ADPD188GG_技术文档

Integrated Optical Module with Ambient

Light Rejection and Two LEDs

Data Sheet

ADPD188GG

FEATURES

GENERAL DESCRIPTION

3.8 mm × 5.0 mm × 0.9 mm module with integrated optical

components

2 green LEDs, 2 PDs with IR cut filter

2 external sensor inputs

3, 370 mA LED drivers

20-bit burst accumulator enabling 20 bits per sample period

On-board sample to sample accumulator enabling up to

27 bits per data read

Custom optical package made to work under a glass window

Optimized SNR for signal limited cases

I²C or SPI communications

The ADPD188GG is a complete photometric system designed to

measure optical signals from ambient light and from synchronous

reflected light emitting diode (LED) pulses. Synchronous measure-

ment offers best-in-class rejection of ambient light interference,

both dc and ac. The module integrates a highly efficient photo-

metric front end, two LEDs, and two photodiode (PD). All of

these items are housed in a custom package that prevents light

from going directly from the LED to the photodiode without first

entering the subject.

The front end of the application specific integrated circuit (ASIC)

consists of a control block, a 14-bit analog-to-digital converter

(ADC) with a 20-bit burst accumulator, and three flexible,

independently configurable LED drivers. The control circuitry

includes flexible LED signaling and synchronous detection.

The analog front end (AFE) features best-in-class rejection of

signal offset and corruption due to modulated interference

commonly caused by ambient light. The data output and

functional configuration occur over a 1.8 V I²C interface or a

serial peripheral interface (SPI) port.

APPLICATIONS

Optical heart rate monitoring

Reflective PPG measurement

CNIBP measurement

FUNCTIONAL BLOCK DIAGRAM

VDD1

VDD2

PDC

ADPD188GG

PD1

EXT_IN1

CH1

BPF

±1 INTEGRATOR

TIA_VREF

VREF

PD2

1µF

EXT_IN2

CH2

TIA_VREF

PDC

PDET1

PD3

CS

CH3

BPF

±1 INTEGRATOR

SCLK

MOSI

MISO

TIME SLOT A

TIA_VREF

DATA

PDET2

PD4

14-BIT

ADC

CH4

TIME SLOT B

DATA

SDA

SCL

VLED1

TIA_VREF

GREEN

GPIO0

GPIO1

DIGITAL

INTERFACE

AND

LED1/DNC

CONTROL

LED1 DRIVER

LED3 DRIVER

LED3

LED2

LED2 DRIVER

LGND

AGND

DGND

NOTES

1. DNC = DO NOT CONNECT. DO NOT CONNECT TO THIS PIN WHEN USING INTERNAL LEDs.

Figure 1.

Rev. B

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 are the property of their respective owners.

One Technology Way, P.O. Box 9106, Norwood, MA 02062-9106, U.S.A.

Technical Support

www.analog.com

ADPD188GG

Data Sheet

TABLE OF CONTENTS

Features.............................................................................................. 1

Typical Connection Diagram................................................... 22

Land Pattern ............................................................................... 22

Recommended Start-Up Sequence.......................................... 23

Reading Data............................................................................... 23

Clocks and Timing Calibration................................................ 24

Optional Timing Signals Available on GPIO0 and GPIO1 . 25

LED Driver Pins and LED Supply Voltage............................. 26

LED Driver Operation............................................................... 26

Determining the Average Current........................................... 27

Determining C_VLED..................................................................... 27

Using External LEDs ................................................................. 28

Calculating Current Consumption.......................................... 28

Mechanical Considerations for Covering the ADPD188GG ... 31

TIA ADC Mode.......................................................................... 31

Pulse Connect Mode.................................................................. 33

Applications ...................................................................................... 1

General Description......................................................................... 1

Functional Block Diagram .............................................................. 1

Revision History ............................................................................... 2

Specifications .................................................................................... 3

Analog Specifications................................................................... 5

Digital Specifications ................................................................... 6

Timing Specifications .................................................................. 7

Absolute Maximum Ratings ........................................................... 9

Thermal Resistance...................................................................... 9

Recommended Soldering Profile ............................................... 9

ESD Caution.................................................................................. 9

Pin Configuration and Function Descriptions .......................... 10

Typical Performance Characteristics........................................... 11

Theory of Operation ...................................................................... 13

Introduction................................................................................ 13

Optical Components.................................................................. 13

Dual Time Slot Operation......................................................... 14

Time Slot Switch......................................................................... 15

Adjustable Sampling Frequency............................................... 16

External Synchronization for Sampling.................................. 16

State Machine Operation .......................................................... 16

Normal Mode Operation and Data Flow................................ 17

Communications Interface ........................................................... 19

I²C Interface ................................................................................ 19

SPI Port........................................................................................ 20

Applications Information.............................................................. 22

Synchronous ECG and PPG Measurement Using TIA ADC

Mode ............................................................................................ 34

Float Mode .................................................................................. 35

Register Listing ............................................................................... 42

LED Control Registers............................................................... 46

AFE Configuration Registers.................................................... 48

Float Mode Registers ................................................................. 52

System Registers......................................................................... 55

ADC Registers ............................................................................ 59

Data Registers ............................................................................. 60

Outline Dimensions....................................................................... 61

Ordering Guide .......................................................................... 61

REVISION HISTORY

4/2020—Rev. A to Rev. B

10/2018—Rev. 0 to Rev. A

Change to Applications Section ..................................................... 1

Changes to Table 4 and Figure 2.................................................... 7

Change to Reset Column, Table 26.............................................. 42

Added Endnote 1, Table 27........................................................... 46

Changes to Table 32....................................................................... 55

Changes to Figure 24 and Figure 25............................................ 22

Changes to Calibrating the 32 kHz Clock Section..................... 25

Added Improving SNR Using Integrator Chopping Section and

Figure 32; Renumbered Sequentially........................................... 29

Added Table 18; Renumbered Sequentially ............................... 30

Changes to Table 26....................................................................... 42

Changes to Table 29....................................................................... 50

Changes to Table 30....................................................................... 51

Changes to Address 0x58 Description Column, Table 31........ 53

2/2018—Revision 0: Initial Version

Rev. B | Page 2 of 61

Data Sheet

ADPD188GG

SPECIFICATIONS

The voltage applied at the VDD1 and VDD2 pins (V_DD) = 1.8 V, and T_A= full operating temperature range, unless otherwise noted.

Table 1.

Parameter

Test Conditions/Comments

Min

Typ

Max Unit

CURRENT CONSUMPTION

See the Calculating Current Consumption section for the relevant

equations

Peak V_DDSupply Current

V_DDStandby Current

Single-channel (Register 0x3C, Bits[8:3] = 0x38)

4.5

0.3

mA

µA

Average V_DDSupply Current

100 Hz data rate; LED offset = 25 µs; LED pulse period (t_{LED_PERIOD}) =

13 µs; LED peak current = 25 mA

1 Pulse

Time Slot A only

Time Slot B only

Both Time Slot A and Time Slot B

Time Slot A only

Time Slot B only

Both Time Slot A and Time Slot B

LED peak current = 25 mA

50 Hz data rate

100 Hz data rate

200 Hz data rate

53

41

76

107

95

µA

10 Pulses

184

Average V_LEDSupply Current

1 Pulse

3.75

7.5

15

38

75

µA

10 Pulses

50 Hz data rate

100 Hz data rate

200 Hz data rate

150

SATURATION ILLUMINANCE¹

Direct Illumination

Blackbody color temperature (T = 5500 K)², PDET1 and PDET2

multiplexed into a single channel (1.2 mm²active area)

Transimpedance amplifier (TIA) gain = 25 kΩ

TIA gain = 50 kΩ

58.8

29.4

14.7

7.4

kLux

TIA gain = 100 kΩ

TIA gain = 200 kΩ

DATA ACQUISITION

ADC Resolution

Per Sample

Per Data Read

LED PERIOD

Single pulse

14

20

27

19

17

Bits

64 pulses to 255 pulses

64 pulses to 255 pulses; 128 samples averaged

AFE width = 4 µs³

AFE width = 3 µs

Time Slot A or Time Slot B; normal mode; 1 pulse;

SLOTA_LED_OFFSET = 23 µs; SLOTA_PERIOD = 19 µs

13

11

0.122

µs

Sampling Frequency⁴

2000 Hz

Both time slots; normal mode; 1 pulse;

SLOTA_LED_OFFSET = 23 µs; SLOTA_PERIOD = 19 µs

Time Slot A or Time Slot B; normal mode; 8 pulses;

SLOTA_LED_OFFSET = 23 µs; SLOTA_PERIOD = 19 µs

0.122

1600 Hz

1000 Hz

Both time slots; normal mode; 8 pulses;

SLOTA_LED_OFFSET = 23 µs; SLOTA_PERIOD = 19 µs

Rev. B | Page 3 of 61

ADPD188GG

Data Sheet

Parameter

Test Conditions/Comments

Min

Typ

Max Unit

CATHODE PIN (PDC) VOLTAGE

During All Sampling Periods

Register 0x54, Bit 7 = 0x0; Register 0x3C, Bit 9 = 1⁵

Register 0x54, Bit 7 = 0x0; Register 0x3C, Bit 9 = 0

Register 0x54, Bit 7 = 0x1; Register 0x54, Bits[9:8] = 0x0⁵

Register 0x54, Bit 7 = 0x1; Register 0x54, Bits[9:8] = 0x1

Register 0x54, Bit 7 = 0x1; Register 0x54, Bits[9:8] = 0x2

1.8

1.3

1.8

1.3

TIA_VREF +

0.25

0

1.8

1.3

TIA_VREF +

0.25

0

1.8

1.3

1.8

1.3

TIA_VREF +

0.25

V

During Time Slot A Sampling

During Time Slot B Sampling

During Sleep Periods

Register 0x54, Bit 7 = 0x1; Register 0x54, Bits[9:8] = 0x3⁶

Register 0x54, Bit 7 = 0x1; Register 0x54, Bits[11:10] = 0x0⁵

Register 0x54, Bit 7 = 0x1; Register 0x54, Bits[11:10] = 0x1

Register 0x54, Bit 7 = 0x1; Register 0x54, Bits[11:10] = 0x2

V

Register 0x54, Bit 7 = 0x1; Register 0x54, Bits[11:10] = 0x3⁶

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

V

Register 0x54, Bit 7 = 0x1; Register 0x54, Bits[13:12] = 0x3

0

V

LEDs

LED Peak Current Setting

Adjustable via the Register 0x22 through Register 0x25

settings

12

370

mA

Dominant Wavelength⁷

LED1; Green LED

Luminous Intensity

Photodiode

I_F= 40 mA

λ = 525 nm, I_F= 40 mA at 25°C

525

nm

2800

3200 mcd

A/W

Responsivity

Wavelength, λ = 525 nm

0.25

Active Area

Photodiode 1

Photodiode 2

0.4

0.8

mm²

POWER SUPPLY VOLTAGES

V_DD

V_LED1

DC Power Supply Rejection

Ratio (PSRR)

The ADPD188GG does not require a specific power-up sequence

Applied at the VDD1 and VDD2 pins

1.7

4

1.8

4.5

24

1.9

5.0

V

dB

8, 9

At 75% full scale input signal

TEMPERATURE RANGE

Operating

−40

+85

°C

¹Saturation illuminance refers to the amount of ambient light that saturates the ADPD188GG signal. Actual results may vary by factors of up to 2× from typical

specifications. As a point of reference, Air Mass 1.5 (AM1.5) sunlight (brightest sunlight) produces 100 kLux.

²Blackbody color temperature (T = 5800 K) closely matches the light produced by solar radiation (sunlight).

³Minimum LED period = (2 × AFE width) + 5 µs.

⁴The 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.

⁵This mode may induce additional noise and is not recommended unless necessary. The 1.8 V setting uses V_DD, 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.

⁶This setting is not recommended for photodiodes because it causes a 1.3 V forward bias of the photodiode.

⁷I_Fis the forward current of the diode.

⁸Set V_LEDxsuch that the maximum desired LED current is achievable with the turn on voltage of the LEDs that are wired to the LEDx/DNC pins. The LEDx/DNC pins are

connected to the LEDx driver, which can be modeled as current sinks (see Figure 1). When an appropriate V_LEDxis used, the voltage at the LEDx/DNC pins adjusts

automatically to accommodate the LED turn on voltage and the LED current.

⁹See Figure 9 for the current limitation at the minimum VLED supply voltage, V_LED

.

Rev. B | Page 4 of 61

Data Sheet

ADPD188GG

ANALOG SPECIFICATIONS

VDD1 = VDD2 = 1.8 V, and T_A= full operating temperature range, unless otherwise noted.

Table 2.

Parameter

Test Conditions/Comments

Min

Typ

Max Unit

EXT_INx SERIES RESISTANCE (R_IN)¹

Measured from −3 µA to +3 µA

6.5

kΩ

PULSED SIGNAL CONVERSIONS, 3 μs

4 μs wide AFE integration; normal operation,

Register 0x43 and Register 0x45 = 0xADA5

TIA feedback resistor

25 kΩ

WIDE LED PULSE²

ADC Resolution³

3.27

1.64

0.82

0.41

nA/LSB

50 kΩ

100 kΩ

200 kΩ

ADC Saturation Level

TIA feedback resistor

25 kΩ

26.8

13.4

6.7

μA

50 kΩ

100 kΩ

200 kΩ

3.35

Ambient Signal Headroom on Pulsed

Signal

TIA feedback resistor

25 kΩ

50 kΩ

100 kΩ

200 kΩ

23.6

11.8

5.9

μA

2.95

PULSED SIGNAL CONVERSIONS, 2 μs

WIDE LED PULSE²

ADC Resolution³

3 μs wide AFE integration; normal operation, Register 0x43

and Register 0x45 = 0xADA5

TIA feedback resistor

25 kΩ

4.62

2.31

1.15

0.58

nA/LSB

50 kΩ

100 kΩ

200 kΩ

ADC Saturation Level

TIA feedback resistor

25 kΩ

37.84

18.92

9.46

μA

50 kΩ

100 kΩ

200 kΩ

4.73

Ambient Signal Headroom on Pulsed

Signal

TIA feedback resistor

25 kΩ

50 kΩ

100 kΩ

200 kΩ

12.56

6.28

3.14

1.57

μA

FULL SIGNAL CONVERSIONS⁴

TIA Saturation Level Pulsed Signal and

Ambient Level

TIA feedback resistor

25 kΩ

50 kΩ

100 kΩ

200 kΩ

50.4

25.2

12.6

6.3

μA

TIA Linear Range

TIA feedback resistor

25 kΩ

42.8

21.4

10.7

5.4

μA

50 kΩ

100 kΩ

200 kΩ

Rev. B | Page 5 of 61

ADPD188GG

Data Sheet

Parameter

Test Conditions/Comments

Min

Typ

Max Unit

SYSTEM PERFORMANCE

Total Output Noise Floor

Normal mode; per pulse; per channel; no LED; photodiode

capacitance (C_PD) = 25 pF

25 kΩ; referred to ADC input

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

channel⁵

1.0

4.6

3.3

78.3

LSB rms

nA rms

dB

50 kΩ; referred to ADC input

1.1

2.5

1.8

77.4

1.2

LSB rms

nA rms

dB

LSB rms

nA rms

dB

LSB 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 channel⁵

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 channel⁵

200 kΩ; referred to ADC input

1.4

0.98

76.7

1.4

0.81

0.57

75.3

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 channel⁵

¹The R_IN value can be ignored for current source inputs or for PD inputs. This value is important for calculating correct voltages for voltage inputs through a resistor.

²This saturation level applies to the ADC only and, therefore, includes only the pulsed signal. Any nonpulsatile signal is removed prior to the ADC stage.

³ADC resolution is listed per pulse. If using multiple pulses, divide by the number of pulses.

⁴This saturation level applies to the full signal path and, therefore, includes both the ambient signal and the pulsed signal.

⁵The 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

VDD1 = VDD2 = 1.7 V to 1.9 V, unless otherwise noted.

Table 3.

Parameter

LOGIC INPUTS

Input Voltage Level

High

Symbol

Test Conditions/Comments

Min

Typ

Max

Unit

V_IH

V_IL

GPIOx, SCLK, MOSI, CS

SCL, SDA

0.7 × VDDx

VDDx

V

High

Low

3.6

0.3 × VDDx

Input Current Level

High

Low

Input Capacitance

LOGIC OUTPUTS

Output Voltage Level

High

I_IH

I_IL

C_IN

−10

+10

µA

pF

10

GPIOx, MISO

V_OH

V_OL

2 mA high level output current

2 mA low level output current

SDA

2 mA low level output current

SDA

VDDx − 0.5

V

Low

0.5

Output Voltage Level

Low

Output Current Level

Low

V_OL1

I_OL

0.2 × VDDx

V

V_OL1= 0.6 V

6

mA

Rev. B | Page 6 of 61

Data Sheet

ADPD188GG

TIMING SPECIFICATIONS

I²C Timing Specifications

Table 4.

Parameter

Symbol

Min

Typ

Max

Unit

SCL

Frequency

1

0.4

Mb/sec

Minimum Pulse Width

High

Low

t₁

t₂

370

530

ns

START CONDITION

Hold Time

t₃

t₄

t₅

t₉

260

50

ns

Setup Time

SDA SETUP TIME

SDA HOLD TIME

SCL AND SDA

Rise Time

0

t₆

t₇

1000

300

ns

Fall Time

STOP CONDITION

Setup Time

t₈

260

ns

t₅

t₃

t₉

SDA

SCL

t₆

t₁

t₂

t₇

t₄

t₈

Figure 2. I²C Timing Diagram

Rev. B | Page 7 of 61

ADPD188GG

Data Sheet

SPI Timing Specifications

Table 5.

Parameter

SCLK

Symbol

Test Conditions/Comments

Min

Typ

Max

Unit

Frequency

Minimum Pulse Width

High

f_SCLK

10

MHz

t_SCLKPWH

t_SCLKPWL

20

ns

Low

CS

Setup Time

t_CS

CS setup to SCLK rising edge

CS hold from SCLK rising edge

CS pulse width high

10

ns

S

Hold Time

t_CS

H

Pulse Width High

t_CS

PWH

MOSI

Setup Time

Hold Time

ns

t_MOSIS

t_MOSIH

t_MISOD

MOSI setup to SCLK rising edge

MOSI hold from SCLK rising edge

10

MISO OUTPUT DELAY

MISO valid output delay from SCLK

falling edge

21

ns

t_CSH

t_CSS

t_CSPWH

t_SCLKPWL

t_SCLKPWH

CS

SCLK

MOSI

t_MOSIH

t_MOSIS

MISO

t_MISOD

Figure 3. SPI Timing Diagram

Rev. B | Page 8 of 61

Data Sheet

ADPD188GG

ABSOLUTE MAXIMUM RATINGS

RECOMMENDED SOLDERING PROFILE

Table 6.

Figure 4 and Table 8 provide details about the recommended

soldering profile.

Parameter

Rating

VDD1, VDD2 to AGND

VDD1, VDD2 to DGND

EXT_IN1/EXT_IN2

−0.3 V to +2.2 V

−0.3 V to +2.2V

−0.3 V to +2.2 V

−0.3 V to +3.6 V

−0.3 V to +5.0 V

CRITICAL ZONE

P

T

TO T

P

L

T

P

GPIO0/GPIO1 to DGND

MISO/MOSI/SCLK/CS to DGND

LEDx/DNC to LGND

SCL/SDA to DGND

T

L

T

L

SMAX

T

SMIN

VLEDx to LGND¹

t_S

RAMP-DOWN

PREHEAT

Electrostatic Discharge (ESD)

Human Body Model (HBM)

Charged Device Model (CDM)

Machine Model (MM)

Solder Reflow (Pb-Free)

Peak Temperature

Time at Peak Temperature

Temperature Range

Powered

3000 V

1250 V

100 V

t25°C TO PEAK

TIME

Figure 4. Recommended Soldering Profile

260 (+0/−5)°C

<30 sec

Table 8. Recommended Soldering Profile

Profile Feature

Condition (Pb-Free)

Average Ramp Rate (T_Lto T_P)

Preheat

2°C/sec max

−40°C to +85°C

−40°C to +105°C

105°C

Storage

Junction Temperature

Minimum Temperature (T_SMIN

)

150°C

Maximum Temperature (T_SMAX

Time, T_SMINto T_SMAX(t_S)

T_SMAXto T_LRamp-Up Rate

)

200°C

60 sec to 120 sec

2°C/sec max

¹The absolute maximum voltage allowable between VLEDx and LGND is the

voltage that causes the LEDx/DNC pins to reach or exceed their absolute

maximum voltage.

Time Maintained Above Liquidous

Temperature

Liquidous Temperature (T_L)

Time (t_L)

Peak Temperature (T_P)

Time Within 5°C of Actual Peak

Temperature (t_P)

Ramp-Down Rate

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.

217°C

60 sec to 150 sec

260 (+0/−5)°C

<30 sec

3°C/sec max

Time 25°C to Peak Temperature

8 minutes max

THERMAL RESISTANCE

Thermal performance is directly linked to printed circuit board

(PCB) design and operating environment. Careful attention to

PCB thermal design is required.

ESD CAUTION

Table 7. Thermal Resistance

Package Type¹

CE-24-1

ASIC

Supply Pins

θ_JA

Unit

VDD1, VDD2

VLED1

67

°C/W

LED1

156 °C/W

¹Thermal impedance simulated values are based on JEDEC 2S2P and two

thermal vias. See JEDEC JESD51.

Rev. B | Page 9 of 61

ADPD188GG

Data Sheet

PIN CONFIGURATION AND FUNCTION DESCRIPTIONS

19

18

17

16

15

14

13

1

2

3

4

5

6

PDC

EXT_IN2

NIC

CS

SCLK

MOSI

MISO

GPIO1

GPIO0

SDA

ADPD188GG

TOP VIEW

VDD2

VLED1

NIC

(Not to Scale)

NIC

7

NOTES

1. DNC = DO NOT CONNECT. DO NOT CONNECT TO THIS PIN

WHEN USING INTERNAL LEDs.

2. NIC = NO INTERNAL CONNECTION.

Figure 5. Pin Configuration

Table 9. Pin Function Descriptions

Pin No. Mnemonic Type¹

Description

Photodiode Common Cathode Bias.

1

2

3

4

5

6

7

PDC

EXT_IN2

NIC

VDD2

VLED1

NIC

AO

AI

NIC

S

NIC

EXT_IN2 Current Input.

No Internal Connection. This pin is not internally connected.

1.8 V Supply.

Green LED Anode Supply Voltage.

No Internal Connection. This pin is not internally connected.

NIC

8

9

LED1/DNC

LED3

LED2

LGND

SCL

AO/DNC LED1 Driver Current Sink (LED1)/Do Not Connect (DNC). Do not connect to this pin when using internal LEDs.

AO

S

LED3 Driver Current Sink. If not in use, leave this pin floating.

LED2 Driver Current Sink. If not in use, leave this pin floating.

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

LED Driver Ground.

DI

I²C Clock Input.

SDA

DO

DIO

DO

DI

I²C Data Output.

GPIO0

GPIO1

MISO

MOSI

SCLK

General-Purpose Input/Output 0.

General-Purpose Input/Output 1.

SPI Master Input, Slave Output.

SPI Master Output, Slave Input.

DI

SPI Clock Input.

CS

DI

SPI Chip Select (Active Low).

DGND

AGND

VREF

VDD1

EXT_IN1

S

Digital Ground.

S

Analog Ground.

REF

S

Internally Generated ADC Voltage Reference. Connect a 1 µF ceramic capacitor from VREF to ground.

1.8 V Supply.

AI

EXT_IN1 Current Input.

¹AO is analog output, AI is analog input, NIC is not internally connected, S is supply, DNC is do not connect, DI is digital input, DO is digital output, DIO is digital input/output, and

REF is analog reference.

Rev. B | Page 10 of 61

Data Sheet

ADPD188GG

TYPICAL PERFORMANCE CHARACTERISTICS

0.35

FILTERED ADPD188

0.30

0.25

0.20

0.15

0.10

0.05

0

20

15

10

5

0

27.5 28.0 28.5 29.0 29.5 30.0 30.5 31.0 31.5 32.0

FREQUENCY (MHz)

200

300

400

500

600

700

800

900

1000

1100

WAVELENGTH (nm)

Figure 8. 32 MHz Clock Frequency Distribution; Default Settings;

Before User Calibration, Register 0x4D = 0x425E

Figure 6. Typical Photodiode Responsivity

400

30

LED COARSE SETTING = 0xF

350

300

250

200

150

100

25

20

15

10

5

50

LED COARSE SETTING = 0x0

0

–25

–20

–15

–10

–5

0

5

10

15

SAMPLE FREQUENCY DEVIATION FROM NOMINAL (%)

LED DRIVER VOLTAGE (V)

Figure 9. LED Driver Current vs. LED Driver Voltage at Various Coarse Settings

Figure 7. 32 kHz Clock Frequency Distribution; Default Settings;

Before User Calibration, Register 0x4B = 0x2612

Rev. B | Page 11 of 61

ADPD188GG

Data Sheet

0°

20°

0°

20°

HORIZONTAL

VERTICAL

HORIZONTAL

VERTICAL

1.0

0.8

1.0

0.8

40°

60°

80°

0.6

0.4

0.2

0.4

0.6

0.4

0.2

0.4

Figure 10. PD1 Relative Sensitivity vs. Angular Displacement

Figure 12. LED Relative Intensity vs. Angular Displacement

0°

20°

1.0

0.8

HORIZONTAL

VERTICAL

40°

60°

80°

0.6

0.4

0.2

0.4

0.6

Figure 11. PD2 Relative Sensitivity vs. Angular Displacement

Rev. B | Page 12 of 61

Data Sheet

ADPD188GG

THEORY OF OPERATION

This highly integrated system works well in environments where

ambient light is poorly controlled and the signal modulation

ratio is low. As a result, the device produces high SNR for

relatively low LED power.

INTRODUCTION

The ADPD188GG is a complete, integrated, optical module

designed for photoplethysmography (PPG) measurements.

The module contains two optical detectors. Photodiode 1 (PDET1)

has 0.4 mm²of active area and is connected to Channel 3 of the

ASIC. Photodiode 2 (PDET2) has 0.8 mm²of active area and is

connected to Channel 4 of the ASIC. The two photodiodes can be

combined into a single detector with 1.2 mm²of active area.

Both photo-diodes are coated with an infrared (IR) cut filter

that maximizes ambient light rejection without the need for other

light cancellation techniques.

OPTICAL COMPONENTS

Photodiode

The ADPD188GG integrates a 1.2 mm²deep junction photodiode.

The optical sensing area is a dual detector that is connected to

Channel PD3 and Channel PD4 in the ASIC. The photodiodes

are accessible from Time Slot A or Time Slot B. The responsivity

of the ADPD188GG photodiode is shown in Figure 6.

The module combines the dual photodetector with two green

LEDs, and a mixed-signal, photometric, front-end ASIC into a

single compact device for optical measurements. The on-board

ASIC includes an analog signal processing block, an ADC, a

digital signal processing block, an I²C and SPI communication

interface, and three, independently programmable, pulsed LED

current sources.

LEDs

The ADPD188GG module integrates two green LEDs.

Table 10. LED Dominant Wavelength

LED Color

Driver

Typical Wavelength (nm)

Green (2×)

LED1

525

In addition to the integrated LEDs, the ADPD188GG has the

ability to drive external LEDs.

The core circuitry stimulates the LEDs and measures the

corresponding optical return signals in discrete data locations.

Data can be read from output registers directly or through a

first in, first out (FIFO) buffer.

1.67

1.42

PD1

PD2

5.0

Figure 13. Optical Component Locations

Rev. B | Page 13 of 61

ADPD188GG

Data Sheet

The timing parameters in Figure 14 are defined as follows:

DUAL TIME SLOT OPERATION

t_A(µs) = 25 + n_A× 19

The ADPD188GG operates 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 14.

where n_Ais the number of pulses for Time Slot A (Register 0x31,

Bits[15:8]).

t_B(µs) = 25 + n_B× 19

where n_Bis the number of pulses for Time Slot B (Register 0x36,

Bits[15:8]).

t₁= 68 µs, the processing time for Time Slot A

t₂= 20 µs, the processing time for Time Slot B

f

_SAMPLEis the sampling frequency (Register 0x12, Bits[15:0]).

ACTIVE

t_A

t₁

t_B

t₂

n_APULSES

n_BPULSES

SLEEP

TIME SLOT A

TIME SLOT B

1/f

SAMPLE

Figure 14. Time Slot Timing Diagram

Table 11. Recommended AFE and LED Timing Configuration

Address

Register Name

SLOTx_LEDMODE

SLOTx_AFEMODE

Time Slot A

0x30

0x39

Time Slot B

0x35

0x3B

Recommended Setting

0x0319

0x2209

Rev. B | Page 14 of 61

Data Sheet

ADPD188GG

PDC

TIME SLOT SWITCH

PDET2

Multiple configurations of the four input channels are supported,

depending on the settings of Register 0x14. The integrated

photodiodes can either be routed to Channel 3 and Channel 4,

or summed together into Channel 1. The external EXT_IN1

and EXT_IN2 inputs can be routed to Channel 1 and Channel 2,

respectively, or summed into Channel 2. See Figure 15 and

Figure 16 for the supported configurations. In Figure 15 and

Figure 16, PDET1 is Photodiode 1, and PDET2 is Photodiode 2.

PDET1

CH1

See Table 12 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.

EXT_IN1

EXT_IN2

CH2

INPUT CONFIGURATION FOR

REGISTER 0x14[11:8] = 1

REGISTER 0x14[7:4] = 1

EXT_IN1

CH1

Figure 16. Current Summation

EXT_IN2

CH2

PDC

PDET1

CH3

PDC

PDET2

CH4

INPUT CONFIGURATION FOR

REGISTER 0x14[11:8] = 5

REGISTER 0x14[7:4] = 5

Figure 15. PD1 to PD4 Connection

Table 12. Time Slot Switch (Register 0x14)

Address

Bits

Name

Description

0x14

[11:8]

SLOTB_PD_SEL

These bits select the connection of input channels for Time Slot B as shown in Figure 15 and

Figure 16.

0x0: inputs are floating in Time Slot B.

0x1: PDET1 and PDET2 are connected to Channel 1; EXT_IN1 and EXT_IN2 are connected

to Channel 2 during Time Slot B.

0x5: EXT_IN1 is connected to Channel 1, EXT_IN2 is connected to Channel 2, PDET1 is

connected to Channel 3, and PDET2 is connected to Channel 4 during Time Slot B.

Other: reserved.

[7:4]

SLOTA_PD_SEL

These bits select the connection of input channels for Time Slot A as shown in Figure 15 and

Figure 16.

0x0: inputs are floating in Time Slot A.

0x1: PDET1 and PDET2 are connected to Channel 1; EXT_IN1 and EXT_IN2 are connected

to Channel 2 during Time Slot A.

0x5: EXT_IN1 is connected to Channel 1, EXT_IN2 is connected to Channel 2, PDET1 is

connected to Channel 3, and PDET2 is connected to Channel 4 during Time Slot A.

Other: reserved.

Rev. B | Page 15 of 61

ADPD188GG

Data Sheet

Providing an External 32 kHz Clock

ADJUSTABLE SAMPLING FREQUENCY

The ADPD188GG has an option for the user to provide an

external 32 kHz clock to the device 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 only. To enable the 32 kHz external

clock, use the following procedure at startup:

Register 0x12 controls the sampling frequency setting of the

ADPD188GG 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

transition of the internal state machine. The maximum sampling

frequencies for some sample conditions are listed in Table 1. The

maximum sample frequency for all conditions, f_{SAMPLE,_MAX}, is

determined by the following equation:

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 0x1 to Register 0x4F, Bits[6:5] to enable the

GPIO1 pin as an input.

3. Write 0x2 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

device is in program mode to configure the device as

required.

f

_{SAMPLE_MAX}= 1/(t_A+ t₁+ t_B+ t₂+ t_{SLEEP_MIN})

where t_{SLEEP_MIN}is the minimum sleep time required between

samples. See the Dual Time Slot Operation section for the

definitions of t_A, t₁, t_B, and t₂.

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, t_Aand t₁do not add to the sampling period and the

new maximum sampling frequency is calculated as follows:

f

_{SAMPLE_MAX}= 1/(t_B+ t₂+ t_{SLEEP_MIN})

6. Write 0x2 to Register 0x10 to start the normal sampling

EXTERNAL SYNCHRONIZATION FOR SAMPLING

operation

The ADPD188GG provides an option to use an external

synchronization signal to trigger the sampling periods. This

external sample synchronization 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 provides the trigger. To enable the external

synchronization signal feature, use the following procedure:

STATE MACHINE OPERATION

During each time slot, the ADPD188GG operates according to

a state machine. The state machine operates in the sequence

shown in Figure 17.

STANDBY

REGISTER 0x10 = 0x0000

ULTRALOW POWER MODE

NO DATA COLLECTION

ALL REGISTER VALUES ARE RETAINED.

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

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.

PROGRAM

REGISTER 0x10 = 0x0001

SAFE MODE FOR PROGRAMMING REGISTERS

NO DATA COLLECTION

DEVICE IS FULLY POWERED IN THIS MODE.

NORMAL OPERATION

4. Write 0x2 to Register 0x10 to start the sampling

operations.

REGISTER 0x10 = 0x0002

LEDs ARE PULSED AND PHOTODIODES ARE SAMPLED

STANDARD DATA COLLECTION

DEVICE POWER IS CYCLED BY INTERNAL STATE MACHINE.

5. Apply the external synchronization 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. The maximum frequency

constraints also apply in this case.

Figure 17. State Machine Operation Flowchart

Rev. B | Page 16 of 61

Data Sheet

ADPD188GG

The ADPD188GG operates in one of three modes: standby,

program, or normal sampling mode.

NORMAL MODE OPERATION AND DATA FLOW

In normal mode, the ADPD188GG follows a specific pattern set

up by the state machine. This pattern is shown in the corres-

ponding data flow diagram in Figure 18. The pattern, in order,

is as follows:

Standby mode is a power saving mode in which data collection

does not occur. All register values are retained in this mode. To

place the device in standby mode, write 0x0 to Register 0x10,

Bits[1:0]. The device powers up in standby mode.

1. LED pulse and sample. The ADPD188GG pulses external

LEDs. The response of the photodiode to the reflected light

is measured by the ADPD188GG. 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.

Program mode is used for programming registers. Always cycle

the ADPD188GG through program mode when writing registers

or changing modes. Because power cycling does not occur in this

mode, the device may consume higher current in program mode

than in normal operation. To place the device in program mode,

write 0x1 to Register 0x10, Bits[1:0].

In normal operation, the ADPD188GG pulses light and collects

data. Power consumption in this mode depends on the pulse

count and data rate. To place the device in normal sampling

mode, write 0x2 to Register 0x10, Bits[1:0].

[14 + LOG (n × N )] BITS

2

A

UP TO 27 BITS

[14 + LOG (n )] BITS

UP TO 20 BITS

N

2

A

16-BIT CLIP

IF VAL ≤ (2 – 1)

VAL = VAL

16 BITS

16

N

A

16

ELSE VAL = 2 – 1

1

14 BITS

n

A

20-BIT CLIP

IF VAL ≤ (2 – 1)

20

n

A

n

A

VAL = VAL

20

ELSE VAL = 2 – 1

0

1

14-BIT

ADC

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

UP TO 27 BITS

Figure 18. State Machine Operating Sequence (Datapath)

Rev. B | Page 17 of 61

ADPD188GG

Data Sheet

data is still acquired by the AFE at the sampling frequency,

_SAMPLE(see Register 0x12), but new data is written to the

LED Pulse and Sample

f

At each sampling period, the selected LED driver drives a series

of LED pulses, as shown in Figure 19. The magnitude, duration,

and number of pulses are programmable over the communications

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.

registers at the rate of f_SAMPLE/N every N^thsample. 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 setting can reduce the number of FIFO reads

required by the host processor.

After each LED pulse, the photodiode output relating to 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 ADPD188GG

via the communications interface, 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 the factors 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 ADPD188GG offers 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 18, 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.

The data read operations are described in more detail in the

Reading Data section.

When using the averaging feature set up by the register, subse-

quent pulses can be averaged by powers of 2. The user can

select from 2, 4, 8, …, up to 128 samples to be averaged. Pulse

SHOWN WITH f

SAMPLE

= 10Hz

OPTICAL SAMPLING

LOCATIONS

0

0.5

1.0

1.5

TIME (Seconds)

2.0

2.5

3.0

LED

CURRENT

(I

)

LED

NUMBER OF LED PULSES (n_AOR n_B

)

Figure 19. Example of a PPG Signal Sampled at a Data Rate of 10 Hz Using Five Pulses per Sample

Rev. B | Page 18 of 61

Data Sheet

ADPD188GG

COMMUNICATIONS INTERFACE

The ADPD188GG supports both an SPI and I²C serial interface,

although only one can be used at any given time in the actual

application. All internal registers are accessed through the

selected communications interface.

For multiword operations, each pair of data bytes is followed by

an acknowledge (ACK) 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, allowing the user to read without readdressing

each register, thereby reducing the amount of overhead

required to read multiple registers. This auto-increment does

not apply to the register that precedes the FIFO, Register 0x5F,

or the last data register, Register 0x7E.

I²C INTERFACE

The ADPD188GG I²C conforms to the UM10204 I²C-Bus

Specification and User Manual, Rev. 05—9 October 2012,

available from NXP Semiconductors. The device supports fast

mode (400 kbps) data transfer. Register read and write operations

are supported, as shown in Figure 20. The 7-bit I²C slave address

2

CS

for the device is 0x64. If the I C interface is being used, the

pin must be pulled high to disable the SPI port.

All register writes are single-word only and require 16 bits

(one word) of data.

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.

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 13. Definitions of I²C Terminology

Term

Description

SCL

Serial clock.

SDA

Serial address and data.

Master

Slave

Start (S)

Start (Sr)

Stop (P)

ACK

The device that initiates a transfer, generates clock signals, and terminates a transfer.

The device addressed by a master. The ADPD188GG 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 (ACK) or no acknowledge (NACK) clock pulse, the SDA line is pulled low, and it remains low.

During the ACK or NACK clock pulse, the SDA line remains high.

NACK

Slave Address

Read (R)

Write (W)

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.

A 0 indicates a transmission.

2

I

C WRITE

REGISTER WRITE

MASTER START

SLAVE

SLAVE ADDRESS + WRITE

REGISTER ADDRESS

DATA[15:8]

DATA[7:0]

STOP

ACK

2

I

C SINGLE-WORD READ MODE

REGISTER READ

MASTER START

SLAVE

ACK

SLAVE ADDRESS + WRITE

REGISTER ADDRESS

Sr

SLAVE ADDRESS + READ

NACK

STOP

DATA[15:8]

ACK

DATA[7:0]

2

I

C MULTIWORD READ MODE

REGISTER READ

MASTER START

SLAVE

ACK/NACK

SLAVE ADDRESS + WRITE

ACK

DATA[15:8]

DATA[7:0]

DATA TRANSFERRED

NOTES

1. THE SHADED AREAS REPRESENT WHEN THE DEVICE IS LISTENING.

Figure 20. I²C Write and Read Operations

Rev. B | Page 19 of 61

ADPD188GG

Data Sheet

R

Table 15. SPI Address and Write/ Byte Format

SPI PORT

Bit 0

Bit 1

Bit 2

Bit 3

Bit 4

Bit 5

Bit 6

Bit 7

CS

The SPI port uses a 4-wire interface, consisting of the

,

A6

A5

A4

A3

A2

A1

A0

W/R

MOSI, MISO, and SCLK signals, and it is always a slave port.

CS

The

signal goes low at the beginning of a transaction and

Data on the MOSI pin is captured on the rising edge of the

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 14. A timing diagram is shown in Figure 3.

Write all data MSB first.

clock, and data is propagated on the MISO pin on the falling

edge of the clock. The maximum read and write speed for the

SPI slave port is 10 MHz.

A sample timing diagram for a multiple word SPI write operation

to a register is shown in Figure 21. A sample timing diagram of

a single-word SPI read operation is shown in Figure 22. The

MISO pin transitions from being three-state to being driven

following the reception of a valid bit. In this example, Byte 0

contains the address and the W/ bit, and subsequent bytes

carry the data. A sample timing diagram of a multiple word SPI

read operation is shown in Figure 23. In Figure 21 to Figure 23,

rising edges on SCLK are indicated with an arrow, signifying

that the data lines are sampled on the rising edge.

R

Table 14. 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]

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.

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 W/ bit. This bit determines whether the communication is

R

a write (Logic Level 1) or a read (Logic Level 0). This format is

shown in Table 15.

Rev. B | Page 20 of 61

Data Sheet

ADPD188GG

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

Figure 21. 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 22. 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 23. SPI Slave Read Clocking (Burst Read Mode, N Bytes)

Rev. B | Page 21 of 61

ADPD188GG

Data Sheet

APPLICATIONS INFORMATION

0.1µF

TYPICAL CONNECTION DIAGRAM

1.8V

1.0µF

Figure 24 shows the recommended connection diagram for the

ADPD188GG using the SPI communications port. Figure 25

shows a circuit using the I²C port. The desired communications

port, together with the GPIO0 and GPIO1 lines, connects to a

system microprocessor or sensor hub. When using the SPI port,

the I²C interface must be disabled by connecting the SDA and

SCL pins high to 1.8 V. When using the I²C interface, the SPI is

PDC

CS

1.8V

1

2

3

4

5

6

7

19

18

17

16

15

14

13

EXT_IN2

SCLK

MOSI

MISO

NIC

1.8V

0.1µF

CS

disabled by connecting to 1.8 V. Tie the unused inputs, SCLK

VDD2

ADPD188GG

and MOSI, to ground. The EXT_IN1 and EXT_IN 2 pins are

current inputs and can be connected to external sensors. A voltage

source can be connected to the EXT_IN1 and EXT_IN2 pins

through a series resistance, effectively converting the voltage

into a current (see the Using the EXT_IN 1 and EXT_IN 2

Inputs with a Voltage Source section).

VLED1

NIC

V

GPIO1

GPIO0

1.8V

LED1

C

VLED

NIC

10kΩ

SDA

Provide a regulated 1.8 V supply, tied to VDD1 and VDD2. The

VLEDx level uses a standard regulator circuit according to the

peak current requirements specified in Table 1 and calculated

in the Calculating Current Consumption section. Place 0.1 µF

ceramic decoupling capacitors as close as possible to VDD1 and

VDD2; a 1.0 µF ceramic capacitor must be placed as close as

possible to the VREF pin.

SCL

10kΩ

1.8V

Figure 25. I²C Mode Connection Diagram

LAND PATTERN

Figure 26 shows the recommended PCB footprint (land pattern).

Table 8 and Figure 4 provide the recommended soldering profile.

2.0mm

For best noise performance, connect AGND, DGND, and

LGND together at a large conductive surface such as a ground

plane, ground pour, or large ground trace.

0.28mm

0.56mm

0.1µF

1.8V

0.3mm

1.0µF

3.0mm

0.2mm

PDC

1

2

3

4

5

6

7

19

18

17

16

15

14

13

CS

0.18mm

EXT_IN2

SCLK

MOSI

MISO

GPIO1

GPIO0

Figure 26. Land Pattern

NIC

VDD2

ADPD188GG

1.8V

VLED

0.1µF

LED1

VLED1

V

C

NIC

SDA

1.8V

Figure 24. SPI Mode Connection Diagram

Rev. B | Page 22 of 61

Data Sheet

ADPD188GG

Reading Data Using the FIFO

RECOMMENDED START-UP SEQUENCE

The ADPD188GG includes a 128-byte FIFO memory buffer

that can be configured to store data from either or both time

slots. Register 0x11 selects the type 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.

At power-up, the device is in standby mode (Register 0x10 =

0x0), as shown in Figure 17. The ADPD188GG does not

require a particular power-up sequence.

From standby mode, to begin measurement, initiate the

ADPD188GG as follows:

Output Data Rate = f_SAMPLE/Nx

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.

4. Write 0x2 to Register 0x10 to start normal sampling

operation.

where:

f

_SAMPLEis the sampling frequency.

Nx is the averaging factor for each time slot (N_Afor Time Slot A

and N_Bfor Time Slot B). In other words, N_A= N_Bmust 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.

To terminate normal operation, follow this sequence to place

the ADPD188GG 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.

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:

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.

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.

READING DATA

5. Enter normal operation mode by setting Register 0x10 to 0x2.

6. When an interrupt occurs,

The ADPD188GG provides 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.

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.

Rev. B | Page 23 of 61

ADPD188GG

Data Sheet

b. Read a complete packet using one or more multiword

accesses using Register 0x60. Reading the FIFO

automatically frees the space for new samples.

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.

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.

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.

Polling Method

To read data from the FIFO in a polling method, use the

following procedure:

1. In program mode, set the configuration of the time slots as

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.

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

Next, begin the polling operations.

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:

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.

3. Write a 0 to the SLOTA_DATA_HOLD or

SLOTB_DATA_

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:

1. Enter program mode by setting Register 0x10 to 0x1.

2. Write 1 to Register 0x00, Bit 15.

HOLD bits (Register 0x5F, Bit 1 and Bit 2, respectively)

previously set. Sample updates are allowed again.

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:

Because a new sample may arrive while the reads are occurring,

this method prevents the new sample from partially overwriting

the data being read.

CLOCKS AND TIMING CALIBRATION

The ADPD188GG operates using two internal time bases. A

32 kHz clock sets the sample timing, and a 32 MHz clock

controls the timing of 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 (HRM) applications require an accurate

time base to achieve an accurate count of beats per minute. The

ADPD188GG provides 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,

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.

Calibrating the 32 kHz Clock

This procedure calibrates items associated with the output data

rate. Calibration of this clock is important for items where an

accurate data rate is important, such as heart rate measurements.

Rev. B | Page 24 of 61

Data Sheet

ADPD188GG

To calibrate the 32 kHz clock,

OPTIONAL TIMING SIGNALS AVAILABLE ON

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 Bits[2:0] in Register 0x02 and set the

interrupt to occur at the sampling frequency by writing

0x0 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, decrease the CLK32K_ADJUST bits

(Register 0x4B, Bits[5:0]). If the monitored interrupt

frequency is larger than the set sampling frequency,

increase the CLK32K_ADJUST bits.

GPIO0 AND GPIO1

The ADPD188GG provides a number of different timing

signals, available via the GPIO0 and GPIO1 pins, to enable ease

of system synchronization and flexible triggering options. Each

GPIOx pin can be configured as an open-drain output if they

are sharing 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.

Table 16. GPIOx Control Settings

Pin Name

Register, Bits Setting Description

GPIO0

0x02, Bit 0

0x02, Bit 1

0x02, Bit 2

0x02, Bit 8

0x02, Bit 9

0x4F, Bit 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

3. Repeat Step 1 until the monitored interrupt signal

frequency is close to the set sampling frequency.

Calibrating the 32 MHz Clock

This procedure calibrates items associated with the fine timing

within a sample period, such as LED pulse width and spacing,

and assumes that the 32 kHz clock is already calibrated.

GPIO1

To calibrate the 32 MHz clock,

1: drives the bus when asserted

0: disables the GPIO1 pin drive

1: enables the GPIO1 pin drive

1. Write 0x1 to Register 0x5F, Bit 0.

2. Enable the CLK_RATIO calculation by writing 0x1 to

Register 0x50, Bit 5 (CLK32M_CAL_EN). This function

counts the number of 32 MHz clock cycles in two cycles of

the 32 kHz clock. With this function enabled, this value is

stored in Register 0x0A, Bits[11:0] and nominally this ratio

is 2000 (0x07D0).

3. Calculate the 32 MHz clock error as follows:

Clock Error = 32 MHz × (1 − CLK_RATIO/2000)

4. Adjust the frequency by setting Bits[7:0] in Register 0x4D

per the following equation:

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 27 and Figure 28. The time slot settings used to generate

the timing diagrams are described in Table 17.

CLK32M_ADJUST = Clock Error/109 kHz

5. Write 0x0 to Register 0x50, Bit 5 to reset the CLK_RATIO

function.

6. Repeat Step 1 through Step 5 until the desired accuracy is

achieved.

Table 17. ADPD188GG Settings Used for the Timing

Diagrams Shown in Figure 27 and Figure 28

Register Setting Description

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

7. Write 0x1 to Register 0x5F, Bit 0, and set the GPIO0 pin

back to the mode desired for normal operation.

SLEEP

SLOT A

SLOT B

SLOT A

SLOT B

Figure 27. Optional Timing Signals Available on GPIOx—Register 0x0B, Bits[12:8] or Bits[4:0] = 0x02, 0x05, 0x06, 0x07, and 0x0F

Rev. B | Page 25 of 61

ADPD188GG

Data Sheet

SLEEP

SLOT A/B

SLEEP

0x02

0x0C

0x0D

0x0E

Figure 28. Optional Timing Signals Available on GPIOx—Register 0x0B, Bits[12:8] or Bits[4:0] = 0x02, 0x0C, 0x0D, and 0x0E

Interrupt Function

f_S/2 Output

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.

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. The f_S/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.

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.

Logic 0 Output

Setting Register 0x0B, Bits[12:8] or Bits[4:0] = 0x10 configures

the respective pin to provide a Logic 0 output.

Logic 1 Output

Setting Register 0x0B, Bits[12:8] or Bits[4:0] = 0x11 configures

the respective pin to provide a Logic 1 output.

Pulse Outputs

32 kHz Oscillator Output

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.

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.

LED DRIVER PINS AND LED SUPPLY VOLTAGE

The LED driver pins (LED1/DNC, LED2, and LED3) 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 completely cease

proper operation. The voltage of the LED driver pins must not be

confused with the supply voltages for the LEDs themselves (V_LED1

and V_LED2). These are the voltages applied to the anodes of the

internal LEDs connected at VLED1.

Output Data Cycle Signal

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.

LED DRIVER OPERATION

The LED drivers for the ADPD188GG are current sinks. Typical

LED driver current vs. LED driver voltage is shown in Figure 9.

Figure 24 shows the basic schematic of how the ADPD188GG

connects to an LED through the LED driver. The Determining

the Average Current section and the Determining CVLED

section define the requirements for the bypass capacitor (C_VLED

)

and the supply voltages of the LEDs (V_LED).

Rev. B | Page 26 of 61

Data Sheet

ADPD188GG

4.5

4.0

3.5

3.0

2.5

2.0

DETERMINING THE AVERAGE CURRENT

When the ADPD188GG drives an LED, it drives the LED in a

series of short pulses. Figure 29 shows the typical ADPD188GG

configuration of a pulse burst sequence. In this sequence, the

LED pulse width, t_{LED_PULSE}, is 3 µs, and the LED pulse period,

t_{LED_PERIOD}, is 19 µs. The goal of C_VLEDis to buffer the LED between

individual pulses. In the worst case scenario, where the pulse

train shown in Figure 29 is a continuous sequence of short

pulses, the V_LEDsupply must supply the average current.

Therefore, calculate I_{LED_AVERAGE}as follows:

I

_{LED_AVERAGE}= (t_{LED_PULSE}/t_{LED_PERIOD}) × I_{LED_PEAK}

where:

_{LED_AVERAGE}is the average current needed from the V_LEDsupply. It is

also the V_LEDsupply current rating.

_{LED_PEAK}is the peak current setting of the LED.

(1)

LED DRIVER CURRENT (mA)

I

Figure 30. Typical LED Forward-Bias Voltage Drop as a Function of the

LED Driver Current

I

For the C_VLEDcapacitor to be sized correctly, 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 V_LEDbypass capacitor, use the

following equation:

For the numbers shown in Figure 29, I_{LED_AVERAGE}= 3/19 ×

I_{LED_PEAK}. For typical LED timing, the average V_LEDsupply

current is 3/19 × 250 mA = 39.4 mA, indicating that the V_LED

supply must support a dc current of 40 mA.

19µs

t

_{LED_PULSE}×I_{LED _ PEAK}

(2)

C_VLED

=

V_{LED _ MIN}− (V_{FB _ LED _ MAX}+ 0.6)

3µs

where:

_{LED_PULSE}is the LED pulse width.

_{LED_PEAK}is the maximum forward-bias current on the LED used

in operating the device.

t

I

LEAD_PEAK

V_{LED_MIN}is the lowest voltage from the V_LEDsupply with no load.

V

_{FB_LED_MAX}is the maximum forward-bias voltage required on

the LED to achieve I_{LED_PEAK}

.

Figure 29. Typical LED Pulse Burst Sequence Configuration

The numerator of the C_VLEDequation 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 V_LEDsupply and the LED required voltage. The LED

required voltage is the voltage of the anode of the LED such that

the 0.6 V compliance of the LED driver at 200 mA and the

forward-bias voltage of the LED operating at the maximum

current is satisfied. For a typical ADPD188GG example, assume

that the lowest value for the V_LEDsupply is 4.4 V, and that the

peak current is 200 mA for two 525 nm LEDs in parallel. The

minimum value for C_VLEDis then equal to 4 µF.

DETERMINING C_VLED

To determine the C_VLEDcapacitor value, determine the maximum

forward-bias voltage, V_{FB_LED_MAX}, of the LED in operation.

From Figure 30, I_{LED_PEAK}converts to V_{FB_LED_MAX}. For example,

with a 200 mA current, V_{FB_LED_MAX}is 3.65 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 when a 200 mA

current is driven through the resistor. These resistances can be

unnecessary constraints on the V_LEDsupply.

C

_VLED= (3 × 10⁻⁶× 0.20)/(4.4 – (3.65 + 0.6)) = 4.0 µF

(3)

As shown in Equation 3, the minimum supply voltage drops close

to the maximum anode voltage, and the demands on C_VLED

become more stringent, forcing the capacitor value higher. It is

important to plug the correct values into these equations. For

example, using an average value for V_{LED_MIN}instead of the

worst-case value for V_{LED_MIN}can cause a problem; therefore,

adding sufficient margin on C_VLEDis strongly recommended.

Rev. B | Page 27 of 61

ADPD188GG

Data Sheet

t_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 the LED

enabled in that particular time slot.

USING EXTERNAL LEDS

(7)

The ADPD188GG LED driver is also connected to an external

package pin so that the driver can drive external LEDs, if desired.

Figure 31 shows a connection diagram that enables driving

external LEDs.

I

V

= 1.8V

DD

SCALE_X is the scale factor for the LED current drive determined

by Bit 13 of the ILED_COARSE register.

C3

0.1µF

V

NC

LED

LEDx_OFFSET is the pulse start time offset expressed in

seconds.

C

VLED

LEDx_PERIOD is the pulse period expressed in seconds.

PULSE_COUNT is the number of pulses.

C2

1µF

ADPD188GG

LED1

VREF

SCL

TO HOST

PROCESSOR

Note that if either Time Slot A or Time Slot B are disabled, I_{AFE_x}

0 for that respective time slot.

=

SDA

GPIOx

LEDx/DNC

Average V_LEDASupply Current

To calculate the average V_LEDAsupply current, use Equation 8.

_{LED_AVG_A}= SLOTA_LED_WIDTH × I_{LEDA_PK}× DR ×

PULSE_COUNT

where:

I

Figure 31. Using the ADPD188GG LED Drivers to Drive External LEDs

(8)

CALCULATING CURRENT CONSUMPTION

SLOTA_LED_WIDTH is the LED pulse width expressed in

seconds.

The current consumption of the ADPD188GG depends on the

user selected operating configuration, as described in the

following equations.

I_{LEDA_PK}is the peak current, expressed in amps, for the Time Slot A

LED.

Total Power Consumption

Average V_LEDBSupply Current

To calculate the total power consumption, use Equation 4.

To calculate the average V_LEDBsupply current, use Equation 9.

_{LED_AVG_B}= SLOTB_LED_WIDTH × I_{LEDB_PK}× DR ×

PULSE_COUNT

where:

SLOTB_LED_WIDTH is the LED pulse width expressed in

seconds.

Total Power = I_{VDD_AVERAGE}× V_DD+ I_{LED_AVERAGE}× V_LED

where:

_{VDD_AVERAGE}is the average V_DDsupply current (supplied at

VDD1 and VDD2).

_DDis the voltage applied at the VDD1 and VDD2 pins.

_{LED_AVERAGE}is the average LED supply current.

_LEDis the voltage at the VLEDx pins, respectively.

(4)

I

(9)

I

V

I

V

I_{LEDB_PK}is the peak current, expressed in amps, for the Time Slot B

LED.

Average V_DDSupply Current

Optimizing SNR per Watt in a Signal Limited System

To calculate the average V_DDsupply current, use Equation 5.

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_{VDD_AVG}= DR × ((I_{AFE_A}× t_SLOTA) + (I_{AFE_B}× t_SLOTB) +

Q_{PROC_x}) + I_{VDD_STANDBY}

(5)

where:

DR is the data rate in Hz.

_{VDD_STANDBY}= 0.2 µA.

_{PROC_x}is an average charge associated with a processing time,

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 12 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.

I

Q

as follows:

When only Time Slot A is enabled,

Q

_{PROC_A}(C) = 0.35 × 10⁻⁶

When only Time Slot B is enabled,

_{PROC_B}(C) = 0.24 × 10⁻⁶

When Time Slot A and Time Slot B are enabled,

_{PROC_AB}(C) = 0.40 × 10⁻⁶

_{AFE_x}(A) = 3.0 × 10⁻³+ (1.5 × 10⁻³× NUM_CHANNELS) +

(4.6 × 10⁻³× I_{LEDX_PK}/ SCALE_X)

(6)

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.

Q

I

Rev. B | Page 28 of 61

Data Sheet

ADPD188GG

integrator sequence are subtracted and the sequences with a

normal integrator sequence are added. An example diagram of

the integrator chopping sequence is shown in Figure 32.

Tuning the Pulse Count

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.

The result is that any low frequency signal contribution from

the integrator is eliminated, leaving only the integrated signal,

which results in higher SNR, especially at higher numbers of

pulses and at lower TIA gains where the noise contribution of

the integrator becomes more pronounced.

Digital chopping is enabled using the registers and bits detailed

in Table 18. The bit fields define the chopping operation for the

first four pulses. This 4-bit sequence is then repeated for all

subsequent pulses. In Figure 32, a sequence is shown where the

second and fourth pulses are inverted while the first and third

pulses remain in the default polarity (noninverted). This configura-

tion is achieved by setting Register 0x17, Bits[3:0] = 0xA and

Register 0x1D, Bits[3:0] = 0xA for Time Slot A and Time Slot B,

respectively. To complete the operation, the math must be

adjusted using Register 0x58. In this example, set Register 0x58,

Bits[9:8] and Register 0x58, Bits[11:10] to b01 to add the third

pulse and subtract the fourth pulse for Time Slot A and Time

Slot B, respectively. Set Register 0x58, Bits[2:1] and Register

0x58, Bits[6:5] to b01 to add the first pulse and subtract the

second pulse for Time Slot A and Time Slot B, respectively.

This sequence then repeats for every subsequent sequence of

four pulses. An even number of pulses must be used with

integrator chop mode.

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.

When using integrator chop mode, the ADC offset registers

(Register 0x18 through Register 0x1B for Time Slot A, and

Register 0x1E through Register 0x21 for Time Slot B) must be

set to 0. These settings are required because any digital offsets

at the output of the ADC are automatically eliminated when the

math is adjusted to subtract the inverted integration sequences

while the default integration sequences are added. Integrator

chop mode also eliminates the need to manually null the ADC

offsets at startup in a typical application. Note that the elimination

of the offset using chop mode may clip at least half of the noise

signal when no input signal is present, which makes measuring

the noise floor during characterization of the system difficult.

For this reason, perform noise floor characterization of the system

either with chop mode disabled or with chop mode enabled but

with a minimal signal present at the input that increases the

noise floor enough such that it is no longer clipped.

Improving SNR Using Integrator Chopping

The last stage in the analog front end that is integrated into the

ADPD188GG data path is a charge integrator. The integrator

uses an on and off integration sequence, synchronized to the

emitted light pulse, which acts as an additional high-pass filter

to remove offsets, drifts, and low frequency noise from the

previous stages. However, the integrating amplifier can itself

introduce low frequency signal content at a low level. The

ADPD188GG has an integrator chop mode that enables

additional chopping in the digital domain to remove this signal.

This chopping is achieved by using even numbers of pulses per

sample and inverting the integration sequence for half of those

sequences. In the calculation to combine the digitized result of

each of the pulses of the sample, the sequences with an inverted

Rev. B | Page 29 of 61

ADPD188GG

Data Sheet

PULSE 1

PULSE 2

PULSE 3

PULSE 4

LED

BAND-PASS

FILTER

OUTPUT

INTEGRATOR

SEQUENCE

+

–

+

–

+

ADC

+

–

+

–

Figure 32. Diagram of Integrator Chopping Sequence

Table 18. Register Settings for Integrator Chop Mode

Hex

Addr.

Data

Bit(s)

Bit Name

Description

0x17

0x1D

0x58

[3:0]

INTEG_ORDER_A Integration sequence order for Time Slot A. Each bit corresponds to the polarity of the integration

sequence of a single pulse in a four-pulse sequence. Bit 0 controls the integration sequence of

Pulse 1, Bit 1 controls Pulse 2, Bit 2 controls Pulse 3, and Bit 3 controls Pulse 4. After four pulses, the

sequence repeats.

0: normal integration sequence.

1: reversed integration sequence.

[3:0]

INTEG_ORDER_B Integration sequence order for Time Slot B. Each bit corresponds to the polarity of the integration

sequence of a single pulse in a four-pulse sequence. Bit 0 controls the integration sequence of

Pulse 1, Bit 1 controls Pulse 2, Bit 2 controls Pulse 3, and Bit 3 controls Pulse 4. After four pulses, the

sequence repeats.

0: normal integration sequence.

1: reversed integration sequence

[11:10]

FLT_MATH34_B

FLT_MATH34_A

FLT_MATH12_B

FLT_MATH12_A

Time Slot B control for adding and subtracting Sample 3 and Sample 4 in a four-pulse sequence (or

any multiple of four pulses, for example, Sample 15 and Sample 16 in a 16-pulse sequence).

00: add third and fourth.

01: add third and subtract fourth.

10: subtract third and add fourth.

11: subtract third and fourth.

[9:8]

[6:5]

[2:1]

Time Slot A control for adding and subtracting Sample 3 and Sample 4 in a four-pulse sequence

(or any multiple of four pulses, for example, Sample 15 and Sample 16 in a 16-pulse sequence).

00: add third and fourth.

01: add third and subtract fourth.

10: subtract third and add fourth.

11: subtract third and fourth.

Time Slot B control for adding and subtracting Sample 1 and Sample 2 in a four-pulse sequence (or

any multiple of four pulses, for example, Sample 13 and Sample 14 in a 16-pulse sequence).

00: add first and second.

01: add first and subtract second.

10: subtract first and add second.

11: subtract first and second.

Time Slot A control for adding and subtracting Sample 1 and Sample 2 in a four-pulse sequence

(or any multiple of four pulses, for example, Sample 13 and Sample 14 in a 16-pulse sequence).

00: add first and second.

01: add first and subtract second.

10: subtract first and add second.

11: subtract first and second.

Rev. B | Page 30 of 61

Data Sheet

ADPD188GG

from the photodiode increases, the ADC output decreases

toward 0.

MECHANICAL CONSIDERATIONS FOR COVERING

THE ADPD188GG

When configuring the integrator as a buffer, there is the option

of either using a gain of 1 or a gain of 0.7. Using the gain of 0.7

increases the usable dynamic range at the input to the TIA. The

buffer gain is set using Register 0x42, Bit 9 for Time Slot A and

Register 0x44, Bit 9 for Time Slot B. Setting this bit to 0 (default)

sets a gain of 1. Setting this bit to 1 configures the buffer with a

gain of 0.7.

In some applications, it may be necessary to cover the

ADPD188GG to protect it from moisture. The ADPD188GG is

designed with this requirement in mind. The unique cross section

of the device, as shown in Figure 13, prevents light from going

directly from the LED to the detector even with a reasonably thick

window. It is recommended that the window thickness be

<0.7 mm and the air gap between the module and the window

be kept to <0.5 mm for optimal operation.

The ADC output (ADC_OUT) is calculated as follows:

TIA ADC MODE

ADC_OUT= 8192 (((2 × TIA_VREF – 2 × i × R_F− 1.8 V)/

146 µV/LSB) × SLOTx_BUF_GAIN)

(11)

Figure 33 shows a way to put the ADPD188GG into a mode

that effectively runs the TIA directly into the ADC without

using the analog band-pass filter (BPF) and integrator. 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 measure the amount

of background and ambient light. This mode can also measure

other dc input currents, such as leakage resistance.

where:

TIA_VREF is the bias voltage for the TIA (the default value is

1.265 V).

i is the input current to the TIA.

R_Fis the TIA feedback resistor.

SLOTx_BUF_GAIN is either 0.7 or 1, based on the setting of

Register 0x42, Bit 9 and Register 0x44, Bit 9.

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.

V

ADPD188GG

IN

OPTIONAL

BUFFER

R

S

R_IN

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.

TIA

ADC

TIA_VREF

Figure 33. TIA ADC Mode Block Diagram

When the devices are in TIA ADC mode, the BPF 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.

Protecting Against TIA Saturation in Normal Operation

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 ADPD188GG 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 ADPD188GG is 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

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 BPF and the integrator. With the

ADC offset register(s) for the desired channel set to 0 and the

TIA_VREF set to 1.265 V, 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.

The recommended TIA ADC mode is one in which the BPF is

bypassed and the integrator is configured as a buffer. This

mode is enabled by writing 0xAE65 to Register 0x43 (Time Slot

A) and/or Register 0x45 (Time Slot B) to bypass the BPF.

Additionally, to configure the integrator as a buffer, set Bit 7 of

Register 0x42 (Time Slot A) and/or Register 0x44 (Time Slot B)

to 1, and set Bit 7 of Register 0x58 to 1. With the ADC offset

register(s) for the desired channel set to 0 and TIA_VREF set to

1.265 V, the output of the ADC is at ~13,000 codes for a single

pulse and a zero input current condition. As the input current

photosignal is subtracted from itself, causing the output signal

to decrease when the effective light signal increases.

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 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

Rev. B | Page 31 of 61

ADPD188GG

Data Sheet

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).

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 V_DD. 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.

To ensure that the TIA does not saturate, a safe operating

region is typically at ¾ full scale and lower. Use Table 19 to

determine how the output 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 PCB Parasitic Input Resistance

During the process of mounting the ADPD188GG, undesired

resistance can develop on the inputs through assembly errors or

debris on the PCB. These resistances can form between the anode

and cathode, or between the anode and some other supply or

ground. In normal operation, the ambient rejection feature of the

ADPD188GG 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.

Using the EXT_IN 1 and EXT_IN 2 Inputs with a Voltage

Source

The ADPD188GG can be used for voltage inputs. Voltage

inputs can be measured in normal mode or in TIA ADC mode.

If these inputs are not a result of stimulation from the LED

driver, TIA ADC mode is preferred. To understand the

conversion gain from a voltage through a series resistor, R_S, the

current can be determined by following the schematic in Figure 34.

Measuring Shunt Resistance on the Photodiode

A shunt resistor across the photodiode/EXT_INx 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 1 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.

V

IN

ADPD188GG

OPTIONAL

BUFFER

R

S

R_IN

EXT_INx

TIA

ADC

TIA_VREF

Figure 34. ADPD188GG Used for Voltage Inputs

Input Current = (V_IN− TIA_VREF)/(R_S+ R_IN)

Values for R_IN are listed in Table 2. R_IN is not needed for

photodiode or other current inputs because the current of these

inputs are not a function of the input resistance. Conversion from

input current in amps to ADC codes (LSBs) follows Table 19 in

TIA ADC mode. Current conversion in normal mode is listed

in Table 2. The offset level shown in Table 19 represents the

expected code value with zero current input. The conversion

gain in nA/LSB can be added onto this for nonzero input currents.

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

Table 19. Analog Specifications for TIA ADC Mode and Digital Integrate Mode

Parameter

Test Conditions/Comments

Typ

Unit

TIA ADC Offset Level

Floating input (Input current = 0A); Register 0x43 and Register 0x45 = 0xAE65; Register 0x42

and Register 0x44, Bit 7 = 1, Register 0x58, Bit 7 = 1

TIA_VREF = Register 0x42 and Register 0x44, Bits[5:4] = 0 (1.14 V)

TIA_VREF = Register 0x42 and Register 0x44, Bits[5:4] = 1 (1.01 V)

TIA_VREF = Register 0x42 and Register 0x44, Bits[5:4] = 2 (0.89 V)

TIA_VREF = Register 0x42 and Register 0x44, Bits[5:4] = 3 (1.27 V); recommended for PD inputs

11400

9700

8100

LSB

13200

Rev. B | Page 32 of 61

Data Sheet

ADPD188GG

Parameter

Test Conditions/Comments

Typ

Unit

TIA ADC Saturation

Levels¹

Values expressed per channel, per sample; buffer gain = 1

25 kΩ

50 kΩ

100 kΩ

38.32

19.16

9.58

μA

200 kΩ

4.79

TIA ADC Resolution

Values expressed per channel, per sample; buffer gain = 1

25 kΩ

50 kΩ

100 kΩ

200 kΩ

2.92

1.5

0.73

0.37

nA/LSB

¹TIA linear dynamic range is 85% of listed saturation levels

Table 20. Configuration Registers to Switch Between Normal Sample Mode and TIA ADC Mode

Data

Address Bits

Normal

Mode Value

TIA ADC

Mode Value

Bit Name

Description

0x42

[15:10] SLOTA_AFE_MODE

0x07

Not applicable In normal mode, this setting configures the integrator

block for optimal operation. This setting is not important

for TIA ADC mode.

[9]

[7]

SLOTA_BUF_GAIN

0x0

0: buffer gain = 1.0.

1: buffer gain = 0.7.

SLOTA_INT_AS_BUF 0x0

0x1

0: normal integrator configuration.

1: convert integrator to buffer amplifier in TIA ADC mode

(required for 0x43 = 0xAE65).

0x43

0x44

[15:0]

[15:10

SLOTA_AFE_CFG

0xADA5

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.

SLOTB_AFE_MODE

SLOTB_BUF_GAIN

0x07

0x0

Not applicable In normal mode, this setting configures the integrator

block for optimal operation. This setting is not important

for TIA ADC mode.

[9]

[7]

0x0

0: buffer gain = 1.0.

1: buffer gain = 0.7.

SLOTB_INT_AS_BUF 0x0

0x1

0: normal integrator configuration.

1: convert integrator to buffer amplifier (required for 0x45 =

0xAE65).

0x45

0x58

[15:0]

[7]

SLOTB_AFE_CFG

ENA_INT_AS_BUF

0xADA5

0xAE65

0x1

Time Slot B AFE connection.

0xAE65: bypasses the band-pass filter.

0xB065: can also be used in TIA ADC mode. This setting

bypasses the band-pass filter (BPF) and the integrator.

0x0

Enables the ability to configure the integrator as a buffer in

TIA ADC mode

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, ECG).

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

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.

Rev. B | Page 33 of 61

ADPD188GG

Data Sheet

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 35 shows a plot of a synchronized ECG

and PPG measurement using the AD8233 with the

ADPD188GG.

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.

10000

The circuit shown in Figure 36 shows a synchronous ECG and

PPG measurement using the AD8233 and the ADPD188GG.

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

the EXT_IN1 current input of the ADPD188GG through a

200 kΩ resistor to convert the voltage output of the AD8233

into a current.

52000

9500

9000

51500

PPG

8500

8000

51000

ECG

7500

7000

50500

The ADPD188GG is configured to alternately measure the

photodiode signal and the ECG signal from the AD8233 on

consecutive time slots to provide fully synchronized PPG and

ECG measurements. Data can be read out of the on-chip FIFO

or straight from data registers. The ADPD188GG channel used

to process the ECG signal is set up in TIA ADC mode and the

6500

50000

6000

SAMPLE RATE (ms)

Figure 35. Plot of Synchronized ECG and PPG Waveforms

1.8V

4.7µF

10MΩ

HPDRIVE

+IN

HPSENSE

IAOUT

10MΩ

1.8V

180kΩ

LA

RA

–IN

REFIN

10MΩ

0.1µF

4.7µF

+V

S

RLDFB

RLD

10MΩ

0.1µF

1nF

360kΩ

RL

GND

AD8233

1.8V

SW

FR

AC/DC

SDN

1MΩ

OPAMP+

REFOUT

OPAMP–

OUT

100kΩ 6.8nF

250kΩ

TO DIGITAL

INTERFACE

RLD SDN

LOD

2.7nF

1MΩ

1.8V

ADPD188GG

200kΩ

5V

EXT_IN1

VLED1

VDD1

VDD2

0.1µF

AGND

DGND

LGND

CVLED

1.8V

10kΩ 10kΩ

LED1/DNC

VREF

SCL

SDA

TO DIGITAL

INTERFACE

1µF

GPIO0

GPIO1

Figure 36. Synchronized PPG and ECG Measurement Using the ADPD188GG with the AD8233

Rev. B | Page 34 of 61

Data Sheet

ADPD188GG

Float Mode Measurement Cycle

FLOAT MODE

Figure 37 shows the float mode measurement cycle timing

diagram, and the following details the points shown:

The ADPD188GG has a unique operating mode, float mode,

that allows excellent SNR at low power in low light situations.

In float mode, the photodiode is first preconditioned to a

known state and then the photodiode anode is disconnected

from the receive path of the ADPD188GG for a preset amount

of float time. During the float time, light falls on the

photodiode, either from ambient light, pulsed LED light, or a

combination of the two depending on the operating mode.

Charge from the sensor is stored directly on the capacitance of

the sensor. At the end of the float time, the photodiode switches

back into the receive path of the ADPD188GG and an inrush of

the accumulated charge occurs, which is subsequently integrated

by the integrator of the ADPD188GG, allowing the maximum

amount of charge to be processed per pulse with the minimum

amount of noise added by the signal path. The charge is

integrated externally on the capacitance of the photodiode for

as long as it takes to acquire maximum charge, independent of

the amplifiers of the signal path, which adds noise to the signal.

•

The precondition period is shown prior to Point A. The

photodiode is connected to the TIA, and the photocurrent

flows into the TIA. The photodiode anode is held at 0.9 V

(Register 0x42 and Register 0x44, Bits[5:4] = 0x2 sets TIA_

VREF = 0.9 V). The photodiode is reverse biased to a

maximum reverse bias of ~250 mV by setting Register 0x54,

Bit 7 = 1 and Register 0x54, Bits[9:8] = 0x2 (for Time Slot

A). At this point, the output of the TIA (TIA_OUT) =

TIA_VREF − (I_PD× R_F), where I_PDis the current flowing

from the PD into the ADPD1080/ADPD1081 input, and

the integrator is off.

At Point A, the photodiode is disconnected from the

receive path. Light continues to fall on the photodiode

producing a charge that accumulates directly on the

photodiode capacitance. As the charge accumulates, the

voltage at the floating photo-diode anode rises. The TIA is

disconnected from the input to the ADPD188GG so that

no current flows through the TIA, and the TIA output is at

TIA_VREF. Just prior to Point B, the integrator resets to 0.

In the Float Mode for Synchronous LED Measurements

section, the LED pulses during the time period between

Point A and Point D. Float times of <4 µs are not allowed.

At Point B, the integrator begins its positive integration

phase. Small dc offsets between the TIA output and the

integrator reference causes the integrator output to ramp

up for positive offsets or ramp down for negative offsets.

The photodiode continues to accumulate charge during

this period.

Amplifier and ADC noise values are constant for a given

measurement. For optimal SNR, it is desirable to have a greater

amount of signal (charge) per measurement. In normal mode,

because the pulse time is fixed, the charge per measurement can

be increased only by increasing the LED drive current. For high

light conditions, this is sufficient. In low light conditions,

however, there is a limit to the available current. In addition,

high current pulses can cause ground noise in some systems.

Green LEDs have lower efficiency at high currents, and many

battery designs do not deliver high current pulses as efficiently.

Float mode allows the user the flexibility to increase the

amount of charge per measurement by either increasing the

LED drive current or by increasing the float time. This

•

At Point C, the integrator begins its negative integration

phase. This reversal in polarity begins to cancel any signal

caused by offsets. This offset cancellation continues through

Point F, where all offsets are cancelled completely.

At Point D, the photodiode switches into the receive path

where all the charge that has accumulated on the photodiode

capacitance during the float time is dumped into the TIA. The

typical charge dump time is less than 2 µs. As the current

flows through the TIA, the output of the TIA responds

with a large negative signal. Because the integrator is in the

negative integration phase at this point, the output of the

integrator rises as the input current to the device integrates

back to total charge. Between Point D and Point E, any

light incident on the photodiode produces additional

photocurrent, which is immediately integrated by the

integrator as charge.

flexibility is especially useful in low current transfer ratio (CTR)

conditions, for example, 10 nA/mA, where normal mode

requires multiple pulses to achieve an acceptable level of SNR.

In float mode, the signal path bypasses the BPF and uses only

the TIA and integrator. In normal mode, the shape of the pulse

is known (typically either 2 µs or 3 µs) and is consistent across

devices and conditions. The shape of the signal coming through

the BPF is also predictable, which allows a user to align the

integrator timing with the zero crossing of the filtered signal. In

float mode, the shape of the signal produced by the charge

dump can differ across devices and conditions. A filtered signal

cannot be reliably aligned; therefore, the BPF cannot be used.

In float mode, the entire charge dump is integrated in the

negative cycle of the integrator and the positive cycle cancels

any offsets.

•

At Point E, the TIA disconnects from the receive path and

the TIA output returns to TIA_VREF. Between Point E

and Point F, the integrator completes the negative

integration phase and cancellation of the offsets.

At Point F, the integrator output is held until sampled by the

ADC.

Rev. B | Page 35 of 61

ADPD188GG

Data Sheet

25.0µs

30.0µs

35.0µs

A

D

E

CONNECT

FLOAT

F

C

B

+ PHASE

PHASE

INTEGRATOR

CHARGE ON PD

TIA OUTPUT

INTEGRATOR

RESPONSE

DON’T CARE

INTEGRATOR

RESET

ADC READ

Figure 37. Float Mode Measurement Cycle Timing Diagram

Float Mode Limitations

V is the amount of voltage change across the photodiode before

the photodiode becomes nonlinear.

When using float mode, the limitations of the mode must be

well understood. For example, there is a finite amount of

charge that can accumulate on the capacitance of the

For a typical discrete optical design using a 7 mm²photodiode

with 70 pF capacitance and 450 mV of headroom, the maximum

amount of charge that can be stored on the photodiode

capacitance is 31.5 pC.

photodiode, and there is also a maximum amount of charge that

can be integrated by the integrator. Based on an initial reverse

bias of 250 mV on the photodiode and assuming that the

photodiode begins to become nonlinear at ~200 mV of forward

bias, there is ~450 mV of headroom for the anode voltage to

increase from its starting point at the beginning of the float time

before the charge ceases to accumulate in a linear fashion. It is

desirable to operate only in the linear region of the photodiode

(see Figure 38). To verify that float mode is operating in the

linear region of the diode, the user can perform a simple check.

Record data at a desired float time, and then record data at half

the float time. It is recommended that the ratio of the two received

signals be 2:1. If this ratio does not hold true, the diode is likely

beginning to forward bias at the longer float time and becomes

nonlinear.

In addition, consider the maximum amount of charge the

integrator of the ADPD188GG can integrate. The integrator

can integrate up to 7.6 pC. When this charge is referred back to

the input, consider the TIA gain. When the TIA gain is at

200 kΩ, the input referred charge is at a 1:1 ratio to the

integrated charge on the integrator. For 100 kΩ gain, it is 2:1;

for 50 kΩ gain, it is 4:1; and for 25 kΩ gain, it is 8:1. For the

previous example using a photodiode with 70 pF capacitance,

use 50 kΩ TIA gain and set the float timing such that, for a

single pulse, the output of the ADC is at 70% of full scale, which

is a typical operating condition. Under these operating conditions,

5.3 pC integrates per pulse by the integrator for 21.2 pC of

charge accumulated on the photodiode capacitance. For small

CTR, however, it can take a long time to accumulate 21.2 pC of

charge on the photodiode capacitance, in which case, use higher

TIA gains according to how much charge can be accumulated in a

given amount of time. Ultimately, float times are determined by

the type of measurement being made (ambient or pulsed LED),

the photodiode capacitance, and the CTR of the system.

PD BEGINS TO

FORWARD BIAS

RECOMMENDED

FLOAT MODE

OPERATING REGION

Float Mode for Ambient Light Measurements

Float mode is used for ambient light measurements where the

background light is sufficiently small. Use TIA ADC mode for

ambient light measurements of higher intensities. Small

amounts of light can be measured with adequate float times,

allowing the incoming charge to accumulate to levels large

enough to be measured above the noise floor of the system. The

source of this light can be any combination of synchronous

light (for example, from a pulsed LED) and asynchronous light

(that is, background). If there is no system generated light

source, the measurement is simply a measure of the

background light.

FLOAT TIME (µs)

Figure 38. Transfer Function of Integrated Charge on the Photodiode vs.

Float Time

The maximum amount of charge that can be stored on the

photodiode capacitance and remain in the linear operating

region of the sensor can be estimated by

Q = CV

where:

Q is the integrated charge.

C is the capacitance of the photodiode.

Rev. B | Page 36 of 61

Data Sheet

ADPD188GG

Use a two pulse differential measurement technique to cancel

out electrical drifts and offsets. Take two measurements, each

of a different float time. The first float time is considerably

shorter than the second pulse. After the two measurements are

taken, Measurement 1 is subtracted from Measurement 2,

which effectively cancels out any offset and drift common to

both measurements. What is left is an ambient light

measurement based on an amount of charge that is integrated

over a time that is the difference of the first and second float

times. For example, if Float Time 1 is 6 µs and Float Time 2 is

26 µs, the ambient light measurement is based on 20 µs of

charge integrated on the photodiode capacitance with any

offset and drift removed. In float mode for ambient light, the

number of pulses must be set to two to cancel drifts and offsets

because only the first pulse can be short. More than two pulses

can be used; however, pulses two through n are always the same

length. If drift cancellation is not required, any number of

pulses can be used and added together. Figure 39 shows an

example of float ambient mode timing, and Table 21 details the

relevant registers that must be configured.

0x59[12:8]/0x5E[12:8]

0x30[12:8]/0x35[12:8]

PRECONDITION TIME

(FLT_PRECON_x)

CHARGE DUMP TIME

(SLOTx_LED_WIDTH)

FLOAT 1 TIME

CONNECT/FLOAT

0x30[7:0]/0x35[7:0]

0x31[7:0]/0x36[7:0]

TIMETO FIRST CHARGE DUMP

FLOAT 2 TIME (SLOTx_PERIOD)

(SLOTx_LED OFFSET)

INTEGRATOR SEQUENCE

0x39[15:11]/0x3B[15:11]

INTEGRATION TIME (SLOTx_AFE_WIDTH)

ACCUMULATED

CHARGE ON PD

Figure 39. Example of Float Ambient Mode Timing

Table 21. Float Ambient Mode Registers

Register

Group

Register Name

Time Slot A

Time Slot B

Float Mode Description

Float Mode

Operation

SLOTx_LED_SEL

0x14, Bits[1:0]

0x14, Bits[3:2]

Set to 0 to enable float mode.

FLT_EN_x

0x5E, Bits[14:13] 0x59, Bits[14:13] Set to 3 to enable float between connect pulses.

FLT_MATH12_x

SLOTx_AFE_CFG

SLOTx_TIA_VREF

SLOTx_V_CATHODE

0x58, Bits[2:1]

0x43, Bits[15:0]

0x42, Bits[5:4]

0x54, Bits[9:8]

0x58, Bits[6:5]

0x45, Bits[15:0]

0x44, Bits[5:4]

Set to 2 to subtract first pulse and add second pulse.

Set to 0xAE65 for TIA and integrator, bypass BPF.

Set to 2 for TIA_VREF = 0.9 V.

0x54, Bits[11:10] Set to 2 for 250 mV reverse bias on the photodiode at

the precondition.

REG54_VCAT_ENABLE 0x54, Bit 7

0x54, Bit 7

Set to 1 to override Register 0x3C cathode voltage

settings.

Float Mode

Timing

FLT_PRECON_x

0x5E, Bits[12:8]

0x59, Bits[12:8]

Precondition time (to start of Float 1 time).

SLOTx_PERIOD

SLOTx_LED_WIDTH

0x31, Bits[7:0]

0x37, Bits[1:0]

0x30, Bits[12:8]

0x36, Bits[7:0]

0x37, Bits[9:8]

0x35, Bits[12:8]

8 LSBs of float period in µs; Float 2 time = SLOTx_PERIOD

2 MSBs of float period.

Connect time in µs; this is the amount of time given to

dump the accumulated charge from the photodiode

capacitance; typically, this is set to 2 µs.

SLOTx_LED_OFFSET

0x30, Bits[7:0]

0x35, Bits[7:0]

Time to first charge dump; Float 1 time =

(SLOTx_LED_OFFSET + SLOTx_LED_WIDTH) −

FLT_PRECONx.

SLOTx_AFE_WIDTH

SLOTx_AFE_OFFSET

0x39, Bits[15:11] 0x3B, Bits[15:11]

Integration time in µs; set to FLT_CONNx + 1.

Integrator start time in 31.25 ns increments; set to

(SLOTx_LED_OFFSETx − SLOTx_AFE_WIDTH − 9.25) µs.

0x39, Bits[10:0]

0x3B, Bits[10:0]

SLOTx_PULSES

0x31, Bits[15:8]

0x36, Bits[15:8]

Number of pulses; set to 2 for float ambient mode.

Rev. B | Page 37 of 61

ADPD188GG

Data Sheet

The return from the LED + ambient + offset is present in one of

the pulses. In the other, only the ambient light and offset is

present. A subtraction of the two pulses is made that eliminates

ambient light as well as any offset and drift. It is recommended

to use groups of four pulses for measurement where the LED is

flashed on Pulse 2 and Pulse 3. The accumulator adds Pulse 2

and Pulse 3 and then subtract Pulse 1 and Pulse 4. To gain

additional SNR, use multiple groups of four pulses.

Float Mode for Synchronous LED Measurements

In float LED mode, photocurrent is generated from ambient

light and pulsed LED light during the float time. Float LED

mode is desirable in low signal conditions where the CTR is

<10 nA/mA. In addition, float mode is a good option in

situations where the user wants to limit the LED drive current

of the green LEDs in a heart rate measurement to keep the

forward voltage drop of the green LED to a level that allows the

elimination of a boost converter for the LED supply. For

example, the LED current can be limited to 10 mA to ensure

that the LED voltage drop is ~3 V so that it can operate directly

from the battery without the need of a boost converter. Float

mode accumulates the received charge during longer LED

pulses without adding noise from the signal path, effectively

yielding the highest SNR and/or photon attainable.

The settings of FLT_LED_FIRE_x, Register 0x5A, Bits[15:8]

determine if the LED fires in which pulse position. Which pulse

positions are added or subtracted is configured in the FLT_

MATH12x and FLT_MATH34x bits of Register 0x58. These

sequences are repeated in groups of four pulses. The value

written to the FIFO or data registers is dependent on the total

number of pulses per sample period. For example, if the device

is setup for 32 pulses, the 4 pulse sequence, as defined in FLT_

LED_FIRE_x and FLT_MATHxxx, repeats eight times and a

single register or FIFO write of the final value based on 32 pulses

executes. Table 22 details the relevant registers for float LED mode.

As with float ambient mode, multiple pulses cancel electrical

offsets and drifts; however, in float LED mode, the ambient

light must also be cancelled because only the reflected return

from the LED pulses is desired. To achieve this, use an even

number of equal length pulses. For every pair of pulses, the

LED flashes in one of the pulses and does not flash in the other.

Table 22. Float LED Mode Registers

Register Address

Time Slot A Time Slot B

0x14, Bits[1:0] 0x14, Bits[3:2]

0x5E, Bits[14:13] 0x59, Bits[14:13] Set to 3 to enable float between connect pulses.

Group

Register Name

SLOTx_LED_SEL

FLT_EN_x

Float Mode Description

Float Mode

Operation

Set to 0 to enable float mode.

FLT_MATH12_x

FLT_MATH34_x

SLOTx_AFE_CFG

SLOTx_TIA_VREF

SLOTx_V_CATHODE

0x58, Bits[2:1]

0x58, Bits[9:8]

0x43, Bits[15:0]

0x42, Bits[5:4]

0x54, Bits[9:8]

0x58, Bits[6:5]

Set to 2 to subtract first pulse and add second pulse.

0x58, Bits[11:10] Set to 1 to add third pulse and subtract fourth pulse.

0x45, Bits[15:0]

0x44, Bits[5:4]

Set to 0xAE65 for TIA + integrator, bypass BPF.

Set to 2 for TIA_VREF = 0.9 V.

0x54, Bits[11:10] Set to 2 for 250 mV reverse bias on the photodiode at the

precondition.

REG54_VCAT_ENABLE

FLT_LED_SELECT_x

0x54, Bit 7

Set to 1 to override Register 0x3C cathode voltage settings.

0x3E, Bits[15:14] 0x3F[15:14]

LED selection for float LED mode.

00 = no LED.

01 = LED1.

10 = LED2.

11 = LED3.

Float Mode

Timing

FLT_PRECON_x

SLOTx_PERIOD

0x5E, Bits[12:8]

0x31, Bits[7:0]

0x59, Bits[12:8]

0x36, Bits[7:0]

Precondition time (to start of float 1 time).

8 LSBs of float period in μs. Float 2 time = SLOTx_PERIOD.

Float 2 time is valid for every pulse subsequent to the first

pulse. Float 1 time must be set equal to Float 2 time in

float LED mode.

SLOTx_PERIOD

0x37, Bits[1:0]

0x30, Bits[12:8]

0x37, Bits[9:8]

0x35, Bits[12:8]

2 MSBs of float period.

SLOTx_LED_WIDTH

Connect time in μs, which is the amount of time given to

dump the accumulated charge from the photodiode

capacitance. Typically, it is set to 2 μs.

SLOTx_LED_OFFSET

0x30, Bits[7:0]

0x35, Bits[7:0]

Time to first charge dump. Float 1 time =

(SLOTx_LED_OFFSET + SLOTx_LED_WIDTH) –

FLT_PRECONx. Float 1 time must be equal to Float 2 time

for float LED mode.

Rev. B | Page 38 of 61

Data Sheet

ADPD188GG

Register Address

Time Slot A Time Slot B

0x39,Bits[15:11] 0x3B, Bits[15:11]

Group

Register Name

Float Mode Description

SLOTx_AFE_WIDTH

SLOTx_AFE_OFFSET

Integration time in μs. set to FLT_CONN + 1.

Integrator start time in 31.25 ns increments. Set to

(SLOTx_LED_OFFSET – SLOTx_AFE_WIDTH − 9.25) μs.

0x39, Bits[10:0]

0x3B, Bits[10:0]

SLOTx_PULSES

0x31, Bits[15:8]

0x36, Bits[15:8]

Number of pulses; must be set in multiples of 2,

minimum 2.

FLT_LED_WIDTH_x

FLT_LED_OFFSET_x

FLT_LED_FIRE_x

0x3E, Bits[12:8]

0x3E, Bits[7:0]

0x5A, Bits[11:8]

0x3F, Bits[12:8]

0x3F, Bits[7:0]

0x5A, Bits[15:12]

LED pulse width for float LED mode in μs.

Time of first LED pulse in float LED mode.

In any given sequence of four pulses, fire the LED in the

selected position. Selections are active low (that is, fire LED

if 0). For example, in a sequence of four pulses on Time Slot B,

Register 0x5A, Bit 12 is the first pulse, and Register 0x5A,

Bit 15 is the fourth pulse. For a sequence of four pulses, fire

the LED in the second and third pulses by writing 0x9 to

Register 0x5A, Bits[15:12].

A timing diagram for a four pulse float LED sequence for

Time Slot B is shown in Figure 40. In this example, the device is

set up for LED pulses of 12 μs that fall within a float period of

16 μs, 2 μs of which are used for dumping of the accumulated

charge on the photodiode. The integration time is set to 3 μs, which

is 1 μs more than the charge dump time to allow for timing margin

when integrating the incoming charge. Note, there is a 9 μs offset

built into the integration start time. Take this offset into account

when setting the SLOTx_AFE_OFFSET value. As shown in

Figure 40, the time of the first charge dump is set to 30 μs. SLOTx_

AFE_OFFSET is set to 0x238 (17.75 μs), taking into account the

3 μs integration time, the 9 μs offset, and an additional 250 ns

for edge placement margin.

To calculate SLOTx_AFE_OFFSET, use the following equation:

SLOTx_AFE_OFFSET = SLOTx_LED_OFFSET –

SLOTx_AFE_WIDTH − 9.25 μs

Placement of the integration period is such that the negative

phase of the integration is centered on the charge dump phase.

The TIA is an inverting stage, therefore, placing the negative

phase of the integration during the dumping of the charge from

the photodiode causes the integrator to increase with the

negative going output signal from the TIA.

The LED flashes in the second and third pulses of the four pulse

sequence. Setting Register 0x58, Bits[6:5] = 2 and Register 0x58,

Bits[11:10] = 1 forces the device to add the second and third

pulses while subtracting the first and fourth pulses, effectively

cancelling out the ambient light and electrical offsets and drift.

Rev. B | Page 39 of 61

ADPD188GG

Data Sheet

PRECONDITION

0x35[12:8] = 0x2

CHARGE DUMPTIME = 2µs

0x36[7:0] = 0x10

FLOAT PERIOD = 16µs

CONNECT/FLOAT

MASKED

LED PULSE

FLASH LED

MASKED LED PULSE

LED PULSES

0x5A[15:12] = 0x9

MASK PULSE 1 AND MASK PULSE 4

FLASH PULSE 2 AND FLASH PULSE 3

0x3F[12:8] = 0xC

LED PW = 12µs

ACCUMULATED

CHARGE ON PD

INTEGRATOR

OUTPUT

INTEGRATOR

SEQUENCE

0x3B[15:11] = 0x3

INTEGRATION

TIME = 3µs

INTEGRATOR

RESET

ADC READ

t = 0

t = 16µs, 0x59[12:8] = 0x10

PRECONDITIONING TIME

t = 17µs, 0x3F[7:0] = 0x11

LED PULSE OFFSET

t = 26.75µs, 0x3B[10:0] = 0x238

START OF INTEGRATION TIME

t = 30µs, 0x35[7:0] = 0x1E

TIME OF FIRST CHARGE DUMP

Figure 40. Example Timing Diagram of Four Pulse Float LED Mode Sequence

A comparison of float ambient mode vs. float LED mode is shown in Table 23 and Table 24.

Table 23. Float Ambient Mode—Measure Ambient Light Level

Pulse Float Time

Integrated Charge

Calculation

Result

Ambient Measurement = Ambient 2 − Ambient 1 (offset cancels)

1

2

3

4

Shorter

Longer

Offset, Ambient 1 (shorter time) Subtract

Offset, Ambient 1 (shorter time) Add

Not applicable Not applicable

Not applicable

Table 24. Float LED Mode—Measurement Synchronous Reflected Light from LED

Pulse Float Time Integrated Charge Calculation Result

1

2

3

4

Equal

Offset + Ambient

Subtract

Add

Sync LED response = reflected LED return (offset and ambient cancel)

Offset + Ambient + LED

Offset + Ambient

Subtract

Rev. B | Page 40 of 61

Data Sheet

ADPD188GG

The user sets an ambient level threshold in the BG_THRESH

register, which is the threshold by which the ADC result of the

subtract cycles in float LED mode are compared against. The

subtract cycles in float LED mode are the positions in the pulse

sequence in which the LED pulse is masked; therefore, it is the

background level measurement. The ADC result is equal to the

raw ADC output minus the contents of the ADC offset register

(Register 0x18 to Register 0x1B and Register 0x1E to

Register 0x21). In the BG_COUNT register, the user sets a limit

on the number of cycles that BG_THRESH is exceeded by the

ADC result before the BG_STATUS bit is set for any particular

channel. Every time the BG_THRESH value is exceeded by the

ADC result during a subtract cycle, an internal counter

Monitoring Ambient Light Levels in Float LED Mode

In real-world applications, it is common for the ambient light

levels to change constantly. When using float LED mode,

increasing amounts of ambient light can approach levels where

it uses an unacceptable amount of the dynamic range of the

charge that can stored on the photodiode capacitance. For this

reason, it is required that the ambient light level is monitored

so that configuration changes can be made when necessary, for

example, float time, TIA gain, and operating mode. There are

two ways to monitor ambient light levels. One way is to use

TIA ADC mode in the alternate time slot and continuously

monitor the ambient light level. The other way is to use a

feature of the ADPD188GG where the ambient light level is

automatically monitored in the background during float mode

operation and is compared against a user-defined threshold. If

the ambient light level exceeds this threshold by some user-

defined number of times, a flag is set by the device that can be

read by the user or can be output to a GPIO. Table 25 lists all

the registers used to monitor the ambient light level while in

float LED mode.

increments. Each channel has its own counter. When this count

exceeds the limit set in the BG_COUNT register, the

BG_STATUS bit is set for the channel. The user can periodically

monitor the BG_STATUS register to check for asserted bits.

Alternatively, a GPIOx pin can be asserted if a BG_STATUS flag

is set. See Table 25 for the various logical combinations of

BG_STATUS flags and interrupts that can be brought out on a

GPIOx.

Table 25. Registers for Monitoring the Ambient Light Level in Float LED Mode

Register

Float Mode Register Name

Time Slot A

Time Slot B

Description

BG_STATUS_x

0x04, Bits[3:0]

0x04, Bits[7:4]

Status of comparison between background light level and background

threshold value (BG_THRESH). A 1 in any bit location means the threshold

has been crossed BG_COUNT number of times. This register is cleared once

it is read.

Bit 0: Time Slot A, Channel 1 exceeded threshold count.

Bit 1: Time Slot A, Channel 2 exceeded threshold count.

Bit 2: Time Slot A, Channel 3 exceeded threshold count.

Bit 3: Time Slot A, Channel 4 exceeded threshold count.

Bit 4: Time Slot B, Channel 1 exceeded threshold count.

Bit 5: Time Slot B, Channel 2 exceeded threshold count.

Bit 6: Time Slot B, Channel 3 exceeded threshold count.

Bit 7: Time Slot B, Channel 4 exceeded threshold count.

BG_THRESH_x

BG_COUNT_x

0x16, Bits[13:0] 0x1C[13:0]

0x16, Bits[15:14] 0x1C[15:14]

The background threshold that is compared against the ADC result during

the subtract cycles during float mode. If the ADC result exceeds the value

in this register, BG_COUNT is incremented.

This is the number of times the ADC value exceeds the BG_THRESH_x value

during the float mode subtract cycles before the BG_STATUS bit is set.

0x0: never set BG_STATUS.

0x1: set when BG_THRESH_x is exceeded 1 time.

0x02: set when BG_THRESH_x is exceeded 4 times.

0x03: set when BG_THRESH_x is exceeded 16 times.

GPIO0 asserts for the following conditions:

0x10: logical OR of BG_STATUS_x, Bits[3:0].

0x1A: logical OR of BG_STATUS_x, Bits[7:4].

0x1B: logical OR of BG_STATUS_x, Bits[7:0].

0x1C: logical OR of BG_STATUS_x, Bits[7:0] and INT.

GPIO1 asserts for the following conditions:

0x10: logical OR of BG_STATUS_x, Bits[3:0].

0x1A: logical OR of BG_STATUS_x, Bits[7:4].

0x1B: logical OR of BG_STATUS_x, Bits[7:0].

0x1C: logical OR of BG_STATUS_x, Bits[7:0] and INT.

GPIO0_ALT_CFG

GPIO1_ALT_CFG

0x0B[4:0]

0x0B[12:8]

Rev. B | Page 41 of 61

ADPD188GG

Data Sheet

REGISTER LISTING

The recommended values are not shown. Only power-on reset values are shown in Table 26. The recommended values are largely

dependent on use case.

Table 26. Numeric Register Listing

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

R/W

0x00 Status

[15:8]

FIFO_SAMPLES[7:0]

0x0000 R/W

[7:0] Reserved

[15:8]

SLOTB_INT

SLOTA_INT

Reserved

0x01 INT_MASK

Reserved

FIFO_INT_

MASK

0x00FF R/W

[7:0] Reserved

SLOTB_INT_ SLOTA_INT_

MASK MASK

Reserved

0x02 GPIO_

DRV

[15:8]

[7:0]

Reserved

GPIO1_DRV GPIO1_POL

GPIO0_DRV GPIO0_POL

0x0000 R/W

GPIO0_ENA

0x04 BG_

STATUS

[15:8]

[7:0]

Reserved

BG_STATUS_B[3:0]

BG_STATUS_A[3:0]

FIFO_THRESH[5:0]

0x06 FIFO_

THRESH

[15:8]

[7:0]

Reserved

0x08 DEVID

[15:8]

[7:0]

REV_NUM[7:0]

DEV_ID[7:0]

0x0A16

0x00C8 R/W

0x0000

R

0x09 I2CS_ID

[15:8]

[7:0]

ADDRESS_WRITE_KEY[7:0]

SLAVE_ADDRESS[6:0]

Reserved

0x0A CLK_

RATIO

[15:8]

[7:0]

Reserved

CLK_RATIO[11:8]

R

CLK_RATIO[7:0]

0x0B GPIO_

CTRL

[15:8]

[7:0]

Reserved

GPIO1_ALT_CFG[4:0]

GPIO0_ALT_CFG[4:0]

0x0000 R/W

0x0D SLAVE_

ADDRESS_

KEY

[15:8]

[7:0]

SLAVE_ADDRESS_KEY[15:8]

SLAVE_ADDRESS_KEY[7:0]

0x0F SW_RESET

[15:8]

[7:0]

Reserved

0x0000 R/W

0x1000 R/W

SW_RESET

Mode[1:0]

0x10 Mode

[15:8]

[7:0]

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

0x0000 R/W

[7:0]

0x14 PD_LED_

SELECT

[15:8]

Reserved

SLOTB_PD_SEL[3:0]

[7:0]

SLOTA_PD_SEL[3:0]

Reserved

SLOTB_LED_SEL[1:0]

SLOTA_LED_SEL[1:0]

SLOTB_NUM_AVG[2:0]

0x15 NUM_

AVG

[15:8]

[7:0] Reserved

SLOTA_NUM_AVG[2:0]

Reserved

0x16 BG_

MEAS_A

[15:8]

[7:0]

BG_COUNT_A[1:0]

BG_THRESH_A[13:8]

BG_THRESH_A[7:0]

0x17 INT_SEQ_A

[15:8]

[7:0]

Reserved

0x0000 R/

W

Reserved

INTEG_ORDER_A[3:0]

0x18 SLOTA_

CH1_

[15:8]

[7:0]

SLOTA_CH1_OFFSET[15:8]

SLOTA_CH1_OFFSET[7:0]

0x2000 R/W

OFFSET

0x19 SLOTA_

CH2_

[15:8]

[7:0]

SLOTA_CH2_OFFSET[15:8]

SLOTA_CH2_OFFSET[7:0]

0x2000 R/W

0x0000 R/W

OFFSET

0x1A SLOTA_

CH3_

[15:8]

[7:0]

SLOTA_CH3_OFFSET[15:8]

SLOTA_CH3_OFFSET[7:0]

OFFSET

0x1B SLOTA_

CH4_

[15:8]

[7:0]

SLOTA_CH4_OFFSET[15:8]

SLOTA_CH4_OFFSET[7:0]

OFFSET

0x1C BG_MEAS_B

[15:8

BG_COUNT_B[1:0]

BG_THRESH_B[13:8]

BG_THRESH_B[7:0]

Rev. B | Page 42 of 61

Data Sheet

ADPD188GG

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

R/W

0x1D INT_SEQ_B

[15:8]

[7:0]

Reserved

0x0000 R/

W

Reserved

INTEG_ORDER_B[3:0]

0x1E SLOTB_CH1_ [15:8]

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]

0x2000 R/W

OFFSET

[7:0]

0x1F SLOTB_CH2_ [15:8]

0x2000 R/W

0x3000 R/W

0x630C R/W

0x0320 R/W

0x0818 R/W

0x0000 R/W

OFFSET

[7:0]

0x20 SLOTB_CH3_ [15:8]

OFFSET

[7:0]

0x21 SLOTB_CH4_ [15:8]

OFFSET

[7:0]

0x22 ILED3_

COARSE

[15:8]

Reserved

ILED3_SCALE

ILED3_SLEW[2:0]

ILED1_SCALE

Reserved

[7:0] Reserved

[15:8]

ILED3_COARSE[3:0]

Reserved

0x23 ILED1_

COARSE

Reserved

[7:0] Reserved

[15:8]

ILED1_SLEW[2:0]

ILED2_SCALE

ILED1_COARSE[3:0]

Reserved

0x24 ILED2_

COARSE

[7:0] Reserved

[15:8]

ILED2_SLEW[2:0]

ILED3_FINE[4:0]

Reserved

ILED2_COARSE[3:0]

ILED2_FINE[4:2]

ILED1_FINE[4:0]

0x25 ILED_FINE

[7:0]

ILED2_FINE[1:0]

Reserved

0x30

SLOTA_LED_ [15:8]

SLOTA_LED_WIDTH[4:0]

SLOTA_LED_OFFSET[7:0]

PULSE

[7:0]

0x31 SLOTA_

NUMPULSES

[15:8]

[7:0]

SLOTA_PULSES[7:0]

SLOTA_PERIOD[7:0]

0x34 LED_

DISABLE

[15:8]

Reserved

SLOTB_

LED_DIS

SLOTA_

LED_DIS

[7:0]

Reserved

0x35 SLOTB_

LED_PULSE

[15:8]

[7:0]

Reserved

SLOTB_LED_WIDTH[4:0]

0x0320 R/W

0x0818 R/W

0x0000 R/W

SLOTB_LED_OFFSET[7:0]

SLOTB_PULSES[7:0]

SLOTB_PERIOD[7:0]

CH2_DISABLE[12:10]

0x36 SLOTB_

NUMPULSES

[15:8]

[7:0]

0x37 ALT_PWR_

DN

[15:8]

[7:0]

CH34_DISABLE[15:13]

SLOTB_PERIOD[9:8]

SLOTA_PERIOD[9:8]

Reserved

0x38 EXT_SYNC_

STARTUP

[15:8]

[7:0]

EXT_SYNC_STARTUP[15:8]

EXT_SYNC_STARTUP[7:0]

0x000

R/W

0x39 SLOTA_AFE_ [15:8]

SLOTA_AFE_WIDTH[4:0]

SLOTA_AFE_OFFSET[10:8]

0x22FC R/W

WINDOW

[7:0]

SLOTA_AFE_OFFSET[7:0]

SLOTB_AFE_OFFSET[7:0]

0x3B SLOTB_

AFE_

[15:8]

[7:0]

SLOTB_AFE_WIDTH[4:0]

SLOTB_AFE_OFFSET[10:8]

WINDOW

0x3C AFE_PWR_

CFG1

[15:8]

Reserved

V_CATHODE AFE_

0x3006 R/W

POWER-

DOWN[5]

[7:0]

AFE_POWERDOWN[4:0]

Reserved

FLT_LED_WIDTH_A[4:0]

0x3E SLOTA_

FLOAT_LED

[15:8]

[7:0]

FLT_LED_SELECT_A[1:0]

Reserved

0x0320 R/W

0x1C38 R/W

FLT_LED_OFFSET_A[7:0]

0x3F SLOTB_

FLOAT_LED

[15:8]

[7:0]

FLT_LED_SELECT_B[1:0]

Reserved

FLT_LED_WIDTH_B[4:0]

FLT_LED_OFFSET_B[7:0]

0x42 SLOTA_

TIA_CFG

[15:8]

SLOTA_AFE_MODE[5:0]

SLOTA_

BUF_GAIN

SLOTA_TIA_GAIN[1:0]

Reserved

[7:0] SLOTA_INT_ SLOTA_TIA_

AS_BUF IND_EN

SLOTA_TIA_VREF[1:0]

Reserved (write 0x1)

0x43 SLOTA_

AFE_CFG

[15:8]

[7:0]

SLOTA_AFE_CFG[15:8]

SLOTA_AFE_CFG[7:0]

SLOTB_AFE_MODE[5:0]

0xADA5 R/W

0x1C38 R/W

0x44 SLOTB_

TIA_CFG

[15:8]

SLOTB_

BUF_GAIN

Reserved

[7:0] SLOTB_INT_ SLOTB_

AS_BUF TIA_IND_EN

SLOTB_TIA_VREF[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

0x2612 R/W

0x4B SAMPLE_

CLK

[15:8]

CLK32K_

BYP

[7:0] CLK32K_EN Reserved

CLK32K_ADJUST[5:0]

Rev. B | Page 43 of 61

ADPD188GG

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

R/W

0x4D CLK32M_

ADJUST

[15:8]

[7:0]

Reserved

0x0098 R/W

0x2090 R/W

0x0000 R/W

CLK32M_ADJUST[7:0]

Reserved

0x4F EXT_SYNC_

SEL

[15:8]

[7:0] Reserved

[15:8]

GPIO1_OE

GPIO1_IE

Reserved

EXT_SYNC_SEL[1:0]

Reserved

GPIO0_IE

Reserved

0x50 CLK32M_

CAL_EN

[7:0] Reserved

GPIO1_CTRL CLK32M_

CAL_EN

Reserved

0x54 AFE_PWR_

CFG2

[15:8]

Reserved

SLEEP_V_CATHODE[1:0]

SLOTA_TIA_GAIN_4[1:0]

SLOTB_V_CATHODE[1:0]

Reserved

SLOTA_V_CATHODE[1:0]

0x0020 R/W

[7:0] REG54_

VCAT_

ENABLE

0x55 TIA_INDEP_

GAIN

[15:8]

Reserved

SLOTB_TIA_GAIN_2[1:0]

Reserved

FLT_MATH12_B[1:0]

SLOTB_TIA_GAIN_4[1:0]

SLOTA_TIA_GAIN_3[1:0]

FLT_MATH34_B[1:0]

SLOTB_TIA_GAIN_3[1:0]

SLOTA_TIA_GAIN_2[1:0]

FLT_MATH34_A[1:0]

0x0000 R/W

[7:0]

[15:8]

[7:0]

0x58 Math

ENA_INT_

AS_BUF

Reserved

Reserved FLT_MATH12_A[1:0]

Reserved

0x59 FLT_

CONFIG_B

[15:8]

[7:0]

Reserved

FLT_EN_B[1:0]

FLT_LED_FIRE_B[3:0]

FLT_EN_A[1:0]

FLT_PRECON_B[4:0]

0x0808 R/W

0x0010 R/W

0x0808 R/W

0x0000 R/W

Reserved

0x5A FLT_

LED_FIRE

[15:8]

[7:0]

FLT_LED_FIRE_A[3:0]

FLT_PRECON_A[4:0]

Reserved (write 0x10)

0x5E FLT_

CONFIG_A

[15:8]

[7:0]

Reserved

0x5F

DATA_

ACCESS_CTL

[15:8]

[7:0]

Reserved

SLOTB_

DATA_

HOLD

SLOTA_

DATA_

HOLD

DIGITAL_

CLOCK_

ENA

0x60 FIFO_

ACCESS

[15:8]

[7:0]

FIFO_DATA[15:8]

0x0000

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]

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]

Rev. B | Page 44 of 61

Data Sheet

ADPD188GG

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

R/W

0x78

B_PD1_LOW [15:8]

[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]

SLOTB_CH3_HIGH[15:8]

SLOTB_CH3_HIGH[7:0]

SLOTB_CH4_HIGH[15:8]

SLOTB_CH4_HIGH[7:0]

0x0000

R

0x79 B_PD2_LOW [15:8]

0x0000

R

[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]

0x7E B_PD3_HIGH [15:8]

[7:0]

0x7F B_PD4_HIGH [15:8]

[7:0]

Rev. B | Page 45 of 61

ADPD188GG

Data Sheet

LED CONTROL REGISTERS

Table 27. 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 the Time Slot Switch

section for detailed descriptions.

[7:4]

[3:2]

0x4

0x0

R/W

SLOTA_PD_SEL

SLOTB_LED_SEL

PDx connection selection for Time Slot A. See the Time Slot Switch

section for detailed descriptions.

Time Slot B LED configuration. These bits determine which LED is

associated with Time Slot B.

0x0: pulse PDx connection to AFE. Float mode and pulse connect

mode enable.

0x1: LEDX1 pulses during Time Slot B.

0x2: LEDX2 pulses during Time Slot B.

0x3: LEDX3 pulses during Time Slot B.

[1:0]

0x1

R/W

SLOTA_LED_SEL

Time Slot A LED configuration. These bits determine which LED is

associated with Time Slot A.

0x0: pulse PDx connection to AFE. Float mode and pulse connect

mode enable.

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

Reserved

ILED3_SCALE

LEDX3 current scale factor.

1: 100% strength.

0: 10% strength; sets the LEDX3 driver in low power mode.

LEDX3 Current Scale = 0.1 + 0.9 × (Register 0x22, Bit 13).

Write 0x1.

12

0x1

0x0

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]¹

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.

LED3_PEAK= LED3_COARSE× LED3_FINE× LED3_SCALE

where:

LED3_PEAKis the LEDX3 peak target value (mA).

LED3_COARSE= 44.5 + 17.8 × (Register 0x22, Bits[3:0]).

LED3_FINE= 0.73 + 0.022 × (Register 0x25, Bits[15:11]).

LED3_SCALE= 0.1 + 0.9 × (Register 0x22, Bit 13).

0x23

[15:14]

13

0x0

0x1

R/W

Reserved

Write 0x0.

ILED1_SCALE

LEDX1 current scale factor.

1: 100% strength.

0: 10% strength; sets the LEDX1 driver in low power mode.

LEDX1 Current Scale = 0.1 + 0.9 × (Register 0x23, Bit 13).

12

0x1

0x0

R/W

Reserved

Write 0x1.

Write 0x0.

[11:7]

Rev. B | Page 46 of 61

Data Sheet

ADPD188GG

Default

Value

Address Data Bit

Access

Name

Description

[6:4]

0x0

R/W

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]¹

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.

LED1_PEAK= LED1_COARSE× LED1_FINE× LED1_SCALE

where:

LED1_PEAKis the LEDX1 peak target value (mA).

LED1_COARSE= 44.5 + 17.8 × (Register 0x23, Bits[3:0]).

LED1_FINE= 0.73 + 0.022 × (Register 0x25, Bits[4:0]).

LED1_SCALE= 0.1 + 0.9 × (Register 0x23, Bit 13).

0x24

[15:14]

13

0x0

0x1

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 = 0.1 + 0.9 × (Register 0x24, Bit 13)

12

0x1

0x0

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]¹

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.

LED2_PEAK= LED2_COARSE× LED2_FINE× LED2_SCALE

where:

LED2_PEAKis the LEDX2 peak target value (mA).

LED2_COARSE= 44.5 + 17.8 × (Register 0x24, Bits[3:0]).

LED2_FINE= 0.73 + 0.022 × (Register 0x25, Bits[10:6]).

LED2_SCALE= 0.1 + 0.9 × (Register 0x24, Bit 13).

0x25¹

[15:11]

[10:6]

0xC

R/W

ILED3_FINE

ILED2_FINE

LEDX3 fine adjust. Current adjust multiplier for LED3.

LEDX3 fine adjust = 0.73 + 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 = 0.73 + 0.022 × (Register 0x25, Bits[10:6]).

See Register 0x24, Bits[3:0], for the full LED2 formula.

Write 0x0.

5

0x0

0xC

R/W

Reserved

[4:0]

ILED1_FINE

LEDX1 fine adjust. Current adjust multiplier for LED1.

LEDX1 Fine Adjust = 0.73 + 0.022 × (Register 0x25, Bits[4:0]).

See Register 0x23, Bits[3:0], for the full LED1 formula.

Rev. B | Page 47 of 61

ADPD188GG

Data Sheet

Default

Value

Address Data Bit

Access

R/W

Name

Description

0x30

[15:13]

[12:8]

[7:0]

0x0

Reserved

Write 0x0.

0x3

R/W

SLOTA_LED_WIDTH

SLOTA_LED_OFFSET

SLOTA_PULSES

LED pulse width (in 1 µs step) for Time Slot A.

LED offset width (in 1 µs step) for Time Slot A.

0x20

0x08

R/W

0x31

0x34

[15:8]

R/W

LED Time Slot A pulse count. n_A: number of LED pulses in

Time Slot A.

[7:0]

[15:10]

9

0x18

0x00

0x0

R/W

SLOTA_PERIOD

Reserved

8 LSBs of LED Time Slot A pulse period (in 1 µs step).

Write 0x0.

SLOTB_LED_DIS

Time Slot B LED disable. 1: disables the LED assigned to Time 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

Reserved

Write 0x00.

0x35

0x36

[15:13]

[12:8]

[7:0]

Reserved

Write 0x0.

0x3

SLOTB_LED_WIDTH

SLOTB_LED_OFFSET

SLOTB_PULSES

SLOTB_PERIOD

LED pulse width (in 1 µs step) for Time Slot B.

LED offset width (in 1 µs step) for Time Slot B.

LED Time Slot B pulse count. n_B: number of LED pulses in Time Slot B.

8 LSBs of LED Time Slot B pulse period (in 1 µs step).

0x20

0x08

0x18

[15:8]

[7:0]

R/W

¹LED equations provide an estimate of the LED current to within 20% at 25°C when applying 5.0 V on the VLEDx pin. This current varies with the VLEDx supply level

and temperature.

AFE CONFIGURATION REGISTERS

Table 28. 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.

8 MSBs of LED Time Slot B pulse period.

Write 0x00

[12:10]

0x0

R/W

CH2_DISABLE

[9:8]

[7:2]

[1:0]

[15:14]

[13:11]

10

0x0

0x00

0x0

0x6

0x0

R/W

SLOTB_PERIOD

Reserved

SLOTA_PERIOD

Reserved

8 MSBs of LED Time Slot A pulse period.

Write 0x0.

0x3C

Reserved

Write 0x6.

Reserved

Reserved.

9

V_CATHODE

0x0: 1.3 V (identical to anode voltage).

0x1: 1.8 V (reverse bias photodiode by 550 mV). This

setting may add noise.

Rev. B | Page 48 of 61

Data Sheet

ADPD188GG

Default

Value

Address

Data Bit

Access

Name

Description

[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]

0x6

0x0

R/W

Reserved

Write 0x6.

Write 0x0.

0x54

[15:14]

[13:12]

Reserved

SLEEP_V_CATHODE

If Bit 7 = 1; this setting is applied to the cathode voltage

while the device is in sleep mode.

0x0: V_DD.

0x1:AFE VREF during idle, V_DDduring sleep.

0x2: floating.

0x3: 0.0 V.

[11:10]

[9:8]

7

0x0

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 determined by Register 0x44, Bits[5:4].

0x0: V_DD(1.8 V).

0x1: equal to PD anode voltage.

0x2: sets a reverse PD bias of ~250 mV (recommended

setting).

0x3: 0.0 V (this forward biases a diode at the input).

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 determined by Register 0x42, Bits[5:4].

0x0: V_DD(1.8 V).

0x1: equal to PD anode voltage.

0x2: sets a reverse PD bias of ~250 mV (recommended

setting).

0x3: 0.0 V (this forward biases a diode at the input).

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

0x0

R/W

Reserved

Reserved.

Write 0x0.

0x55

[15:12]

[11:10]

Reserved

0x0

SLOTB_TIA_GAIN_4

TIA gain for Time Slot B, Channel 4 when Register 0x44,

Bit 6 = 1.

0: 200 kΩ.

1: 100 kΩ.

2: 50 kΩ.

3: 25 kΩ.

[9:8]

0x0

R/W

SLOTB_TIA_GAIN_3

TIA gain for Time Slot B, Channel 3 when Register 0x44,

Bit 6 = 1.

0: 200 kΩ

1: 100 kΩ.

2: 50 kΩ.

3: 25 kΩ.

Rev. B | Page 49 of 61

ADPD188GG

Data Sheet

Default

Value

Address

Data Bit

Access

Name

Description

[7:6]

[5:4]

[3:2]

[1:0]

0x0

R/W

SLOTB_TIA_GAIN_2

TIA gain for Time Slot B, Channel 2 when Register 0x44,

Bit 6 = 1.

0: 200 kΩ

1: 100 kΩ.

2: 50 kΩ.

3: 25 kΩ.

R/W

SLOTA_TIA_GAIN_4

SLOTA_TIA_GAIN_3

SLOTA_TIA_GAIN_2

TIA gain for Time Slot A, Channel 4 when Register 0x42,

Bit 6 = 1.

0: 200 kΩ

1: 100 kΩ.

2: 50 kΩ.

3: 25 kΩ.

TIA gain for Time Slot A, Channel 3 when Register 0x42,

Bit 6 = 1.

0: 200 kΩ

1: 100 kΩ.

2: 50 kΩ.

3: 25 kΩ.

TIA gain for Time Slot A, Channel 2 when Register 0x42,

Bit 6 = 1.

0: 200 kΩ

1: 100 kΩ.

2: 50 kΩ.

3: 25 kΩ.

Table 29. AFE Configuration Registers, Time Slot A

Data

Address Bit

Default

Value

Access Name

Description

0x17

[15:4]

[3:0]

0x000

0x0

R/W

Reserved

INTEG_ORDER_A

Write 0x000

Integration sequence order for Time Slot A. Each bit corresponds to the

polarity of the integration sequence of a single pulse in a four-pulse

sequence. Bit 0 controls the integration sequence of Pulse 1, Bit 1 controls

Pulse 2, Bit 2 controls Pulse 3, and Bit 3 controls Pulse 4. After four pulses,

the sequence repeats.

0: normal integration sequence.

1: reversed integration sequence.

0x39

0x42

[15:11] 0x4

R/W

SLOTA_AFE_WIDTH AFE integration window width (in 1 µs step) for Time Slot A.

SLOTA_AFE_OFFSET AFE integration window offset for Time Slot A in 31.25 ns steps.

[10:0]

0x2FC

[15:10] 0x07

SLOTA_AFE_MODE

SLOTA_BUF_GAIN

Set to 0x07.

9

0x0

0: integrator as buffer gain = 1.

1: integrator as buffer gain = 0.7.

Set to 0.

8

7

0x0

R/W

Reserved

SLOTA_INT_AS_BUF 0: normal integrator configuration.

1: converts integrator to buffer amplifier (used in TIA ADC mode only).

6

0x0

R/W

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.

[5:4]

0x3

R/W

SLOTA_TIA_VREF

Set V_REFof the TIA for Time Slot A.

0: 1.14 V.

1: 1.01 V.

2: 0.90 V.

3: 1.27 V (default recommended).

Rev. B | Page 50 of 61

Data Sheet

ADPD188GG

Data

Address Bit

Default

Value

Access Name

Description

[3:2]

[1:0]

0x2

0x0

R/W

Reserved

SLOTA_TIA_GAIN

Reserved. Write 0x1.

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.

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 (must set Register 0x42, Bit 7 = 1 and Register

0x58, Bit 7 = 1).

0xB065: TIA ADC mode (if Register 0x42, Bit 7 = 0).

Others: reserved.

Table 30. AFE Configuration Registers, Time Slot B

Data

Address Bit

Default

Value

Access Name

Description

0x1D

[15:4]

[3:0]

0x000

0x0

R/W

Reserved

INTEG_ORDER_B

Write 0x000

Integration sequence order for Time Slot B. Each bit corresponds to the

polarity of the integration sequence of a single pulse in a four-pulse

sequence. Bit 0 controls the integration sequence of Pulse 1, Bit 1 controls

Pulse 2, Bit 2 controls Pulse 3, and Bit 3 controls Pulse 4. After four pulses,

the sequence repeats.

0: normal integration sequence.

1: reversed integration sequence

0x3B

0x44

[15:11] 0x04

[10:0] 0x17

[15:10] 0x07

R/W

SLOTB_AFE_WIDTH AFE integration window width (in 1 µs step) for Time Slot B.

SLOTB_AFE_OFFSET AFE integration window offset for Time Slot B in 31.25 ns steps.

SLOTB_AFE_MODE

SLOTB_BUF_GAIN

Set to 0x07.

9

0x0

0: integrator as buffer gain = 1.

1: integrator as buffer gain = 0.7.

Set to 0.

8

7

0x0

R/W

Reserved

SLOTB_INT_AS_BUF 0: normal integrator configuration.

1: convert integrator to buffer amplifier (used in TIA ADC mode only).

6

0x0

R/W

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.

[5:4]

0x3

R/W

SLOTB_TIA_VREF

Reserved

Set VREF of the TIA for Time Slot B.

0: 1.14 V.

1: 1.01 V.

2: 0.90 V.

3: 1.27 V (default recommended).

Write 0x1.

[3:2]

0x2

R/W

Rev. B | Page 51 of 61

ADPD188GG

Data Sheet

Data

Address Bit

Default

Value

Access Name

R/W SLOTB_TIA_GAIN

Description

[1:0]

0x0

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

[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 (must set Register 0x44, Bit 7 = 1 and Register

0x58, Bit 7 = 1).

0xB065: TIA ADC mode (if Register 0x44, Bit 7 = 0).

Others: reserved.

FLOAT MODE REGISTERS

Table 31. Float Mode Registers

Default

Value

Address Data Bit

Access

Name

Description

0x04

[15:8]

[7:4]

0x0

R

Reserved

BG_STATUS_B

Not applicable.

0x0

Status of comparison between background light level and

background threshold value for Time Slot B (BG_THRESH_B). A 1

in any bit location means the threshold has been crossed

BG_COUNT_B number of times. This register is cleared once it is

read.

Bit 4: Time Slot B, Channel 1 exceeded threshold count.

Bit 5: Time Slot B, Channel 2 exceeded threshold count.

Bit 6: Time Slot B, Channel 3 exceeded threshold count.

Bit 7: Time Slot B, Channel 4 exceeded threshold count.

[3:0]

0x0

R

BG_STATUS_A

Status of comparison between background light level and

background threshold value for Time Slot A (BG_THRESH_A).

A 1 in any bit location means the threshold has been crossed

BG_COUNT_A number of times. This register is cleared once it is

read.

Bit 0: Time Slot A, Channel 1 exceeded threshold count.

Bit 1: Time Slot A, Channel 2 exceeded threshold count.

Bit 2: Time Slot A, Channel 3 exceeded threshold count.

Bit 3: Time Slot A, Channel 4 exceeded threshold count.

0x16

[15:14]

[13:0]

0x0

R/W

BG_COUNT_A

BG_THRESH_A

For Time Slot A, this is the number of times the ADC value

exceeds the BG_THRESH_A value during the float mode subtract

cycles before the BG_STATUS_A bit is set.

0: never set BG_STATUS_A.

1: set when BG_THRESH_A is exceeded 1 time.

2: set when BG_THRESH_A is exceeded 4 times.

3: set when BG_THRESH_A is exceeded 16 times.

The background threshold for Time Slot A that is compared

against the ADC result during the subtract cycles during float

mode. If the ADC result exceeds the value in this register,

BG_COUNT_A is incremented.

Rev. B | Page 52 of 61

Data Sheet

ADPD188GG

Default

Value

Address Data Bit

Access

Name

Description

0x1C

[15:14]

0x0

R/W

BG_COUNT_B

For Time Slot B, this is the number of times the ADC value

exceeds the BG_THRESH_B value during the float mode subtract

cycles before the BG_STATUS_B bit is set.

0: never set BG_STATUS_B.

1: set when BG_THRESH_B is exceeded 1 time.

2: set when BG_THRESH_B is exceeded 4 times.

3: set when BG_THRESH_B is exceeded 16 times.

[13:0]

0x0

R/W

BG_THRESH_B

The background threshold for Time Slot B that is compared

against the ADC result during the subtract cycles during float

mode. If the ADC result exceeds the value in this register,

BG_COUNT_B is incremented.

0x3E

0x3F

0x58

[15:14]

FLT_LED_SELECT_A

Time Slot A LED selection for float LED mode.

0: no LED selected.

1: LED1.

2: LED2.

3: LED3.

13

0

R/W

Reserved

Write 0x0.

[12:8]

[7:0]

0x03

0x20

0x0

FLT_LED_WIDTH_A

FLT_LED_OFFSET_A

FLT_LED_SELECT_B

Time Slot A LED pulse width for LED float mode in 1 µs steps.

Time to first LED pulse in float mode for Time Slot A.

[15:14]

Time Slot B LED selection for float LED mode.

0: no LED selected.

1: LED1.

2: LED2.

3: LED3.

13

0

R/W

Reserved

Write 0x0.

[12:8]

[7:0]

0x03

0x20

0x0

FLT_LED_WIDTH_B

FLT_LED_OFFSET_B

Reserved

Time Slot B LED pulse width for LED float mode in 1 µs steps.

Time to first LED pulse in Float mode for Time Slot A.

Reserved.

[15:12]

[11:10]

FLT_MATH34_B

Time Slot B control for adding and subtracting Sample 3 and

Sample 4 in a four-pulse sequence (or any multiple of four pulses,

for example, Sample 15 and Sample 16 in a 16-pulse sequence).

00: add third and fourth.

01: add third and subtract fourth.

10: subtract third and add fourth.

11: subtract third and fourth.

[9:8]

0x0

R/W

FLT_MATH34_A

Time Slot A control for adding and subtracting Sample 3 and

Sample 4 in a four-pulse sequence (or any multiple of four pulses,

for example, Sample 15 and Sample 16 in a 16-pulse sequence).

00: add third and fourth.

01: add third and subtract fourth.

10: subtract third and add fourth.

11: subtract third and fourth.

7

0x0

R/W

ENA_INT_AS_BUF

FLT_MATH12_B

Set to 1 to enable the configuration of the integrator as a buffer in

TIA ADC mode.

[6:5]

Time Slot B control for adding and subtracting Sample 1 and

Sample 2 in a four-pulse sequence (or any multiple of four pulses,

for example, Sample 13 and Sample 14 in a 16-pulse sequence).

00: add first and second.

01: add first and subtract second.

10: subtract first and add second.

11: subtract first and second.

Write 0x0.

[4:3]

0x0

R/W

Reserved

Rev. B | Page 53 of 61

ADPD188GG

Data Sheet

Default

Value

Address Data Bit

Access

Name

Description

[2:1]

0x0

R/W

FLT_MATH12_A

Time Slot A control for adding and subtracting Sample 1 and

Sample 2 in a four-pulse sequence (or any multiple of four pulses,

for example, Sample 13 and Sample 14 in a 16-pulse sequence).

00: add first and second.

01: add first and subtract second.

10: subtract first and add second.

11: subtract first and second.

Write 0x0.

0

0x0

R/W

Reserved

FLT_EN_B

0x59

15

Write 0x0.

[14:13]

0: default setting, float disabled for Time Slot B.

1: reserved.

2: reserved.

3: enable float mode.

[12:8]

0x08

R/W

FLT_PRECON_B

Float mode preconditioning time for Time Slot B. Time to start of

first float time, which is typically 16 µs.

[7:0]

0x08

0x0

R/W

Reserved

Write 0x08.

0x5A

[15:12]

FLT_LED_FIRE_B

In any given sequence of four pulses, fire the LED in the selected

position by writing a zero into that pulse position. Mask the LED

pulse (that is, do not fire LED) by writing a 1 into that position. In a

sequence of four pulses on Time Slot B, Register 0x5A, Bit 12, is

the first pulse, Bit 13 is the second pulse, Bit 14 is the third pulse,

and Bit 15 is the fourth pulse.

[11:8]

0x0

R/W

FLT_LED_FIRE_A

In any given sequence of four pulses, fire the LED in the selected

position by writing a zero into that pulse position. Mask the LED

pulse (that is, do not fire LED) by writing a 1 into that position. In a

sequence of four pulses on Time Slot A, Register 0x5A, Bit 8, is the

first pulse, Bit 9 is the second pulse, Bit 10 is the third pulse, and

Bit 11 is the fourth pulse.

[7:0]

15

0x10

0x0

R/W

Reserved

FLT_EN_A

Write 0x10.

0x5E

Write 0x0.

[14:13]

0x0

0: default setting, float disabled for Time Slot A.

1: reserved

2: reserved

3: enable float mode in Time Slot A.

[12:8]

[7:0]

0x08

R/W

FLT_PRECON_A

Reserved

Float mode preconditioning time for Time Slot A. Time to start of

first float time, which is typically 16 µs.

Write 0x08.

Rev. B | Page 54 of 61

Data Sheet

ADPD188GG

SYSTEM REGISTERS

Table 32. System Registers

Default

Value

Address Data Bit

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

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

0x1

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

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]

[15:10]

9

0x1F

0x00

0x0

R/W

Reserved

GPIO1_DRV

Write 0x1F.

0x02

Write 0x0000.

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

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

R/W

GPIO0_DRV

GPIO0_POL

GPIO0 polarity.

0: the GPIO0 pin is active high.

1: the GPIO0 pin is active low.

Rev. B | Page 55 of 61

ADPD188GG

Data Sheet

Default

Value

Address Data Bit

Access

R/W

Name

Description

0x06

[15:14]

[13:8]

0x0

Reserved

FIFO_THRESH

Write 0x0.

0x00

R/W

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

0x0A

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]

[11:0]

0x0

R

Reserved

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]

[12:8]

0x0

R/W

Reserved

Write 0x0.

0x00

GPIO1_ALT_CFG

Alternate configuration for the GPIO1 pin.

0x00: GPIO1 is backward compatible to the ADPD103 PDSO pin

functionality.

0x01: interrupt function provided on GPIO1, as defined in Register 0x01.

0x02: asserts at the start of the first time slot, deasserts at end of last

time slot.

0x05: Time Slot A pulse output.

0x06: Time Slot B pulse output.

0x07: pulse output of both time slots.

0x0C: output data cycle occurred for Time Slot A.

0x0D: output data cycle occurred for Time Slot B.

0x0E: output data cycle occurred.

0x0F: 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.

Write 0x0.

[7:5]

[4:0]

0x0

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. B | Page 56 of 61

Data Sheet

ADPD188GG

Default

Value

Address Data Bit

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 R/W

Reserved

Mode

Write 0x000.

0x0

R/W

Determines the operating mode of the ADPD1080/ADPD1081.

0x0: standby.

0x1: program.

0x2: normal operation.

0x11

[15:14]

13

0x0

R/W

Reserved

Reserved.

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

R/W

Reserved

Reserved.

SLOTB_FIFO_MODE

Time Slot B FIFO data format.

0: no data to FIFO.

1: 16-bit sum of all four channels.

2: 32-bit sum of all four channels.

4: four channels of 16-bit sample data for Time Slot B.

6: four channels of 32-bit extended sample data for Time Slot B.

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

R/W

SLOTB_EN

Time Slot B enable. 1: enables Time Slot B.

Time Slot A FIFO data format.

[4:2]

SLOTA_FIFO_MODE

0: no data to FIFO.

1: 16-bit sum of all four channels.

2: 32-bit sum of all four channels.

4: four channels of 16-bit sample data for Time Slot A.

6: four channels of 32-bit extended sample data for Time Slot A.

Others: reserved.

1

0x0

R/W

Reserved

Write 0x0.

0

SLOTA_EN

Time Slot A enable. 1: enables Time Slot A.

Write 0x4000 when EXT_SYNC_SEL is 01 or 10. Otherwise, write 0x0.

Write 0x13.

0x38

0x4B

[15:0]

[15:9]

8

0x0000 R/W

EXT_SYNC_STARTUP

Reserved

0x13

0x0

R/W

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

6

0x0

R/W

CLK32K_EN

Reserved

Sample clock power-up. Enables the data sample clock.

0x0: clock disabled.

0x1: normal operation.

Write 0x0.

Rev. B | Page 57 of 61

ADPD188GG

Data Sheet

Default

Value

Address Data Bit

Access

Name

Description

[5:0]

0x12

R/W

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

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.

Write 0x20.

0x4F

[15:8]

0x20

0x1

0x0

0x1

0x0

R/W

Reserved

7

Reserved

Write 0x1.

6

GPIO1_OE

GPIO1_IE

GPIO1 pin output enable.

GPIO1 pin input enable.

Write 0x1.

5

4

Reserved

[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

GPIO0_IE

Reserved

GPIO1_CTRL

GPIO0 pin input enable.

Write 0x0.

0

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

CLK32M_CAL_EN

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.

[4:0]

[15:3]

2

0x00

Reserved

Write 0x0.

0x5F

0x0000 R/W

0x0 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.

Rev. B | Page 58 of 61

Data Sheet

ADPD188GG

Default

Value

Address Data Bit

Access

Name

Description

1

0x0

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.

0

0x0

R/W

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.

ADC REGISTERS

Table 33. ADC Registers

Default

Address Data Bits Value

Access Name

Description

0x12

[15:0]

0x0028

R/W

FSAMPLE

Sampling frequency: f_SAMPLE= 32 kHz/(Register 0x12, Bits[15:0] × 4).

For example, 100 Hz = 0x0050; 200 Hz = 0x0028.

Write 0x0.

0x15

[15:11]

[10:8]

0x00

0x6

R/W

Reserved

SLOTB_NUM_AVG

Sample sum/average for Time Slot B. Specifies the averaging factor,

N_B, 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

R/W

Reserved

[6:4]

SLOTA_NUM_AVG

Sample sum/average for Time Slot A. N_A: 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

0x1F

0x20

0x21

[15:0]

0x2000

SLOTA_CH1_OFFSET Time Slot A Channel 1 ADC offset. The value to subtract from the