MLX90314
Programmable Sensor Interface
OFFSET
DAC_OFFSET (new value) ~ OF[9:0]+[OFTCi* dT]
Different Modes
Analog Mode
The parameters OF and GN represent, respectively,
offset correction and span control, while OFTCi and
GNTCi represent their temperature coefficients
(thermal zero shift and thermal span shift). After reset,
the firmware continuously calculates the offset and
gain DAC settings as follows: The EEPROM holds
parameters GN, OF, OFTCi and GNTCi, where “i” is
the gap number and can be 1 < i < 4. The transfer
function is described below.
OF[9:0] = Fixed Gain, bytes 4 and 17 in EEPROM.
OFTCi = Offset for a given temperature
segment I. OFTCiL and OFTCiH in
EEPROM table.
dT = Temp. change within the appropriate gap.
Calculation of the offset for a given temperature seg-
ment is performed the same way as for the gain.
[mV/V]
OF[9:0]
(1.83−−1.57)*
−1.57 = DAC_OFFSET
1023
Vout
=
FG
*
DAC_GAIN
*
CSGN[2:0]
*
{Vin+DAC_OFFSET+CSOF}
Digital Mode
Iout = FG * DAC_GAIN * CSGN[1:0] *
The MLX90314 firmware provides the capability of
digitally processing the sensor signal in addition to the
analog processing. This capability allows for signal
correction.
{Vin+DAC_OFFSET+CSOF} * 8.85mA/V
FG = Hardware Gain (~72V/V). Part of the hardware
design, and not changeable.
CSGN = Course Gain, part of byte 2 in EEPROM.
CSOF = Coarse Offset, part of byte 2 in
EEPROM.
Signal Correction
While in digital mode the firmware can perform signal
correction. This is an adjustment to the output level
based on the input signal level. Adjustment
coefficients can be set for five different signal ranges.
The output is obtained by the following formula:
GAIN
DAC_GAIN (new value) ~ GN[9:0] + [GNTCi * dT]
GN[9:0] = Fixed Gain, bytes 3 and 17 in EEPROM.
GNTCi = Gain TC for a given temperature
segment I. GNTCiL and GNTCiH in
EEPROM table.
Output = (Signal – Pi) * Pci + Poff where
Signal = input signal measurement;
Poff = Pressure ordinate
Pi = Pressure signal point (I = 2,3,4,5)
Pci = programmed coefficient.
dT = Temp. change within the appropriate gap.
How to calculate gain in the first temp. gap?:
DAC_GAIN = GN[9:0] - GNTC1 * (T1 – Temp_f1)
How to calculate gain in the other temp. gaps?:
The PCi coefficients are coded on 12 bits: one bit for
the sign, one for the unity, and the rest for the
decimals. The Pi are coded on 10 bits (0-3FFh) in
high-low order.
PNB_TNB: contains the number of signal points,
coded on the four MSB’s. The four LSB’s are reserved
for the number of temperature points. See Table 4 and
Table 5.
2nd gap: DAC_GAIN = GN[9:0] + GNTC2 *
(Temp_f2 – T1)
3th gap: DAC_GAIN = DAC_GAIN2 + GNTC3 *
(Temp_f3 – T2)
4th gap: DAC_GAIN = DAC_GAIN3 + GNTC4 *
(Temp_f4 – T3)
Where:
Compensation Trade-Offs
A compromise must be made between temperature
compensation and pressure correction. The EEPROM
space where the signal coefficients are stored is
shared with the temperature coefficients, with the
result that an EEPROM byte can be used either for a
temperature coefficient or for a signal coefficient, but
not both. Table 6 presents the possibilities among the
maximum number of temperature gaps and the
maximum number of signal gaps.
Temp_f = Filtered temp. (previously described).
If GNTC1 > 2047 => DAC_GAIN
↑
If GNTC2,3,4 > 2047 => DAC_GAIN
↓
[V/V]
GN[9: 0]
(0.97 − 0.48)*
+ 0.48 = DAC _ GAIN
1023
MLX902xx Name of Sensor Rev Y.X 22/Aug/98
3901090314
Page 11
Page 11
Nov/04
Rev 007
MELEXIS [ Melexis Microelectronic Systems ]
