AD9545 datasheet(129/157 Pages) AD

129 / 157 page

Data Sheet

AD9545

Rev. A | Page 129 of 157

Therefore,

T3 Exponent = C3_ExpVal = −28

= 0xE4 (hexadecimal)

T3 Significand = round(C3 × 215 – C3_ExpVal)

= round(2.2493907 × 10−9 × 215 − (−28))

= 19,786

= 0x4D4A (hexadecimal)

Using C4 as an example, convert Coefficient C4 = 2.7943570 ×

10−11 to its corresponding bit field values. First, check that C4 is

greater than the quantization limit.

C4_ExpVal =

log













log

7943570

log













−

= −35

This is above the quantization limit of −127, therefore

T4 Exponent = C4_ExpVal = −35

= 0xDD (hexadecimal)

T4 Significand = round(C4 × 215 − C4_ExpVal)

= round(2.7943570 × 10−11 × 215 − (−35))

= 31,462

= 0x7AE6 (hexadecimal)

Using C5 as an example, convert Coefficient C5 = −2.4817310 ×

10−13 to its corresponding bit field values. First, check that C5 is

above the quantization limit.

C5_ExpVal =

log













log

4817310

log













−

= −41

This value is greater than the quantization limit of −127.

Therefore,

T5 Exponent = C5_ExpVal = −41

= 0xD7 (hexadecimal)

T5 Significand = round(C5 × 215 − C5_ExpVal)

= round(−2.4817310 × 10−13 × 215 − (−41))

= −17883

= 0xBA25 (hexadecimal)

Use the following formulas to verify the value of the bit field

computations closely matches the desired coefficient value;

= Constant compensation value × 2−45

(26)

Cx|x ≠ D = Tx significand × 2Tx exponent – 15

(27)

Note that there may be a slight difference between the values

produced by Equation 26 and Equation 27 and the original

coefficient values due to quantization effects.

Example (coefficient verification)

C0 = 35,349,513,305 × 2−45

= 1.004694 × 10-3 (verified)

C1 = −24,849 × 2−18−15

= −2.892804 × 10−6 (verified)

C2 = −26,265 × 2−22−15

= −1.911030 × 10−14 (verified)

C3 = 19,786 × 2−28−15

= 2.249408 × 10−9 (verified)

C4 = 31,462 × 2−35−15

= 2.794387 × 10−11 (verified)

C5 = −17,883 × 2−41−15

= −2.481765 × 10−13 (verified)

Digital Filter

The CF1 values produced by the polynomial calculator block in

Figure 99 depend on temperature measurements (internal or

external) as an input parameter. Any noise associated with those

measurements is a potential noise source on CF1, which can

lead to a degradation of phase noise performance. To mitigate

potential noise injection by the compensation mechanism,

Compensation Method 1 employs a filter that applies a

smoothing function to the raw CF1 values.

The user controls the filter bandwidth via the 3-bit unsigned

coefficient output filter cutoff bit field in Bits[D2:D0] of

colocated poles that yield a 3 dB bandwidth per Table 54.

Table 54. Digital Filter Bandwidth

Binary Value

3 db Bandwidth (Hz)

000

156

001

010

011

100

101

110

111

The device automatically bypasses the digital filter under the

following conditions:

•

The system clock is unstable

•

The difference between the input and output of the digital

filter exceeds ~125 ppm.

None of these conditions occur during normal operation; thus,

the filter is nearly always present.

AD9545 Datasheet(PDF) 129 Page - Analog Devices

AD9545 Datasheet(HTML) 129 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 Download

Link URL