1 Features

• 14-Bit, dual channel 1.5-GSPS ADC

• Noise spectral density:

– NSD = -153 dBFS/Hz (no AVG)

– NSD = -156 dBFS/Hz (2x AVG)

– NSD = -159 dBFS/Hz (4x AVG)

• Single core (non-interleaved) ADC architecture

• Aperture jitter: 50 fs

• Low close-in residual phase noise:

– -133 dBc/Hz at 10 kHz offset

• Spectral performance (fIN = 900 MHz, -4 dBFS):

– 2x internal averaging

– SNR: 66.8 dBFS

– SFDR HD2,3: 74 dBc

– SFDR worst spur: 90 dBFS

• Input fullscale: 1.0/1.1 Vpp (1/1.8 dBm)

• Code error rate (CER): 10-15

• Full power input bandwidth (-3 dB): 1.6 GHz

• JESD204B serial data interface

– Maximum lane rate: 13 Gbps

– Supports subclass 1 deterministic latency

• Digital down-converters

– Up to four DDC per ADC channel

– Complex output: 4x, 16x to 128x decimation

– 48-bit NCO phase coherent frequency hopping

– Fast frequency hopping: < 1 us

• Power consumption: 1.8 W/channel (2x AVG)

• Power supplies: 1.8 V, 1.2 V

2 Applications

• Phased array radar

• Spectrum analyzer

• Software defined radio (SDR)

• Electronic warfare

• High-speed digitizer

• Cable infrastructure

• Communications infrastructure

3 Description

The ADC32RF52 is a single core 14-bit, 1.5 GSPS,

dual channel analog to digital converter (ADC) that

supports RF sampling with input frequencies up to

2 GHz. The design maximizes signal-to-noise ratio

(SNR) and delivers a noise spectral density of -153

dBFS/Hz. Using additional internal ADCs along with

on-chip signal averaging, the noise density improves

to -156 dBFS/Hz (2x AVG) and -159 dBFS/Hz (4x

AVG).

Each ADC channel can be connected to a quad-band

digital down-converter (DDC) using a 48-bit NCO

which supports phase coherent frequency hopping.

Using the GPIO pins for NCO frequency control,

frequency hopping can be achieved in less than 1 μs.

The ADC32RF52 supports the JESD204B serial data

interface with subclass 1 deterministic latency using

data rates up to 13 Gbps.

The power efficient ADC architecture consumes 1.4

W/ch at 1.5 Gsps and provides power scaling with

lower sampling rates.