12-29-2020 02:51 PM
Dear Bosch Community
I'm successfuly using a BME680 with the bsec library 1.6.1480 on an ESP32 microcontroller (configuration stored on SPI Flash) based on the basic_config_state.ino example. Everything works fine, the sensor is reaching accurcy 3 as it should and the config is written routinely and read at sartup as normal. When I transported the system to a different place (several occasions), the eCO2 value was never lower than about 2000 afterwards and the eVOC values was at 10'000 even after a longer time period. It was only after erasing the configuration that the sensor was working properly again.
I don't understand why I this happens and what had gone wrong or what problematic handling I have done. The only way to solve this problem is to erase the configuration and then everything works as expected and normal. What happens during transportation when the sensor is not powered on? Does anyone have an idea? This is critical when I want to ship the device with a stored configuration.
Best Regards, JuergL
Solved! Go to Solution.
12-30-2020 03:07 AM - last edited on 01-12-2021 02:49 PM by Saskia
After sensor is moved to another position, BESC algorithm need create a new baseline, and at least keep half an hour to fit this new environment.
Did you use a new sensor? The sensor with the defferent mode need the different initial time after first power on.
How long has the sensor been keeping the new position?
12-30-2020 10:53 AM
I'm using the sensor in LP mode. It was for at least 3 days in one place, with static accuracy reached level 3, and then moved to a different place and finally back again with the sensor turned off. It was the same sensor that was in the same place as when it acquired the initial configuration. The sensor was running for at least 6 hours with these high values.
I'm a little bit confused as I understand well that the sensor need to calibrate himself and this takes time. But when I move the sensor with its configuration stored in the microcontroller to a different place I expect the sensor to give similar results as before under similar conditions. In my cases it was just constant at a maximum value. This is doesn't sound logic to me. If a sensor indicates such high values it would mean that we should immediately leave the room. This happened when the window was open and everything fine, controlled with separate CO2 and TVOC sensors.
12-31-2020 03:08 AM
Could you tell me the below inquiries?
What state was the senosr when you moved the sensor to a different place? Power on or off? How long was the sensor stayed at another place?
When was the sensor configuration stored? after or before moved to another place or before finally back again with the sensor turned off?
I need more information to ask for our internal experts' help.
12-31-2020 04:49 PM
I used two sensors and both were stable and had static accuracy 3 after about 3 days continuous recording. The state is saved every 6 hours when accuracy 3 is reached and it was saved at least once prior to power down. The power was turned off and the sensors were moved away. The next day one sensor was powered up again on a different place and showed this behavior. The second sensor was returned that date to the original place and left there (not powered on). When I heared about the problems about two days later I powered the other sensor in my office and wanted to cheque if I have the same problem: Unfortunately yes, I had this extremly high values too. They remained for at least 6 hours. But I'm not sure what the accuracy was at this moment. I have seen this behavior before and think I could reproduce it if necessary.
You are welcome, JuergL