We see a high variation in LPS33HW outputs. These are factory calibrated to 0.1 hPa, but we are seeing +/-3 hPa. Do we have to do anything specific at power up to apply the factory trimming values?

[JMola.1]

We are using two sensors. One is attached to a patient breathing circuit and the other is set to ambient pressure. We display the differential.

[Eleon BORLINI]

Hi @JMola.1​ ,

do you see the same variation on both sensors you are using? Or the +/-3 hPa is the differential pressure between the two sensors?

The factory calibration values are loaded in register just after the device power-up, and after that you have to

/*
 * Restore default configuration
 */
 lps33hw_reset_set(&dev_ctx, PROPERTY_ENABLE);
 do {
 lps33hw_reset_get(&dev_ctx, &rst);
 } while (rst);
 /*
 * Enable Block Data Update
 */
 lps33hw_block_data_update_set(&dev_ctx, PROPERTY_ENABLE);
 /*
 * Set Output Data Rate
 */
 lps33hw_data_rate_set(&dev_ctx, LPS33HW_ODR_10_Hz);

I suggest you to check the read_data_simple.c file on Github repository.

To increase the accuracy, and if you are interested in differential pressure (i.e. the pressure variation with respect to a reference pressure), you may use the Autozero function described in the datasheet, p.34, and implemented in C code in the github drivers (lps33hw_reg.c).

-Eleon

[JMola.1]

Eleon,
Thank you for the response.
None of the sensors read accurate barometric pressure in our testing against a NIST calibrated instrument. Individually they read +/-3 hPa. When paired we have seen differences as large as 5 hPa.
We will look at the Git Hub information and the code you show below and get back to you.
Regards,
John