3–322
Motorola Sensor Device Data
For More Information On This Product,
Go to: www.freescale.com
Figure 5. Two Op–Amp Signal–Conditioning Stage with Variable Gain and Negative Dc Level Shift Adjust
VIN2
VIN1
VREF
NODE 1
R1
R2
R4
R3
VCC
NODE 2
U1
VO
′
VO
U1
RG
ROFF
To derive the general transfer function, nodal analysis and
superposition are used:
VO
=
R4
R3
R4
R3
R2R4
R1R3
R4
RG
R4
RG
R2R4
R3RG
R2R4
R1R3
1 VIN2
–
R2R4
R3RG
R4
ROFF
VIN1
+
VREF
+
(VIN2 – VCC)
(14)
As before, defining the sensor’s differential output as
V
SENSOR
, defining V
IN2
as S+ for pressure sensor
applications, and using the simplification that
R1 = R4
and
R2 = R3
obtains the following simplified transfer function:
VO =
R4
R3
R4
ROFF
2R4
RG
1
(VSENSOR) + VREF
+
(S+ – VCC)
(15)
The gain is
G =
R4
R3
+
2R4
RG
+ 1
(16)
To adjust the gain, refer to the guidelines presented in the
section on Two Op–Amp Gain Stage with Variable Gain
.
V
REF
is the positive dc level shift, and the negative dc level
shift is:
V–shift =
R4
ROFF
(S+ – VCC)
(17)
The following guidelines will help design the circuitry for the
negative dc voltage level shift:
To establish a stable negative dc level shift, VCC should be
regulated; otherwise, the amount of negative level shift will
vary as VCC varies.
ROFF should be the only resistor varied to adjust the
negative level shift. Varying R4 will change the gain of the
two op–amp circuit and reduce the common mode rejec-
tion.
To determine the value of ROFF:
1. Determine the amount of negative dc level shifting re-
quired (defined here as V–shift).
2. R4 already should have been determined to set the gain
for the desired signal–conditioned sensor output.
3. Although V–shift is dependent on S+, S+ changes only
slightly over the entire pressure range. With Motorola’s
MPX10 powered at a 5 V supply, S+ will have a value of
approximately 2.51 V at zero pressure and will increase
as high as 2.53 V at full–scale pressure. This error over
the full–scale pressure span of the device is negligible
when considering that many applications use an 8–bit
A/D converter to segment the pressure range. Using an
8–bit A/D, the 20 mV (0.02 V) error corresponds to only
1 bit of error over the entire pressure range (1 bit / 255
bits x 100% = 0.4% error).
4. ROFF is then calculated by the following equation:
ROFF =
S
+
– VCC
V
–
shift
R4
(18)
An alternative to using this equation is to use a
potentiometer for ROFF that has a resistance range on the
order of R4 (perhaps 1 to 5 times the value of R4). Use the
potentiometer to fine tune the negative dc level shift, while
monitoring the zero pressure offset output voltage, VO. As
before, if a fixed resistor is preferable, then measure the
potentiometer’s resistance and replace the potentiometer
with the closest 1% resistor value.
Important note: The common mode rejection of this amplifier
topology will be low and perhaps unacceptable in some
applications. (A SPICE model of this amplifier topology
showed the common mode rejection to be 28 dB.) However,
this circuit is presented as a solution for applications where
only two operational amplifiers are available and the common
mode rejection is not critical when considering the required
F
Freescale Semiconductor, Inc.
n
.