Markus Robert Kessler  wrote:
> Hi there,
>
> for monitoring air pressure, temperature and (relative) air humidity, I
> purchased a few "breakout boards" carrying Bosch BME280 sensor.
>
> There are several models out there with different board layout, most of
> them from China.
>
> So, to monitor those envirmonment values I have several Raspberries where
> each one is connected to a BME280 via I2C. Same worked with Arduino
> formerly, but I want them to be connected via internet / wifi.
>
> Nothing special. Yes?
>
> Well, what I am wondering about is, that temp. and pressure are very
> similar on every board, but not so the relative humidity.
>
> Most significant "outlayer" is produced by Adafruit BME280, which claims
> that the humidity value is around 10 percent points higher than most
> other sensors (i.e., 40% China noname and 50% Adafruit). There was no
> difference whether connected one after the other, or both, Adafruit and
> China noname on the same bus with different addresses.
>
> I think, assuming that Adafruit's values are the most precise ones, just
> because this is the most expensive one, is a little too naive :-)
>
> Any experiences with that stuff, or knowing ways to test / verify /
> calibrate those modules?
>
> Best regards,
>
> Markus

Ideally, you’d measure the humidity at three or more points, say, the
minimum likely value, the maximum likely value, and the midpoint. This
would provide for linearity correction as well as calibration.  Since
calibration often varies with temperature, similar variation of temperature
is also needed if accuracy is important.

The humidity can be measured directly from wet-bulb and dry-bulb
temperature readings, assuming a sufficient volume of air to make the
measurements.

For a single-point calibration, I’d use the air in the room, and compare
with a wet/dry-bulb psychrometer.

--
-michael - NadaNet 3.1 and AppleCrate II:  http://michaeljmahon.com

--- SoupGate-Win32 v1.05