MLX90323 Datasheet, PDF (15/25 Pages) Melexis Microelectronic Systems – 4 – 20 mA Loop Sensor Interface with Signal Conditioning and EEPROM

MLX90323
4 – 20 mA Loop Sensor Interface
with Signal Conditioning and EEPROM
9 Calibration
9.1 Baseline Calibration
The baseline calibration involves setting the operating modes and temperature gain along with the coarse
and fixed gain and offset.
The operating modes are turbo mode, internal or external temperature sensor, and checksum test. The
checksum test, if used, should be enabled after all other settings are complete.
The invert signal bit swaps the differential inputs to the signal path. This has the same effect as swapping the
connections to the VBP and VBN pins of the chip.
Temperature amplifier gain, GNTP, should be set such that the analog converter doesn’t under-run or over-
run at the temperature extremes. When using the internal temperature sensor over the full temperature span
of the device (-40C to +125°C) typically a GNTP setting of 2 will work. When using an external temperature
sensor the voltage on the TMP pin must stay within the ranges as described in the datasheet. The more of
the voltage range used the greater the temperature compensation adjustment range will be.
The coarse gain and fixed gain should be set before the coarse offset and fixed offset. Gain and offset are
inter-related, the offset is multiplied by the gain. It is much easier to program the gain first then offset. It may
be necessary to make some minor adjustment to the coarse offset settings before adjusting the gain. This is
only needed if the output clips at either high end or low end. It is difficult to precisely calculate the offset and
gain values. The amplifier circuitry within the chip uses resistors implemented in silicon. These resistors
have around a 20% tolerance, thus the gain and offset will vary from chip to chip. Each chip is tested to
provide a gain and offset adjustment capability within a specified range. For calculations a typical value can
be taken.
9.2 Temperature calibration
The temperature compensation capability of the MLX90323 is piece-wise with first order compensation in
each segment (gap). The compensation is based on the difference between the current digitized filtered
temperature and the appropriate temperature point. The equations have been previously described. The first
temperature gap is slightly different than the other three. The first gap uses the temperature difference
between the current temperature and the T1 temperature point (the upper end of the temperature gap). The
second temperature gap also uses the T1 point for determining the temperature differences (the low end of
the second gap). The third and fourth gaps also use the temperature point at the low end of their gap. This
means that programming the temperature compensation for the first gap is slightly different than the other
gaps. The compensation coefficients for the first gap (OFTC1 and GNTC1) apply to digitized filtered
temperature values from T1 down to zero. The fourth gap coefficients apply to digitized filtered temperature
range from T3 to 1023 (decimal).
The temperature points T1 thru T3 should always be in increasing order from T1 to T2 to T3. If the
temperature sensor has increasing signal with increasing temperature then the compensation procedure is
intuitively easy. This is the case with the internal temperature sensor. If the temperature sensor has
decreasing signal with increasing temperature then the compensation will start at a hotter temperature and go
towards cold. The procedure below is written with regard to the filtered digitized temperature not the real
physical temperature.
3901090323
Rev 003
Page 15 of 25
Data Sheet
Feb/12