Data Sheet

ADA4352-2

analog.com

Rev 0

42 of 47

Total output DC offset error (at 25°C) = VOS,OUT + (IB+ x RIN+)

where, IB+ is the input bias current at the noninverting input.

RIN+ is the source resistance at the noninverting input.

If the noninverting resistance is kept at a minimum, the IB+ × RIN+ term is insignificant.

There are other sources of DC error while using the ADA4352-2 : the gain error, and error due to CMRR and PSRR.

The error contributions from CMRR and PSRR can be reduced using accurate supplies and calibration, and are

typically negligible at DC for a TIA application.

There is also a gain error term because of the resistor tolerance. The equation for this error term is given by IIN x

ΔRFX, where, IIN is the photodiode current and ΔRFX is the deviation from the nominal value of RFx in ohms.

Combining ADA4352-2 with a 16-Bit SAR ADC

The ADA4352-2 can be used to directly drive a successive approximation register (SAR) ADC. The slew rate of the

ADA4352-2 allows for fast output settling within the acquisition time specification of the ADC. The ADA4352-2 also

draws low supply current and can thus be paired with low power, high resolution ADCs for a low power precision

signal chain.

Typical ADA4352-2 Optical Signal Chain Application

Figure 84 shows a typical single-supply application using the AD4696, a high accuracy, low power, 16-channel, 16-

bit SAR ADC to measure an optical signal level using a photodiode. Between the output of the ADA4352-2 and the

analog front-end of the ADC is an external low-pass filter formed by REXT and CEXT.

Figure 84. Optical Signal Chain with ADA4352-2 and AD4696

External RC Filter Design Considerations for SAR ADC Driving

As seen in previous sections, the RC filter at the output is used for several purposes. For ADC driving applications,

this external RC filter also serves to limit both wideband noise into the ADC, and nonlinear kickback from the ADC

inputs into the amplifier output.

Selecting the RC filter between the PGTIA output and ADC input is an iterative process, and the best combination

depends on the intended application. For example, in lower frequency applications, opt for a higher value RC to

ADA4352-2

GAIN RESISTOR NETWORK

KELVIN

SWITCH

NETWORK

VOUT

REXT

CEXT

VADCIN

7pF

40.2kΩ

2.5pF

450kΩ

33pF

315Ω

26.5pF

3.5kΩ

AVDD

+IN

–IN

–VB VCM

AVSS

ADC

16-BIT

MSPS

0.5/1

AD4695/AD4696