CS5317 PERFORMANCE

The CS5317 features 100% tested dynamic per-

formance. The following section is included to

illustrate the test method used for the CS5317.

FFT Tests and Windowing

The CS5317 is tested using Fast Fourier Trans-

form (FFT) techniques to analyze the converter’s

dynamic performance. A pure sine wave is ap-

plied to the CS5317 and a "time record" of 1024

samples is captured and processed. The FFT algo-

rithm analyzes the spectral content of the digital

waveform and distributes its energy among 512

"frequency bins". Assuming an ideal sinewave,

distribution of energy in bins outside of the fun-

damental and dc can only be due to quantization

effects and errors in the CS5317.

If sampling is not synchronized to the input sine-

wave it is highly unlikely that the time record will

contain an exact integer number of periods of the

input signal. However, the FFT assumes that the

signal is periodic, and will calculate the spectrum

of a signal that appears to have large discontinui-

ties,

thereby

yielding

a

severely

distorted

spectrum. To avoid this problem, the time record

is multiplied by a window function prior to per-

forming the FFT. The window function smoothly

forces the endpoints of the time record to zero,

removing the discontinuities. The effect of the

"window" in the frequency domain is to convo-

lute the spectrum of the window with that of the

actual input.

The quality of the window used for harmonic

analysis is typically judged by its highest side-

lobe level. The Blackman-Harris window used to

test the CS5317 has a maximum side-lobe level

of -92 dB.

Figure 7 shows an FFT plot of a typical CS5317

with a 1 kHz sinewave input generated by an "ul-

tra-pure" sine wave generator and the output

multiplied by a Blackman-Harris window. Arti-

facts of windowing are discarded from the

signal-to-noise calculation using the assumption

that quantization noise is white. All FFT plots in

this data sheet were derived by averaging the FFT

results from ten time records. This filters the

spectral variability that can arise from capturing

finite time records, without disturbing the total

energy outside the fundamental. All harmonics

and the -92 dB side-lobes from the Blackman-

Harris window are therefore clearly visible in the

plots.

CLKIN

(Hz)

Mode

N

CLKOUT

(MHz)

R * (k

Ω)C * (nF)

7200

CLKG2

512

1.8432

1.0

2262

911

11.6

187

9600

CLKG2

512

2.4576

1.0

3016

1215

15.5

106

14400

CLKG1

256

1.8432

1.0

4524

1822

11.6

94

19200

CLKG1

256

2.4576

1.0

6032

2430

15.5

52

* The values for R and C are as calculated using the described method. Component tolerances have

not been allowed for. Notice that Ko and Kd can vary over a wide range, so using tight tolerances

for R and C is not justified. Use the nearest conveniently available value.

Table 2 Example PLL Loop Filter R and C values

Figure 7. CS5317 Dynamic Performance

Signal

Amplitude

Relative to

Full Scale

dc

Input Frequency

0dB

-20dB

-40dB

-60dB

-80dB

-100dB

-120dB

9.6 kHz

1 kHz

Sampling Rate: 19.2 kHz

Full Scale:

S/(N+D): 81.39 dB

+ 2.75 V

_

CS5317

DS27F4

15