ADPD103BCPZ datasheet(20/53 Pages) AD

20 / 53 page

Data Sheet

ADPD103

Rev. B | Page 19 of 52

AFE Integration Offset Starting Point

The starting point of this offset, as expressed in microseconds, is set

such that the falling edge of the integration window aligns with the

falling edge of the LED.

LED_FALLING_EDGE = LED_OFFSET + LED_WIDTH

and,

AFE_INTEGRATION_FALLING_EDGE = 9 +

AFE_OFFSET + AFE_WIDTH

If both falling edges are set equal to each other, solve for

AFE_OFFSET to obtain the following equation:

AFE_OFFSET_STARTING_POINT = LED_OFFSET +

LED_WIDTH − 9 − AFE_WIDTH

Setting the AFE offset to any point in time earlier than the

starting point is equivalent to setting the integration in the

future; the AFE cannot integrate the result from an LED pulse

that has not yet occurred.As a result, anAFE_OFFSET value less

than theAFE_OFFSET_STARTING_POINT is an erroneous

setting. Such a result may indicate that current in the TIA is

operating in the reverse direction from the intended schematic,

where the LED pulse is causing the current to leave the TIA

rather than enter it.

Because, for most setups, the AFE_WIDTH is 1 µs wider than

the LED_WIDTH, the AFE_OFFSET_STARTING_POINT

value is typically 10 µs less than the LED_OFFSET value. Any

value less than LED_OFFSET – 10 is erroneous. The optimal

AFE offset is some time after the AFE_OFFSET_STARTING_

POINT. The band-pass filter response, LED response, and

photodiode response each add some delay. In general, the com-

ponent choice, board layout, LED_OFFSET, and LED_WIDTH are

the variables that can change the AFE_OFFSET. After a specific

design is set, the AFE_OFFSET can be locked down and does

not need to be optimized further.

Sweeping the AFE Position

The AFE offsets for Time Slot A and Time Slot B are controlled

by Bits[10:0] of Register 0x39 and Register 0x3B, respectively.

Each LSB represents one cycle of the 32 MHz clock, or 31.25 ns.

The register can be thought of as 211−1 of these 31.25 ns steps, or

it can be broken into an AFE_COARSE setting using Bits[10:5]

to represent 1 µs steps and Bits[4:0] to represent 31.25 ns steps.

Sweeping the AFE position from the starting point to find a

local maximum is the recommended way to optimize the AFE

offset. The setup for this test is to allow the LED light to fall on

the photodiode in a static way. This is typically done with a

reflecting surface at a fixed distance. The AFE position can then

be swept to look for changes in the output level. When adjusting

the AFE position, it is important to sweep the position using the

31.25 ns steps. Typically, a local maximum is found within 2 µs

of the starting point for most systems. Figure 18 shows an example

of anAFE sweep, where 0 on the x-axis represents theAFE starting

point defined previously. Each data point in the plot corresponds to

one 31.25 ns step of the AFE_OFFSET. The optimal location for

AFE_OFFSET in this example is 0.687 µs from the AFE starting

point.

0.687

100

AFE OFFSET FROM STARTING POINT (µs)

0.15

0.30

0.45

0.60

0.75

0.90

1.05

1.20

1.35

1.50

Figure 18. AFE Sweep Example

Table 14 lists some typical LED andAFE values after optimization.

In general, it is not recommended to use the AFE_OFFSET

numbers in Table 14 without first verifying them against theAFE

sweep method. Repeat this method for every new LED width and

with every new set of hardware made with theADPD103. For

maximum accuracy, it is recommended that the 32 MHz clock be

calibrated prior to sweeping theAFE.

Table 14. AFE Window Settings

LED Register 0x30 or Register 0x35

AFE Register 0x39 or Register 0x3B

Comment

0x0219

0x19FB

2 µs LED pulse, 3 µs AFE width, 25 µs LED delay

0x0319

0x21F4

3 µs LED pulse, 4 µs AFE width, 25 µs LED delay

ADPD103BCPZ датащи(PDF) 20 Page - Analog Devices

ADPD103BCPZ датащи(HTML) 20 Page - Analog Devices

Html Pages

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53

датащи скачать

ссылки URL