03-20-2019 12:41 PM
We are having an issue with an entire lot of BME280 sensors that have been reflowed onto our design. During reflow we followed the exact recommended curve for the BME280. We have now noticed the RH to be wildly off from actual RH. About 13% to 15% off. We attempted to recondition the sensors using the reconditioning procedures in the datasheet. This did not work for us. It seemed to only improve the sensors by 2%.
Are there any other methods to reconditoning these sensors? Is there any known eratta or other issues not listed in the data sheet?
05-10-2019 06:08 PM
Hi, I noticed what appears to be errors in the current bme280.c code listed on GitHub for BoschSensortec. It is an obvious repition of a comment line, but also swapped readings of data_msb and data_lsb starting at line 718: existing 718: /* Store the parsed register values for temperature data */ data_lsb = (uint32_t)reg_data[6] << 8; data_msb = (uint32_t)reg_data[7]; uncomp_data->humidity = data_msb | data_lsb;
edited 718: /* Store the parsed register values for humidity data */ data_msb = (uint32_t)reg_data[6] << 8; data_lsb = (uint32_t)reg_data[7]; uncomp_data->humidity = data_msb | data_lsb; |
|
05-20-2019 07:32 PM
Thank-you for reporting this, we have forwarded this feedback to the responsible team.
The result of the binary operation will actually be the same in both cases, but the variable names should be renamed to prevent confusion.
08-29-2019 07:06 AM - edited 08-29-2019 07:07 AM
Hello!
I have the same problem.
I have two BME280 sensors soldered on the little PCBs, which i bought on Aliexpress.
Both of them are accurate in temperature and pressure, but show different RH compared to each other and compared to the calibrated Honeywell's HIH4000 analog RH sensor.
One have a constant offset about 20% and another 7% relatively to HIH4000 measurements.
Is it possible that the producer of the PCBs damaged sensors during reflow? Can reconditioning procedure restore the calibration of the sensors?
08-29-2019 08:22 AM