Datasheet

15
Signal Chain Design Guide
Sensor Calibration/Compensation
Sensor Characteristics
Sensor characteristics vary, both for device to device as
well as for a given device over the operating conditions. To
optimize system operation, this sensor variation may
require some compensation. This compensation may
simply address device to device variation, or be more
dynamic to also address the variations of the device over
the operating conditions. The system voltage and
temperature may effect the sensor output characteristics
such as output voltage offset and linearity. This
conditioning circuit can also be used to optimize the range
of the sensors conditioned signal into the Analog-to-Digital
conversion circuit.
Depending on the sensor, the sensor’s output may either
be voltage or a current. A possible compensation circuit
for each output type will be discussed.
In this first case, the sensor generates an output voltage.
Temperature sensors are typical sensors that generate a
voltage output which varies unit to unit.
Voltage Control
A simple voltage control circuit (see figure below) can
ensure that the sensors output voltage is optimized to
the input range of the next stage in the signal chain. This
circuit is a gain amplifier, where the R
1
and R
2
resistances
determine the amplifier’s gain. The amplifier’s output
voltage range is limited by the Vdd and Vss voltages.
Controlling the Vos voltage can optimize the Vout voltage
profile, based on the sensor’s output voltage (Vsen).
Inverting Amplifier (Voltage Gain)
Either a DAC or a Digital Potentiometer can be used to
control the voltage at Vos. This device can be a non-
volatile version so that at system power up the Vos
voltage is at the calibrated voltage, programmed during
manufacturing test, to address the sensor’s device to
device variation. If dynamic control is desired, the DAC or
Digital Potentiometer can be interfaced to a microcontroller
so that dynamic changes to the Vos voltage compensate
for the system conditions and non-linearity of the sensor.
Analog-to-Digital
Conversion
Conditioning
Circuit
(Optimizes Sensor’s
Output)
Sensor
V
OUT
R2
R1
VOS
V
SEN VIN
VDD
Op Amp
+
–
VDD
Typically during the manufacturing stage the test system
will write this compensation data into some non-volatile
memory in the system which the microcontroller will use
during normal operation to adjust the Vos voltage.
In this second case, the sensor generates an output
current. Photodiodes are a typical sensors that generate a
current output, and can vary ±30% at +25°C (unit-to-unit).
Current to Voltage
A simple current to voltage converter circuit (see Figure
below), is used to create a voltage on the output of the op
amp (V
1
), which can then be compensated. In this circuit,
the photodiodes Ipd current times the Rf resistance equals
the voltage at the op amps output (V
1
). The Rf resistance
needs to be selected so that at the minimum Ipd(max)
current, the Vout voltage is at the maximum input voltage
for the next stage of the signal chain. Typically this will
be done when the DAC or Digital Potentiometer is at Full
Scale (so Vout ≈ V
1
). For photodiodes where the Ipd(max)
current exceeds the minimum Ipd(max) current (increasing
the V
1
voltage), the DAC or Digital Potentiometer Wiper
code be programmed to attenuate the that V
1
voltage to
the desired Vout max voltage. This then compensates for
the variation of the photodiode‘s Ipd current.
Photodiode Calibration (Trans-Impedance Amplifier)
This device can be a non-volatile version so that at
system power up the voltage attenuation is at the level,
programmed during manufacturing test, to address the
sensor’s device to device variation. If dynamic control
is desired, the DAC or Digital Potentiometer can be
interfaced to a microcontroller so that dynamic changes
to the voltage attenuation compensate for the system
conditions and non-linearity of the sensor. Typically during
the manufacturing stage the test system will write this
compensation data into some non-volatile memory in the
system which the microcontroller will use during normal
operation to adjust the voltage attenuation.
Cf may be used to stabilize the op amp. Additional
information on Amplifying High-Impedance Sensors is
available in Application Note AN951.
RF
VPD
Op Amp
+
–
C
F
RAB
V1
A
B
CN
V
OUT
IPD
C
(1)