ADL5310

Rev. A | Page 15 of 20

A 10 nF capacitor on each VSUM pin (20 nF parallel equivalent)

combined with the 16 kΩ source resistance yields a 500 Hz pole,

which is sufficiently below the bandwidth for the minimum

input current of 3 nA.

Residual crosstalk disturbance is particularly problematic at the

lowest currents for two reasons. First, the loop is unable to reject

summing node disturbances beyond the limited bandwidth.

Second, the settling response at the lowest currents to any

residual disturbance is significantly slower than that for input

currents even one or two decades higher (see Figure 18).

–6

–3

0

3

6

9

12

0

0.2

0.4

0.6

0.8

1.0

1.2

0

0.5

1.0

1.5

2.0

2.5

TIME (ms)

ACTIVE CHANNEL OUTPUT PULSE, 1-DECADE STEP

3

µA TO 30µA

INACTIVE CHANNEL RESPONSE

Figure 36. Crosstalk Pulse Response for Various Input Current Values

Figure 36 shows the measured response of an inactive channel

(dc input) to a 1-decade current step on the input of the active

channel for several inactive channel dc current values. Addi-

tional system considerations may be necessary to ensure

adequate settling time following a known transient when one or

both channels are operating at very low input currents.

RELATIVE AND ABSOLUTE POWER

MEASUREMENTS

When properly calibrated, the ADL5310 provides two inde-

pendent channels capable of accurate absolute optical power

measurements. Often, it is desirable to measure the relative

gain or absorbance across an optical network element, such as

an optical amplifier or variable attenuator. If each channel has

identical logarithmic slopes and intercepts, this can easily be

done by differencing the output signals of each channel. In

reality, channel mismatch can result in significant errors over a

wide range of input levels if left uncompensated. Postprocessing

of the signal can be used to account for individual channel

characteristics. This requires a simple calculation of the

expected input level for a measured log voltage, followed by

differencing of the two signal levels in the digital domain for a

relative gain or absorbance measurement. A more straight-

forward analog implementation includes the use of a current

mirror, as shown in Figure 37. The current mirror is used to

feed an opposite polarity replica of the cathode photocurrent of

PD2 into Channel 2 of the ADL5310. This allows one channel to

be used as an absolute power meter for the optical signal

incident on PD2, while the opposite channel is used to directly

compute the log ratio of the two input signals.

5V

TEMPERATURE

COMPENSATION

BIAS

GENERATOR

1k

Ω

2M

Ω

4.7nF

1k

Ω

4.7nF

0.1

µF

1k

Ω

4.7nF

1nF

1nF

0.1

µF

1nF

1nF

VNEG

COMM

COMM

VREF

VRDZ

VPOS

VSUM

INP1

PD1

InGaAs PIN

1k

Ω

4.7nF

PD2

InGaAs PIN

IRF1

OUT1

SCL1

BIN1

LOG1

TEMPERATURE

COMPENSATION

COMM

log

log

log

log

5V

VSUM

5V

INP2

IRF2

OUT2

Φ

α

*

Φ

SCL2

BIN2

LOG2

100pA

**

α

ADL5310

Figure 37. Absolute and Relative Power Measurement Application

Using Modified Wilson Current Mirror

The presented current mirror is a modified Wilson mirror.

Other current mirror implementations would also work, though

the modified Wilson mirror provides fairly constant perfor-

mance over temperature. It is essential to use matched pair

transistors when designing the current mirror to minimize the

effects of temperature gradients and beta mismatch.