offset is correctly compensated per the AFE Operation section. If using multiple pulses, divide by the number of pulses.
3 This saturation level applies to the full signal path and, therefore, includes both the ambient signal and the pulsed signal.
4 The noise term of the saturation SNR value refers to the receive noise only and does not include photon shot noise or any noise on the LED signal itself.
DIGITAL SPECIFICATIONS
DVDD = 1.7 V to 1.9 V, unless otherwise noted.
Table 5.
Parameter
Symbol
Test Conditions/Comments
Min
Typ
Max
Unit
LOGIC INPUTS (GPIOx, SCL, SDA,
SCLK, MOSI, CS)
Input Voltage Level
High
Low
VIH
VIL
0.7 × DVDD
3.6
0.3 × DVDD
V
V
Input Current Level
High
Low
Input Capacitance
LOGIC OUTPUTS
Output Voltage Level
High
IIH
IIL
CIN
−10
−10
+10
+10
μA
μA
pF
10
GPIOx, MOSI
VOH
VOL
2 mA high level output current
2 mA low level output current
SDA
2 mA low level output current
SDA
DVDD − 0.5
V
V
Low
0.5
Output Voltage Level
Low
Output Current Level
Low
VOL1
IOL
0.2 × DVDD
V
VOL1 = 0.6 V
6
mA
Rev. A | Page 9 of 66
ADPD105/ADPD106/ADPD107
Data Sheet
TIMING SPECIFICATIONS
Table 6. I2C Timing Specifications
Parameter
I2C PORT1
Symbol
Test Conditions/Comments
Min
Typ
Max
Unit
See Figure 3
SCL
Frequency
400
kHz
Minimum Pulse Width
High
Low
t1
t2
600
1300
ns
ns
Start Condition
Hold Time
Setup Time
SDA Setup Time
SCL and SDA
Rise Time
t3
t4
t5
600
600
100
ns
ns
ns
t6
t7
1000
300
ns
ns
Fall Time
Stop Condition
Setup Time
t8
600
ns
1 Guaranteed by design.
t3
t5
t3
SDA
t6
t1
SCL
t2
t7
t4
t8
Figure 3. I2C Timing Diagram
Rev. A | Page 10 of 66
Data Sheet
ADPD105/ADPD106/ADPD107
Table 7. SPI Timing Specifications
Parameter
SPI PORT
SCLK
Symbol
Test Conditions/Comments
Min
Typ
Max
Unit
Frequency
Minimum Pulse Width
High
fSCLK
10
MHz
tSCLKPWH
tSCLKPWL
20
20
ns
ns
Low
CS
Setup Time
tCS
CS setup to SCLK rising edge
CS hold from SCLK rising edge
CS pulse width high
10
10
10
ns
ns
ns
S
Hold Time
tCS
H
Pulse Width High
tCS
PWH
MOSI
ns
ns
Setup Time
Hold Time
MISO Output Delay
tMOSIS
tMOSIH
tMISOD
MOSI setup to SCLK rising edge
MOSI hold from SCLK rising edge
MISO valid output delay from SCLK
falling edge
10
10
20
ns
tCSH
tCSS
tCSPWH
tSCLKPWL
tSCLKPWH
CS
SCLK
MOSI
tMOSIH
tMOSIS
MISO
tMISOD
Figure 4. SPI Timing Diagram
Rev. A | Page 11 of 66
ADPD105/ADPD106/ADPD107
Data Sheet
ABSOLUTE MAXIMUM RATINGS
RECOMMENDED SOLDERING PROFILE
Table 8.
Figure 5 and Table 10 provide details about the recommended
soldering profile.
Parameter
Rating
AVDD to AGND
−0.3 V to +2.2 V
−0.3 V to +2.2 V
−0.3 V to +2.2 V
−0.3 V to +2.2 V
−0.3 V to +3.6 V
−0.3 V to +3.9 V
−0.3 V to +3.9 V
150°C
DVDD to DGND
GPIO0 to DGND
GPIO1 to DGND
LEDXx to LGND
SCL to DGND
SDA to DGND
CRITICAL ZONE
tP
T
TO T
P
L
T
P
RAMP-UP
T
L
tL
T
SMAX
T
SMIN
Junction Temperature
ESD
tS
RAMP-DOWN
PREHEAT
Human Body Model (HBM)
Charged Device Model (CDM)
Machine Model (MM)
1500 V
500 V
100 V
t25°C TO PEAK
TIME
Figure 5. Recommended Soldering Profile
Stresses at or above those listed under Absolute Maximum
Ratings may cause permanent damage to the product. This is a
stress rating only; functional operation of the product at these
or any other conditions above those indicated in the operational
section of this specification is not implied. Operation beyond
the maximum operating conditions for extended periods may
affect product reliability.
Table 10. Recommended Soldering Profile
Profile Feature
Condition (Pb-Free)
Average Ramp Rate (TL to TP)
Preheat
3°C/sec max
Minimum Temperature (TSMIN
)
150°C
Maximum Temperature (TSMAX
Time (TSMIN to TSMAX) (tS)
TSMAX to TL Ramp-Up Rate
)
200°C
60 sec to 180 sec
3°C/sec maximum
THERMAL RESISTANCE
Time Maintained Above Liquidous
Temperature
Liquidous Temperature (TL)
Time (tL)
Thermal performance is directly linked to printed circuit board
(PCB) design and operating environment. Close attention to
PCB thermal design is required.
217°C
60 sec to 150 sec
+260 (+0/−5)°C
<30 sec
Table 9. Thermal Resistance
Package Type
Peak Temperature (TP)
Time Within 5°C of Actual Peak
Temperature (tP)
Ramp-Down Rate
Time from 25°C to Peak Temperature
θJA
54.9
60
Unit
°C/W
°C/W
°C/W
CP-28-5
CB-16-18
CB-17-1
6°C/sec maximum
8 minutes maximum
60
ESD CAUTION
Rev. A | Page 12 of 66
Data Sheet
ADPD105/ADPD106/ADPD107
PIN CONFIGURATIONS AND FUNCTION DESCRIPTIONS
21 NIC
20 NIC
19 NIC
GPIO0 1
GPIO1 2
DVDD 3
AGND 4
VREF 5
AVDD 6
PD1 7
ADPD105
18 NIC
17 NIC
16 NIC
15 PD8
TOP VIEW
(Not to Scale)
NOTES
1. NIC = NOT INTERNALLY CONNECTED (NONBONDED PAD).
THIS PIN CAN BE GROUNDED.
2. EXPOSED PAD (DIGITAL GROUND). CONNECT THE
EXPOSED PAD TO GROUND.
Figure 6. 28-Lead LFCSP Pin Configuration (ADPD105)
Table 11. 28-Lead LFCSP Pin Function Descriptions
Pin No.
Mnemonic
GPIO0
GPIO1
DVDD
AGND
VREF
AVDD
PD1
PD2
PD3
PD4
PDC
PD5
PD6
PD7
PD8
NIC
LEDX1
LEDX3
LEDX2
LGND
SCL
Type1
DIO
DIO
S
S
REF
S
AI
AI
AI
AI
AO
AI
AI
AI
AI
Description
1
2
3
4
5
6
7
8
General-Purpose Input/Output (I/O). This pin is used for interrupts and various clocking options.
General-Purpose I/O. This pin is used for interrupts and various clocking options.
1.8 V Digital Supply.
Analog Ground.
Internally Generated ADC Voltage Reference. Buffer this pin with a 1 μF capacitor to AGND.
1.8 V Analog Supply.
Photodiode Current Input (Anode). If not in use, leave this pin floating.
Photodiode Current Input (Anode). If not in use, leave this pin floating.
Photodiode Current Input (Anode). If not in use, leave this pin floating.
Photodiode Current Input (Anode). If not in use, leave this pin floating.
Photodiode Common Cathode Bias.
Photodiode Current Input (Anode). If not in use, leave this pin floating.
Photodiode Current Input (Anode). If not in use, leave this pin floating.
Photodiode Current Input (Anode). If not in use, leave this pin floating.
Photodiode Current Input (Anode). If not in use, leave this pin floating.
Not Internally Connected (Nonbonded Pad). This pin can be grounded.
LED Driver 1 Current Sink. If not in use, leave this pin floating.
LED Driver 3 Current Sink. If not in use, leave this pin floating.
LED Driver 2 Current Sink. If not in use, leave this pin floating.
LED Driver Ground.
9
10
11
12
13
14
15
16 to 22
23
24
25
26
27
28
R
AO
AO
AO
S
DI
DIO
S
I2C Clock Input.
I2C Data Input/Output.
Exposed Pad (Digital Ground). Connect the exposed pad to ground.
SDA
EPAD (DGND)
1 DIO means digital input/output, S means supply, REF means voltage reference, AI means analog input, AO means analog output, R means reserved, and DI means
digital input.
Rev. A | Page 13 of 66
ADPD105/ADPD106/ADPD107
Data Sheet
ADPD105
TOP VIEW, BALL SIDE DOWN
(Not to Scale)
1
2
3
A
B
C
D
E
F
LGND
LEDX2
SDA
DVDD
AGND
AVDD
PD3-4
LEDX3
SCL
LEDX1
GPIO0
DGND
VREF
GPIO1
PD1-2
PDC
Figure 7. ADPD105 Pin Configuration
Table 12. ADPD105 Pin Function Descriptions
Pin No.
A1
A2
B1
B2
B3
Mnemonic
LGND
LEDX2
LEDX3
LEDX1
SDA
Type1 Description
S
LED Driver Ground.
AO
AO
AO
DIO
DI
LED Driver 2 Current Sink. If not in use, leave this pin floating.
LED Driver 3 Current Sink. If not in use, leave this pin floating.
LED Driver 1 Current Sink. If not in use, leave this pin floating.
I2C Data Input/Output (I/O).
C1
SCL
I2C Clock Input.
C2
C3
D2
D3
E1
E2
E3
F1
GPIO0
DVDD
DGND
AGND
GPIO1
VREF
DIO
S
S
General Purpose I/O. This pin is used for interrupts and various clocking options.
1.8 V Digital Supply.
Digital Ground.
S
Analog Ground.
DIO
REF
S
General Purpose I/O. This pin is used for interrupts and various clocking options.
Internally Generated ADC Voltage Reference. Buffer this pin with a 1 μF capacitor to AGND.
1.8 V Analog Supply.
Photodiode Combined Current Input of Photodiode 1 (PD1) and Photodiode 2 (PD2). If not in use,
leave this pin floating.
AVDD
PD1-2
AI
F2
F3
PDC
PD3-4
AO
AI
Photodiode Common Cathode Bias.
Photodiode Combined Current Input of Photodiode 3 (PD3) and Photodiode 4 (PD4). If not in use,
leave this pin floating.
1 S means supply, AO means analog output, DIO means digital input/output, DI means digital input, REF means voltage reference, AI means analog input, and AO
means analog output.
Rev. A | Page 14 of 66
Data Sheet
ADPD105/ADPD106/ADPD107
ADPD106
TOP VIEW, BALL SIDE DOWN
(Not to Scale)
1
2
3
A
B
C
D
E
F
LEDX2
LGND
GPIO0
DGND
SCLK
VREF
PD1-2
LEDX3
GPIO1
LEDX1
MISO
MOSI
AGND
CS
AVDD
PDC
Figure 8. 16-Ball WLCSP Pin Configuration (ADPD106)
Table 13. 16-Ball WLCSP Pin Function Descriptions
Pin No.
A1
A2
B1
B2
B3
C1
C2
C3
D2
D3
E1
Mnemonic
LEDX2
LGND
LEDX3
LEDX1
GPIO0
GPIO1
MISO
Type1 Description
AO
S
LED Driver 2 Current Sink. If not in use, leave this pin floating.
LED Driver Ground
AO
AO
DIO
DIO
DO
S
LED Driver 3 Current Sink. If not in use, leave this pin floating.
LED Driver 1 Current Sink. If not in use, leave this pin floating.
General Purpose I/O. This pin is used for interrupts and various clocking options.
General Purpose I/O. This pin is used for interrupts and various clocking options.
SPI Data Output.
Digital Ground.
DGND
MOSI
SCLK
DI
SPI Data Input.
DI
SPI Clock Input.
CS
DI
SPI Chip Select, Active Low.
E2
E3
F1
F2
AGND
VREF
AVDD
PDC
S
REF
S
AO
AI
Analog Ground.
Internally Generated ADC Voltage Reference. Buffer this pin with a 1 μF capacitor to AGND.
1.8 V Analog Supply.
Photodiode Common Cathode Bias.
F3
PD1-2
Photodiode Combined Current Input of PD1 and PD2. If not in use, leave this pin floating.
1 AO means analog output, S means supply, DIO means digital input/output, DO means digital output, DI means digital input, REF means voltage reference, and
AI means analog input.
Rev. A | Page 15 of 66
ADPD105/ADPD106/ADPD107
Data Sheet
ADPD107
TOP VIEW, BALL SIDE DOWN
(Not to Scale)
1
2
3
A
B
C
D
E
F
LEDX2
LGND
GPIO0
DGND
SCLK
VREF
PD3-4
LEDX3
GPIO1
LEDX1
MISO
MOSI
AGND
CS
AVDD
PD1-2
PDC
Figure 9. ADPD107 Pin Configuration
Table 14. ADPD107 Pin Function Descriptions
Pin No.
A1
A2
B1
B2
B3
C1
C2
C3
D1
D2
D3
E1
Mnemonic
LEDX2
LGND
LEDX3
LEDX1
GPIO0
GPIO1
MISO
Type1 Description
AO
S
LED Driver 2 Current Sink. If not in use, leave this pin floating.
LED Driver Ground.
AO
AO
DIO
DIO
DO
S
LED Driver 3 Current Sink. If not in use, leave this pin floating.
LED Driver 1 Current Sink. If not in use, leave this pin floating.
General Purpose I/O. This pin is used for interrupts and various clocking options.
General Purpose I/O. This pin is used for interrupts and various clocking options.
Master Input, Slave Output.
DGND
CS
Digital Ground.
DI
SPI Chip Select. Active low.
MOSI
SCLK
DI
Master Output, Slave Input.
DI
SPI Clock Input.
AVDD
AGND
VREF
PD1-2
PDC
S
1.8 V Analog Supply.
E2
E3
F1
F2
S
Analog Ground.
REF
AI
AO
AI
Internally Generated ADC Voltage Reference. Buffer this pin with a 1 μF capacitor to AGND.
Photodiode Combined Current Input of PD1 and PD2. If not in use, leave this pin floating.
Photodiode Common Cathode Bias.
F3
PD3-4
Photodiode Combined Current Input of PD3 and PD4. If not in use, leave this pin floating.
1 AO means analog output, S means supply, DIO means digital input/output, DO means digital output, DI means digital input, REF means voltage reference, and AI
means analog input.
Rev. A | Page 16 of 66
Data Sheet
ADPD105/ADPD106/ADPD107
TYPICAL PERFORMANCE CHARACTERISTICS
70
3.0
2.5
2.0
1.5
1.0
0.5
0
60
50
40
30
20
10
0
NOISE vs.
NUMBER OF CHANNELS
SQUARE ROOT
OF CHANNEL COUNT
32
33
34
35
36
37
38
39
40
41
1
2
3
4
FREQUENCY (Hz)
NUMBER OF CHANNELS
Figure 10. 32 kHz Clock Frequency Distribution
Figure 13. Noise Multiple of Channel Sum vs. Number of Channels
(Default Settings, Before User Calibration: Register 0x4B = 0x2612)
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0
60
50
40
30
20
10
0
0
50
100
150
200
32.0 32.5 33.0 33.5 34.0 34.5 35.0 35.5 36.0
FREQUENCY (Hz)
TIA GAIN (kΩ)
Figure 11. 32 MHz Clock Frequency Distribution
Figure 14. Referred to Input (RTI) Noise vs. TIA Gain
(Default Settings, Before User Calibration: Register 0x4D = 0x0098)
4.0
400
350
300
250
200
150
100
50
N
= 0.24 × √C + 0.81
LED COARSE SETTING = 0xF
RMS
PD
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0
200k
N
N
= 0.11 × √C + 0.97
RMS
PD
100k
50k
= 0.04 × √C + 1.13
RMS
PD
25k
N
= 0.02 × √C + 1.16
PD
RMS
LED COARSE SETTING = 0x0
LED DRIVER VOLTAGE (V)
0
0
20
40
60
80
100
120
140
160
PHOTODIODE CAPACITANCE (pF)
Figure 15. LED Driver Current vs. LED Driver Voltage at Various Coarse Settings
Figure 12. RMS Noise vs. Photodiode Capacitance
Rev. A | Page 17 of 66
ADPD105/ADPD106/ADPD107
Data Sheet
45
40
35
30
25
20
360
340
320
300
280
260
240
220
0
1
2
3
4
5
6
7
8
9
10 11 12 13 14 15
0
1
2
3
4
5
6
7
8
9
10 11 12 13 14 15
LED FINE SETTING, REGISTER 0x25
LED FINE SETTING, REGISTER 0x25
Figure 16. LED Driver Current vs. LED Fine Setting (Coarse Setting = 0x0)
Figure 17. LED Driver Current vs. LED Fine Setting (Coarse Setting = 0xF)
Rev. A | Page 18 of 66
Data Sheet
ADPD105/ADPD106/ADPD107
THEORY OF OPERATION
INTRODUCTION
DUAL TIME SLOT OPERATION
The ADPD105/ADPD106/ADPD107 operate as a complete
optical transceiver stimulating up to three LEDs and measuring
the return signal on up to two separate current inputs. The core
consists of a photometric front end coupled with an ADC, digital
block, and three independent LED drivers. The core circuitry
stimulates the LEDs and measures the return in the analog block
through one to eight photodiode inputs, storing the results in
discrete data locations. The two inputs can be configured to drive
four simultaneous input channels. Data can be read directly by a
register, or through a FIFO. This highly integrated system includes
an analog signal processing block, digital signal processing block,
I2C communication interface on the ADPD105 or an SPI port on
the ADPD107, and programmable pulsed LED current sources.
The ADPD105/ADPD106/ADPD107 operate in two independent
time slots, Time Slot A and Time Slot B, which are carried out
sequentially. The entire signal path from LED stimulation to data
capture and processing is executed during each time slot. Each
time slot has a separate datapath that uses independent settings
for the LED driver, AFE setup, and the resulting data. Time Slot A
and Time Slot B operate in sequence for every sampling period, as
shown in Figure 18.
The timing parameters are defined as follows:
tA (μs) = SLOTA_LED_OFFSET + nA × SLOTA_LED_PERIOD
where nA is the number of pulses for Time Slot A (Register 0x31,
Bits[15:8]).
The LED driver is a current sink and is agnostic to LED supply
voltage and LED type. The photodiode (PDx) inputs can
accommodate any photodiode with an input capacitance of less
than 100 pF. The ADPD105/ADPD106/ADPD107 are designed
to produce a high SNR for relatively low LED power while greatly
reducing the effect of ambient light on the measured signal.
tB (μs) = SLOTB_LED_OFFSET + nB × SLOTB_LED_PERIOD
where nB is the number of pulses for Time Slot B (Register 0x36,
Bits[15:8]).
Calculate the LED period using the following equation:
LED_PERIOD, minimum = 2 × AFE_WIDTH + 11
t1 and t2 are fixed and based on the computation time for each
slot. If a slot is not in use, these times do not add to the total
active time. Table 15 defines the values for these LED and
sampling time parameters.
ACTIVE
ACTIVE
tA
t1
tB
t2
nA PULSES
nB PULSES
SLEEP
TIME SLOT A
TIME SLOT B
1/f
SAMPLE
Figure 18. Time Slot Timing Diagram
Table 15. LED Timing and Sample Timing Parameters
Parameter
Register
Bits
Test Conditions/Comments
Min Typ Max Unit
SLOTA_LED_OFFSET1
0x30
0x35
0x31
0x36
[7:0] Delay from power-up to LEDA rising edge
[7:0] Delay from power-up to LEDB rising edge
[7:0] Time between LED pulses in Time Slot A; SLOTx_AFE_WIDTH = 4 ꢀs 19
[7:0] Time between LED pulses in Time Slot B; SLOTx_AFE_WIDTH = 4 ꢀs 19
Compute time for Time Slot A
23
23
63
63
63
63
μs
μs
μs
μs
μs
μs
μs
SLOTB_LED_OFFSET1
SLOTA_LED_PERIOD2
SLOTB_LED_PERIOD2
t1
t2
tSLEEP
68
20
Compute time for Time Slot B
Sleep time between sample periods
222
1 Setting the SLOTx_LED_OFFSET below the specified minimum value may cause failure of ambient light rejection for large photodiodes.
2 Setting the SLOTx_LED_PERIOD below the specified minimum value can cause invalid data captures.
Rev. A | Page 19 of 66
ADPD105/ADPD106/ADPD107
Data Sheet
TIME SLOT SWITCH
ADPD105 LFCSP Input Configurations
PD1
PD2
PD3
PD4
PD5
PD6
PD7
PD8
CH1
CH2
CH3
CH4
Up to eight photodiodes (PD1 to PD8) can be connected to the
ADPD105 in the LFCSP package. The photodiode anodes are
connected to the PD1 to PD8 input pins; the photodiode
cathodes are connected to the cathode pin, PDC. The anodes are
assigned in three different configurations depending on the
settings of Register 0x14 (see Figure 22, Figure 23, and Figure 21).
A switch sets which photodiode group is connected during Time
Slot A and Time Slot B. See Table 17 for the time slot switch
registers. It is important to leave any unused inputs floating for
proper operation of the devices. The photodiode inputs are
current inputs and as such, these pins are also considered to be
voltage outputs. Tying these inputs to a voltage may saturate the
analog block.
Register 0x14, PD1 to PD8 Input Configurations
PD1
CH1
PD2
INPUT CONFIGURATION FOR
REGISTER 0x14[11:8] = 4
REGISTER 0x14[7:4] = 4
Figure 20. PD5 to PD8 Connection
PD3
PD1
CH2
PD4
CH1
PD2
PD5
PD3
CH3
PD6
CH2
PD4
PD7
PD5
CH4
PD8
CH3
PD6
INPUT CONFIGURATION FOR
REGISTER 0x14[11:8] = 5
REGISTER 0x14[7:4] = 5
PD7
Figure 19. PD1 to PD4 Connection
CH4
PD8
INPUT CONFIGURATION FOR
REGISTER 0x14[11:8] = 1
REGISTER 0x14[7:4] = 1
Figure 21. 2 to 1 PD Current Summation
Rev. A | Page 20 of 66
Data Sheet
ADPD105/ADPD106/ADPD107
Table 16. Time Slot Switch (Register 0x14)
Address
Bits
Name
Description
0x14
[11:8]
SLOTB_PD_SEL
Selects connection of photodiode for Time Slot B as shown in Figure 22, Figure 23,
and Figure 21.
0x0: inputs are floating in Time Slot B.
0x1: all PDx pins (PD1 to PD8) are connected during Time Slot B.
0x4: PD5 to PD8 are connected during Time Slot B.
0x5: PD1 to PD4 are connected during Time Slot B.
Other: reserved.
[7:4]
SLOTA_PD_SEL
Selects connection of photodiode for Time Slot A as shown in Figure 22, Figure 23,
and Figure 21.
0x0: inputs are floating in Time Slot A.
0x1: all PDx pins (PD1 to PD8) are connected during Time Slot A.
0x4: PD5 to PD8 are connected during Time Slot A.
0x5: PD1 to PD4 are connected during Time Slot A.
Other: reserved.
WLCSP Input Configurations
Up to two photodiodes can be connected to four channels of the
ADPD105 and ADPD107 WLCSP models. The ADPD106
accommodates a single photodiode that can be connected to
Channel 1 and Channel 2. The photodiode anodes are connected
to the PD1-2 and PD3-4 input pins; the photodiode cathodes are
connected to the cathode pin, PDC. The anodes are assigned in
two different configurations depending on the settings of
Register 0x14 (see Figure 22 and Figure 23).
PD1-2
CH1
CH2
CH3
CH4
Register 0x14, PD1 to PD4 Input Configurations
Figure 22 shows the configuration where each of the PD inputs
is connected to two channels. This configuration is the high
dynamic range mode used for large photodiode currents. Figure 23
shows the configuration where each of the inputs to the device
is connected to a single channel. This mode allows the user to
maximize SNR for situations where lower photodiode currents are
expected.
PD3-4
A switch sets which photodiode group is connected during Time
Slot A and Time Slot B. See Table 17 for the time slot switch
registers. It is important to leave any unused inputs floating for
proper operation of the devices. The photodiode inputs are
current inputs and, as such, these pins are also considered to be
voltage outputs. Tying these inputs to a voltage may saturate the
analog block.
INPUT CONFIGURATION FOR
REGISTER 0x14[11:8] = 5
REGISTER 0x14[7:4] = 5
Figure 22. PD1 to PD4 Connection
PD1-2
CH1
Note that the ADPD106 only includes the options shown for the
PD1-2 input pin shown in Figure 22 and Figure 23.
PD3-4
CH2
INPUT CONFIGURATION FOR
REGISTER 0x14[11:8] = 1
REGISTER 0x14[7:4] = 1
Figure 23. Current Summation—Two Photodiodes Summed into One Current
Rev. A | Page 21 of 66
ADPD105/ADPD106/ADPD107
Data Sheet
Table 17. Time Slot Switch (Register 0x14)
Address
Bits
Name
Description
0x14
[11:8]
SLOTB_PD_SEL
Selects connection of photodiode forTime Slot B as shown in Figure 22 and Figure 23.
0x0: inputs are floating in Time Slot B.
0x1: PD1-2 is connected to Channel 1; PD3-4 is connected to Channel 2 during Time Slot B.
0x5: PD1-2 is connected to Channel 1 and Channel 2; PD3-4 is connected to Channel 3
and Channel 4 during Time Slot B.
Other: reserved.
[7:4]
SLOTA_PD_SEL
Selects connection of photodiode for Time Slot A as shown in Figure 22 and Figure 23.
0x0: inputs are floating in Time Slot A.
0x1: PD1-2 is connected to Channel 1; PD3-4 is connected to Channel 2 during Time Slot A.
0x5: PD1-2 is connected to Channel 1 and Channel 2; PD3-4 is connected to Channel 3
and Channel 4 during Time Slot A.
Other: reserved.
appropriate input buffer using Register 0x4F, Bit 1, for the
GPIO0 pin, or Register 0x4F, Bit 5, for the GPIO1 pin.
3. Write 0x4000 to Register 0x38.
4. Write 0x2 to Register 0x10 to start the sampling operations.
5. Apply the external sync signal on the selected pin at the
desired rate; sampling occurs at that rate. As with normal
sampling operations, read the data using the FIFO or the
data registers.
ADJUSTABLE SAMPLING FREQUENCY
Register 0x12 controls the sampling frequency setting of the
ADPD105/ADPD106/ADPD107 and Register 0x4B, Bits[5:0]
further tunes this clock for greater accuracy. The sampling
frequency is governed by an internal 32 kHz sample rate clock
that also drives the trans-ition of the internal state machine. The
maximum sampling frequencies for some sample conditions
are listed in Table 3. The maximum sample frequency for all
conditions is determined by the following equation:
The maximum frequency constraints also apply in this case.
f
SAMPLE, MAX = 1/(tA + t1 + tB + t2 + tSLEEP, MIN)
Providing an External 32kHz Clock
where tSLEEP, MIN is the minimum sleep time required between
samples.
The ADPD105/ADPD106/ADPD107 have an option for the
user to provide an external 32 kHz clock to the devices for
system synchronization or for situations where a clock with
better accuracy than the internal 32 kHz clock is required. The
external 32 kHz clock is provided on the GPIO1 pin. To enable
the 32 kHz external clock, use the following procedure at startup:
If a given time slot is not in use, elements from that time slot do
not factor into the calculation. For example, if Time Slot A is
not in use, tA and t1 do not add to the sampling period and the
new maximum sampling frequency is calculated as follows:
1. Drive the GPIO1 pin to a valid logic level or with the
desired 32 kHz clock prior to enabling the GPIO1 pin as an
input. Do not leave the pin floating prior to enabling it.
2. Write 01 to Register 0x4F, Bits[6:5] to enable the GPIO1 pin as
an input.
f
SAMPLE, MAX = 1/(tB + t2 + tSLEEP, MIN)
See the Dual Time Slot Operation section for the definitions of
tA, t1, tB, and t2.
External Sync for Sampling
3. Write 10 to Register 0x4B, Bits[8:7] to configure the devices to
use an external 32 kHz clock. This setting disables the
internal 32 kHz clock and enables the external 32 kHz clock.
4. Write 0x1 to Register 0x10 to enter program mode.
5. Write additional control registers in any order while the
devices are in program mode to configure the devices as
required.
The ADPD105/ADPD106/ADPD107 provide an option to use
an external sync signal to trigger the sampling periods. This
external sample sync signal can be provided either on the
GPIO0 pin or the GPIO1 pin. This functionality is controlled
by Register 0x4F, Bits[3:2]. When enabled, a rising edge on
the selected input specifies when the next sample cycle occurs.
When triggered, there is a delay of one to two internal sampling
clock (32 kHz) cycles, and then the normal start-up sequence
occurs. This sequence is the same as when the normal sample
timer providesthe trigger. To enable the external sync signal
feature, use the following procedure:
6. Write 0x2 to Register 0x10 to start the normal sampling
operation.
1. Write 0x1 to Register 0x10 to enter program mode.
2. Write the appropriate value to Register 0x4F, Bits[3:2] to
select whether the GPIO0 pin or the GPIO1 pin specifies
when the next sample cycle occurs. Also, enable the
Rev. A | Page 22 of 66
Data Sheet
ADPD105/ADPD106/ADPD107
when writing registers or changing modes. Because no power
cycling occurs in this mode, the devices may consume higher
current in program mode than in normal operation. To place
the devices in program mode, write 0x1 to Register 0x10, Bits[1:0].
STATE MACHINE OPERATION
During each time slot, the ADPD105/ADPD106/ADPD107
operate according to a state machine. The state machine
operates in the following sequence, shown in Figure 24.
In normal operation, the ADPD105/ADPD106/ADPD107 pulse
light and collect data. Power consumption in this mode depends
on the pulse count and data rate. To place the devices in normal
sampling mode, write 0x2 to Register 0x10, Bits[1:0].
STANDBY
REGISTER 0x10 = 0x0000
ULTRALOW POWER MODE
NO DATA COLLECTION
ALL REGISTER VALUES ARE RETAINED.
NORMAL MODE OPERATION AND DATA FLOW
In normal mode, the ADPD105/ADPD106/ADPD107 follow a
specific pattern set up by the state machine. This pattern is shown
in the corresponding data flow in Figure 25. The pattern is as
follows:
PROGRAM
REGISTER 0x10 = 0x0001
SAFE MODE FOR PROGRAMING REGISTERS
NO DATA COLLECTION
DEVICE IS FULLY POWERED IN THIS MODE.
1. LED pulse and sample. The ADPD105/ADPD106/ADPD107
pulse external LEDs. The response of a photodiode or photo-
diodes to the reflected light is measured by the ADPD105/
ADPD106/ADPD107. Each data sample is constructed from
the sum of n individual pulses, where n is user configurable
between 1 and 255.
2. Intersample averaging. If desired, the logic can average n
samples, from 2 to 128 in powers of 2, to produce output
data. New output data is saved to the output registers every
N samples.
3. Data read. The host processor reads the converted results
from the data register or the FIFO.
4. Repeat. The sequence has a few different loops that enable
different types of averaging while keeping both time slots
close in time relative to each other.
NORMAL OPERATION
REGISTER 0x10 = 0x0002
LEDs ARE PULSED AND PHOTODIODES ARE SAMPLED
STANDARD DATA COLLECTION
DEVICE POWER IS CYCLED BY INTERNAL STATE MACHINE.
Figure 24. State Machine Operation Flowchart
The ADPD105/ADPD106/ADPD107 operate in one of three
modes: standby, program, and normal sampling mode.
Standby mode is a power saving mode in which no data collection
occurs. All register values are retained in this mode. To place
the devices in standby mode, write 0x0 to Register 0x10,
Bits[1:0]. The devices power up in standby mode.
Program mode is used for programming registers. Always cycle
the ADPD105/ADPD106/ADPD107 through program mode
[14 + LOG
(
n
× N )] BITS
A A
2
UP TO 27 BITS
[14 + LOG (n )] BITS
UP TO 20 BITS
N
2
A
A
16-BIT CLIP
IF VAL ≤ (2 – 1)
VAL = VAL
16 BITS
16
N
A
16
ELSE VAL = 2 – 1
1
14 BITS
14 BITS
n
A
20-BIT CLIP
IF VAL ≤ (2 – 1)
20
n
A
n
A
VAL = VAL
20
ELSE VAL = 2 – 1
14-BIT
ADC
0
1
1
REGISTER
0x11[13]
[14 + LOG (n )] BITS
2
A
16-BIT
DATA
REGISTERS
32-BIT DATA
REGISTERS
UP TO 22 BITS
FIFO
ADC OFFSET
SAMPLE 1: TIME SLOT A
SAMPLE 1: TIME SLOT B
0
1
SAMPLE N : TIME SLOT A
A
SAMPLE N : TIME SLOT B
B
TIME SLOT A
TIME SLOT B
N
B
16-BIT CLIP
16
IF VAL ≤ (2 – 1)
NOTES
1. n AND n = NUMBER OF LED PULSES FOR TIME SLOT A AND TIME SLOT B.
N
B
VAL = VAL
A
B
[14 + LOG (n )] BITS
UP TO 20 BITS
16
16 BITS
2
B
ELSE VAL = 2 – 1
1
2. N AND N = NUMBER OF AVERAGES FOR TIME SLOT A AND TIME SLOT B.
A
B
[14 + LOG (n × N )] BITS
2 B B
UP TO 27 BITS
Figure 25. ADPD105/ADPD106/ADPD107 Datapath
Rev. A | Page 23 of 66
ADPD105/ADPD106/ADPD107
Data Sheet
LED Pulse and Sample
data is still acquired by the AFE at the sampling frequency,
f
SAMPLE (Register 0x12), but new data is written to the registers at
At each sampling period, the selected LED driver drives a series
of LED pulses, as shown in Figure 26. The magnitude, duration,
and number of pulses are programmable over the I2C interface.
Each LED pulse coincides with a sensing period so that the sensed
value represents the total charge acquired on the photodiode in
response to only the corresponding LED pulse. Charge, such as
ambient light, that does not correspond to the LED pulse is
rejected.
the rate of fSAMPLE/N every Nth sample. This new data consists of
the sum of the previous N samples. The full 32-bit sum is stored
in the 32-bit registers. However, before sending this data to the
FIFO, a divide by N operation occurs. This divide operation
maintains bit depth to prevent clipping on the FIFO.
Use this between sample averaging to lower the noise while
maintaining 16-bit resolution. If the pulse count registers are
kept to 8 or less, the 16-bit width is never exceeded. Therefore,
when using Register 0x15 to average subsequent pulses, many
pulses can be accumulated without exceeding the 16-bit word
width. This averaging can reduce the number of FIFO reads
required by the host processor.
After each LED pulse, the photodiode output relating the pulsed
LED signal is sampled and converted to a digital value by the
14-bit ADC. Each subsequent conversion within a sampling
period is summed with the previous result. Up to 255 pulse values
from the ADC can be summed in an individual sampling period.
There is a 20-bit maximum range for each sampling period.
Data Read
Averaging
The host processor reads output data from the ADPD105/
ADPD107 via the I2C protocol on the ADPD105 or the SPI port
on the ADPD107. Data is read from the data registers or from
the FIFO. New output data is made available every N samples,
where N is the user configured averaging factor. The averaging
factors for Time Slot A and Time Slot B are configurable
independently of each other. If they are the same, both time slots
can be configured to save data to the FIFO. If the two averaging
factors are different, only one time slot can save data to the FIFO;
data from the other time slot can be read from the output
registers.
The ADPD105/ADPD106/ADPD107 offer sample accumulation
and averaging functionality to increase signal resolution.
Within a sampling period, the AFE can sum up to 256 sequential
pulses. As shown in Figure 25, samples acquired by the AFE are
clipped to 20 bits at the output of the AFE. Additional resolution,
up to 27 bits, can be achieved by averaging between sampling
periods. This accumulated data of N samples is stored as 27-bit
values and can be read out directly by using the 32-bit output
registers or the 32-bit FIFO configuration.
When using the averaging feature set up by Register 0x15,
subsequent pulses can be averaged by powers of 2. The user can
select from 2, 4, 8 … up to 128 samples to be averaged. Pulse
The data read operations are described in more detail in the
Reading Data section.
SHOWN WITH f
= 10 Hz
SAMPLE
OPTICAL SAMPLING
LOCATIONS
0
0.5
1.0
1.5
2.0
2.5
3.0
TIME (s)
LED
CURRENT
(I
)
LED
NUMBER OF LED PULSES (nA OR nB
)
Figure 26. Example of a Photoplethysmography (PPG) Signal Sampled at a Data Rate of 10 Hz Using Five Pulses per Sample
Rev. A | Page 24 of 66
Data Sheet
ADPD105/ADPD106/ADPD107
important to verify proper AFE position for every new hardware
design or the LED width.
AFE OPERATION
The timing within each pulse burst is important for optimizing
the operation of the ADPD105/ADPD106/ADPD107. Figure 27
shows the timing waveforms for a single time slot as an LED pulse
response propagates through the analog block of the AFE. The
first graph, shown in green, shows the ideal LED pulsed output.
The filtered LED response, shown in blue, shows the output of
the analog integrator. The third graph, shown in orange, illustrates
an optimally placed integration window. When programmed to
the optimized value, the full signal of the filtered LED response
can be integrated. The AFE integration window is then applied to
the output of the band-pass filter (BPF) and the result is sent to
the ADC and summed for N pulses. If the AFE window is not
correctly sized or located, all of the receive signal is not properly
reported and system performance is not optimal; therefore, it is
AFE INTEGRATION OFFSET ADJUSTMENT
The AFE integration width must be equal or larger than the LED
width. As AFE width increases, the output noise increases and
the ability to suppress high frequency content from the environ-
ment decreases. It is therefore desirable to keep the AFE integration
width small. However, if the AFE width is too small, the LED
signal is attenuated. With most hardware selections, the AFE
width produces the optimal SNR at 1 ꢀs more than the LED
width. After setting LED width, LED offset, and AFE width, the
ADC offset can then be optimized. The AFE offset must be
manually set such that the falling edge of the first segment of the
integration window matches the zero crossing of the filtered LED
response.
LED WIDTH
REGISTER 0x30 [12:8]
REGISTER 0x35 [12:8]
N LED PULSES
REGISTER 0x31 [15:8]
LED PERIOD
REGISTER 0x36 [15:8]
REGISTER 0x31 [7:0]
LED DRIVE STRENGTH
LED OFFSET
REGISTER 0x36 [7:0]
REGISTER 0x30 [7:0]
REGISTER 0x35 [7:0]
LED PULSE
REGISTER 0x22, REGISTER 0x23,
REGISTER 0x24, REGISTER 0x25
FOR N PULSES
FOR N PULSES
CONTROLLED BY:
TIA SETTINGS
AFE SETTINGS
REGISTER 0x42,
REGISTER 0x43,
REGISTER 0x44,
REGISTER 0x45
FILTERED LED
RESPONSE
9µs + AFE OFFSET
REGISTER 0x39 [10:0]
REGISTER 0x3B [10:0]
AFE INTEGRATION
WINDOW
FOR N PULSES
AFE
WIDTH
REGISTER 0x39 [15:11]
REGISTER 0x3B [15:11]
AFE
WIDTH
ADC
CONVERSION
ADC
CONVERSION
+
TIME (µs)
Figure 27. AFE Operation Diagram
Rev. A | Page 25 of 66
ADPD105/ADPD106/ADPD107
Data Sheet
AFE Integration Offset Starting Point
The register can be thought of as 211 − 1 of these 31.25 ns steps, or
it can be broken into an AFE coarse setting using Bits[10:5] to
represent 1 μs steps and Bits[4:0] to represent 31.25 ns steps.
Sweeping the AFE position from the starting point to find a
local maximum is the recommended way to optimize the AFE
offset. The setup for this test is to allow the LED light to fall on
the photodiode in a static way. This test is typically done with a
reflecting surface at a fixed distance. The AFE position can then
be swept to look for changes in the output level. When adjusting
the AFE position, it is important to sweep the position using the
31.25 ns steps. Typically, a local maximum is found within 2 μs
of the starting point for most systems. Figure 28 shows an example
of an AFE sweep, where 0 on the x-axis represents the AFE starting
point defined previously. Each data point in Figure 28 corresponds
to one 31.25 ns step of the SLOTx_AFE_OFFSET. The optimal
location for SLOTx_AFE_OFFSET in this example is 0.687 μs
from the AFE starting point.
The starting point of this offset, as expressed in microseconds, is set
such that the falling edge of the integration window aligns with the
falling edge of the LED.
LED_FALLING_EDGE = SLOTx_LED_OFFSET +
SLOTx_LED_WIDTH
and,
AFE_INTEGRATION_FALLING_EDGE = 9 +
SLOTx_AFE_OFFSET + SLOTx_AFE_WIDTH
If both falling edges are set equal to each other, solve for
SLOTx_AFE_OFFSET to obtain the following equation:
AFE_OFFSET_STARTING_POINT = SLOTx_LED_
OFFSET + SLOTx_LED_WIDTH − 9 – SLOTx_AFE_
WIDTH
Setting the AFE offset to any point in time earlier than the
starting point is equivalent to setting the integration in the
future; the AFE cannot integrate the result from an LED pulse
that has not yet occurred. As a result, a SLOTx_AFE_OFFSET
value less than the AFE_OFFSET_STARTING_POINT value is an
erroneous setting. Such a result may indicate that current in the
TIA is operating in the reverse direction from intended, where
the LED pulse is causing the current to leave the TIA rather
than enter it.
100
0.687
95
90
85
80
75
Because, for most setups, the SLOTx_AFE_WIDTH is 1 μs
wider than the SLOTx_LED_WIDTH, the AFE_OFFSET_
STARTING_POINT value is typically 10 μs less than the
SLOTx_LED_OFFSET value. Any value less than SLOTx_LED_
OFFSET – 10 is erroneous. The optimal AFE offset is some time
after the AFE_OFFSET_STARTING_POINT value. The band-
pass filter response, LED response, and photodiode response
each add some delay. In general, the component choice, board
layout, SLOTx_LED_OFFSET, and SLOTx_LED_WIDTH are
the variables that can change the SLOTx_AFE_OFFSET value.
After a specific design is set, the SLOTx_AFE_OFFSET value
can be locked down and does not need to be optimized further.
0
0.15 0.30 0.45 0.60 0.75 0.90 1.05 1.20 1.35 1.50
AFE OFFSET FROM STARTING POINT (µs)
Figure 28. AFE Sweep Example
Table 18 lists some typical LED and AFE values after optimization.
In general, it is not recommended to use the SLOTx_AFE_OFFSET
numbers in Table 18 without first verifying them against the AFE
sweep method. Repeat this method for every new LED width and
with every new set of hardware made with the ADPD105/
ADPD107. For maximum accuracy, it is recommended that the
32 MHz clock be calibrated prior to sweeping the AFE.
Sweeping the AFE Position
The AFE offsets for Time Slot A and Time Slot B are controlled
by Bits[10:0] of Register 0x39 and Register 0x3B, respectively.
Each LSB represents one cycle of the 32 MHz clock, or 31.25 ns.
Table 18. AFE Window Settings
LED Register 0x30 or Register 0x35
AFE Register 0x39 or Register 0x3B
Comment
0x0219
0x0319
0x1A08
0x21FE
2 μs LED pulse, 3 μs AFE width, 25 μs LED delay
3 μs LED pulse, 4 μs AFE width, 25 μs LED delay
Rev. A | Page 26 of 66
Data Sheet
ADPD105/ADPD106/ADPD107
I2C SERIAL INTERFACE
4. Write the desired SLAVE_ADDRESS bits using Register 0x09,
Bits[7:1]. While writing to Register 0x09, Bits[7:1], write
0xAD to Register 0x09, Bit[15:8]. Register 0x09 must be
written to immediately after writing to Register 0x0D.
5. Repeat Step 1 to Step 4 for all the devices that need
SLAVE_ADDRESS changed.
The ADPD105 supports an I2C serial interface via the SDA
(data) and SCL (clock) pins. All internal registers are accessed
through the I2C interface. The ADPD105 is an I2C only device
and does not support an SPI.
The ADPD105 conforms to the UM10204 I2C-Bus Specification
and User Manual, Rev. 05—9 October 2012, available from NXP
Semiconductors. It supports a fast mode (400 kbps) data transfer.
Register read and write are supported, as shown in Figure 29.
Figure 3 shows the timing diagram for the I2C interface.
6. Set the GPIO0 and GPIO1 pins as desired for normal
operation using the new SLAVE_ADDRESS for each device.
I2C Write and Read Operations
Figure 29 illustrates the ADPD105 I2C write and read operations.
Single-word and multiword read operations are supported. For
a single register read, the host sends a no acknowledge (NACK)
after the second data byte is read and a new register address is
needed for each access.
Slave Address
The default 7-bit I2C slave address for the device is 0x64,
2
W
followed by the R/ bit. For a write, the default I C slave
address is 0xC8; for a read, the default I2C address is 0xC9. The
slave address is configurable by writing to Register 0x09, Bits[7:1].
When multiple ADPD105 devices are on the same bus lines, the
GPIO0 and GPIO1 pins can be used to select specific devices
for the address change. Register 0x0D can be used to select a
key to enable address changes in specific devices. Use the
following procedure to change the slave address when multiple
ADPD105 devices are connected to the same I2C bus lines:
For multiword operations, each pair of data bytes is followed by
an acknowledge from the host until the last byte of the last word
is read. The host indicates the last read word by sending a no
acknowledge. When reading from the FIFO (Register 0x60), the
data is automatically advanced to the next word in the FIFO
and the space is freed. When reading from other registers, the
register address is automatically advanced to the next register,
except at Register 0x5F or Register 0x7F, where the address does
not increment. This autoincrementing allows lower overhead
reading of sequential registers.
1. Using Register 0x4F, enable the input buffer of the GPIO1 pin,
the GPIO0 pin, or both, depending on the key being used.
2. For the device identified as requiring an address change,
set the GPIO0 and/or GPIO1 pins high or low to match the
key being used.
All register writes are single word only and require 16 bits (one
word) of data.
3. Write the SLAVE_ADDRESS_KEY bits using Register 0x0D,
Bits[15:0] to match the desired function. The allowed keys
are shown in Table 33.
The software reset (Register 0x0F, Bit 0) returns an acknowledge.
The device then returns to standby mode with all registers in
the default state.
Table 19. Definition of I2C Terminology
Term
Description
SCL
Serial clock.
SDA
Serial address and data.
Master
Slave
Start (S)
Start (Sr)
Stop (P)
ACK
The master is the device that initiates a transfer, generates clock signals, and terminates a transfer.
The slave is the device addressed by a master. The ADPD105 operates as a slave device.
A high to low transition on the SDA line while SCL is high; all transactions begin with a start condition.
Repeated start condition.
A low to high transition on the SDA line while SCL is high. A stop condition terminates all transactions.
During the acknowledge or no acknowledge clock pulse, the SDA line is pulled low and remains low.
During the acknowledge or no acknowledge clock pulse, the SDA line remains high.
After a start (S), a 7-bit slave address is sent, which is followed by a data direction bit (read or write).
A 1 indicates a request for data.
NACK
Slave Address
Read (R)
Write (W)
A 0 indicates a transmission.
Rev. A | Page 27 of 66
ADPD105/ADPD106/ADPD107
Data Sheet
2
I
C WRITE
REGISTER WRITE
MASTER START
SLAVE
SLAVE ADDRESS + WRITE
REGISTER ADDRESS
DATA[15:8]
DATA[7:0]
STOP
ACK
ACK
ACK
ACK
2
I
C SINGLE WORD READ MODE
REGISTER READ
MASTER START
SLAVE
ACK
SLAVE ADDRESS + WRITE
REGISTER ADDRESS
REGISTER ADDRESS
Sr
Sr
SLAVE ADDRESS + READ
SLAVE ADDRESS + READ
NACK
STOP
STOP
ACK
ACK
DATA[15:8]
DATA[15:8]
ACK
ACK
ACK
ACK
DATA[7:0]
2
I
C MULTIWORD READ MODE
REGISTER READ
MASTER START
SLAVE
ACK/NACK
SLAVE ADDRESS + WRITE
ACK
DATA[7:0]
DATA TRANSFERRED
n (DATA[15:8]+ACK+DATA[7:0] + ACK/NACK)
NOTES
1. THE SHADED AREAS REPRESENT WHEN THE DEVICE IS LISTENING.
Figure 29. I2C Write and Read Operations
SPI PORT
The first byte written in a SPI transaction is a 7-bit address,
which is the location of the address being accessed, followed by
The ADPD106 and ADPD107 are SPI only devices. They do not
support the I2C interface. The SPI port uses a 4-wire interface,
the W/ bit. This bit determines whether the communication is
R
CS
consisting of the , MOSI, MISO, and SCLK signals, and it is
a write (Logic Level 1) or a read (Logic Level 0). This format is
shown in Table 21.
CS
always a slave port. The
signal goes low at the beginning of a
transaction and high at the end of a transaction. The SCLK signal
latches MOSI on a low to high transition. The MISO data is shifted
out of the device on the falling edge of SCLK and must be clocked
into a receiving device, such as a microcontroller, on the SCLK
rising edge. The MOSI signal carries the serial input data, and
the MISO signal carries the serial output data. The MISO signal
remains three state until a read operation is requested, which
allows other SPI-compatible peripherals to share the same MISO
line. All SPI transactions have the same basic format shown in
Table 20. A timing diagram is shown in Figure 4. Write all data
MSB first.
R
Bit 4
Table 21. SPI Address and Write/ Byte Format
Bit 0
Bit 1
Bit 2
Bit 3
Bit 5
Bit 6
Bit 7
A6
A5
A4
A3
A2
A1
A0
W/R
The format for the SPI communications slave port is commonly
known as SPI Mode 3, where clock polarity (CPOL) = 1 and
clock phase (CPHA) = 1 (see Figure 30). The base value of the
clock is 1. Data is captured on the rising edge of the clock, and
data is propagated on the falling edge.
The maximum read and write speed for the SPI slave port is
10 MHz.
Table 20. Generic Control Word Sequence
Byte 0
Byte 1
Byte 2
Subsequent Bytes
Address[6:0], Data[15:8]
W/R
Data[7:0]
Data[15:8], Data[7:0]
SCLK
(CPOL = 1)
CS
1
1
1
2
2
2
3
3
3
4
4
4
5
5
5
6
6
6
7
7
7
8
8
8
CYCLE #
MISO
Z
Z
Z
Z
(CPHA = 1)
MOSI
Figure 30. Clock Polarity and Phase for the SPI Slave Port (CPOL = 1, CPHA = 1)
Rev. A | Page 28 of 66
Data Sheet
ADPD105/ADPD106/ADPD107
A sample timing diagram for a multiple word SPI write operation
to a register is shown in Figure 31. A sample timing diagram of
a single-word SPI read operation is shown in Figure 32. The
MISO pin transitions from being three-state to being driven
read operation is shown in Figure 33. In Figure 31 to Figure 33,
rising edges on SCLK are indicated with an arrow, signifying
that the data lines are sampled on the rising edge.
When performing multiple word reads or writes, the data
address is automatically incremented to the next consecutive
address for subsequent transactions except for Address 0x5F,
Address 0x60 (FIFO), and Address 0x7F.
R
following the reception of a valid bit. In this example, Byte 0
R
contains the address and the W/ bit, and subsequent bytes
carry the data. A sample timing diagram of a multiple word SPI
0
1
2
3
4
5
6
7
8
9 10 11 12 13 14 15 16 17 18 19 20 21 22
CS
SCLK
MOSI
ADDRESS[6:0]
DATA BYTE 1
DATA BYTE 2
DATA BYTE N
W/R
Figure 31. SPI Slave Write Clocking (Burst Write Mode, N Bytes)
0
1
2
3
4
5
6
7
8
9 10 11 12 13 14 15 16 17 18 19 20 21 22 23
CS
SCLK
MOSI
MISO
ADDRESS[6:0]
W/R
DATA BYTE 1
DATA BYTE 2
Figure 32. SPI Slave Read Clocking (Single-Word Mode, Two Bytes)
0
1
2
3
4
5
6
7
8
9 10 11 12 13 14 15 16 17 18 19 20 21 22
CS
SCLK
MOSI
MISO
ADDRESS[6:0]
W/R
DATA BYTE 1
DATA BYTE 2
DATA BYTE N
Figure 33. SPI Slave Read Clocking (Burst Read Mode, N Bytes)
Rev. A | Page 29 of 66
ADPD105/ADPD106/ADPD107
Data Sheet
1.8V
TYPICAL CONNECTION DIAGRAM
ADPD105
Figure 34 shows a typical circuit used for wrist-based heart rate
measurement with the ADPD105 WLCSP using a green LED.
The 1.8 V I2C communication lines, SCL and SDA, along with
the GPIO0 and GPIO1 lines, connect to a system
DVDD
PDC
0.1µF
1.8V
AVDD
0.1µF
PD1-2
PD3-4
AGND
DGND
LGND
VLED
microprocessor or sensor hub. The I2C signals can have pull-up
resistors connected to a 1.8 V or a 3.3 V power supply. The
GPIO0 and GPIO1 signals are only compatible with a 1.8 V
supply and may need a level translator. The circuit shown in
Figure 34 is identical for the ADPD107, except the I2C interface
is replaced by an SPI. There are multiple ways to connect
photodiodes to the 8-channel ADPD105 LFCSP, as shown in
Table 22 and Figure 37. The photodiode anodes are connected
to the PD1 to PD8 input pins and the photodiode cathodes are
connected to the cathode pin, PDC.
10kΩ 10kΩ
4.7µF
SCL
SDA
TO DIGITAL
INTERFACE
LEDX1
VREF
GPIO0
GPIO1
1µF
Figure 34. Typical Wrist-Based HRM Measurement
1
2
3
A
B
LGND
LEDX2
Provide the 1.8 V supply, VDD, to AVDD and DVDD. The LED
supply uses a standard regulator circuit according to the peak
current requirements specified in Table 3 and calculated in the
LED Driver Pins and LED Supply Voltage section.
SDA
LEDX3
SCL
LEDX1
GPIO0
DGND
VREF
C
D
E
F
DVDD
AGND
For best noise performance, connect AGND, DGND, and
LGND together at a large conductive surface, such as a ground
plane, a ground pour, or a large ground trace.
The number of photodiodes or LEDs used varies depending on
the application as well as the dynamic range and SNR required.
For example, in an application where a single, large photodiode
is used, the dynamic range can be increased by splitting the current
between multiple inputs. This is achieved by connecting the anode
of the photodiode to multiple channels so that the current can
be split evenly among the number of channels connected,
effectively increasing the dynamic range over a single channel
configuration. Alternatively, in situations where the photodiode is
small or the signal is greatly attenuated, SNR can be maximized
by connecting the anode of the photodiode to just a single channel.
It is important to leave the unused input floating for proper device
operation.
GPIO1
PD1-2
AVDD
PD3-4
PDC
Figure 35. ADPD105 Connection and PCB Layout Diagram (Top View)
1
2
3
A
B
LEDX2
LGND
Figure 35 and Figure 36 show the recommended connection
diagram and printed circuit board (PCB) layout for the
ADPD105 and ADPD107, respectively. The current input pins,
PD1-2 and PD3-4, have a typical voltage of 1.3 V during the
sampling period. During the sleep period, these pins are
connected to the cathode pin. The cathode and anode voltages
are listed in Table 3.
GPIO0
DGND
SCLK
VREF
PD3-4
LEDX3
GPIO1
LEDX1
MISO
C
MOSI
D
E
CS
AGND
AVDD
PD1-2
F
PDC
Figure 36. ADPD107 Connection and PCB Layout Diagram, Dashed Line
Traces From Blind Vias (Top View)
Rev. A | Page 30 of 66
Data Sheet
ADPD105/ADPD106/ADPD107
PD1
7
PD1
PD1
PD1
PD8
7
7
7
15
14
15
14
15
14
15
14
8
8
8
8
PD1
7
PD8
7
15
14
15
14
PD1
PD8
7
15
14
8
8
8
Figure 37. Photodiode Configuration Options for the ADPD105 LFCSP
Table 22. Typical Photodiode Anode to Input Channel Connections for the ADPD105 LFCSP1, 2
Input Channel
Photodiode Anode Configuration
PD1
D1
NC
D1
NC
D1
D1
D1
NC
D1
PD2
NC
NC
D1
NC
NC
D1
D2
NC
D2
PD3
NC
NC
D1
NC
NC
D1
D3
NC
D3
PD4
NC
NC
D1
NC
NC
D1
D4
NC
D4
PD5
NC
D1
NC
D1
D2
D2
NC
D1
D5
PD6
NC
NC
NC
D1
NC
D2
NC
D2
D6
PD7
NC
NC
NC
D1
NC
D2
NC
D3
D7
PD8
NC
NC
NC
D1
NC
D2
NC
D4
D8
Single Photodiode (PD1)
Two Photodiodes (PD1, PD2)
Four Photodiodes (PD1 to PD4)
Eight Photodiodes (PD1 to PD8)
1 Dx refers to the diode connected to the specified channel.
2 NC means do not connect under the conditions provided in Table 21. Leave all unused inputs floating.
requirements for the bypass capacitor (CVLED) and the supply
voltages of the LEDs (VLEDx).
LED DRIVER PINS AND LED SUPPLY VOLTAGE
The LEDX1, LEDX2, and LEDX3 pins have an absolute maximum
voltage rating of 3.6 V. Any voltage exposure over this rating affects
the reliability of the device operation and, in certain circumstances,
causes the device to cease proper operation. The voltage of the
LEDx pins must not be confused with the supply voltages for the
LED themselves (VLEDx). VLEDx is the voltage applied to the anode of
the external LED, whereas the LEDXx pin is the input of the
internal current driver, and the pins are connected to the
cathode of the external LED.
ADPD105/
ADPD106/
V
LEDx
SUPPLY
ADPD107
C
VLED
LGND
LEDx
Figure 38. VLEDx Supply Schematic
DETERMINING THE AVERAGE CURRENT
The ADPD105/ADPD106/ADPD107 drive an LED in a series of
short pulses. Figure 39 shows the typical ADPD105/ADPD106/
ADPD107 configuration of a pulse burst sequence.
19µs
LED DRIVER OPERATION
The LED driver for the ADPD105/ADPD106/ADPD107 is a
current sink. The compliance voltage, measured at the driver
pin with respect to ground, required to maintain the
3µs
programmed LED current level is a function of the current
required. Figure 15 shows the typical compliance voltages
required at the various coarse LED settings. Figure 38 shows the
basic schematic of how the ADPD105/ADPD106/ADPD107
connect to an LED through the LED driver. The Determining the
Average Current and the Determining CVLED sections define the
I
LED_MAX
Figure 39. Typical LED Pulse Burst Sequence Configuration
Rev. A | Page 31 of 66
ADPD105/ADPD106/ADPD107
Data Sheet
In this example, the LED pulse width, tLED_PULSE, is 3 μs, and the
LED pulse period, tLED_PERIOD, is 19 μs. The LED being driven is a
pair of green LEDs driven to a 250 mA peak. The goal of CVLED
is to buffer the LED between individual pulses. In the worst case
scenario, where the pulse train shown in Figure 39 is a continuous
sequence of short pulses, the VLEDx supply must supply the
average current. Therefore, calculate ILED_AVERAGE as follows:
To correctly size the CVLED capacitor, do not deplete it during the
pulse of the LED to the point where the voltage on the capacitor is
less than the forward bias on the LED. To calculate the minimum
value for the VLEDx bypass capacitor, use the following equation:
tLED_PULSE ILED_ MAX
(2)
CVLED
VLED_ MIN (VFB _ LED_ MAX 0.2)
where:
LED_PULSE is the LED pulse width.
LED_MAX is the maximum forward-biased current on the LED
used in operating the device.
I
LED_AVERAGE = (tLED_PULSE/tLED_PERIOD) × ILED_MAX
where:
LED_AVERAGE is the average current needed from the VLEDx supply
(1)
t
I
I
during the pulse period, and it is also the VLEDx supply current
rating.
LED_MAX is the peak current setting of the LED.
For the numbers shown in Equation 1, ILED_AVERAGE = 3/19 ×
LED_MAX. For typical LED timing, the average VLEDx supply
V
V
LED_MIN is the lowest voltage from the VLEDx supply with no load.
FB_LED_MAX is the maximum forward-biased voltage required on
I
the LED to achieve ILED_MAX
.
The numerator of the CVLED equation sets up the total discharge
amount in coulombs from the bypass capacitor to satisfy a
single programmed LED pulse of the maximum current. The
denominator represents the difference between the lowest voltage
from the VLEDx supply and the LED required voltage. The LED
required voltage is the voltage of the anode of the LED such that
the 0.2 V compliance of the LED driver and the forward-biased
voltage of the LED operating at the maximum current is satisfied.
For a typical ADPD105/ADPD106/ADPD107 example, assume
that the lowest value for the VLEDx supply is 4.4 V, and that the
peak current is 250 mA for two 528 nm LEDs in parallel. The
minimum value for CVLED is then equal to 3 μF.
I
current is 3/19 × 250 mA = 39.4 mA, indicating that the VLEDx
supply must support a dc current of 40 mA.
DETERMINING CVLED
To determine the CVLED capacitor value, determine the maximum
forward-biased voltage, VFB_LED_MAX, of the LED in operation.
The LED current, ILED_MAX, converts to VFB_LED_MAX as shown in
Figure 40. In this example, 250 mA of current through two green
LEDs in parallel yields VFB_LED_MAX = 3.95 V. Any series resistance
in the LED path must also be included in this voltage. When
designing the LED path, keep in mind that small resistances can
add up to large voltage drops due to the LED peak current being
very large. In addition, these resistances can be unnecessary
constraints on the VLEDx supply.
C
VLED = (3 × 10−6 × 0.250)/(4.4 – (3.95 + 0.2)) = 3 μF
(3)
As shown in the Equation 3, as the minimum supply voltage drops
close to the maximum anode voltage, the demands on CVLED
become more stringent, forcing the capacitor value higher. It is
important to insert the correct values into Equation 2, Equation 2,
4.5
TWO 528nm LEDs
ONE 850nm LED
4.0
3.5
3.0
2.5
2.0
1.5
1.0
and Equation 3. For example, using an average value for VLED
_
MIN
instead of the worst case value for VLED_MIN can cause a serious
design deficiency, resulting in a CVLED value that is too small and
that causes insufficient optical power in the application. Therefore,
adding a sufficient margin on CVLED is strongly recommended. Add
additional margin to CVLED to account for derating of the capacitor
value over voltage, bias, temperature, and other factors over the life
of the component.
LED INDUCTANCE CONSIDERATIONS
The LED drivers (LEDXx) on the ADPD105/ADPD106/ADPD107
have configurable slew rate settings (Register 0x22, Bits[6:4],
Register 0x23, Bits[6:4], and Register 0x24, Bits[6:4]). These slew
rates are defined in Table 3. Even at the lowest setting, careful
consideration must be taken in board design and layout. If a large
series inductor, such as a long PCB trace, is placed between the
LED cathode and one of the LEDXx pins, voltage spikes from
the switched inductor can cause violations of absolute maximum
and minimum voltages on the LEDXx pins during the slew
portion of the LED pulse.
0
50
100
150
200
250
LED DRIVER CURRENT SETTING (mA)
Figure 40. Example of the Average LED Forward-Bias Voltage Drop as a
Function of the Driver Current
Rev. A | Page 32 of 66
Data Sheet
ADPD105/ADPD106/ADPD107
To verify that there are no voltage spikes on the LEDXx pins due
to parasitic inductance, use an oscilloscope on the LEDXx pins to
monitor the voltage during normal operation. Any positive spike
>3.6 V may damage the devices.
READING DATA
The ADPD105/ADPD106/ADPD107 provide multiple methods
for accessing the sample data. Each time slot can be independently
configured to provide data access using the FIFO or the data
registers. Interrupt signaling is also available to simplify timely data
access. The FIFO is available to loosen the system timing
requirements for data accesses.
In addition, a negative spike <−0.3 V may also damage the devices.
RECOMMENDED START-UP SEQUENCE
At power-up, the device is in standby mode (Register 0x10 =
0x0), as shown in Figure 24. The ADPD105/ADPD106/
ADPD107 do not require a particular power-up sequence.
Reading Data Using the FIFO
The ADPD105/ADPD106/ADPD107 include a 128-byte FIFO
memory buffer that can be configured to store data from either
or both time slots. Register 0x11 selects the kind of data from
each time slot to be written to the FIFO. Note that both time
slots can be enabled to use the FIFO, but only if their output
data rate is the same.
From standby mode, to begin measurement, initiate the
ADPD105/ADPD106/ADPD107 as follows:
1. Set the CLK32K_EN bit (Register 0x4B, Bit 7) to start the
sample clock (32 kHz clock). This clock controls the state
machine. If this clock is off, the state machine is not able to
transition as defined by Register 0x10.
2. Write 0x1 to Register 0x10 to force the device into program
mode. Step 1 and Step 2 can be swapped, but the actual
state transition does not occur until both steps occur.
3. Write additional control registers in any order while the
device is in program mode to configure the devices as
required.
Output data rate = fSAMPLE/N
where:
f
SAMPLE is the sampling frequency.
N is the averaging factor for each time slot (NA for Time Slot A
and NB for Time Slot B). In other words, NA = NB must be true
to store data from both time slots in the FIFO.
Data packets are written to the FIFO at the output data rate. A
data packet for the FIFO consists of a complete sample for each
enabled time slot. Data for each photodiode channel can be stored
as either 16 or 32 bits. Each time slot can store 2, 4, 8, or 16 bytes of
data per sample, depending on the mode and data format. To
ensure that data packets are intact, new data is only written to
the FIFO if there is sufficient space for a complete packet. Any
new data that arrives when there is not enough space is lost.
The FIFO continues to store data when sufficient space exists.
Always read FIFO data in complete packets to ensure that data
packets remain intact.
4. Write 0x2 to Register 0x10 to start normal sampling
operation.
To terminate normal operation, follow this sequence to place
the ADPD105/ADPD106/ADPD107 in standby mode:
1. Write 0x1 to Register 0x10 to force the devices into
program mode.
2. Write to the registers in any order while the devices are in
program mode.
3. Write 0x00FF to Register 0x00 to clear all interrupts. If
desired, clear the FIFO as well by writing 0x80FF to
Register 0x00.
4. Write 0x0 to Register 0x10 to force the devices into standby
mode.
5. Optionally, stop the 32 kHz clock by resetting the CLK32K_
EN bit (Register 0x4B, Bit 7). Register 0x4B, Bit 7 = 0 is the
only write that must be written when the device is in standby
mode (Register 0x10 = 0x0). If 0 is written to this bit while in
program mode or normal mode, the devices become
unable to transition into any other mode, including standby
mode, even if they are subsequently written to do so. As a
result, the power consumption in what appears to be standby
mode is greatly elevated. For this reason, and due to the very
low current draw of the 32 kHz clock while in operation, it
is recommended from an ease of use perspective to keep the
32 kHz clock running after it is turned on.
The number of bytes currently stored in the FIFO is available in
Register 0x00, Bits[15:8]. A dedicated FIFO interrupt is also
available and automatically generates when a specified amount
of data is written to the FIFO.
Interrupt-Based Method
To read data from the FIFO using an interrupt-based method,
use the following procedure:
1. In program mode, set the configuration of the time slots as
desired for operation.
2. Write Register 0x11 with the desired data format for each
time slot.
3. Set FIFO_THRESH in Register 0x06, Bits[13:8] to the
interrupt threshold. A recommended value for this is the
number of 16-bit words in a data packet, minus 1. This
causes an interrupt to generate when there is at least one
complete packet in the FIFO.
4. Enable the FIFO interrupt by writing a 0 to the FIFO_
INT_MASK in Register 0x01, Bit 8. Also, configure the
interrupt pin (GPIO0) by writing the appropriate value to
the bits in Register 0x02.
Rev. A | Page 33 of 66
ADPD105/ADPD106/ADPD107
Data Sheet
5. Enter normal operation mode by setting Register 0x10 to 0x2.
6. When an interrupt occurs
write 0 to Bit 5. To enable the interrupt for Time Slot B,
write 0 to Bit 6. Either or both interrupts can be set.
2. Configure the interrupt pin (GPIOx) by writing the
appropriate value to the bits in Register 0x02.
3. An interrupt generates when the data registers are updated.
4. The interrupt handler must perform the following:
a. Read Register 0x00 and observe Bit 5 or Bit 6 to confirm
which interrupt has occurred. This step is not required if
only one interrupt is in use.
a. There is no requirement to read the FIFO_SAMPLES bits,
because the interrupt is generated only if there is one or
more full packets. Optionally, the interrupt routine can
check for the presence of more than one available
packet by reading these bits.
b. Read a complete packet using one or more multiword
accesses using Register 0x60. Reading the FIFO
automatically frees the space for new samples.
b. Read the data registers before the next sample can be
written. The system must have interrupt latency and
service time short enough to respond before the next
data update, based on the output data rate.
c. Write a 1 to Bit 5 or Bit 6 in Register 0x00 to clear the
interrupt.
The FIFO interrupt automatically clears immediately upon
reading any data from the FIFO and is set again only when the
FIFO is written and the number of words is above the threshold.
Polling Method
To read data from the FIFO in a polling method, use the
following procedure:
If both time slots are in use, it is possible to use only the
Time Slot B interrupt to signal when all registers can be read. It
is recommended to use the multiword read to transfer the data
from the data registers.
1. In program mode, set the configuration of the time slots as
desired for operation.
2. Write Register 0x11 with the desired data format for each
time slot.
3. Enter normal operation mode by setting Register 0x10 to 2.
Reading Data from Registers Without Interrupts
If the system interrupt response is not fast or predictable enough to
use the interrupt method, or if the interrupt pin (GPIOx) is not
used, it is possible to obtain reliable data access by using the data
hold mechanism. To guarantee that the data read from the registers
is from the same sample time, it is necessary to prevent the
update of samples while reading the current values. The method
for doing register reads without interrupt timing is as follows:
Next, begin the polling operations.
1. Wait for the polling interval to expire.
2. Read the FIFO_SAMPLES bits (Register 0x00, Bits[15:8]).
3. If FIFO_SAMPLES ≥ the packet size, read a packet using
the following steps:
a. Read a complete packet using one or more multiword
accesses via Register 0x60. Reading the FIFO
automatically frees the space for new samples.
b. Repeat Step 1.
1. Write a 1 to SLOTA_DATA_HOLD or SLOTB_DATA_
HOLD (Register 0x5F, Bit 1 and Bit 2, respectively) for the
time slot requiring access (both time slots can be accessed).
This setting prevents sample updates.
2. Read the registers as desired.
When a mode change is required, or any other disruption to
normal sampling is necessary, the FIFO must be cleared. Use
the following procedure to clear the state and empty the FIFO:
3. Write a 0 to the SLOTA_DATA_HOLD or SLOTB_DATA_
HOLD bits (Register 0x5F, Bit 1 and Bit 2, respectively)
previously set. Sample updates are allowed again.
1. Enter program mode by setting Register 0x10 to 0x1.
2. Write 1 to Register 0x00, Bit 15.
Because a new sample may arrive while the reads are occurring,
this method prevents the new sample from partially overwriting
the data being read.
Reading Data from Registers Using Interrupts
The latest sample data is always available in the data registers and is
updated simultaneously at the end of each time slot. The data value
for each photodiode channel is available as a 16-bit value in Regis-
ter 0x64 through Register 0x67 for Time Slot A, and Register 0x68
through Register 0x6B for Time Slot B. If allowed to reach their
maximum value, Register 0x64 through Register 0x6B clip. If
Register 0x64 through Register 0x6B saturate, the unsaturated (up
to 27 bits) values for each channel are available in Register 0x70
through Register 0x77 for Time Slot A and Register 0x78 through
Register 0x7F for Time Slot B. Sample interrupts are available to
indicate when the registers are updated and can be read. To use
the interrupt for a given time slot, use the following procedure:
CLOCKS AND TIMING CALIBRATION
The ADPD105/ADPD106/ADPD107 operate using two internal
time bases: a 32 kHz clock sets the sample timing, and a 32 MHz
clock controls the timing of the internal functions such as LED
pulsing and data capture. Both clocks are internally generated
and exhibit device to device variation of approximately 10%
(typical).
Heart rate monitoring applications require an accurate time
base to achieve an accurate count of beats per minute. The
ADPD105/ADPD106/ADPD107 provide a simple calibration
procedure for both clocks.
1. Enable the sample interrupt by writing a 0 to the appropriate
bit in Register 0x01. To enable the interrupt for Time Slot A,
Rev. A | Page 34 of 66
Data Sheet
ADPD105/ADPD106/ADPD107
Calibrating the 32 kHz Clock
OPTIONAL TIMING SIGNALS AVAILABLE ON
GPIO0 AND GPIO1
Calibrating the 32 kHz clock also calibrates items associated
with the output data rate. Calibration of this clock is important
for applications where an accurate data rate is important, such
as heart rate measurements.
The ADPD105/ADPD106/ADPD107 provide a number of
different timing signals, available via the GPIO0 and GPIO1 pins,
to enable ease of system synchronization and flexible triggering
options. Each of the GPIOx pins can be configured as an open-
drain output if they are to share the bus with other drivers, or
they can be configured to always drive the bus. Both outputs
also have polarity control in situations where a timing signal
must be inverted from the default.
To calibrate the 32 kHz clock,
1. Set the sampling frequency to the highest the system can
handle, such as 2000 Hz. Because the 32 kHz clock controls
sample timing, its frequency is readily accessible via the
GPIO0 pin. Configure the interrupt by writing the appropriate
value to the bits in Register 0x02 and set the interrupt to occur
at the sampling frequency by writing 0 to Register 0x01,
Bit 5 or Bit 6. Monitor the GPIO0 pin. The interrupt
frequency must match the set sample frequency.
2. If the monitored interrupt frequency is less than the set
sampling frequency, increase the CLK32K_ADJUST bit
(Register 0x4B, Bits[5:0]). If the monitored interrupt
frequency is larger than the set sampling frequency,
decrease the CLK32K_ADJUST bits.
Table 23. GPIOx Control Settings
Pin Name
Register[Bits] Setting Description
GPIO0
0x02[0]
0x02[1]
0x02[2]
0x02[8]
0x02[9]
0x4F[6]
0: polarity active high
1: polarity active low
0: always drives the bus
1: drives the bus when asserted
0: disables the GPIO0 pin drive
1: enables the GPIO0 pin drive
0: polarity active high
1: polarity active low
0: always drives the bus
GPIO1
3. Repeat Step b until the monitored interrupt signal
frequency is close enough to the set sampling frequency.
1: drives the bus when asserted
0: disables the GPIO1 pin drive
1: enables the GPIO1 pin drive
Calibrating the 32 MHz Clock
Calibrating the 32 MHz clock also calibrates items associated
with the fine timing within a sample period, such as LED pulse
width and spacing, assuming that the 32 kHz clock has been
calibrated.
The various available timing signals are controlled by the
settings in Register 0x0B. Bits[12:8] of this register control the
timing signals available on GPIO1, and Bits[4:0] control the
timing signals available on GPIO0. All of the timing signals
described in this data sheet are available on either (or both) of
the GPIO0 and GPIO1 pins. Timing diagrams are shown in
Figure 41 and Figure 42. The time slot settings used to generate
the timing diagrams are described in Table 24.
To calibrate the 32 MHz clock,
1. Write 0x1 to Register 0x5F, Bit 0.
2. Enable the CLK_RATIO calculation by writing 0x1 to
Register 0x50, Bit 5. This function counts the number of
32 MHz clock cycles in two cycles of the 32 kHz clock.
With this function enabled, this cycle value is stored in
Register 0x0A, Bits[11:0] and nominally this ratio is 2000
(0x7D0).
Table 24. ADPD105/ADPD106/ADPD107 Settings Used for
Timing Diagrams Shown in Figure 41 and Figure 42
Register Setting Description
3. Calculate the 32 MHz clock error as follows:
0x31
0x36
0x15
0x0118 Time Slot A: 1 LED pulse
0x0418 Time Slot B: 4 LED pulses
0x0120 Time Slot A decimation = 4, Time Slot B
decimation = 2
Clock Error = 32 MHz × (1 − CLK_RATIO/2000)
4. Adjust the frequency by setting Bits[7:0] in Register 0x4D
per the following equation:
CLK32M_ADJUST = Clock Error/109 kHz
5. Write 0x0 to Register 0x50, Bit 5 to reset the CLK_RATIO
function.
Repeat Step 2 through Step 5 until the desired accuracy is
achieved.
Write 0x0 to Register 0x5F, Bit 0. Also, set the GPIO0 pin back
to the mode desired for normal operation.
Rev. A | Page 35 of 66
ADPD105/ADPD106/ADPD107
Data Sheet
SLEEP
SLOT A
SLOT B
SLOT A
SLOT B
0x02
0x05
0x06
0x07
0x0F
Figure 41. Optional Timing Signals Available on GPIOx—Register 0x0B, Bits[12:8] or Bits[4:0] = 0x02, 0x05, 0x06, 0x07, and 0x0F
SLEEP
SLEEP
SLOT A/B
SLEEP
SLOT A/B
SLOT A/B
SLOT A/B
SLOT A/B
SLOT A/B
SLEEP
SLEEP
0x02
0x0C
0x0D
0x0E
Figure 42. Optional Timing Signals Available on GPIOx—Register 0x0B, Bits[12:8] or Bits[4:0] = 0x02, 0x0C, 0x0D, and 0x0E
ADPD103 Backward Compatibility
Pulse Outputs
Setting Register 0x0B = 0 provides backward compatibility to
the ADPD103. The GPIO0 pin mirrors the functionality of the
ADPD103 INT pin. The GPIO1 pin mirrors the functionality of
the ADPD103 PDSO pin.
Three options are available to provide a copy of the LED pulse
outputs. Setting Register 0x0B, Bits[12:8] or Bits[4:0] = 0x05
provides a copy of the Time Slot A LED pulses on the respective
pin. A setting of 0x06 provides the Time Slot B pulses, and a
setting of 0x07 provides the pulse outputs of both time slots.
Interrupt Function
Output Data Cycle Signal
Setting Register 0x0B, Bits[12:8] or Bits[4:0] = 0x01 configures
the respective pin to perform the interrupt function as defined
by the settings in Register 0x01.
There are three options available to provide a signal that
indicates when the output data is written to the output data
registers or to the FIFO. Setting Register 0x0B, Bits[12:8] or
Bits[4:0] = 0x0C provides a signal that indicates that a data
value is written for Time Slot A. A setting of 0x0D provides a
signal that indicates that a data value is written for Time Slot B,
and 0x0E provides a signal to indicate that a value is written for
either time slot. The signal asserts at the end of the time slot,
when the output data is already written, and deasserts at the
start of the subsequent sample. This timing signal is especially
useful in situations where the FIFO is being used. For example,
one of the GPIOx pins can be configured to provide an interrupt
after the FIFO reaches the FIFO threshold set in Register 0x06,
Bits[13:8], while the other GPIOx pin can be configured to
provide the output data cycle signal. This signal can be used to
trigger a peripheral device, such as an accelerometer, so that
time aligned signals are provided to the processor.
Sample Timing
Setting Register 0x0B, Bits[12:8] or Bits[4:0] = 0x02 configures
the respective pin to provide a signal that asserts at the beginning
of the first time slot of the current sample and deasserts at the
end of the last time slot of the current sample. For example, if
both time slots are enabled, this signal asserts at the beginning
of Time Slot A and deasserts at the end of Time Slot B. If only a
single time slot is enabled, the signal asserts at the beginning of
the enabled time slot and deasserts at the end of this same time slot.
Rev. A | Page 36 of 66
Data Sheet
ADPD105/ADPD106/ADPD107
fS/2 Output
CALCULATING CURRENT CONSUMPTION
Setting Register 0x0B, Bits[12:8] or Bits[4:0] = 0x0F configures
the respective pin to provide a signal that toggles at half the
sampling rate. This timing signal is useful in, for example,
situations where more than two LEDs per sample are required.
This signal can be used as a select signal to a multiplexer being
used to mux two LEDs into a single LED driver, providing the
ability to drive up to four separate LEDs per sample period. In
such a case, the ADPD105/ADPD106/ADPD107 are operated at
2× the sampling rate and the LED settings can be reconfigured
during the sleep period between samples. If identical LED settings
(current and timing) are used for the LEDs being muxed, then
up to four LEDs can be sampled per sampling period without
host inter-vention. An example of this configuration is shown in
Figure 43.
The current consumption of the ADPD105/ADPD106/ADPD107
depends on the user selected operating configuration, as
determined by the following equations.
Total Power Consumption
To calculate the total power consumption, use Equation 4.
Total Power = IVDD_AVG × VDD + ILEDA_AVG × VLEDA
LEDB_AVG × VLEDB
Average VDD Supply Current
To calculate the average VDD supply current, use Equation 5.
VDD_AVG = DR × ((IAFE_A × tSLOTA) + (IAFE_B × tSLOTB) +
+
I
(4)
(5)
I
QPROC_X) + IVDD_STANDBY
where:
DR is the data rate in Hz.
VDD_STANDBY = 0.2 μA.
PROC_X is an average charge associated with a processing time,
The fS/2 timing signal always starts in an active low state when
the device switches from standby mode to normal operating
mode and transitions to a high state at the completion of the
first sample.
I
Q
as follows:
When only Time Slot A is enabled,
Q
PROC_A (C) = 0.135 × 10−6 + (186 × 10−6 × (2.69 × 10−3 +
(6.2 × 10−3/SCALE_A × ILEDA_PK))
When only Time Slot B is enabled,
Q
PROC_B (C) = 0.135 × 10−6 + (134 × 10−6 × (2.69 × 10−3 +
(6.2 × 10−3/SCALE_A × ILEDA_PK))
When Time Slot A and Time Slot B are enabled,
Q
PROC_AB (C) = 0.135 × 10−6 + (206 × 10−6 × (2.69 × 10−3 +
(6.2 × 10−3/SCALE_A × ILEDA_PK))
AFE_x (A) = 3.0 × 10−3 + (1.5 × 10−3 × NUM_CHANNELS) +
Figure 43. Example Using the fS/2 Timing Signal
I
Logic 0 Output
(5.7 × 10−3/SCALE_X × ILEDX_PK
SLOTx (sec) = LEDx_OFFSET + LEDx_PERIOD ×
PULSE_COUNT
where:
NUM_CHANNELS is the number of active channels.
LEDX_PK is the peak LED current, expressed in amps, for whichever
LED is enabled in that particular timeslot. Note that in the QPROC
calculations, the processing charge only scales as a function of
the LED current setting for the LED configured to operate in
Time Slot A, regardless of whether or not Time Slot A is enabled.
For this reason, be sure to minimize the LED peak current for
the Time Slot A LED to save power when Time Slot A is not
being used.
)
(6)
Setting Register 0x0B, Bits[12:8] or Bits[4:0] = 0x10 configures
the respective pin to provide a Logic 0 output.
t
(7)
Logic 1 Output
Setting Register 0x0B, Bits[12:8] or Bits[4:0] = 0x11 configures
the respective pin to provide a Logic 1 output.
I
32 kHz Oscillator Output
Setting Register 0x0B, Bits[12:8] or Bits[4:0] = 0x13 configures
the respective pin to provide a copy of the on-board 32 kHz
oscillator.
SCALE_X is the scale factor for the LED current drive determined
by Bit 13 of the LEDX_DRV registers: Register 0x22, Register 0x23,
and Register 0x24.
LEDx_OFFSET is the pulse start time offset expressed in
seconds.
LEDx_PERIOD is the pulse period expressed in seconds.
PULSE_COUNT is the number of pulses.
Rev. A | Page 37 of 66
ADPD105/ADPD106/ADPD107
Data Sheet
Note that if either Time Slot A or Time Slot B are disabled, IAFE_x = 0
for that respective time slot. Additionally, if operating in digital
integrate mode, power savings can be realized by setting
Register 0x3C, Bits[8:3] = 010010. This setting disables the
band-pass filters that are bypassed in digital integrate mode,
changing the AFE power contribution calculation to
optimized, the user can then increase the number of pulses to
achieve the desired system SNR.
Optimizing SNR per Watt in a Signal Limited System
In practice, optimizing for peak SNR is not always practical.
One scenario in which the PPG signal has a poor SNR is the
signal limited regime. In this scenario, the LED current reaches
an upper limit before the desired dc return level is achieved.
I
AFE_x (mA) = 3.0 × 10−3 + (1.0 × 10−3 × NUM_CHANNELS) +
(5.7 × 10−3/SCALE_X × ILEDX_PK
)
(8)
Tuning in this case starts where the peak SNR tuning stops. The
starting point is nominally a 50k gain, as long as the lowest LED
current setting of 8 mA does not saturate the photodiode and the
50k gain provides enough protection against intense background
light. In these cases, use a 25k gain as the starting point.
Average VLEDA Supply Current
To calculate the average VLEDA supply current, use Equation 9.
I
LED_AVG_A = SLOTA_LED_WIDTH × ILEDA_PK × DR ×
PULSE_COUNT
where:
(9)
The goal of the tuning process is to bring the dc return signal to a
specific ADC range, such as 50% or 60%. The ADC range choice is
a function of the margin of headroom needed to prevent saturation
as the dc level fluctuates over time. The SNR of the PPG waveform
is always some percentage of the dc level. If the target level cannot
be achieved at the base gain, increase the gain and repeat the
procedure. The tuning system may need to place an upper limit
on the gain to prevent saturation from ambient signals.
SLOTA_LED_WIDTH is the LED pulse width expressed in
seconds.
I
LEDA_PK is the peak current, expressed in amps, for whichever
LED is selected for Time Slot A.
Average VLEDB Supply Current
To calculate the average VLEDB supply current, use Equation 10.
LED_AVG_B = SLOTB_LED_WIDTH × ILEDB_PK × DR ×
PULSE_COUNT
where:
I
Tuning the Pulse Count
(10)
After the LED peak current and TIA gain are optimized,
increasing the number of pulses per sample increases the SNR
by the square root of the number of pulses. There are two ways to
increase the pulse count. The pulse count registers (Register 0x31,
Bits[15:8], and Register 0x36, Bits[15:8]) change the number of
pulses per internal sample. Register 0x15, Bits[6:4] and Bits[10:8],
controls the number of internal samples that are averaged together
before the data is sent to the output. Therefore, the number of
pulses per sample is the pulse count register multiplied by the
number of subsequent samples being averaged. In general, the
internal sampling rate increases as the number of internal
sample averages increase to maintain the desired output data
rate. The SNR/watt is most optimal with pulse count values of
16 or less. Above pulse count values of 16, the square root
relationship does not hold in the pulse count register. However,
this relationship continues to hold when averaged between
samples using Register 0x15.
SLOTB_LED_WIDTH is the LED pulse width expressed in
seconds.
I
LEDB_PK is the peak current, expressed in amps, for whichever
LED is selected for Time Slot B.
OPTIMIZING SNR PER WATT
The ADPD105/ADPD106/ADPD107 offer a variety of
parameters that the user can adjust to achieve the best signal.
One of the key goals of system performance is to obtain the best
system SNR for the lowest total power. This goal is often referred
to as optimizing SNR/watt. Even in systems where only the SNR
matters and power is a secondary concern, there may be a lower
power or a high power means of achieving the same SNR.
Optimizing for Peak SNR
The first step in optimizing for peak SNR is to find a TIA gain
and LED level that gives the best performance where the
number of LED pulses remains constant. If peak SNR is the
goal, the noise section of Table 4 can be used as a guide. It is
important to note that the SNR improves as a square root of the
number of pulses averaged together, whereas the increase in the
LED power consumed is directly proportional to the number of
LED pulses. In other words, for every doubling of the LED pulse
count, there is a doubling of the LED power consumed and a 3 dB
SNR improvement. As a result, avoid any change in the gain con-
figuration that provides less than 3 dB of improvement for a 2×
power penalty; any TIA gain configuration that provides more
than 3 dB of improvement for a 2× power penalty is a good
choice. If peak SNR is the goal and there is no issue saturating
the photodiode with LED current at any gain, the 50k TIA gain
setting is an optimal choice. After the SNR per pulse per channel is
Note that increasing LED peak current increases SNR almost
directly proportional to LED power, whereas increasing the
number of pulses by a factor of n results in only a nominal√(n)
increase in SNR.
When using the sample sum/average function (Register 0x15),
the output data rate decreases by the number of summed
samples. To maintain a static output data rate, increase the
sample frequency (Register 0x12) by the same factor as that
selected in Register 0x15. For example, for a 100 Hz output data
rate and a sample sum/average of four samples, set the sample
frequency to 400 Hz.
Rev. A | Page 38 of 66
Data Sheet
ADPD105/ADPD106/ADPD107
Dual-Channel AFE Mode
OPTIMIZING POWER BY DISABLING UNUSED
CHANNELS AND AMPLIFIERS
Single-Channel AFE Mode
In situations where two of the four channels are in use, the
other two channels can be disabled. Enable Channel 1 and
Channel 2 (with Channel 3 and Channel 4 disabled) by writing
0x7 to Register 0x37, Bits[15:13]. Operate Channel 3 and Channel 4
in dual channel mode (with Channel 1 and Channel 2 disabled)
by writing 0x7 to both Register 0x3C, Bits[5:3] and Register 0x37,
Bits[12:10].
When using a single photodiode in an application, with that
photodiode connected to a single AFE channel (either Channel 1
or Channel 2), the ADPD105/ADPD106/ADPD107 have an
option to power down the unused channels, placing the device in
single AFE channel mode. Because three of the four AFE channels
are turned off in this mode, the power consumption is considerably
reduced.
Three-channel mode can also be achieved with the appropriate
settings. See Table 25 for the settings required to power down
different combinations of channels. Refer to the Time Slot
Switch section to determine the different combinations of the
PDx inputs and enabled channels required to optimize the
system configuration for maximum SNR and lowest power.
If only Channel 1 is being used, disable Channel 2, Channel 3, and
Channel 4 by writing 0x7 to Register 0x3C, Bits[8:6]. If only
Channel 2 is being used, disable Channel 1 by writing 0x7 to
Register 0x3C, Bits[5:3], and disable Channel 3 and Channel 4
by writing 0x7 to Register 0x37, Bits[15:13].
Table 25. Channel Power-Down Settings
Register 0x3C,
Bits[8:6]
Register 0x3C,
Bits[5:3]
Register 0x37,
Bits[15:13]
Register 0x37,
Bits[12:10]
Number of Channels Channels Enabled
1
1
2
2
3
4
Channel 1
Channel 2
Channel 1, Channel 2
Channel 3, Channel 4
Channel 2, Channel 3, Channel 4 0x0
All channels 0x0
0x7
0x0
0x0
0x0
0x0
0x7
0x0
0x7
0x7
0x0
Not applicable
Not applicable
0x7
0x7
0x0
0x0
0x0
0x0
0x0
0x7
0x0
0x0
Rev. A | Page 39 of 66
ADPD105/ADPD106/ADPD107
Data Sheet
mode from the ADPD103 that is kept in the ADPD105/
ADPD106/ADPD107 for backward compatibility.
Powering Down Individual Amplifiers for Additional
Power Savings
Each channel includes a TIA, a BPF, and an integrator, which
can also be configured as a buffer (see Figure 44).Options are
built into the devices to power down individual amplifiers in the
signal path. For example, in digital integrate mode, the BPF is
bypassed but left powered up by default. The BPF can be disabled
completely, which saves 1/3 of the power dissipated by the AFE
during the sampling phase. See the descriptions for Register 0x3C
and Register 0x37 in Table 30 for information on how to disable
the individual amplifiers.
The recommended TIA ADC mode is one in which the band-
pass filter is bypassed and the integrator is configured as an
inverting buffer. This mode is enabled by writing 0xAE65 to
Register 0x43 (Time Slot A) and/or Register 0x45 (Time Slot B)
to bypass the band-pass filter. Additionally, Bit 7 of Register 0x42
(Time Slot A) and/or Register 0x44 (Time Slot B) must be set to
1 to configure the integrator as a buffer. With the ADC offset
register(s) for the desired channel set to 0, the output of the
ADC is at ~3000 codes for a single pulse and a zero input
current condition. As the input current from the photodiode
increases, the ADC output increases toward 16,384.
BPF
TIA
The ADC output (ADCOUT) is calculated as follows:
ADCOUT = 8192 ((2 VBIAS − 2iRF − 1.8 V)/146 μV/LSB) (11)
where:
±1 INTEGRATOR
V
BIAS
Figure 44. Signal Path Block Diagram
V
BIAS is the bias voltage for the TIA (the default value is 1.265 V).
i is the input current to the TIA.
RF is the TIA feedback resistor.
It is important to leave any unused input channels floating for
proper device operation.
In Equation 11, use + for the inverting configuration and use −
when using the noninverting configuration with the buffer.
TIA ADC MODE
There is a way to put the devices into a mode that effectively runs
the TIA directly into the ADC without using the analog band-pass
filter and integrator as shown in Figure 45. This mode is referred to
as TIA ADC mode. There are two basic applications of TIA ADC
mode. In normal operation, all of the background light is blocked
from the signal chain, and therefore cannot be measured. TIA_
ADC mode can be used to measure the amount of background/
ambient light. This mode can also be used to measure other dc
input currents, such as leakage resistance.
Equation 11 is an approximation and does not account for
internal offsets and gain errors. The calculation also assumes
that the ADC offset registers are set to 0
One time slot can be used in TIA ADC mode at the same time
the other time slot is being used in normal pulsed mode. This
capability is useful for monitoring ambient and pulsed signals at
the same time. The ambient signal is monitored during the time
slot configured for TIA ADC mode, while the pulsed signal,
with the ambient signal rejected, is monitored in the time slot
configured for normal mode.
OPTIONAL
BUFFER
–1
ADC
TIA
Protecting Against TIA Saturation in Normal Operation
V
One of the reasons to monitor TIA ADC mode is to protect
against environments that may cause saturation. One concern
when operating in high light conditions, especially with larger
photodiodes, is that the TIA stage may become saturated while
the ADPD105/ADPD106/ADPD107 continue to communicate
data. The resulting saturation is not typical. The TIA, based on its
settings, can only handle a certain level of photodiode current.
Based on the way the ADPD105/ADPD106/ADPD107 are
configured, if there is a current level from the photodiode that is
larger than the TIA can handle, the TIA output during the LED
pulse effectively extends the current pulse, making it wider. The
AFE timing is then violated because the positive portion of the
band-pass filter output extends into the negative section of the
integration window. Thus, the photosignal is subtracted from
itself, causing the output signal to decrease when the effective
light signal increases.
BIAS
Figure 45. TIA ADC Mode Block Diagram
When the devices are in TIA ADC mode, the band-pass filter
and the integrator stage are bypassed. This bypass effectively
wires the TIA directly into the ADC. At the set sampling
frequency, the ADC samples Channel 1 through Channel 4 in
sequential order, and each sample is taken at 1 μs intervals.
There are two modes of operation in TIA ADC mode. One
mode is an inverting configuration where TIA ADC mode
directly drives the ADC. This mode is enabled by setting
Register 0x43 (Time Slot A) and/or Register 0x45 (Time Slot B)
to 0xB065, which bypasses the band-pass filter and the
integrator. With the ADC offset register(s) for the desired
channel set to 0, the output of the ADC is at ~13,000 codes for a
single pulse and a zero input current condition. As the input
current from the photodiode increases, the ADC output
decreases toward 0. This configuration is a legacy TIA ADC
To measure the response from the TIA and verify that this stage
is not saturating, place the device in TIA ADC mode and slightly
modify the timing. Specifically, sweep SLOTx_AFE_OFFSET
Rev. A | Page 40 of 66
Data Sheet
ADPD105/ADPD106/ADPD107
until two or three of the four channels reach a minimum value
(note that TIA is in an inverting configuration). All four channels
do not reach this minimum value because, typically, 3 μs LED
pulse widths are used and the ADC samples the four channels
sequentially at 1 μs intervals. This procedure aligns the ADC
sampling time with the LED pulse to measure the total amount of
light falling on the photodetector (for example, background light +
LED pulse).
between the anode and cathode, or between the anode and
some other supply or ground. In normal operation, the ambient
rejection feature of the ADPD105/ADPD106/ADPD107 masks
the primary effects of these resistances, making it very difficult to
detect them. However, even at 1 MΩ to 10 MΩ, such resistance
can impact performance significantly through added noise or
decreased dynamic range. TIA ADC mode can be used to screen
for these assembly issues.
If this minimum value is above 0 LSB, the TIA is not saturated.
However, take care, because even if the result is not 0 LSB,
operating the device near saturation can quickly result in
saturation if light conditions change. A safe operating region
is typically at ¾ full scale and lower. Use Table 26 to determine
how the input codes map to ADC levels on a per channel per
pulse basis. These codes are not the same as in normal mode
because the band-pass filter and integrator are not unity-gain
elements.
Measuring Shunt Resistance on the Photodiode
A shunt resistor across the photodiode does not generally affect
the output level of the device in operation because the effective
impedance of the TIA is very low, especially if the photodiode is
held to 0 V in operation. However, such resistance can add noise
to the system, degrading performance. The best way to detect
photodiode leakage, also called photodiode shunt resistance, is
to place the device in TIA ADC mode in the dark and vary the
operation mode cathode voltage. Setting the cathode to 1.3 V
places 0 V across the photodiode because the anode is always at
1.3 V while in operation. Setting the cathode to 1.8 V places
0.5 V across the photodiode. Using the register settings in Table 3
to control the cathode voltage, measure the TIA ADC value at
both voltages. Next, divide the voltage difference of 0.5 V by the
difference of the ADC result after converting it to a current. This
result is the approximate shunt resistance. Values greater than
10 MΩ may be difficult to measure, but this method is useful in
identifying gross failures.
Coarse Ambient Light Measurement
Using the typical values in Table 26, TIA ADC mode can be used
to measure or quantify the amount of background or ambient light
present on the photodetector. The settings are the same in the
method described in the Protecting Against TIA Saturation in
Normal Operation section, except the timing used in the
normal operating mode is sufficient for this mode. There is no
need to sweep SLOTx_AFE_OFFSET. If SLOTx_AFE_OFFSET is
in the same place as the normal mode operation, the TIA ADC
mode does not return the same value, regardless of whether the
LED is on or off.
Measuring TIA Input Shunt Resistance
A resistance to develop between the TIA input and another
supply or ground on the PCB is an example of another problem
that can occur. These resistances can force the TIA into saturation
prematurely. This premature saturation, in turn, takes away
dynamic range from the device in operation and adds a Johnson
noise component to the input. To measure these resistances, place
the device in TIA ADC mode in the dark and start by measuring
the TIA ADC offset level with the photodiode inputs disconnected
(Register 0x14, Bits[11:8] = 0 or Register 0x14, Bits[7:4] = 0).
From this, subtract the value of TIA ADC mode with the
darkened photodiode connected and convert the difference
into a current. If the value is positive, and the ADC signal
decreased, the resistance is to a voltage higher than 1.3 V, such
as VDD. Current entering the TIA causes the output to drop. If
the output difference is negative due to an increase of codes at
the ADC, current is being pulled out of the TIA and there is a
shunt resistance to a lower potential than 1.3 V, such as ground.
In TIA_MODE, the dark level is a high level near 13,000 LSBs
per channel per pulse (see Table 26). To measure this value, select
no photodiode by writing a 0x0 to Register 0x14, Bits[11:8] for
Time Slot B or Register 0x14, Bits[7:4] for Time Slot A. This
setting internally opens the photodiode connection and gives
a baseline LSB value that coincides with a zero signal input.
After Register 0x14 is restored to its normal value, while
connecting the photodiode to the TIA, this TIA ADC result can
be subtracted from the open photodiode case to yield a
background light measurement. Use Table 26 to translate this
measurement into an input photocurrent. Use this result for
coarse absolute measurements only, because it is typically only
accurate to within 10%.
Measuring PCB Parasitic Input Resistance
During the process of mounting the ADPD105/ADPD106/
ADPD107, undesired resistance can develop on the inputs through
assembly errors or debris on the PCB. These resistances can form
Rev. A | Page 41 of 66
ADPD105/ADPD106/ADPD107
Data Sheet
Table 26. Analog Specifications for TIA ADC and Digital Integrate Modes
Parameter
Test Conditions/Comments
Typ
Unit
TIA ADC/Digital Integration Saturation Levels
Values expressed per channel, per sample; TIA feedback resistor
25 kΩ
50 kΩ
100 kΩ
200 kΩ
38.32
19.16
9.58
ꢀA
ꢀA
ꢀA
ꢀA
4.79
TIA ADC Resolution
Values expressed per channel, per sample; TIA feedback resistor
25 kΩ
50 kΩ
100 kΩ
200 kΩ
2.92
1.5
0.73
0.37
nA/LSB
nA/LSB
nA/LSB
nA/LSB
Output with No Input Photocurrent
ADC offset (Register 0x18 to Register 0x21) = 0x0
13,000 LSB
The other writes required to switch to digital integration mode
are listed in Table 27.
DIGITAL INTEGRATE MODE
Digital integrate mode is built into the ADPD105/ADPD106/
ADPD107 and allows the device to accommodate longer
LED/AFE pulse widths and different types of sensors at the
input. The analog integration mode described in the AFE
Operation section is ideally suited for applications requiring a
large LED duty cycle, or applications that require customization
of the sampling scheme. Digital integrate mode allows the
integration function to be performed after the ADC in the
digital domain. This mode enables the device to handle a much
wider range of sensors at the input.
When using digital integrate mode, up to two photodiodes can
be connected to the ADPD105/ADPD106/ADPD107 inputs; one
photodiode per PDx input group (PD1/PD2/PD3/PD4 or PD5/
PD6/PD7/PD8). Never connect the same photodiode across the
two PDx groups. In digital integrate mode, there are options to
connect the photodiode to all four AFE channels (PD1/PD2/PD3/
PD4 or PD5/PD6/PD7/PD8), or just a single AFE channel (PD1
or PD5). When connecting to a single AFE channel, write 0x1
to Register 0x54, Bit 14 for Time Slot A, or, for Time Slot B,
write 0x1 to Register 0x54, Bit 15.
In digital integrate mode, the ADC performs a conversion every
1 μs during the integration window. During the integration
window, the digital engine either adds to or subtracts from
the previous sample. The band-pass filter is bypassed and the
integrator is converted to a voltage buffer, allowing the digital
engine to perform the integration function. In this mode, after
the timing is optimized, the output of the ADC increases as the
light level on the photodiode increases.
When connecting to a single AFE channel, there is also an
option to turn off Channel 2, Channel 3, and Channel 4 (and to
save power) by writing 0x7 to Register 0x55, Bits[15:13]. When
connecting to all four channels (PD1/PD2/PD3/PD4 or PD5/
PD6/PD7/PD8), write 0x0 (default)to Register 0x54, Bit 14 for
Time Slot A, or write 0x0 (default) to Register 0x54, Bit 15 for
Time Slot B. Ensure that all AFE channels are powered up by
writing 0x0 to Register 0x55, Bits[15:13].
The integration window is a combination of negative and positive
windows where the duration of these windows is set by SLOTx_
AFE_WIDTH. At the end of the digital integration window, the
resulting sum is sent to the decimate unit as the sample for that
LED pulse. There is one sample per time slot for every sample
cycle. Table 27 lists the registers required for placing the device
in digital integrate mode.
Connecting the single photodiode to a single AFE channel offers
the best SNR performance in cases where signal is limited, whereas
connecting the single photodiode to all four AFE channels offers
the best dynamic range in cases where signal is large.
Digital Integration Sampling Modes
There are two sampling modes that can be used while the
device is in digital integration mode. These modes are single-
sample pair mode and double-sample pair mode.
There may also be changes needed in the SLOTx_AFE_OFFSET
registers and FIFO configuration register (0x11). To read the
final value through the FIFO, set the appropriate values in Regis-
ter 0x11, Bits[4:2] for Time Slot A, and Register 0x11, Bits[8:6]
for Time Slot B. Alternatively, the final output is also available
through the data registers; Register 0x64, Register 0x70, and
Register 0x74 for Time Slot A, and Register 0x68, Register 0x78,
and Register 0x7C for Time Slot B.
In single-sample pair mode, there is a single negative sample
region and a single positive sample region, shown in Figure 48
and Figure 49. To use single-sample pair mode, write 0x1 to
Register 5A, Bit 5 for Time Slot A, or Register 5A, Bit 6 for
Time Slot B. The negative sample region starts at SLOTx_AFE_
OFFSET + 9 and its duration (the number of samples taken) is set
by SLOTx_AFE_WIDTH. The positive sample region starts at
SLOTx_AFE_OFFSET + 9 + SLOTx_AFE_WIDTH, and its
duration is also set by SLOTx_AFE_WIDTH. Set the timing
such that the negative sample region falls entirely in the flat
To put the ADPD105/ADPD106/ADPD107 into digital
integration mode during Time Slot A, write 0x1 to Register 0x58,
Bit 12. To put the ADPD105/ADPD106/ADPD107 into digital
integration mode in Time Slot B, write 0x1 to Register 0x58, Bit 13.
Rev. A | Page 42 of 66
Data Sheet
ADPD105/ADPD106/ADPD107
(dark) portion of the LED response, whereas the positive
sample region falls in the pulsed region of the LED response.
Placing the LED pulse offset, SLOTx_LED_OFFSET, at the
beginning of SLOTx_AFE_OFFSET + 9 + SLOTx_AFE_WIDTH
achieves this timing. The output is the difference of the signals
in the two regions.
Background Values
In digital integrate mode, the digital integration background
value, DI_BACKGROUND, or dark values are also stored and
available as output data. This is in addition to the output value
during the LED pulse, DI_OUTPUT, which has the dark value
subtracted. DI_BACKGROUND is the sum of the negative
region samples.
Double-sample pair mode is another way to sample. In this
mode, there are two negative sample regions and one long
positive sample region (see Figure 46 and Figure 47). To use
double-sample pair mode, write 0x0 to Register 0x5A, Bit 5 for
Time Slot A, or Bit 6 for Time Slot B. The first negative sample
region starts at SLOTx_AFE_OFFSET + 9 and its duration is set
by SLOTx_AFE_WIDTH. The positive sample region starts at
SLOTx_AFE_OFFSET + 9+ SLOTx_AFE_WIDTH and its
duration is twice the SLOTx_AFE_WIDTH. After this, there is
another negative sample region that starts at SLOTx_AFE_OFFSET
+ 9 + 3 × SLOTx_AFE_WIDTH, and its duration is SLOTx_AFE_
WIDTH. Set the timing such that both of the negative sample
regions fall in the flat (dark) portion of the LED response and
the positive sample region falls in the pulsed portion of the LED
response. Placing the LED pulse offset, SLOTx_LED_OFFSET
at the beginning of SLOTx_AFE_OFFSET + 9 + SLOTx_AFE_
WIDTH achieves this timing. The output is calculated by
summing the response of all the regions in a negative/positive/
negative manner. The double-sample pair mode is useful for cases
when the background light is not constant because it has better
background rejection, but it also uses more power than single-
sample pair mode.
To include these values in the FIFO, set Register 0x11, Bits[4:2]
for Time Slot A, and Register 0x11, Bits[8:6] for Time Slot B. For
16-bit data, set this value to 0x3; for 32-bit data, set this value to
0x04. These settings are also available through the data registers;
Register 0x65, Register 0x71, and Register 0x75 for Time Slot A,
and Register 0x69, Register 0x79, and Register 0x7D for Time
Slot B. It is recommended that the channel offsets (Register 0x18 to
Register 0x21) be set to 0x1F00 when including the background
values in the FIFO in digital integration mode. These channel
offsets do not affect the sample values, but do provide more
headroom for the background values.
Saturation Detection in Digital Integrate Mode
In normal operation, when using the band-pass filter and the
integrator, the ADC almost always saturates before the TIA.
Unlike in normal operation, saturation of the TIA or the ADC
cannot be detected solely by looking at the signal value where
the signal value is the positive sample region minus the reference
region in digital integrate mode. This is because the integrated
value does not by itself contain any information indicating
whether one of the ADC conversions during the integration
period exceeded the ADC output range. As a result, the real-
time output may have saturated only for a fraction of the ADC
conversions within a sample and the final accumulated sum may
not reflect this. To detect TIA saturation in digital integration
mode, both the background values, DI_BACKGROUND, and
the signal values, DI_OUTPUT, must be collected. Refer to the
Background Values section for the correct settings for
Sample Timing Modes
There are two options for timing the sample regions: gapped
mode and continuous mode.
In gapped timing mode, there is a space between the negative and
positive sample regions. The width of this region is specified by
SLOTA_AFE_FOFFSET for Time Slot A and SLOTB_AFE_
FOFFSET for Time Slot B in 31.25 ns steps. To enable this feature,
write 0x1 to Register 0x5A, Bit 7. This bit enables gapped timing
for the time slot (or time slots) that are in digital integrate mode.
This mode is helpful when there are unwanted transients in the
LED response that must be ignored for an accurate output.
Register 0x11 that provide these values.
For single-sample pair mode, saturation has occurred when
(DI_OUTPUT/(min(SLOTx_LED_WIDTH,
SLOTx_AFE_WIDTH)) + DI_BACKGROUND/
AFE_WIDTH)/NUM_PULSES > 0x3FFF
If there are no concerns about LED response transients, select
continuous timing mode. In this mode, there is no space
between the negative and positive sample regions. Write 0x0 to
Register 0x5A, Bit 7 for continuous timing of the sample regions.
For double-sample pair mode, saturation has occurred when
(DI_OUTPUT/(min(SLOTx_LED_WIDTH, 2 ×
SLOTx_AFE_WIDTH)) + DI_BACKGROUND/(2 ×
SLOTx_AFE_WIDTH))/NUM_PULSES > 0x3FFF
Both gapped and continuous sample timing modes can be used
with single-sample pair or double-sample pair mode. Example
timing diagrams are shown in Figure 46, Figure 47, Figure 48,
and Figure 49.
Rev. A | Page 43 of 66
ADPD105/ADPD106/ADPD107
Data Sheet
LED
SAMPLE
SLOTx_AFE_OFFSET + 9
NEGATIVE
SAMPLE
REGION
POSITIVE SAMPLE REGION
NEGATIVE
SAMPLE
REGION
Figure 46. Digital Integration Mode in Double-Sample Pair Mode with Continuous Sample Timing
LED
SAMPLE
SLOTx_AFE_OFFSET + 9
NEGATIVE
SAMPLE
REGION
POSITIVE SAMPLE
REGION
NEGATIVE
SAMPLE
REGION
SLOTx_AFE_FOFFSET
SLOTx_AFE_FOFFSET
Figure 47. Digital Integration Mode in Double-Sample Pair Mode with Gapped Sample Timing
LED
SAMPLE
SLOTx_AFE_OFFSET + 9
NEGATIVE
SAMPLE
REGION
POSITIVE
SAMPLE
REGION
SLOTx_AFE_FOFFSET
Figure 48. Digital Integration Mode in Single-Sample Pair Mode with Gapped Sample Timing
LED
SAMPLE
SLOTx_AFE_OFFSET + 9
NEGATIVE
SAMPLE
REGION
POSITIVE
SAMPLE
REGION
Figure 49. Digital Integration Mode in Single-Sample Pair Mode with Continuous Sample Timing
Rev. A | Page 44 of 66
Data Sheet
ADPD105/ADPD106/ADPD107
Table 27. Configuration Registers to Switch Between the Normal Sample Mode, TIA ADC Mode, and Digital Integration Mode
Normal
Mode
Value
Digital
Integration
Mode Value
Data
Address Bits
TIA ADC
Mode Value
Bit Name
Description
0x42
[15:8] SLOTA_AFE_MODE
0x1C
Not applicable 0x1D
In normal mode, this setting configures the
integrator block for optimal operation. In
digital integration mode, this setting
configures the integrator block as a buffer.
This setting is not important for TIA ADC
mode.
7
SLOTA_INT_AS_BUF
0x0
0x1
Not applicable 0: normal integrator configuration.
1: convert integrator to buffer amplifier (this is
done automatically in digital integrate mode).
0x43
0x44
[15:0] SLOTA_AFE_CFG
[15:8] SLOTB_AFE_MODE
0xADA5
0xAE65
0xAE65
Time Slot A AFE connection.
0xAE65 bypasses the band-pass filter.
0xB065 can also be used in TIA ADC mode.
This setting bypasses the BPF and the
integrator.
0x1C
Not applicable 0x1D
In normal mode, this setting configures the
integrator block for optimal operation. In
digital integration mode, this setting
configures the integrator block as a buffer.
This setting is not important for TIA ADC
mode.
7
SLOTB_INT_AS_BUF
0x0
0x1
Not applicable 0: normal integrator configuration.
1: convert integrator to buffer amplifier (this is
done automatically in digital integrate mode).
0x45
[15:0] SLOTB_AFE_CFG
0xADA5
0xAE65
0xAE65
Time Slot B AFE connection.
0xAE65 bypasses the band-pass filter.
0xB065 can also be used in TIA ADC mode.
This setting bypasses the BPF and the
integrator.
0x4E
0x58
[15:0] ADC_TIMING
Not
applicable
Not applicable 0x0040
Set ADC clock to 1 MHz in TIA ADC mode.
13
SLOTB_DIGITAL_INT_EN 0x0
0x0
0x1
Digital integrate mode enable, Time Slot B.
0: disable.
1: enable.
12
SLOTA_DIGITAL_INT_EN 0x0
0x0
0x1
Digital integrate mode enable, Time Slot A.
0: disable.
1: enable.
0x5A
[15:0] DIG_INT_CFG
Not
applicable
Not applicable Variable
Configuration of digital integration depends
on the use case. This register is ignored for
other modes.
To enable pulse connect mode, the device is configured identically
to normal mode, except that Register 0x14, Bits[3:2] = 0 for
Time Slot B, and Register 0x14, Bits[1:0] = 0 for Time Slot A.
PULSE CONNECT MODE
In pulse connect mode, the photodiode input connections are
pulsed according to the timing set up in the LED pulse timing
registers. In this mode, if the LED pulse timing is set up to provide
a 2 μs LED pulse, the device pulses the connection to the photo-
diode input for 2 μs instead of providing a 2 μs LED pulse. This
mode is an alternate to TIA ADC mode, allowing the entire
signal path, including the band-pass filter and integrator, to be
used to measure ambient light as well as other types of measure-
ments with different types of sensors (for example, ECGs).
SYNCHRONOUS ECG AND PPG MEASUREMENT
USING TIA ADC MODE
In wearable devices developed for monitoring the health care of
patients, it is often necessary to have synchronized measurements
of biomedical signals. For example, a synchronous measurement of
patient ECG and PPG can be used to determine the pulse wave
transit time (PWTT), which can then be used to estimate blood
pressure.
Rev. A | Page 45 of 66
ADPD105/ADPD106/ADPD107
Data Sheet
The circuit shown in Figure 50 shows a synchronous ECG and
PPG measurement using the AD8233 and the ADPD105. The
AD8233 implements a two-pole high-pass filter with a cutoff
frequency at 0.3 Hz, and a two-pole low-pass filter with a cutoff
frequency of 37 Hz. The output of the AD8233 is fed to one of
the current inputs of the ADPD105 through a 50 kꢁ resistor to
convert the voltage output of the AD8233 into a current.
ECG measurements. Data can be read out of the on-chip FIFO
or straight from data registers. The ADPD105 channel used to
process the ECG signal is set up in TIA ADC mode and the input
bias voltage must be set to the 0.90 V setting using Bits[5:4] of
Register 0x42 if the ECG signal is on Time Slot A, or Register 0x44
on Time Slot B. The TIA gain setting can be set to optimize the
dynamic range of the signal path. The channel used to process
the PPG signal is configured in its normal operating mode.
Figure 51 shows a plot of a synchronized ECG and PPG
measurement using the AD8233 with the ADPD105.
The ADPD105 is configured to alternately measure the
photodiode signal and the ECG signal from the AD8233 on
consecutive timeslots to provide fully synchronized PPG and
1.8V
4.7µF
10Mꢀ
HPDRIVE
+IN
HPSENSE
IAOUT
10Mꢀ
10Mꢀ
1.8V
180kꢀ
180kꢀ
LA
RA
–IN
REFIN
10Mꢀ
0.1µF
4.7µF
+V
S
10Mꢀ
RLDFB
RLD
0.1µF
1nF
360kꢀ
RL
GND
AD8233
1.8V
SW
FR
AC/DC
SDN
1Mꢀ
1Mꢀ
OPAMP+
REFOUT
OPAMP–
OUT
100kꢀ 6.8nF
250kꢀ
TO DIGITAL
INTERFACE
RLD SDN
LOD
2.7nF
1Mꢀ
1.8V
ADPD105
50kꢀ
DVDD
AVDD
PD1-2
0.1µF
0.1µF
VLED
PDC
AGND
DGND
LGND
1.8V
PD3-4
10kꢀ 10kꢀ
LEDX1
SCL
SDA
VREF
TO DIGITAL
INTERFACE
1µF
GPIO0
GPIO1
Figure 50. Synchronized PPG and ECG Measurement Using ADPD105 with the AD8233
10000
52000
9500
9000
51500
PPG
8500
8000
51000
ECG
7500
7000
50500
6500
50000
6000
SAMPLE RATE (ms)
Figure 51. Plot of Synchronized ECG and PPG Waveforms
Rev. A | Page 46 of 66
Data Sheet
ADPD105/ADPD106/ADPD107
Measuring Voltages Using the Current Inputs
voltage source through a series resistance, RIN must be taken
into consideration. The conversion gain from a voltage into a
current can be determined by following the schematic in Figure 52.
In some situations, as in a case where a synchronized ECG and
PPG measurement is being attempted, a voltage measurement
must be made using the current inputs of the ADPD105/
ADPD106/ADPD107. First, convert the voltage source to a
current by placing a series resistor, RS, in series between the
voltage source and the current input of the device, as shown in
Figure 52.
IIN = (VIN − TIA_VREF)/(RS + RIN)
Typically, RIN is approximately 13 kꢁ. The value of RIN varies as
a function of supply voltage. This variability is shown in Figure 53
45
+85°C
+25°C
0°C
40
–40°C
ADPD105/ADPD106/
ADPD107
35
30
25
20
15
10
5
R
R
IN
S
V
IN
I
IN
ADC
TIA
TIA_VREF
Figure 52. Using the ADPD105/ADPD106/ADPD107 Inputs with Voltage Sources
Secondly, there is a switch resistance, RIN, that must be taken
into consideration when converting a voltage source to a
current. This switch resistance is not a factor in a typical
photodiode application since the device is only dealing with
input currents and these currents are not a function of the input
resistance. However, when driving the device inputs from a
0
1.70
1.75
1.80
1.85
1.90
1.95
V
SUPPLY VOLTAGE (V)
DD
Figure 53. Variability of Switch Input Resistance as a Function of VDD Supply
Voltage
Rev. A | Page 47 of 66
ADPD105/ADPD106/ADPD107
Data Sheet
REGISTER LISTING
Table 28. Numeric Register Listing1
Bit 15
Bit 14
Bit 6
Bit 13
Bit 5
Bit 12
Bit 4
Bit 11
Bit 3
Bit 10
Bit 2
Bit 9
Bit 1
Bit 8
Bit 0
Hex.
Addr. Name
Bits
Bit 7
Reset
RW
0x00 Status
[15:8]
FIFO_SAMPLES[7:0]
0x0000 R/W
[7:0] Reserved
[15:8]
SLOTB_INT
SLOTA_INT
Reserved
Reserved
0x01 INT_MASK
Reserved
FIFO_INT_ 0x00FF R/W
MASK
[7:0] Reserved
SLOTB_INT_
MASK
SLOTA_INT_
MASK
0x02 GPIO_DRV
[15:8]
[7:0]
Reserved
Reserved
GPIO1_DRV GPIO1_POL 0x0000 R/W
GPIO0_ENA GPIO0_DRV GPIO0_POL
0x06 FIFO_
THRESH
[15:8]
[7:0]
Reserved
FIFO_THRESH[5:0]
0x0000 R/W
0x0516
0x00C8 R/W
0x0000
Reserved
REV_NUM[7:0]
0x08 DEVID
[15:8]
[7:0]
R
DEV_ID[7:0]
0x09 I2CS_ID
0x0A CLK_RATIO
[15:8]
[7:0]
ADDRESS_WRITE_KEY[7:0]
SLAVE_ADDRESS[6:0]
Reserved
[15:8]
[7:0]
Reserved
CLK_RATIO[11:8]
R
CLK_RATIO[7:0]
0x0B GPIO_CTRL [15:8]
[7:0]
Reserved
Reserved
GPIO1_ALT_CFG[4:0]
GPIO0_ALT_CFG[4:0]
0x0000 R/W
0x0000 R/W
0x0D SLAVE_
ADDRESS_
[15:8]
[7:0]
SLAVE_ADDRESS_KEY[15:8]
SLAVE_ADDRESS_KEY[7:0]
KEY
0x0F SW_RESET
[15:8]
[7:0]
Reserved
Reserved
0x0000 R/W
0x0000 R/W
0x1000 R/W
SW_RESET
Mode[1:0]
0x10 Mode
[15:8]
[7:0]
Reserved
Reserved
0x11 SLOT_EN
[15:8]
Reserved
RDOUT_
MODE
FIFO_OVRN_
PREVENT
Reserved
SLOTB_
FIFO_
MODE[2]
[7:0]
SLOTB_FIFO_MODE[1:0]
SLOTB_EN
SLOTA_FIFO_MODE[2:0]
Reserved
SLOTA_EN
0x12 FSAMPLE
[15:8]
FSAMPLE[15:8]
FSAMPLE[7:0]
0x0028 R/W
0x0541 R/W
0x0600 R/W
0x2000 R/W
0x2000 R/W
0x2000 R/W
0x2000 R/W
0x2000 R/W
0x2000 R/W
0x2000 R/W
0x2000 RW
[7:0]
0x14 PD_LED_
SELECT
[15:8]
Reserved
SLOTB_PD_SEL[3:0]
SLOTB_LED_SEL[1:0] SLOTA_LED_SEL[1:0]
SLOTB_NUM_AVG[2:0]
Reserved
[7:0]
SLOTA_PD_SEL[3:0]
Reserved
0x15 NUM_AVG
[15:8]
[7:0] Reserved
SLOTA_NUM_AVG[2:0]
0x18 SLOTA_CH1_ [15:8]
SLOTA_CH1_OFFSET[15:8]
SLOTA_CH1_OFFSET[7:0]
SLOTA_CH2_OFFSET[15:8]
SLOTA_CH2_OFFSET[7:0]
SLOTA_CH3_OFFSET[15:8]
SLOTA_CH3_OFFSET[7:0]
SLOTA_CH4_OFFSET[15:8]
SLOTA_CH4_OFFSET[7:0]
SLOTB_CH1_OFFSET[15:8]
SLOTB_CH1_OFFSET[7:0]
SLOTB_CH2_OFFSET[15:8]
SLOTB_CH2_OFFSET[7:0]
SLOTB_CH3_OFFSET[15:8]
SLOTB_CH3_OFFSET[7:0]
SLOTB_CH4_OFFSET[15:8]
SLOTB_CH4_OFFSET[7:0]
OFFSET
[7:0]
0x19 SLOTA_CH2_ [15:8]
OFFSET
[7:0]
0x1A SLOTA_CH3_ [15:8]
OFFSET
[7:0]
0x1B SLOTA_CH4_ [15:8]
OFFSET
[7:0]
0x1E SLOTB_CH1_ [15:8]
OFFSET
[7:0]
0x1F SLOTB_CH2_ [15:8]
OFFSET
[7:0]
0x20 SLOTB_CH3_ [15:8]
OFFSET
[7:0]
0x21 SLOTB_CH4_ [15:8]
OFFSET
[7:0]
Rev. A | Page 48 of 66
Data Sheet
ADPD105/ADPD106/ADPD107
Bit 15
Bit 7
Bit 14
Bit 6
Bit 13
Bit 12
Bit 4
Bit 11
Bit 3
Bit 10
Bit 2
Bit 9
Bit 1
Bit 8
Bit 0
Hex.
Addr. Name
Bits
Bit 5
Reset
RW
0x22 ILED3_
COARSE
[15:8]
Reserved
ILED3_SCALE
ILED3_SLEW[2:0]
ILED1_SCALE
ILED1_SLEW[2:0]
Reserved
0x3000 R/W
0x3000 R/W
0x3000 R/W
0x630C R/W
0x0320 R/W
0x0818 R/W
0x0000 R/W
[7:0] Reserved
[15:8]
ILED3_COARSE[3:0]
Reserved
0x23 ILED1_
COARSE
Reserved
[7:0] Reserved
[15:8] Reserved
[7:0] Reserved
[15:8]
ILED1_COARSE[3:0]
0x24 ILED2_
COARSE
ILED2_SCALE Reserved
ILED2_SLEW[2:0]
ILED3_FINE[4:0]
ILED2_COARSE[3:0]
0x25 ILED_FINE
ILED2_FINE[4:2]
ILED1_FINE[4:0]
SLOTA_LED_WIDTH[4:0]
[7:0]
ILED2_FINE[1:0]
Reserved
Reserved
0x30
SLOTA_LED_ [15:8]
PULSE
[7:0]
SLOTA_LED_OFFSET[7:0]
SLOTA_LED_NUMBER[7:0]
SLOTA_LED_PERIOD[7:0]
0x31 SLOTA_
NUMPULSES
[15:8]
[7:0]
0x34 LED_
DISABLE
[15:8]
Reserved
SLOTB_
LED_DIS
SLOTA_
LED_DIS
[7:0]
Reserved
0x35 SLOTB_LED_ [15:8]
Reserved
SLOTB_LED_WIDTH[4:0]
0x0320 R/W
0x0818 R/W
0x0000 R/W
PULSE
[7:0]
SLOTB_LED_OFFSET[7:0]
SLOTB_LED_NUMBER[7:0]
SLOTB_LED_PERIOD[7:0]
0x36 SLOTB_
NUMPULSES
[15:8]
[7:0]
0x37 ALT_PWR_
DN
[15:8]
[7:0]
CH34_DISABLE[15:13]
CH2_DISABLE[12:10]
Reserved
Reserved
0x38 EXT_SYNC_ [15:8]
EXT_SYNC_STARTUP[15:8]
EXT_SYNC_STARTUP[7:0]
0x000
R/W
STARTUP
[7:0]
0x39 SLOTA_AFE_ [15:8]
SLOTA_AFE_WIDTH[4:0]
SLOTA_AFE_OFFSET[2:0]
SLOTB_AFE_WIDTH[4:0]
SLOTB_AFE_OFFSET[2:0]
Reserved
SLOTA_AFE_OFFSET[5:3]
SLOTA_AFE_FOFFSET[4:0]
SLOTB_AFE_OFFSET[5:3]
SLOTB_AFE_FOFFSET[4:0]
0x22FC R/W
0x22FC R/W
0x3006 R/W
WINDOW
[7:0]
0x3B SLOTB_AFE_ [15:8]
WINDOW
[7:0]
0x3C AFE_PWR_
CFG1
[15:8]
Reserved
Reserved
V_CATHODE AFE_
POWER-
DOWN[5]
[7:0]
AFE_POWERDOWN[4:0]
SLOTA_AFE_MODE[7:0]
SLOTA_TIA_VBIAS[1:0]
Reserved
0x42 SLOTA_
TIA_CFG
[15:8]
0x1C38 R/W
[7:0] SLOTA_INT_
AS_BUF
SLOTA_TIA_
IND_EN
Reserved (write 0x1)
SLOTA_TIA_GAIN[1:0]
0x43 SLOTA_
AFE_CFG
[15:8]
[7:0]
SLOTA_AFE_CFG[15:8]
SLOTA_AFE_CFG[7:0]
SLOTB_AFE_MODE[7:0]
0xADA5 R/W
0x1C38 R/W
0x44 SLOTB_
TIA_CFG
[15:8]
[7:0] SLOTB_INT_
AS_BUF
SLOTB_
TIA_IND_EN
SLOTB_TIA_VBIAS[1:0]
Reserved (write 0x1)
SLOTB_TIA_GAIN[1:0]
0x45 SLOTB_
AFE_CFG
[15:8]
[7:0]
SLOTB_AFE_CFG[15:8]
SLOTB_AFE_CFG[7:0]
Reserved
0xADA5 R/W
0x4B SAMPLE_
CLK
[15:8]
CLK32K_ 0x2612 R/W
BYP
[7:0] CLK32K_EN
Reserved
CLK32K_ADJUST[5:0]
0x4D CLK32M_
ADJUST
[15:8]
[7:0]
Reserved
0x0098 R/W
0x0060 R/W
CLK32M_ADJUST[7:0]
ADC_TIMING[15:8]
ADC_TIMING[7:0]
Reserved
0x4E ADC_CLOCK [15:8]
[7:0]
0x4F EXT_SYNC_ [15:8]
SEL
0x2090 R/W
Reserved
[7:0] Reserved
GPIO1_OE
GPIO1_IE
Reserved
EXT_SYNC_SEL[1:0] GPIO0_IE
Reserved
0x50 CLK32M_
CAL_EN
[15:8]
Reserved
0x0000 R/W
[7:0] Reserved
GPIO1_CTRL
CLK32M_
CAL_EN
0x54 AFE_PWR_
CFG2
[15:8] SLOTB_
SLOTA_
SINGLE_CH_
DIG_INT
SLEEP_V_CATHODE [1:0]
SLOTB_V_CATHODE[1:0] SLOTA_V_CATHODE[1:0] 0x0020 R/W
SINGLE_CH_
DIG_INT
[7:0] REG54_VCAT_
ENABLE
Reserved
0x55 TIA_INDEP_ [15:8]
DIGINT_POWER[2:0]
Reserved
SLOTB_TIA_GAIN_4[1:0]
SLOTA_TIA_GAIN_3[1:0]
SLOTB_TIA_GAIN_3[1:0] 0x0000 R/W
SLOTA_TIA_GAIN_2[1:0]
GAIN
[7:0]
SLOTB_TIA_GAIN_2[1:0]
SLOTA_TIA_GAIN_4[1:0]
Rev. A | Page 49 of 66
ADPD105/ADPD106/ADPD107
Data Sheet
Bit 15
Bit 7
Bit 14
Bit 6
Bit 13
Bit 5
Bit 12
Bit 4
Bit 11
Bit 3
Bit 10
Bit 2
Bit 9
Bit 1
Bit 8
Bit 0
Hex.
Addr. Name
Bits
Reset
RW
0x58 DIGITAL_
INT_EN
[15:8]
Reserved
SLOTB_
DIGITAL_
INT_EN
SLOTA_
DIGITAL_INT_
EN
Reserved
0x0000 R/W
0x0000 R/W
0x0000 R/W
[7:0]
Reserved
Reserved
0x5A DIG_
INT_CFG
[15:8]
[7:0] DIG_INT_
GAPMODE
SLOTB_DIG_
INT_SAMPLE_ INT_SAMPLE_
MODE
SLOTA_DIG_
Reserved
MODE
0x5F DATA_
[15:8]
[7:0]
Reserved
ACCESS_CTL
Reserved
SLOTB_
DATA_
HOLD
SLOTA_
DATA_
HOLD
DIGITAL_
CLOCK_
ENA
0x60 FIFO_
ACCESS
[15:8]
[7:0]
FIFO_DATA[15:8]
0x0000
0x0000
0x0000
0x0000
0x0000
0x0000
0x0000
0x0000
0x0000
0x0000
0x0000
0x0000
0x0000
0x0000
0x0000
0x0000
0x0000
0x0000
0x0000
0x0000
0x0000
0x0000
0x0000
R
R
R
R
R
R
R
R
R
R
R
R
R
R
R
R
R
R
R
R
R
R
R
FIFO_DATA[7:0]
0x64 SLOTA_
PD1_16BIT
[15:8]
[7:0]
SLOTA_CH1_16BIT[15:8]
SLOTA_CH1_16BIT[7:0]
SLOTA_CH2_16BIT[15:8]
SLOTA_CH2_16BIT[7:0]
SLOTA_CH3_16BIT[15:8]
SLOTA_CH3_16BIT[7:0]
SLOTA_CH4_16BIT[15:8]
SLOTA_CH4_16BIT[7:0]
SLOTB_CH1_16BIT[15:8]
SLOTB_CH1_16BIT[7:0]
SLOTB_CH2_16BIT[15:8]
SLOTB_CH2_16BIT[7:0]
SLOTB_CH3_16BIT[15:8]
SLOTB_CH3_16BIT[7:0]
SLOTB_CH4_16BIT[15:8]
SLOTB_CH4_16BIT[7:0]
SLOTA_CH1_LOW[15:8]
SLOTA_CH1_LOW[7:0]
SLOTA_CH2_LOW[15:8]
SLOTA_CH2_LOW[7:0]
SLOTA_CH3_LOW[15:8]
SLOTA_CH3_LOW[7:0]
SLOTA_CH4_LOW[15:8]
SLOTA_CH4_LOW[7:0]
SLOTA_CH1_HIGH[15:8]
SLOTA_CH1_HIGH[7:0]
SLOTA_CH2_HIGH[15:8]
SLOTA_CH2_HIGH[7:0]
SLOTA_CH3_HIGH[15:8]
SLOTA_CH3_HIGH[7:0]
SLOTA_CH4_HIGH[15:8]
SLOTA_CH4_HIGH[7:0]
SLOTB_CH1_LOW[15:8]
SLOTB_CH1_LOW[7:0]
SLOTB_CH2_LOW[15:8]
SLOTB_CH2_LOW[7:0]
SLOTB_CH3_LOW[15:8]
SLOTB_CH3_LOW[7:0]
SLOTB_CH4_LOW[15:8]
SLOTB_CH4_LOW[7:0]
SLOTB_CH1_HIGH[15:8]
SLOTB_CH1_HIGH[7:0]
SLOTB_CH2_HIGH[15:8]
SLOTB_CH2_HIGH[7:0]
0x65 SLOTA_
PD2_16BIT
[15:8]
[7:0]
0x66 SLOTA_
PD3_16BIT
[15:8]
[7:0]
0x67 SLOTA_
PD4_16BIT
[15:8]
[7:0]
0x68 SLOTB_
PD1_16BIT
[15:8]
[7:0]
0x69 SLOTB_
PD2_16BIT
[15:8]
[7:0]
0x6A SLOTB_
PD3_16BIT
[15:8]
[7:0]
0x6B SLOTB_
PD4_16BIT
[15:8]
[7:0]
0x70 A_PD1_
LOW
[15:8]
[7:0]
0x71 A_PD2_
LOW
[15:8]
[7:0]
0x72 A_PD3_
LOW
[15:8]
[7:0]
0x73 A_PD4_
LOW
[15:8]
[7:0]
0x74 A_PD1_
HIGH
[15:8]
[7:0]
0x75 A_PD2_
HIGH
[15:8]
[7:0]
0x76 A_PD3_
HIGH
[15:8]
[7:0]
0x77 A_PD4_
HIGH
[15:8]
[7:0]
0x78 B_PD1_
LOW
[15:8]
[7:0]
0x79 B_PD2_
LOW
[15:8]
[7:0]
0x7A B_PD3_
LOW
[15:8]
[7:0]
0x7B B_PD4_
LOW
[15:8]
[7:0]
0x7C B_PD1_
HIGH
[15:8]
[7:0]
0x7D B_PD2_
HIGH
[15:8]
[7:0]
Rev. A | Page 50 of 66
Data Sheet
ADPD105/ADPD106/ADPD107
Bit 15
Bit 7
Bit 14
Bit 6
Bit 13
Bit 5
Bit 12
Bit 4
Bit 11
Bit 3
Bit 10
Bit 2
Bit 9
Bit 1
Bit 8
Bit 0
Hex.
Addr. Name
Bits
Reset
RW
0x7E B_PD3_
HIGH
[15:8]
[7:0]
[15:8]
[7:0]
SLOTB_CH3_HIGH[15:8]
SLOTB_CH3_HIGH[7:0]
SLOTB_CH4_HIGH[15:8]
SLOTB_CH4_HIGH[7:0]
0x0000
R
0x7F B_PD4_
HIGH
0x0000
R
1 The recommended values are not shown. Only power-on reset values are shown in Table 28. The recommended values are largely dependent on use case. See Table 29
to Table 35 for the recommended values.
Rev. A | Page 51 of 66
ADPD105/ADPD106/ADPD107
Data Sheet
LED CONTROL REGISTERS
Table 29. LED Control Registers
Default
Value
Address Data Bit
Access
R/W
Name
Description
0x14
[15:12]
[11:8]
0x0
Reserved
Write 0x0 to these bits for proper operation.
0x5
R/W
SLOTB_PD_SEL
PDx connection selection for Time Slot B. See Figure 22 and Figure 23
0x0: all photodiode inputs are floating.
0x1: all photodiode inputs are connected during Time Slot B.
0x5: PD1/PD2/PD3/PD4 are connected during Time Slot B.
Other: reserved.
[7:4]
[3:2]
[1:0]
0x4
0x0
0x1
R/W
R/W
R/W
SLOTA_PD_SEL
SLOTB_LED_SEL
SLOTA_LED_SEL
PDx connection selection for Time Slot A. See Figure 22 and
Figure 23.
0x0: all photodiode inputs are floating.
0x1: all photodiode inputs are connected during Time Slot A.
0x5: PD1/PD2/PD3/PD4 are connected during Time Slot A.
Other: reserved.
Time Slot B LED configuration. These bits determine which LED is
associated with Time Slot B.
0x0: pulse PDx connection to AFE.
0x1: LEDX1 pulses during Time Slot B.
0x2: LEDX2 pulses during Time Slot B.
0x3: LEDX3 pulses during Time Slot B.
Time Slot A LED configuration. These bits determine which LED is
associated with Time Slot A.
0x0: pulse PDx connection to AFE.
0x1: LEDX1 pulses during Time Slot A.
0x2: LEDX2 pulses during Time Slot A.
0x3: LEDX3 pulses during Time Slot A.
Write 0x0.
0x22
[15:14]
13
0x0
0x1
R/W
R/W
Reserved
ILED3_SCALE
LEDX3 current scale factor.
1: 100% strength.
0: 40% strength; sets the LEDX3 driver in low power mode.
LEDX3 Current Scale = 0.4 + 0.6 × (Register 0x22, Bit 13).
Write 0x1.
12
0x1
0x0
0x0
R/W
R/W
R/W
Reserved
[11:7]
[6:4]
Reserved
Write 0x0.
ILED3_SLEW
LEDX3 driver slew rate control. The slower the slew rate, the safer
the performance in terms of reducing the risk of overvoltage of the
LED driver.
0x0: the slowest slew rate.
…
0x7: the fastest slew rate.
[3:0]
0x0
R/W
ILED3_COARSE
LEDX3 coarse current setting. Coarse current sink target value of
LEDX3 in standard operation.
0x0: lowest coarse setting.
…
0xF: highest coarse setting.
LED3PEAK = LED3COARSE × LED3FINE × LED3SCALE
where:
LED3PEAK is the LEDX3 peak target value (mA).
LED3COARSE = 50.3 + 19.8 × (Register 0x22, Bits[3:0]).
LED3FINE = 0.74 + 0.022 × (Register 0x25, Bits[15:11]).
LED3SCALE = 0.4 + 0.6 × (Register 0x22, Bit 13).
Rev. A | Page 52 of 66
Data Sheet
ADPD105/ADPD106/ADPD107
Default
Value
Address Data Bit
Access
R/W
Name
Description
0x23
[15:14]
13
0x0
0x1
Reserved
ILED1_SCALE
Write 0x0.
R/W
LEDX1 current scale factor.
1: 100% strength.
0: 40% strength; sets the LEDX1 driver in low power mode.
LEDX1 Current Scale = 0.4 + 0.6 × (Register 0x23, Bit 13).
12
0x1
0x0
0x0
R/W
R/W
R/W
Reserved
Write 0x1.
Write 0x0.
[11:7]
[6:4]
Reserved
ILED1_SLEW
LEDX1 driver slew rate control. The slower the slew rate, the safer
the performance in terms of reducing the risk of overvoltage of the
LED driver.
0: the slowest slew rate.
…
7: the fastest slew rate.
[3:0]
0x0
R/W
ILED1_COARSE
LEDX1 coarse current setting. Coarse current sink target value of
LEDX1 in standard operation.
0x0: lowest coarse setting.
…
0xF: highest coarse setting.
LED1PEAK = LED1COARSE × LED1FINE × LED1SCALE
where:
LED1PEAK is the LEDX1 peak target value (mA).
LED1COARSE = 50.3 + 19.8 × (Register 0x23, Bits[3:0]).
LED1FINE = 0.74 + 0.022 × (Register 0x25, Bits[4:0]).
LED1SCALE = 0.4 + 0.6 × (Register 0x23, Bit 13).
0x24
[15:14]
13
0x0
0x1
R/W
R/W
Reserved
Write 0x0.
ILED2_SCALE
LEDX2 current scale factor.
1: 100% strength.
0: 40% strength; sets the LEDX2 driver in low power mode.
LED2 Current Scale = 0.4 + 0.6 × (Register 0x24, Bit 13)
12
0x1
0x0
0x0
R/W
R/W
R/W
Reserved
Write 0x1.
Write 0x0.
[11:7]
[6:4]
Reserved
ILED2_SLEW
LEDX2 driver slew rate control. The slower the slew rate, the safer
the performance in terms of reducing the risk of overvoltage of the
LED driver.
0: the slowest slew rate.
…
7: the fastest slew rate.
[3:0]
0x0
R/W
ILED2_COARSE
LEDX2 coarse current setting. Coarse current sink target value of
LEDX2 in standard operation.
0x0: lowest coarse setting.
…
0xF: highest coarse setting.
LED2PEAK = LED2COARSE × LED2FINE × LED2SCALE
where:
LED2PEAK is the LEDX2 peak target value (mA).
LED2COARSE = 50.3 + 19.8 × (Register 0x24, Bits[3:0]).
LED2FINE = 0.74 + 0.022 × (Register 0x25, Bits[10:6]).
LED2SCALE = 0.4 + 0.6 × (Register 0x24, Bit 13).
Rev. A | Page 53 of 66
ADPD105/ADPD106/ADPD107
Data Sheet
Default
Value
Address Data Bit
Access
Name
Description
0x25
[15:11]
0xC
R/W
ILED3_FINE
LEDX3 fine adjust. Current adjust multiplier for LED3.
LEDX3 fine adjust = 0.74 + 0.022 × (Register 0x25, Bits[15:11]).
See Register 0x22, Bits[3:0], for the full LED3 formula.
LEDX2 fine adjust. Current adjust multiplier for LED2.
LEDX2 fine adjust = 0.74 + 0.022 × (Register 0x25, Bits[10:6]).
See Register 0x24, Bits[3:0], for the full LED2 formula.
Write 0x0.
[10:6]
0xC
R/W
ILED2_FINE
5
0x0
0xC
R/W
R/W
Reserved
[4:0]
ILED1_FINE
LEDX1 fine adjust. Current adjust multiplier for LED1.
LEDX1 fine adjust = 0.74 + 0.022 × (Register 0x25, Bits[4:0]).
See Register 0x23, Bits[3:0], for the full LED1 formula.
Write 0x0.
0x30
0x31
[15:13]
[12:8]
[7:0]
0x0
R/W
R/W
R/W
R/W
Reserved
0x3
SLOTA_LED_WIDTH
SLOTA_LED_OFFSET
SLOTA_LED_NUMBER
LED pulse width (in 1 ꢀs step) for Time Slot A.
LED offset width (in 1 ꢀs step) for Time Slot A.
0x20
0x08
[15:8]
LED Time Slot A pulse count. nA: number of LED pulses in Time
Slot A. This is typically LED1. Adjust in the application. A setting of
six pulses (0x06) is typical.
[7:0]
[15:10]
9
0x18
0x00
0x0
R/W
R/W
R/W
SLOTA_LED_PERIOD
Reserved
LED Time Slot A pulse period (in 1 ꢀs step).
Write 0x0.
0x34
SLOTB_LED_DIS
Time Slot B LED disable. 1: disables the LED assigned toTime Slot B.
Register 0x34 keeps the drivers active and prevents them from
pulsing current to the LEDs. Disabling both LEDs via this register is
often used to measure the dark level.
Use Register 0x11 instead to enable or disable the actual time slot
usage and not only the LED.
8
0x0
R/W
SLOTA_LED_DIS
Time Slot A LED disable. 1: disables the LED assigned to Time Slot A.
Use Register 0x11 instead to enable or disable the actual time slot
usage and not only the LED.
[7:0]
0x00
0x0
R/W
R/W
Reserved
Write 0x00.
0x35
0x36
[15:13]
[12:8]
[7:0]
Reserved
Write 0x0.
0x3
SLOTB_LED_WIDTH
SLOTB_LED_OFFSET
LED pulse width (in 1 ꢀs step) for Time Slot B.
LED offset width (in 1 ꢀs step) for Time Slot B.
0x20
0x08
[15:8]
R/W
R/W
SLOTB_LED_NUMBER LED Time Slot B pulse count. nB: number of LED pulses in Time Slot B.
This is typically LED2. A setting of six pulses (0x06) is typical.
[7:0]
0x18
SLOTB_LED_PERIOD
LED Time Slot B pulse period (in 1 ꢀs step).
AFE GLOBAL CONFIGURATION REGISTERS
Table 30. AFE Global Configuration Registers
Default
Value
Address
Data Bit
Access
Name
Description
0x37
[15:13]
0x0
R/W
CH34_DISABLE
Power-down options for Channel 3 and Channel 4 only.
Bit 13: power down Channel 3, Channel 4 TIA op amp.
Bit 14: power down Channel 3, Channel 4 BPF op amp.
Bit 15: power down Channel 3, Channel 4 integrator op amp.
Bit 10: power down Channel 2 TIA op amp.
Bit 11: power down Channel 2 BPF op amp.
Bit 12: power down Channel 2 integrator op amp.
Write 0x000.
[12:10]
[9:0]
0x0
R/W
R/W
CH2_DISABLE
Reserved
0x000
Rev. A | Page 54 of 66
Data Sheet
ADPD105/ADPD106/ADPD107
Default
Value
Address
Data Bit
Access
R/W
Name
Description
0x3C
[15:14]
[13:11]
10
0x0
0x6
0x0
0x0
Reserved
Reserved
Reserved
V_CATHODE
Write 0x0.
R/W
Write 0x6.
R/W
Reserved.
9
R/W
0x0: 1.3 V (identical to anode voltage); recommended setting.
0x1: 1.8 V (reverse bias photodiode by 550 mV). This
setting may add noise.
[8:3]
0x00
R/W
AFE_POWERDOWN
AFE channels power-down select.
0x0: keeps all channels on.
Bit 3: power down Channel 1 TIA op amp.
Bit 4: power down Channel 1 BPF op amp.
Bit 5: power down Channel 1 integrator op amp.
Bit 6: power down Channel 2, Channel 3, and Channel 4
TIA op amp.
Bit 7: power down Channel 2, Channel 3, and Channel 4
BPF op amp.
Bit 8: power down Channel 2, Channel 3, and Channel 4
integrator op amp.
[2:0]
15
0x6
0x0
R/W
R/W
Reserved
Write 0x6.
0x54
SLOTB_SINGLE_CH_DIG_INT
0: in Time Slot B, use all four photodiode channels in
parallel for digital integration (default setting for highest
dynamic range).
1: in Time Slot B, use only Channel 1 for digital integration.
This limits connection to PD1 or PD5.
14
0x0
0x0
R/W
R/W
SLOTA_SINGLE_CH_DIG_INT
SLEEP_V_CATHODE
0: in Time Slot A, use all four photodiode channels in
parallel for digital integration (default setting for highest
dynamic range)
1: in Time Slot A, use only Channel 1 for digital integration.
This limits connection to PD1 or PD5.
[13:12]
If Bit 7 = 1; this setting is applied to the cathode voltage
while the device is in sleep mode. The anode voltage is
always set to the cathode voltage in sleep mode.
0x0: VDD (1.8 V).
0x1: 1.3 V.
0x2: 1.55 V.
0x3: 0.0 V.
[11:10]
0x0
0x0
R/W
R/W
SLOTB_V_CATHODE
SLOTA_V_CATHODE
REG54_VCAT_ENABLE
If Bit 7 = 1; this setting is applied to the cathode voltage
while the device is in Time Slot B operation. The anode
voltage is always 1.3 V in Time Slot B mode.
0x0: VDD (1.8 V).
0x1: 1.3 V.
0x2: 1.55 V.
0x3: 0.0 V (this forward biases a diode at the input).
[9:8]
If Bit 7 = 1; this setting is applied to the cathode voltage
while the device is in Time Slot A operation. The anode
voltage is always 1.3 V in Time Slot A mode.
0x0: VDD (1.8 V).
0x1: 1.3 V.
0x2: 1.55 V.
0x3: 0.0 V (this forward biases a diode at the input).
7
0x0
R/W
R/W
0: use the cathode voltage settings defined by Register 0x3C,
Bit 9.
1: override Register 0x3C, Bit 9 with cathode settings
defined by Register 0x54, Bits[13:8].
[6:0]
0x20
Reserved
Reserved.
Rev. A | Page 55 of 66
ADPD105/ADPD106/ADPD107
Data Sheet
Default
Value
Address
Data Bit
[15:14]
13
Access
R/W
Name
Description
0x58
0x0
Reserved
Reserved.
0x0
R/W
SLOTB_DIGITAL_INT_EN
0x0: Time Slot B operating in normal mode.
0x1: Time Slot B operating in digital integration mode.
0x0: Time Slot A operating in normal mode.
0x1: Time Slot A operating in digital integration mode.
Reserved.
12
0x0
R/W
SLOTA_DIGITAL_INT_EN
[11:0]
[15:8]
7
0x000
0x00
0x0
R/W
R/W
R/W
Reserved
0x5A
Reserved
Write 0x0.
DIG_INT_GAPMODE
Digital integrate gapped mode enable.
0: no gap between negative and positive sample regions.
1: use SLOTA_AFE_FOFFSET for Time Slot A or SLOTB_AFE_
FOFFSET for Time Slot B to specify the gap in ꢀs.
6
0x0
R/W
R/W
R/W
SLOTB_DIG_INT_SAMPLE_
MODE
Digital integrate single sample pair mode for Time Slot B.
0: double sample pair mode.
1: single sampled pair mode.
5
0x0
SLOTA_DIG_INT_SAMPLE_
MODE
Digital integrate single sample pair mode for Time Slot A.
0: double sample pair mode.
1: single sampled pair mode.
[4:0]
0x00
Reserved
Write 0x00.
Table 31. AFE Configuration Registers, Time Slot A
Data
Bit
Default
Value
Address
Access Name
Description
0x39
[15:11]
[10:5]
0x4
R/W
R/W
SLOTA_AFE_WIDTH
AFE integration window width (in 1 ꢀs step) for Time Slot A.
0x17
SLOTA_AFE_OFFSET
SLOTA_AFE_FOFFSET
SLOTA_AFE_MODE
AFE integration window coarse offset (in 1 ꢀs step) for
Time Slot A.
[4:0]
0x1C
0x1C
R/W
R/W
AFE integration window fine offset (in 31.25 ns step) for
Time Slot A.
0x42
[15:8]
0x1C: Time Slot A AFE setting for normal mode. All four blocks
of the signal chain are in use during normal mode (the TIA, the
BPF, followed by the integrator, and finally the ADC).
0x1D: Time Slot A AFE setting for digital integrate mode.
0: normal integrator configuration.
1: converts integrator to buffer amplifier (this is done
automatically in digital integrate mode).
7
6
0x0
0x0
R/W
R/W
SLOTA_INT_AS_BUF
SLOTA_TIA_IND_EN
Enable Time Slot A TIA gain individual settings. When it is
enabled, the Channel 1 TIA gain is set via Register 0x42,
Bits[1:0], and the Channel 2 through Channel 4 TIA gain is set
via Register 0x55, Bits[5:0].
0: disable TIA gain individual setting.
1: enable TIA gain individual setting.
Set VBIAS of the TIA for Time Slot A.
0: 1.14 V.
[5:4]
0x3
R/W
SLOTA_TIA_VBIAS
1: 1.01 V.
2: 0.90 V.
3: 1.27 V (default recommended).
Reserved. Write 0x1.
[3:2]
[1:0]
0x2
0x0
R/W
R/W
Reserved
SLOTA_TIA_GAIN
Transimpedance amplifier gain for Time Slot A. When
SLOTA_TIA_IND_EN is enabled, this value is for Time Slot B,
Channel 1 TIA gain. When SLOTA_TIA_IND_EN is disabled, it is
for all four Time Slot A channel TIA gain settings.
0: 200 kΩ.
1: 100 kΩ.
2: 50 kΩ.
3: 25 kΩ.
Rev. A | Page 56 of 66
Data Sheet
ADPD105/ADPD106/ADPD107
Data
Bit
Default
Value
Address
Access Name
Description
0x43
[15:0]
0xADA5
R/W
SLOTA_AFE_CFG
AFE connection in Time Slot A.
0xADA5: analog full path mode (TIA_BPF_INT_ADC).
0xAE65: TIA ADC mode (if Register 0x42, Bit 7 = 1).
0xB065: TIA ADC mode (if Register 0x42, Bit 7 = 0).
0xAE65: digital integration mode.
Others: reserved.
0x55
[15:13]
0x0
R/W
DIGINT_POWER
Power-down for Channel 2, Channel 3, and Channel 4 in digital
integration mode.
0: keep all channels powered up.
7: powers down Channel 2, Channel 3, and Channel 4.
[12]
0x0
0x0
R/W
R/W
Reserved
Write 0x0.
[11:10]
SLOTB_TIA_GAIN_4
TIA gain for Time Slot B, Channel 4 (PD4).
0: 200 kΩ
1: 100 kΩ.
2: 50 kΩ.
3: 25 kΩ.
[9:8]
[7:6]
[5:4]
[3:2]
[1:0]
0x0
0x0
0x0
0x0
0x0
R/W
R/W
R/W
R/W
R/W
SLOTB_TIA_GAIN_3
SLOTB_TIA_GAIN_2
SLOTA_TIA_GAIN_4
SLOTA_TIA_GAIN_3
SLOTA_TIA_GAIN_2
TIA gain for Time Slot B, Channel 3 (PD3).
0: 200 kΩ
1: 100 kΩ.
2: 50 kΩ.
3: 25 kΩ.
TIA gain for Time Slot B, Channel 2 (PD2).
0: 200 kΩ
1: 100 kΩ.
2: 50 kΩ.
3: 25 kΩ.
TIA gain for Time Slot A, Channel 4 (PD4).
0: 200 kΩ
1: 100 kΩ.
2: 50 kΩ.
3: 25 kΩ.
TIA gain for Time Slot A, Channel 3 (PD3).
0: 200 kΩ
1: 100 kΩ.
2: 50 kΩ.
3: 25 kΩ.
TIA gain for Time Slot A, Channel 2 (PD2).
0: 200 kΩ
1: 100 kΩ.
2: 50 kΩ.
3: 25 kΩ.
0x5A
[15:8]
[7]
0x00
0x0
R/W
R/W
Reserved
Write 0x0.
DIG_INT_GAPMODE
Digital integration gapped mode enable.
0: no gap between negative and positive sample regions.
1: use SLOTA_AFE_FOFFSET for Time Slot A or SLOTB_AFE_
FOFFSET for Time Slot B to specify the gap in ꢀs.
[6]
0x0
R/W
R/W
R/W
SLOTB_DIG_INT_SAMPLEMODE
SLOTA_DIG_INT_SAMPLEMODE
Reserved
Digital integration single-sample pair mode for Time Slot B.
0: double-sample pair mode.
1: single-sampled pair mode.
[5]
0x0
Digital integration single-sample pair mode for Time Slot A.
0: double-sample pair mode.
1: single-sampled pair mode.
[4:0]
0x00
Write 0x0.
Rev. A | Page 57 of 66
ADPD105/ADPD106/ADPD107
Data Sheet
Table 32. AFE Configuration Registers, Time Slot B
Data
Address Bit
DefaultValue Access
Name
Description
0x3B
[15:11] 0x04
R/W
SLOTB_AFE_WIDTH
SLOTB_AFE_OFFSET
AFE integration window width (in 1 ꢀs step) forTime Slot B.
[10:5]
0x17
0x1C
0x1C
R/W
R/W
R/W
AFE integration window coarse offset (in 1 ꢀs step) for
Time Slot B.
[4:0]
SLOTB_AFE_FOFFSET
SLOTB_AFE_MODE
AFE integration window fine offset (in 31.25 ns step) for
Time Slot B.
0x44
[15:8]
0x1C: Time Slot B AFE setting for normal mode
(TIA_BPF_INT_ADC).
0x1D: Time Slot B AFE setting for digital integrate mode.
0: normal integrator configuration.
1: convert integrator to buffer amplifier (this is done
automatically in digital integrate mode).
7
6
0x0
0x0
R/W
R/W
SLOTB_INT_AS_BUF
SLOTB_TIA_IND_EN
Enable Time Slot B TIA gain individual settings. When it is
enabled, the Channel 1 TIA gain is set via Register 0x44,
Bits[1:0], and the Channel 2 through Channel 4 TIA gain is
set via Register 0x55, Bits[11:6].
0: disable TIA gain individual setting.
1: enable TIA gain individual setting.
Set VBIAS of the TIA for Time Slot B.
0: 1.14 V.
[5:4]
0x3
R/W
SLOTB_TIA_VBIAS
1: 1.01 V.
2: 0.90 V.
3: 1.27 V (default recommended).
Write 0x1.
[3:2]
[1:0]
0x2
0x0
R/W
R/W
Reserved
SLOTB_TIA_GAIN
Transimpedance amplifier gain for Time Slot B. When
SLOTB_TIA_IND_EN is enabled, this value is for Time Slot B,
Channel 1 TIA gain. When SLOTB_TIA_IND_EN is disabled,
it is for all four Time Slot B channel TIA gain settings.
0: 200 kΩ.
1: 100 kΩ.
2: 50 kΩ.
3: 25 kΩ.
0x45
0x58
[15:0]
0xADA5
R/W
SLOTB_AFE_CFG
AFE connection in Time Slot B.
0xADA5: analog full path mode (TIA_BPF_INT_ADC).
0xAE65: TIA ADC mode (if Register 0x44, Bit 7 = 1).
0xB065: TIA ADC mode (if Register 0x44, Bit 7 = 0).
0xAE65: digital integration mode.
Others: reserved.
[15:14] 0x0
R/W
R/W
Reserved
Write 0x0.
13
0x0
SLOTB_DIGITAL_INT_EN
Digital integration mode, enable Time Slot B.
0: disable.
1: enable.
12
0x0
R/W
R/W
SLOTA_DIGITAL_INT_EN
Reserved
Digital integration mode, enable Time Slot A.
0: disable.
1: enable.
[11:0]
0x0000
Write 0x0000.
Rev. A | Page 58 of 66
Data Sheet
ADPD105/ADPD106/ADPD107
SYSTEM REGISTERS
Table 33. System Registers
Data
Address Bit
Default
Access Name
Description
0x00
[15:8]
0x00
R/W
FIFO_SAMPLES
FIFO status. Number of available bytes to be read from the FIFO. When
comparing this to the FIFO length threshold (Register 0x06, Bits[13:8]),
note that the FIFO status value is in bytes and the FIFO length
threshold is in words, where one word = two bytes.
Write 1 to Bit 15 to clear the contents of the FIFO.
Write 0x1 to clear this bit to 0x0.
7
6
0x0
0x0
R/W
R/W
Reserved
SLOTB_INT
Time Slot B interrupt. Describes the type of interrupt event. A 1
indicates an interrupt of a particular event type has occurred. Write a 1
to clear the corresponding interrupt. After clearing, the register goes to
0. Writing a 0 to this register has no effect.
5
0x0
R/W
SLOTA_INT
Time Slot A interrupt. Describes the type of interrupt event. A 1 indicates
an interrupt of a particular event type has occurred. Write a 1 to clear
the corresponding interrupt. After clearing, the register goes to 0.
Writing a 0 to this register has no effect
[4:0]
[15:9]
8
0x00
0x00
0x1
R/W
R/W
R/W
Reserved
Write 0x1F to clear these bits to 0x00.
Write 0x00.
0x01
Reserved
FIFO_INT_MASK
Sends an interrupt when the FIFO data length has exceeded the FIFO
length threshold in Register 0x06, Bits[13:8]. A 0 enables the interrupt.
7
6
0x1
0x1
R/W
R/W
Reserved
Write 0x1.
SLOTB_INT_MASK
Sends an interrupt on the Time Slot B sample. Write a 1 to disable the
interrupt. Write a 0 to enable the interrupt.
5
0x1
R/W
SLOTA_INT_MASK
Sends an interrupt on the Time Slot A sample. Write a 1 to disable the
interrupt. Write a 0 to enable the interrupt.
[4:0]
0x1F
R/W
R/W
R/W
Reserved
Reserved
GPIO1_DRV
Write 0x1F.
0x02
[15:10] 0x00
Write 0x0000.
9
0x0
GPIO1 drive.
0: the GPIO1 pin is always driven.
1: the GPIO1 pin is driven when the interrupt is asserted; otherwise, it is
left floating and requires a pull-up or pull-down resistor, depending on
polarity (operates as open drain). Use this setting if multiple devices
must share the GPIO1 pin.
8
0x0
R/W
GPIO1_POL
GPIO1 polarity.
0: the GPIO1 pin is active high.
1: the GPIO1 pin is active low.
Write 0x00
[7:3]
2
0x00
0x0
R/W
R/W
Reserved
GPIO0_ENA
GPIO0 pin enable.
0: disable the GPIO0 pin. The GPIO0 pin floats, regardless of interrupt
status. The status register (Address 0x00) remains active.
1: enable the GPIO0 pin.
GPIO0 drive.
0: the GPIO0 pin is always driven.
1: the GPIO0 pin is driven when the interrupt is asserted; otherwise, it is
left floating and requires a pull-up or pull-down resistor, depending on
polarity (operates as open drain). Use this setting if multiple devices
must share the GPIO0 pin.
1
0
0x0
0x0
R/W
R/W
GPIO0_DRV
GPIO0_POL
GPIO0 polarity.
0: the GPIO0 pin is active high.
1: the GPIO0 pin is active low.
Rev. A | Page 59 of 66
ADPD105/ADPD106/ADPD107
Data Sheet
Data
Address Bit
Default
Access Name
Description
0x06
[15:14] 0x0
R/W
R/W
Reserved
Write 0x0.
[13:8]
0x00
FIFO_THRESH
FIFO length threshold. An interrupt is generated when the number of
data-words in the FIFO exceeds the value in FIFO_THRESH. The interrupt
pin automatically deasserts when the number of data-words available
in the FIFO no longer exceeds the value in FIFO_THRESH.
[7:0]
[15:8]
[7:0]
[15:8]
[7:1]
0
0x00
0x04
0x16
0x00
0x64
0x0
R/W
R
Reserved
REV_NUM
DEV_ID
Write 0x00.
0x08
0x09
Revision number.
Device ID.
R
W
R/W
R
ADDRESS_WRITE_KEY Write 0xAD when writing to SLAVE_ADDRESS. Otherwise, do not access.
SLAVE_ADDRESS
Reserved
I2C slave address.
Do not access.
Write 0x0.
0x0A
0x0B
[15:12] 0x0
R
Reserved
[11:0]
0x000
R
CLK_RATIO
When the CLK32M_CAL_EN bit (Register 0x50, Bit 5) is set, the device
calculates the number of 32 MHz clock cycles in two cycles of the 32 kHz
clock. The result, nominally 2000 (0x07D0), is stored in the CLK_RATIO bits.
[15:13] 0x0
R/W
R/W
Reserved
Write 0x0.
[12:8]
0x00
GPIO1_ALT_CFG
Alternate configuration for the GPIO1 pin.
0x0: GPIO1 is backward compatible to the ADPD103 PDSO pin
functionality.
0x1: interrupt function provided on GPIO1, as defined in Register 0x01.
0x2: asserts at the start of the first time slot, deasserts at end of last
time slot.
0x5: Time Slot A pulse output.
0x6: Time Slot B pulse output.
0x7: pulse output of both time slots.
0xC: output data cycle occurred for Time Slot A.
0xD: output data cycle occurred for Time Slot B.
0xE: output data cycle occurred.
0xF: toggles on every sample, which provides a signal at half the
sampling rate.
0x10: ouput = 0
0x11: output = 1
0x13: 32 kHz oscillator output.
Remaining settings are not supported.
Write 0x0.
[7:5]
[4:0]
0x0
R/W
R/W
Reserved
0x00
GPIO0_ALT_CFG
Alternate configuration for the GPIO0 pin.
0x0: GPIO0 is backward compatible to the ADPD103 INT pin functionality.
0x1: interrupt function provided on GPIO0, as defined in Register 0x01.
0x2: asserts at the start of the first time slot, deasserts at end of last
time slot.
0x5: Time Slot A pulse output.
0x6: Time Slot B pulse output.
0x7: pulse output of both time slots.
0xC: output data cycle occurred for Time Slot A.
0xD: output data cycle occurred for Time Slot B.
0xE: output data cycle occurred.
0xF: toggles on every sample, which provides a signal at half the
sampling rate.
0x10: output = 0.
0x11: output = 1.
0x13: 32 kHz oscillator output.
Remaining settings are not supported.
Rev. A | Page 60 of 66
Data Sheet
ADPD105/ADPD106/ADPD107
Data
Address Bit
Default
Access Name
Description
0x0D
[15:0]
0x0000
R/W
SLAVE_ADDRESS_KEY Enable changing the I2C address using Register 0x09.
0x04AD: enable address change always.
0x44AD: enable address change if GPIO0 is high.
0x84AD: enable address change if GPIO1 is high.
0xC4AD: enable address change if both GPIO0 and GPIO1 are high.
0x0F
0x10
[15:1]
0
0x0000
0x0
R
Reserved
Write 0x0000.
R/W
SW_RESET
Software reset. Write 0x1 to reset the device. This bit clears itself after a
reset. For I2C communications, this command returns an acknowledge
and the device subsequently returns to standby mode with all registers
reset to the default state.
[15:2]
[1:0]
0x0000
0x0
R/W
R/W
Reserved
Mode
Write 0x000.
Determines the operating mode of the ADPD105/ADPD106/ADPD107.
0x0: standby.
0x1: program.
0x2: normal operation.
0x11
[15:14] 0x0
R/W
R/W
Reserved
Reserved.
13
0x0
RDOUT_MODE
Readback data mode for extended data registers.
0x0: block sum of N samples.
0x1: block average of N samples.
12
0x1
R/W
FIFO_OVRN_PREVENT 0x0: wrap around FIFO, overwriting old data with new.
0x1: new data if FIFO is not full (recommended setting).
[11:9]
[8:6]
0x0
0x0
R/W
R/W
Reserved
Reserved.
SLOTB_FIFO_MODE
Time Slot B FIFO data format.
0: no data to FIFO.
1: 16-bit sample in digital integration mode or 16-bit sum of all four
channels when not in digital integration mode.
2: 32-bit sample in digital integration mode or 32-bit sum of all four
channels when not in digital integration mode.
3: 16-bit sample and 16-bit background value in digital integration mode.
4: 32-bit sample and 32-bit background value in digital integration
mode or four channels of 16-bit sample data for Time Slot B when not
in digital integration mode.
6: four channels of 32-bit extended sample data for Time Slot B when
not in digital integration mode.
Others: reserved.
The selected Time Slot B data is saved in the FIFO. Available only if Time
Slot A has the same averaging factor, N (Register 0x15, Bits[10:8] =
Bits[6:4]), or if Time Slot A is not saving data to the FIFO (Register 0x11,
Bits[4:2] = 0).
5
0x0
0x0
R/W
R/W
SLOTB_EN
Time Slot B enable. 1: enables Time Slot B.
Time Slot A FIFO data format.
0: no data to FIFO.
[4:2]
SLOTA_FIFO_MODE
1: 16-bit sample in digital integration mode or 16-bit sum of all
four channels when not in digital integration mode.
2: 32-bit sample in digital integration mode or 32-bit sum of all
four channels when not in digital integration mode.
3: 16-bit sample and 16-bit background value in digital integration mode.
4: 32-bit sample and 32-bit background value in digital integration
mode or four channels of 16-bit sample data for Time Slot B when not
in digital integration mode.
6: four channels of 32-bit extended sample data for Time Slot B when
not in digital integration mode.
Others: reserved.
1
0
0x0
0x0
R/W
R/W
Reserved
Write 0x0.
SLOTA_EN
Time Slot A enable. 1: enables Time Slot A.
Rev. A | Page 61 of 66
ADPD105/ADPD106/ADPD107
Data Sheet
Data
Address Bit
Default
0x0000
0x13
Access Name
Description
0x38
0x4B
[15:0]
R/W
R/W
R/W
EXT_SYNC_STARTUP
Write 0x4000 when EXT_SYNC_SEL is b01 or b10. Otherwise, write 0x0.
Write 0x26.
[15:9]
8
Reserved
0x0
CLK32K_BYP
Bypass internal 32 kHz oscillator.
0x0: normal operation.
0x1: provide external clock on the GPIO1 pin. The user must set
Register 0x4F, Bits[6:5] = 01 to enable the GPIO1 pin as an input.
7
0x0
R/W
CLK32K_EN
Sample clock power-up. Enables the data sample clock.
0x0: clock disabled.
0x1: normal operation.
6
0x0
R/W
R/W
Reserved
Write 0x0.
[5:0]
0x12
CLK32K_ADJUST
Data sampling (32 kHz) clock frequency adjust. This register is used to
calibrate the sample frequency of the device to achieve high precision
on the data rate as defined in Register 0x12. Adjusts the sample master
32 kHz clock by 0.6 kHz per LSB. For a 100 Hz sample rate as defined in
Register 0x12, 1 LSB of Register 0x4B, Bits[5:0], is 1.9 Hz.
Note that a larger value produces a lower frequency. See the Clocks
and Timing Calibration section for more information regarding clock
adjustment.
00 0000: maximum frequency.
10 0010: typical center frequency.
11 1111: minimum frequency.
Write 0x00.
0x4D
[15:8]
[7:0]
0x00
0x98
R/W
R/W
Reserved
CLK32M_ADJUST
Internal timing (32 MHz) clock frequency adjust. This register is used to
calibrate the internal clock of the device to achieve precisely timed LED
pulses. Adjusts the 32 MHz clock by 109 kHz per LSB.
See the Clocks and Timing Calibration section for more information
regarding clock adjustment.
0000 0000: minimum frequency.
1001 1000: default frequency.
1111 1111: maximum frequency.
0x0040: ADC clock speed = 1 MHz.
0x0060: ADC clock speed = 500 kHz.
Write 0x20.
0x4E1
0x4F
[15:0]
0x0060
R/W
ADC_TIMING1
[15:8]
0x20
0x1
0x0
0x0
0x1
0x0
R/W
R/W
R/W
R/W
R/W
R/W
Reserved
7
Reserved
Write 0x1.
6
GPIO1_OE
GPIO1_IE
GPIO1 pin output enable.
GPIO1 pin input enable.
5
4
Reserved
Write 0x1.
[3:2]
EXT_SYNC_SEL
Sample sync select.
00: use the internal 32 kHz clock with FSAMPLE to select sample timings.
01: use the GPIO0 pin to trigger sample cycle.
10: use the GPIO1 pin to trigger sample cycle.
11: reserved.
1
0x0
R/W
R/W
R/W
R/W
GPIO0_IE
Reserved
Reserved
GPIO1_CTRL
GPIO0 pin input enable.
0
0x0
Write 0x0.
0x50
[15:7]
6
0x000
0x0
Write 0x000.
Controls the GPIO1 output when the GPIO1 output is enabled
(GPIO1_OE = 0x1).
0x0: GPIO1 output driven low.
0x1: GPIO1 output driven by the AFE power-down signal.
5
0x0
R/W
R/W
CLK32M_CAL_EN
Reserved
As part of the 32 MHz clock calibration routine, write 1 to begin the
clock ratio calculation. Read the result of this calculation from the
CLK_RATIO bits in Register 0x0A.
Reset this bit to 0 prior to reinitiating the calculation.
Write 0x0.
[4:0]
0x00
Rev. A | Page 62 of 66
Data Sheet
ADPD105/ADPD106/ADPD107
Data
Address Bit
Default
0x0000
0x0
Access Name
Description
0x5F
[15:3]
2
R/W
R/W
Reserved
Write 0x0000.
SLOTB_DATA_HOLD
Setting this bit prevents the update of the data registers corresponding to
Time Slot B. Set this bit to ensure that unread data registers are not
updated, guaranteeing a contiguous set of data from all four photodiode
channels.
1: hold data registers for Time Slot B.
0: allow data register update.
1
0
0x0
0x0
R/W
R/W
SLOTA_DATA_HOLD
Setting this bit prevents the update of the data registers corresponding to
Time Slot A. Set this bit to ensure that unread data registers are not
updated, guaranteeing a contiguous set of data from all four photodiode
channels.
1: hold data registers for Time Slot A.
0: allow data register update.
DIGITAL_CLOCK_ENA Set to 1 in order to enable the 32 MHz clock when calibrating the 32 MHz
clock. Always disable the 32 MHz clock following the calibration by
resetting this bit to 0.
1 Clock speed setting is only relevant during digital integrate mode.
ADC REGISTERS
Table 34. ADC Registers
Data
Address Bits
Default Access Name
0x0028 R/W FSAMPLE
Description
0x12
[15:0]
Sampling frequency: fSAMPLE = 32 kHz/(Register 0x12, Bits[15:0] × 4).
For example, 100 Hz = 0x0050; 200 Hz = 0x0028.
Write 0x0.
0x15
[15:11] 0x00
R/W
R/W
Reserved
[10:8]
0x6
SLOTB_NUM_AVG
Sample sum/average for Time Slot B. Specifies the averaging factor, NB,
which is the number of consecutive samples that is summed and averaged
after the ADC. Register 0x70 to Register 0x7F hold the data sum. Register 0x64
to Register 0x6B and the data buffer in Register 0x60 hold the data average,
which can be used to increase SNR without clipping, in 16-bit registers.
The data rate is decimated by the value of the SLOTB_NUMB_AVG bits.
0: 1.
1: 2.
2: 4.
3: 8.
4: 16.
5: 32.
6: 64.
7: 128.
Write 0x0.
7
0x0
0x0
R/W
R/W
Reserved
[6:4]
SLOTA_NUM_AVG
Sample sum/average for Time Slot A. NA: same as Bits[10:8] but for
Time Slot A. See description in Register 0x15, Bits[10:8].
[3:0]
0x0
R/W
Reserved
Write 0x0.
0x18
0x19
0x1A
0x1B
0x1E
[15:0]
0x2000 R/W
0x2000 R/W
0x2000 R/W
0x2000 R/W
0x2000 R/W
SLOTA_CH1_OFFSET Time Slot A Channel 1 ADC offset. The value to subtract from the raw ADC
value. A value of 0x2000 is typical.
[15:0]
[15:0]
[15:0]
[15:0]
SLOTA_CH2_OFFSET Time Slot A Channel 2 ADC offset. The value to subtract from the raw ADC
value. A value of 0x2000 is typical.
SLOTA_CH3_OFFSET Time Slot A Channel 3 ADC offset. The value to subtract from the raw ADC
value. A value of 0x2000 is typical.
SLOTA_CH4_OFFSET Time Slot A Channel 4 ADC offset. The value to subtract from the raw ADC
value. A value of 0x2000 is typical.
SLOTB_CH1_OFFSET Time Slot B Channel 1 ADC offset. The value to subtract from the raw ADC
value. A value of 0x2000 is typical.
Rev. A | Page 63 of 66
ADPD105/ADPD106/ADPD107
Data Sheet
Data
Address Bits
Default Access Name
Description
0x1F
0x20
0x21
[15:0]
0x2000 R/W
0x2000 R/W
0x2000 R/W
SLOTB_CH2_OFFSET Time Slot B Channel 2 ADC offset. The value to subtract from the raw ADC
value. A value of 0x2000 is typical.
[15:0]
[15:0]
SLOTB_CH3_OFFSET Time Slot B Channel 3 ADC offset. The value to subtract from the raw ADC
value. A value of 0x2000 is typical.
SLOTB_CH4_OFFSET Time Slot B Channel 4 ADC offset. The value to subtract from the raw ADC
value. A value of 0x2000 is typical.
DATA REGISTERS
Table 35. Data Registers
Data
Address Bits
Access Name
FIFO_DATA
Description
0x60
0x64
0x65
0x66
0x67
0x68
0x69
0x6A
0x6B
0x70
0x71
0x72
0x73
0x74
0x75
0x76
0x77
0x78
0x79
0x7A
0x7B
0x7C
0x7D
0x7E
0x7F
[15:0]
R
R
R
R
R
R
R
R
R
R
R
R
R
R
R
R
R
R
R
R
R
R
R
R
R
Next available word in FIFO.
[15:0]
[15:0]
[15:0]
[15:0]
[15:0]
[15:0]
[15:0]
[15:0]
[15:0]
[15:0]
[15:0]
[15:0]
[15:0]
[15:0]
[15:0]
[15:0]
[15:0]
[15:0]
[15:0]
[15:0]
[15:0]
[15:0]
[15:0]
[15:0]
SLOTA_CH1_16BIT
SLOTA_CH2_16BIT
SLOTA_CH3_16BIT
SLOTA_CH4_16BIT
SLOTB_CH1_16BIT
SLOTB_CH2_16BIT
SLOTB_CH3_16BIT
SLOTB_CH4_16BIT
SLOTA_CH1_LOW
SLOTA_CH2_LOW
SLOTA_CH3_LOW
SLOTA_CH4_LOW
SLOTA_CH1_HIGH
SLOTA_CH2_HIGH
SLOTA_CH3_HIGH
SLOTA_CH4_HIGH
SLOTB_CH1_LOW
SLOTB_CH2_LOW
SLOTB_CH3_LOW
SLOTB_CH4_LOW
SLOTB_CH1_HIGH
SLOTB_CH2_HIGH
SLOTB_CH3_HIGH
SLOTB_CH4_HIGH
16-bit value of Channel1 in Time Slot A.
16-bit value of Channel 2 in Time Slot A.
16-bit value of Channel 3 in Time Slot A.
16-bit value of Channel 4 in Time Slot A.
16-bit value of Channel 1 in Time Slot B.
16-bit value of Channel 2 in Time Slot B.
16-bit value of Channel 3 in Time Slot B.
16-bit value of Channel 4 in Time Slot B.
Low data-word for Channel 1 in Time Slot A.
Low data-word for Channel 2 in Time Slot A.
Low data-word for Channel 3 in Time Slot A.
Low data-word for Channel 4 in Time Slot A.
High data-word for Channel 1 in Time Slot A.
High data-word for Channel 2 in Time Slot A.
High data-word for Channel 3 in Time Slot A.
High data-word for Channel 4 in Time Slot A.
Low data-word for Channel 1 in Time Slot B.
Low data-word for Channel 2 in Time Slot B.
Low data-word for Channel 3 in Time Slot B.
Low data-word for Channel 4 in Time Slot B.
High data-word for Channel 1 in Time Slot B.
High data-word for Channel 2 in Time Slot B.
High data-word for Channel 3 in Time Slot B.
High data-word for Channel 4 in Time Slot B.
2. Write 0x0001 to Register 0x10 to enter program mode.
REQUIRED START-UP LOAD PROCEDURE
3. Write to the other registers; the register order is not
important while the device is in program mode.
4. Write 0x0002 to Register 0x10 to start normal sampling
operation.
The required start-up load procedure is as follows:
1. Write to 0x1 to Register 0x4B, Bit 7 to enable the clock that
drives the state machine.
Rev. A | Page 64 of 66
Data Sheet
ADPD105/ADPD106/ADPD107
OUTLINE DIMENSIONS
0.25
0.20
0.15
4.10
4.00 SQ
3.90
PIN 1
INDICATOR
PIN 1
INDICATOR
22
28
1
21
0.40
BSC
EXPOSED
PAD
2.70
2.60 SQ
2.50
7
15
14
8
0.45
0.40
0.35
0.20 MIN
TOP VIEW
BOTTOM VIEW
0.80
0.75
0.70
FOR PROPER CONNECTION OF
THE EXPOSED PAD, REFER TO
THE PIN CONFIGURATION AND
FUNCTION DESCRIPTIONS
0.05 MAX
0.02 NOM
COPLANARITY
0.08
SECTION OF THIS DATA SHEET.
SEATING
PLANE
0.20 REF
COMPLIANT TO JEDEC STANDARDS MO-220-WGGE.
Figure 54. 28-Lead Lead Frame Chip Scale Package [LFCSP]
4 mm × 4 mm Body and 0.75 mm Package Height
(CP-28-5)
Dimensions shown in millimeters
1.44
1.40
1.36
3
2
1
0.225
A
B
C
D
E
F
BALL A1
IDENTIFIER
2.00
REF
2.50
2.46
2.42
0.40
BSC
BOTTOM VIEW
(BALL SIDE UP)
TOP VIEW
(BALL SIDE DOWN)
0.235
0.300
0.330
0.300
0.270
0.560
0.500
0.440
END VIEW
COPLANARITY
0.05
SEATING
PLANE
0.230
0.200
0.170
0.300
0.260
0.220
Figure 55. 16-Ball Wafer Level Chip Scale Package [WLCSP]
(CB-16-18)
Dimensions shown in millimeters
Rev. A | Page 65 of 66
ADPD105/ADPD106/ADPD107
Data Sheet
1.44
1.40
1.36
3
2
1
A
B
C
D
E
F
BALL A1
IDENTIFIER
2.00
REF
2.50
2.46
2.42
0.40
BSC
BOTTOM VIEW
(BALL SIDE UP)
TOP VIEW
(BALL SIDE DOWN)
0.80 REF
0.330
0.300
0.270
0.560
0.500
0.440
END VIEW
COPLANARITY
0.05
SEATING
PLANE
0.230
0.200
0.170
0.300
0.260
0.220
Figure 56. 17-Ball Wafer Level Chip Scale Package [WLCSP]
(CB-17-1)
Dimensions shown in millimeters
ORDERING GUIDE
Model1
ADPD105BCPZ
ADPD105BCPZRL
ADPD105BCBZR7
ADPD106BCBZR7
ADPD107BCBZR7
EVAL-ADPD105Z-GEN
Temperature Range Package Description
Package Option
CP-28-5
CP-28-5
CB-16-18
CB-16-18
CB-17-1
−40°C to +85°C
−40°C to +85°C
−40°C to +85°C
−40°C to +85°C
−40°C to +85°C
28-Lead Lead Frame Chip Scale Package [LFCSP]
28-Lead Lead Frame Chip Scale Package [LFCSP]
16-Ball Wafer Level Chip Scale Package [WLCSP]
16-Ball Wafer Level Chip Scale Package [WLCSP]
17-Ball Wafer Level Chip Scale Package [WLCSP]
Generic ADPD105 Evaluation Board2
1 Z = RoHS Compliant Part.
2 This evaluation board is used for the ADPD105, ADPD106, and ADPD107.
©2017 Analog Devices, Inc. All rights reserved. Trademarks and
registered trademarks are the property of their respective owners.
D14693-0-1/17(A)
Rev. A | Page 66 of 66
相关型号:
©2020 ICPDF网 联系我们和版权申明