English
Language : 

MLX90320_12 Datasheet, PDF (14/33 Pages) Melexis Microelectronic Systems – Automotive sensor interface
MLX90320
Automotive sensor interface
GnOUT = 2.11× (GnIN − AGND) + AGND
OUT = −1.668× GnOUT +1.16725×VDD
Explanation of the parameters used in the global equations:
INP-INM is the differential output from the sensor
IINV is the bit that swaps the MLX90320 inputs INP and INM
POSout, NEGout, Dtsout, CSout, Gnin and Gnout are MLX90320 internal nodes represented in the
schematic of the sensor signal chain (fig 5)
AGND is an analog ground dependent of the supply voltage VDD.
GDIDO, GDTS, GCS form the sensor signal chain coarse gain programmable.
FNOFF and CSOFF are respectively the sensor signal chain fine and coarse offset programmable.
FNGAIN is the sensor signal chain fine gain programmable.
OUT is the application mode output of the MLX90320.
The different sensor chain calibration parameters with their range will be described in the following paragraphs.
6.4.1 The Gain calibration of the sensor signal chain.
Three programmable coarse gain stages allow calibrating a wide range of sensor output spans (1.8mV/V to
63mV/V range) to the desired MLX90320 output span. Amplifier DIDO is a differential input – differential output
amplifier, while amplifier DTS and CS are dual-to-single-ended amplifiers giving a single ended output voltage
referred to the ground.
Each one of these three amplifiers is one bit programmable:
The DIDO gain is 3.25 or 13 depending on the value of the corresponding bit stored on the address 7
of the EEPROM.
The DTS gain is 1.994 or 4.96 depending on the value of the corresponding bit stored on the address 7
of the EEPROM.
The CS gain is 1.238 or 1.934 depending on the value of the corresponding bit stored on the address 7
of the EEPROM.
Besides the three programmable coarse gain stages, there is also a 10 bits programmable fine gain stage within
the range 44.6% to 99%. The fine gain calibration allows an accurate adjustment of the output span. The fine
gain can be calculated by the formula:
FNGAIN = (0.446 + FNGainreal × (0.99 − 0.446))
Equation 1
Explanation of parameters used in equation 1:
FNGAIN is the fine gain used in the signal sensor chain.
FNGainreal is the value of the fine gain in the [0..1] range with 10 bits resolution.
The fine gain calibration allows also a second order compensation of the drift with temperature of the sensor
sensitivity. The value of the fine gain is given by the formula:
FNGainreal = G0 + G1 × (T − T1 ) + G2 × (T − T1 )2
Equation 2
Explanation of parameters used in equation 2:
3901090320
Rev 007
Page 14 of 33
Data Sheet
Feb/12