AD9546SLASHPCBZ datasheet(171/205 Pages) AD

171 / 205 page

Data Sheet

AD9546

Rev. 0 | Page 171 of 205

COMPENSATION METHOD 1

Compensation Method 1 employs the open-loop method of

system clock compensation (see the Open-Loop Method

section). As shown in Figure 116, Compensation Method 1

takes the polynomial coefficients (Cx) and a temperature value

(T) to compute a compensation factor (CF1). Compensation

Method 1 differs slightly from the open-loop method by the

inclusion of a digital filter (see the Digital Filter section).

POLYNOMIAL

CALCULATOR

C0 C1 C2 C3 C4 C5

DIGITAL

FILTER

COEFFICIENT

OUTPUT FILTER

CUTOFF

CF1

Figure 116. Compensation Method Number 1

The coefficient values, C0 to C5, are programmed into the register

map by the user (see the Programming the Coefficients (Cx) for

Compensation Method 1 section). The temperature value comes

from the internal temperature sensor or a temperature register

(see the Temperature Sensor section for details). The CF1

compensation factor applies to the designated NCOs and TDCs

to correct for frequency variations of the system clock source.

CF1 embodies the factor, 1 + FFECOMP, shown in Figure 114.

How to Derive Coefficients from Frequency Data

To compensate for the frequency variation of the system clock

source, knowledge of the frequency vs. temperature profile is a

necessity. Generally, the temperature profile of the source is a

table of temperature values with corresponding values of the

source frequency. Table 96 shows a hypothetical example of a

25 MHz source in the two leftmost columns.

Because the AD9546 implements compensation in terms of

fractional error, it is necessary to convert the frequencies in

Table 96 to FFE values (refer to the System Clock Compensation

Overview section). Table 96 shows the resulting FFE values.

However, Compensation Method 1 generates CF1 as a time

error correction rather than a frequency error correction.

Therefore, the user must convert the FFE data to FPE data

(refer to the System Clock Compensation Overview section for

the conversion formula). Table 96 shows the resulting FPE

values. From the calculated FPE values, find the coefficients

(Cx) that best fit the FPE data, which generally involves

mathematical regression techniques using specialized

mathematical software tools.

Applying such regression techniques to the hypothetical FPE

values in Table 96 yields the following coefficients:

C0 = 1.0046936 × 10−3

C1 = −2.8927765 × 10−6

C2 = −1.9110192 × 10−7

C3 = 2.2493907 × 10−9

C4 = 2.7943570 × 10−11

C5 = −2.4817310 × 10−13

Table 96. Example 25 MHz Source Data

Temperature (°C)

Frequency (MHz)

FFE (×10−6)

FPE (×10−6)

−20

24.9757265855

−970.93658

971.88021

−10

24.9745666538

−1017.33385

1018.36987

24.9751062576

−995.74970

996.74220

24.9758628851

−965.48459

966.41766

24.9780535472

−877.85811

878.62943

24.9789426612

−842.29355

843.00361

24.9809631193

−761.47523

762.05552

24.9810049826

−759.80070

760.37843

24.9800392641

−798.42944

799.06744

24.9777091418

−891.63433

892.43005

24.9742106356

−1031.57458

1032.63982

AD9546/PCBZ Datasheet(PDF) 171 Page - Analog Devices

AD9546/PCBZ Datasheet(HTML) 171 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 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 ...More

Datasheet Descarga

Enlace URL