Reference Section
•
125
Chapter 5
where
τ
s
= R
s
C
m
,
τ
srp
= R
srp
C
m
, where R
srp
is the residual (uncompensated) series
resistance in terms of Prediction, given by R
srp
= R
s
(1 - % Prediction /100), and, in the
frequency domain s = j
w
(
w
is the natural frequency,
w
= 2
π
f), or in the time domain s
is the operator d/dt. Thus, V
cp
= V
c
·
(1 + (R
s
/ R
srp
- 1)e-t/
τ
srp
).
Moreover, the membrane potential, V
m
, is given by V
m
= V
cp
/(1+s
τ
s
) = V
c
/(1+s
τ
srp
), or
V
m
= V
c
(1 - e-t/
τ
srp
). Therefore, advancing the Prediction setting to 80% gives R
srp
of
2 M
Ω
and
τ
srp
of 100
µ
s. That is, the speed with which the membrane potential
responds to a voltage command is improved 5-fold over that which is achieved with
0% Prediction. Prediction of 98% gives R
srp
of 200 k
Ω
and
τ
srp
of 10
µ
s. The
membrane potential will now respond to a step voltage command with a 10-90%
risetime of about 22
µ
s and will settle to within 1% of its final value in less than 50
µ
s.
Saturation Effects
Note that the equation presented above for V
cp
(
i.e.
the command potential plus
Prediction signal) can be used to define the maximum allowable % Prediction
for a given size step voltage command. (This limit should not be confused
with limitations imposed by the stability of the Prediction circuit itself.) The
command plus Prediction signal is attenuated at the headstage by a 10:1
voltage divider. Since the circuitry in the MultiClamp 700A main unit will
saturate at about ±11-12 V, V
cp
is limited in absolute value to about 1.1 to
1.2 V. To be conservative, we will use 1.1 V in the following calculations.
The peak amplitude of V
cp
for a step voltage command, V
c
, is given by V
c
(R
s
/R
srp
)
that can be rewritten as V
c
/ (1 - % Prediction /100). So we may state the limitation on
V
c
as a function of % Prediction as:
V
c
≤
1.1(1 - % Prediction /100)
or the limitation on % Prediction as a function of V
c
as:
%
Prediction
≤
100(1 - V
c
/1.1)
Thus, for example, if it is known that the maximum command step to be used in a
particular experiment is 100 mV, Prediction may be set at 91% without fear of
saturation of V
cp
; this is true regardless of the value of R
s
or C
m
. In fact, this is a rather
conservative estimate since it is derived on the assumption that the signal V
cp
will