ESP8266 Community Forum

This is a waste of time. Interference is also manifested when the CRC is correct. There is problem in the measurement, not in communication.

I also think that the problem is in LUA software. I can't make my DHT22 (AM2302) to work. I 've tired all the software I 've found during two days of searching.
With exact same setup (same connections to nodemcu and sensor) but using SDK examples compiled on Eclipse, my sensor works perfectly.

Try to put a 1000uF capacitor between the 3.3V and GND connection... Possibly with a 100nF ceramic capacitor in parallel.. I am using a DHT22 in a small enclosure right next to a ESP-03 and a small switch mode power supply. No interference detected over a period of 48 hrs... But I am using the 1000uF with 100nF capacitors...

Hi there, I've just got a nodemcu and a dht11.
I spent a few days going with a similar problem of wildly fluctuating temp/humidity readings from the sensors.
I bought two of them and both were all over the place.

It would work sometimes so I figured it must be the timings so I altered all sorts.

Then I realised the CPU could be set to either 80 or 160MHZ - I set it to 160MHZ using (in Lua):

and it's been super stable ever since.

The timings for reading the sensor posted across the internet all assume a clockspeed of 160MHZ - give it a bash and let me know if it works..