Let's Control It

I was running an old version of espeasy, from somewhere around 2021, on an ESP32.
A BME280 was connected to it and had an offset of about 1 degree celcius. No problem.

Now I created a new set: new ESP32, new BME280 and the latest version of espeasy.
Now the reading is almost 5 degrees too high, and also more "spiky".
So to be sure it isn't a faulty BME280, I took the old BME280, connected it to the new ESP32/espeasy but this one now also shows over 4 degrees too high.
So that kinda proofs that something has changed in the software?

In both setups, old and new, the position of the BME280 is at about the same (same wall, same height, simmilar casing with simmilar openings).
Based on small changes in placement, I would accept perhaps a few tenths difference, not over 3 degrees more "wrong".

There have been some changes in the code for sure, but based on your reply I'm afraid there might be something else wrong here.

Back in 2018 or 2019 I have changed the BME280 code to make sure everything is being read in a single I2C call.
The Bosch sensors require this as you otherwise may end up reading from 2 separate readings, which typically show a temperature offset of a few degrees.

Another reason can be that the sensor is being read too frequently.
What is your set interval for this task?
It should be at least 2 sec

I was already thinking about how often to read, so I already changed from 60s to 120s.

Hmm 60s shouldn't be a problem at all.

I have to check to see if something else has changed.
For example changed code regarding putting the sensor to sleep or something like that.

The temperature reading is always too high?

Check the configuration settings in the new version of ESPEasy related to the BME280 sensor. Chcek if the sampling rates, sensor type selection, and calibration options, match the configurations in the old version as closely as possible.

TD-er wrote: ↑19 Dec 2023, 14:52 The temperature reading is always too high?

I think that's the clou.
What I saw that last time that I swapped the BME280 from the new one back to the old one, it first looked fine. Then after a few minutes, it started to change.
That now brings my to one change that I forgot: in the old setup I had the ESP32 in a diffierent housing, now the ESP32 is in the same housing as the BME280.
So most probably it's as simple as heat from the ESP32 that is now close(r) to the BME280.

OK, so no need to look into it further?

TD-er wrote: ↑19 Dec 2023, 15:47 OK, so no need to look into it further?

No, your hint was enough. I take it back and start my own investigation. Thanks