Hi,
I have two SHT25 sensors (product code PR5-8) which I have tried connecting to a Raspberry Pi using the I2C (PR2-3) board. The communication with the sensor appears to work as expected; I can use the sample Python code from the repository and my custom code passes CRC checks on the returned data.
I am running into an issue where the SHT sensors seem to read higher temperatures in ambient room air than other devices. For example, I have a Thermoworks ThermaQ with two different K-probes attached. One probe reads 22.7, another reads 23.6, but the SHT25 reads 24.9 using the Python code. This is in an air conditioned room with the thermostat set to 22.
I have tried checking the user register of the SHT25 and it appears that bit 0x04 is turned off (the user register returns 58 or 0x3A) which seems to eliminate the onboard heater.
Do you have any suggestions or things to try?
Thanks in advance!
i dont see anything wrong with your setup. You might reach out to Sensrion
Info.US@sensirion.com
Thanks for the response; I’ll try reaching out to them in parallel. It seems strange to me that both SHT25 boards I ordered are exhibiting the same behaviour, though.
Is it possible there is thermal spillover from the rest of the board to the SHT25 sensor? I’ve noticed it doesn’t have milled cutouts like Figure 10 on the datasheet (https://sensirion.com/media/documents/66189A49/61642308/Sensirion_Humidity_Sensors_SHT25_Datasheet.pdf). My board from NCD just has four small holes; one on each corner. When I take an IR gun (also, thermoworks) to my board, the back of the PCB where the sensor is mounted does read in at temperatures quite warmer than ambient air (in the range of 24-25 Celsius, which matches the I2C readings). I don’t have a thermal imager, though.
Please let me know what you think.
the on board cutout are are for an external filter cap.
if you think parts might be causing this drift then you can offset the sensor read value in the software .