
Switched-Capacitor Building Blocks
35
The preceding equation indicates that the accuracy of t is determined by the
inherent precisions of R
0
and C
0
, which tend to vary considerably with temperature
and technology tolerance [1]. In practice, the absolute errors of integrated R
0
and C
0
due to temperature variations and fabrication inaccuracies are normally on the order
of 10%. This implies that in a worst-case scenario, the variance of t obtained from
Equation (3.16) may amount to as large as 20%, which is apparently unacceptable to
most signal processing applications.
If R
0
is replaced by, say, the parallel simulation shown in Figure 3.1(c), the ...