Temperature&humidity sensor (High Accuracy&Mini) v1.0 Bad readings

Hey everyone,

I have some trouble getting believable readings of the Temperature&humidity sensor (High Accuracy&Mini) v1.0 in combination with the GrovePi+ and Raspberry 2 B+.

It has the latest Jessie image, the firmware of the GrovePi+ is also updated.

When using the grove_i2c_temp_hum_mini.py and the other script, the readings are all over the place. The temperature is somewhere between 30 C and 222 C (if you breath on it), and the humidity is between -24 and 200 % (again after breathing on the sensor). The changes occur within 1 second, so an example of 10 seconds of running the script gives me

(46.125, 8.0)
(54.125, -8.0)
(254.125,200.0)
(187.65, 123.0)
(110.51, 18.0)
(46.125, -8.0)
(46.125, 4.0)
(54.125, 24.0)
(16.125, 22.0)
(36.75, -4.0)

This readings occur within 1 second.
The sensor is connected to the I2C-2 port, but I have also tried I2C-1 and I2C-3, SERIAL and D4 (and tried different python-scripts, grove_temperature_sensor.py, grove_dht_pro.py)

What can cause this problem?

I guess the sensor is found by the Raspberry Pi, when i2cdetect -y 1 is run,

     0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f
00:          -- -- -- -- -- -- -- -- -- -- -- -- --
10: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
20: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
30: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
40: 40 -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
50: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
60: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
70: -- -- -- -- -- -- -- --

Hi @v.oomen,

It looks like you got the right address in i2cdetect -y 1.


I want to get this thing clear for you and for future reference.
We are talking about this temperature and humidity sensor.

Another thing is that this sensor communicates over I2C, so trying this sensor on other ports is redundant.


Give me a day to check this sensor out and I’ll reach you back.

Thank you!

Hi @v.oomen,

Unknowingly I’ve tested a temperature & humidity sensor that was also found at address 0x40 - this made me jump into conclusions that we have the same sensor.

Without a sensor to test with, it’s pretty hard for me to give any support.


Nevertheless, this is a sensor that has been developed by Seeed and I think a good idea would be to copy-paste this issue over their forums.
Here’s a link to their website: https://www.seeedstudio.com/


I’m sorry for not being able to help you on this matter.
My first impression was that there’s some bad math in the code and there’s nothing wrong with the sensor.

Thank you!

Hi,

Thanks for the quick respons. Indeed, it’s the sensor you linked to, so we’re talking about the same sensor.
And thanks for your efforts of trying to find the problem/solution. I will take this issue to Seeed and when the problem has been solved, I will let it know here as well.

Thanks!