User Guide

Chapter 12 Monte Carlo and sensitivity/worst-case analyses
312
The second .MODEL statement scales the nominal value
of Rb2 by 1.1 to approximately 800 ohms. The gain still
increases with a small increase in R, but a larger increase
in R increases the base voltage so much that it drives the
BJT into saturation and nearly eliminates the gain. The
worst-case analysis is fooled by the sensitivity analysis
into assuming that Rb2 must be minimized to degrade the
gain, but maximizing Rb2 is much worse (see Figure 88).
Note that even an optimizer, which checks the local
gradients to determine how the parameters should be
varied, is fooled by this circuit.
Figure 87 Correct worst-case results.
Figure 88 Incorrect worst-case results.
Consi
d
er a s
l
ig
h
t
l
y
d
i
ff
erent scenario: R
b
2
is set to 720 ohms so that maximizing it is
not enough to saturate the BJT, but Rb1 is
variable also. The true worst case occurs
when Rb2 is maximized and Rb1 is
minimized. Checking their individual
effects is not sufficient, even if the circuit
were simulated four times with each
resistor in turn set to its extreme values.
Output is monotonic wit
h
in t
h
e to
l
erance
range. Sensitivity analysis correctly points
to the minimum value.
Output is non-monotonic wit
h
in t
h
e
tolerance range, thus producing incorrect
worst-case results.
Pspug.book Page 312 Wednesday, November 11, 1998 1:14 PM