search

Sensirion / arduino-i2c-scd4x

Arduino library for Sensirion SCD4x sensors
BSD 3-Clause "New" or "Revised" License
53 stars 21 forks source link

Question about the forced calibration procedure #16

Closed remyhx closed 2 years ago

remyhx commented 2 years ago

Hi,

We are calibrating some SCD41’s together in a fixed CO2 environment (500ppm). After calibration we noticed big differences between the sensors.

In the datasheet:

“ 3.7.1 perform_forced_recalibration Description: To successfully conduct an accurate forced recalibration, the following steps need to be carried out:

  1. Operate the SCD4x in the operation mode later used in normal sensor operation (periodic measurement, low power periodic measurement or single shot) for > 3 minutes in an environment with homogenous and constant CO2 concentration.
  2. Issue stop_periodic_measurement. Wait 500 ms for the stop command to complete.
  3. Subsequently issue the perform_forced_recalibration command and optionally read out the FRC correction (i.e. the magnitude of the correction) after waiting for 400 ms for the command to complete. • A return value of 0xffff indicates that the forced recalibration has failed. Note that the sensor will fail to perform a forced recalibration if it was not operated before sending the command. Please make sure that the sensor is operated at the voltage desired for the application when applying the forced recalibration sequence.”

While calibrating we take a 10 minute period to build up the gas value in the gas chamber, how important is the measurement done in that 10 min period? (While its not stable yet, only after 5 min). Are all those measurements in the 10 min period used for calibration, or only one measurement at the moment measurement is stopped and the forced cal command is issued?

psachs commented 2 years ago

Hi @remyhx

For the force re-calibration the measurement values of the last 3 minutes will be used. Therefore it is best that you measure in a stable environment for at least 3minutes before performing a force re-calibration.

In your case it seems that everything is done correctly. You can start measurement while building up the environment in the gas chamber as long as you make sure that the environment is stable for the last 3+ minutes. Please also check if the re-calibration was successful (didn't return 0xffff).

If you calibrate at 500ppm you can expect a device-to-device variation of +-40 ppm.

I hope this clarify things.

Best regards, Pascal