AD9772AST datasheet(11/30 Pages) AD

11 / 30 page

REV. 0

AD9772

–11–

FUNCTIONAL DESCRIPTION

Figure 22 shows a simplified block diagram of the AD9772.

The AD9772 is a complete, 2

× oversampling, 14-bit DAC that

includes a 2

× interpolation filter, a phase-locked loop (PLL)

clock multiplier and a 1.20 V bandgap voltage reference. While

the AD9772’s digital interface can support input data rates as

high as 150 MSPS, its internal DAC can operate up to 400 MSPS,

thus providing direct IF conversion capabilities. The 14-bit

DAC provides two complementary current outputs whose full-

scale current is determined by an external resistor. The AD9772

features a flexible, low jitter, differential clock input providing

excellent noise rejection while accepting a sine wave input. An

on-chip PLL clock multiplier produces all of the necessary

synchronized clocks from an external reference clock source.

Separate supply inputs are provided for each functional block to

ensure optimum noise and distortion performance. A SLEEP

mode is also included for power savings.

14-BIT DAC

ZERO

STUFF

MUX

FIR

INTERPOLATION

FILTER

EDGE-

TRIGGERED

LATCHES

CLOCK DISTRIBUTION

AND MODE SELECT

2 /4

MUX

CONTROL

FILTER

CONTROL

1 /2

PLL CLOCK

MULTIPLIER

+1.2V REFERENCE

AND CONTROL AMP

AD9772

CLKCOM CLKVDD

MOD0 MOD1

RESET

PLLLOCK

DIV0 DIV1

CLK+

CLK–

DATA

INPUTS

(DB13...DB0)

SLEEP

DCOM

DVDD

ACOM

AVDD

REFLO

PLLCOM

LPF

PLLVDD

IOUTA

IOUTB

REFIO

FSADJ

Figure 22. Functional Block Diagram

Preceding the 14-bit DAC is a 2

× digital interpolation filter that

can be configured for a low pass (i.e., baseband mode) or high

pass (i.e., direct IF mode) response. The input data is latched

into the edge-triggered input latches on the rising edge of the

differential input clock as shown in Figure 1a and then interpo-

lated by a factor of two by the digital filter. For traditional base-

band applications, the 2

× interpolation filter has a low pass

response. For direct IF applications, the filter’s response can be

converted into a high pass response to extract the higher image.

The output data of the 2

× interpolation filter can update the

14-bit DAC directly or undergo a “zero-stuffing” process to

increase the DAC update rate by another factor of two. This

action enhances the relative signal level and passband flatness of

the higher images.

DIGITAL MODES OF OPERATION

The AD9772 features four different digital modes of operation

controlled by the digital inputs, MOD0 and MOD1. MOD0

controls the 2

× digital filter’s response (i.e., low pass or high

pass), while MOD1 controls the “zero-stuffing” option. The

selected mode as shown in Table II will depend on whether the

application requires the reconstruction of a baseband or IF signal.

Table II. Digital Modes

Digital

Zero-

Mode

MOD0

MOD1

Filter

Stuffing

Baseband

Low

Baseband

Low

Yes

Direct IF

High

Direct IF

High

Yes

Applications requiring the highest dynamic range over a wide

bandwidth should consider operating the AD9772 in a baseband

mode. Note, the “zero-stuffing” option can also be used in this

mode although the ratio of signal to image power will be re-

duced. Applications requiring the synthesis of IF signals should

consider operating the AD9772 in a Direct IF mode. In this

case, the “zero-stuffing” option should be considered when

synthesizing and selecting IFs beyond the input data rate, fDATA.

If the reconstructed IF falls below fDATA, the “zero-stuffing”

option may or may not be beneficial. Note, the dynamic range

(i.e., SNR/SFDR) is also optimized by disabling the PLL Clock

Multiplier (i.e., PLLVDD to PLLCOM) and using an external

low jitter clock source operating at the DAC update rate, fDAC.

Interpolation Filter Description

The 2

× interpolation filter is based on a 43-tap half-band sym-

metric FIR topology that can be configured for a low or high

pass response, depending on state of the MOD0 control input.

The low pass response is selected with MOD0 LOW while the

high pass response is selected with MOD0 HIGH. The low pass

frequency and impulse response of the half-band interpolation

filter are shown in Figures 2a and 2b, while Table I lists the

idealized filter coefficients. Note, a FIR filter’s impulse response

is also represented by its idealized filter coefficients.

The 2

× interpolation filter essentially multiplies the input data

rate to the DAC by a factor of two, relative to its original input

data rate, while simultaneously reducing the magnitude of the

1st image associated with the original input data rate occurring

at fDATA – fFUNDAMENTAL. Note, as a result of the 2

× interpola-

tion, the digital filter’s frequency response is uniquely defined

over its Nyquist zone of dc to fDATA, with mirror images occur-

ring in adjacent Nyquist zones.

The benefits of an interpolation filter are clearly seen in Figure

23, which shows an example of the frequency and time domain

representation of a discrete time sine wave signal before and

after it is applied to the 2

× digital interpolation filter in a low

pass configuration. Images of the sine wave signal appear around

multiples of the DAC’s input data rate (i.e., fDATA) as predicted

by sampling theory. These undesirable images will also appear

at the output of a reconstruction DAC, although attenuated by

the DAC’s sin(x)/x roll-off response.

In many bandlimited applications, the images from the recon-

struction process must be suppressed by an analog filter follow-

ing the DAC. The complexity of this analog filter is typically

determined by the proximity of the desired fundamental to the

first image and the required amount of image suppression. Add-

ing to the complexity of this analog filter may be the require-

ment of compensating for the DAC’s sin(x)/x response.

AD9772AST датащи(PDF) 11 Page - Analog Devices

AD9772AST датащи(HTML) 11 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

датащи скачать

ссылки URL