User defined (in OP tabs) COEFFICIENT for calibration of analog inputs

[Didier B]

Hi Sailoog and team !
What about associating to each analog input a user defined (in OP tabs) coefficient to calibrate the measuring chain (errors due to either the sensor or distortions due to the hardware linking the sensor to the RPi).
 -For 1W sensors, the reading of my 2 BMP180 have more than one degree difference. I know an error of 2°C is meaningless in the coolant temperature, but it is quite significant in sea temp. when going for a swim ...
- Same for I2C sensors (pressure, temperature, humidity)
 -For SPI interface, as the input value range of MPC3008 is 0 to 3.3 (or 3.2 ?) V, for a battery voltage supervision (10- 14 V) a divider bridge is required, bringing in more distortion to the reading. To monitor a battery current via a shunt (30 to 50 mV) an amplifier will be needed whose gain will depend on the precision of  a couple of resistors ... Of course for an MPC3008 8 coefficients would be reqired.

Of course the HW used to connect a sensor to an input has to adapt the sensor range to the input range, but a SW coefficient would make the calibration process a breeze.

[Sailoog]

Good point. I see calibration coefficients essential in analog sensors coming from a MPC3008. In digital sensors like I2C and 1W I do not see it essential, but why not?

We can try it on v0.9.0 when SPI analog sensors are implemented.

Thanks!

[Didier B]

Well, I agree it is not essential, but  I'd like to have the possibility to display calibrated pressure, temperature and RH indications...

PS I can't use
neither the quote function, any idea about a reason ?
nor the' get more smilies function ?

[Sailoog]

Sure, it would be a great improvement. Talking about priorities, do you mind if we focus on analog interface and signal k implementation and then we go with this?

Are you talking about "Quick reply" or "new Reply"?