ADMV8526ACCZ-R7 datasheet(13/31 Pages) AD

13 / 31 page

Data Sheet

ADMV8526

THEORY OF OPERATION

analog.com

Rev. 0 | 13 of 31

INTERPOLATION PLOTS

To garner a visual representation of the interpolation functions,

the interpolation coefficients vs. their input (from the interpolation

tables) can be plotted on a scatter plot. Figure 22, Figure 23, and

Figure 24 are the interpolation functions of y, v, and t using the

interpolation coefficients.

Figure 22. Interpolation Function of y = f(x)

Figure 23. Interpolation Function of v = f(y)

Figure 24. Interpolation Function of t = f(y)

INTERPOLATION COEFFICIENT CALIBRATION

Two primary reasons for the need to calibrate the interpolation

coefficients include accounting for chip process variation and when

a different operating bandwidth is required. The calibration of inter-

polation coefficients normally follows a four phase process (see

Figure 26).

In the first calibration phase, the bandwidth and match coefficients,

V1 and T1, are determined for a desired bandwidth. To perform

this calibration phase, the fCENTER load value must be set to 32,

and then the bandwidth and match load values are adjusted. When

satisfied with the results, the V1 and T1 coefficients can be set to

the bandwidth and match load values, respectively.

For the second calibration phase, the bandwidth and match coeffi-

cients, V2 and T2, are determined for a desired bandwidth. To

perform this calibration phase, the fCENTER load value must be set

to a high value (180 is recommended), and then the bandwidth and

match load values are adjusted. When satisfied with the results, the

V2 coefficient can be adjusted so that the computed result of v =

f(y) = f(180) is equal to the bandwidth load value. Similarly, the T2

coefficient can be adjusted so that the computed result of t = f(y) =

f(180) is equal to the match load value.

For the third calibration phase, the bandwidth and match coeffi-

cients, V0 and T0, are determined for a desired bandwidth. To

perform this calibration phase, the fCENTER load value must be set

to a low value (18 is recommended), and then the bandwidth and

match load values are adjusted. When satisfied with the results, the

V0 coefficient can be adjusted so that the computed result of v =

f(y) = f(18) is equal to the bandwidth load value. Similarly, the T0

coefficient can be adjusted so that the computed result of t = f(y) =

f(18) is equal to the match load value.

For the fourth calibration phase, adjustments are made to all of the

y coefficients to ensure the operating fCENTER is as close as possi-

ble to the anticipated fCENTER. To perform this calibration phase, use

Table 9 as a reference for determining the target frequency for each

y coefficient. For each x value listed in Table 9, compute the y, v,

and t functions, and then set the fCENTER, bandwidth, and match

load values, respectively.

FILTER CODE READ BACK

The capacitor codes that are applied to the filter can be read

back from the chip using Register 0x060 to Register 0x062. These

registers represent the actual state of the capacitors on chip.

This information can be useful for debugging purposes or during

interpolation coefficient calibration.

ADMV8526ACCZ-R7 Datasheet(PDF) 13 Page - Analog Devices

ADMV8526ACCZ-R7 Datasheet(HTML) 13 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

Datasheet Download

Link URL