LTC1041
5
1041fa
APPLICATIO S I FOR ATIO
WUUU
R
S
• C
IN
. The ability to fully charge C
IN
from the signal
source during the controller’s active time is critical in
determining errors caused by the input charging current.
For source resistances less than 10kΩ, C
IN
fully charges
and no error is caused by the charging current.
For R
S
> 10kΩ
For source resistances greater than 10kΩ, C
IN
cannot fully
charge, causing voltage errors. To minimize these errors,
an input bypass capacitor, C
S
, should be used. Charge is
shared between C
IN
and C
S
, causing a small voltage error.
The magnitude of this error is A
V
= V
IN
• C
IN
(C
IN
+ C
S
). This
error can be made arbitrarily small by increasing C
S
.
The averaging effect of the bypass capacitor, C
S
, causes
another error term. Each time the input switches cycle
between the plus and minus inputs, C
IN
is charged and
discharged. The average input current due to this is
I
AVG
= V
IN
• C
IN
• f
S
, where f
S
is the sampling frequency.
Because the input current is directly proportional to the
differential input voltage, the LTC1041 can be said to have
an average input resistance of R
IN
= V
IN
/I
AVG
= I/(f
S
• C
IN
).
Since two comparator inputs are connected in parallel, R
IN
is one half of this value (see typical curve of R
IN
versus
Sampling Frequency). This finite input resistance causes
an error due to the voltage divider between R
S
and R
IN
.
The input voltage error caused by both of these effects is
V
ERROR
= V
IN
[2C
IN
/(2C
IN
+ C
S
) + R
S
/(R
S
+ R
IN
)].
Example: assume f
S
= 10Hz, R
S
= 1M, C
S
= 1µF, V
IN
= 1V,
V
ERROR
= 1V(66µV + 660µV) = 726µV. Notice that most of
the error is caused by R
IN
. If the sampling frequency is
reduced to 1Hz, the voltage error from the input
impedance effects is reduced to 136µV.
Figure 2. Equivalent Input Circuit
V
IN
R
S
C
S
LTC1041 • AI01
S1
S2
C
IN
(≈ 33pF)
V
–
LTC1041 DIFFERENTIAL INPUT
+
–
Input Voltage Range
The input switches of the LTC1041 are capable of
switching either to the V
+
supply or ground. Consequently,
the input voltage range includes both supply rails. This is
a further benefit of the sampling input structure.
Error Specifications
The only measurable errors on the LTC1041 are the
deviations from “ideal” of the upper and lower switching
levels (Figure 1b). From a control standpoint, the error in
the SET POINT and deadband is critical. These errors may
be defined in terms of V
U
and V
L
.
SET POINT error
V
SET POINT
deadband error V
U
U
≡
+
≡
()
V
V DELTA
L
L
2
2
–
––•
The specified error limits (see electrical characteristics)
include error due to offset, power supply variation, gain,
time and temperature.
Pulsed Power (V
P-P
) Output
It is often desirable to use the LTC1041 with resistive
networks such as bridges and voltage dividers. The power
consumed by these resistive networks can far exceed that
of the LTC1041 itself.
At low sample rates the LTC1041 spends most of its time
off. A switched power output, V
P-P
, is provided to drive the
input network, reducing its average power as well. V
P-P
is
switched to V
+
during the controller’s active time (≈ 80µs)
and to a high impedance (open circuit) when internal
power is switched off.
Figure 3 shows the V
P-P
output circuit. The V
P-P
output
voltage is not precisely controlled when driving a load
(see typical curve of V
P-P
Output Voltage vs Load Current).
In spite of this, high precision can be achieved in two ways:
(1) driving ratiometric networks and (2) driving fast set-
tling references.
In ratiometric networks all the inputs are proportional to
V
P-P
(Figure 4). Consequently, the absolute value of V
P-P
does not affect accuracy.
