First up, I’ve posted to the Raspberry Pi forum and Adafruit forum on this, but hoping someone here might have the answer.
I’m planning on using a DHT22 connected to a Raspberry Pi to control a fridge and humidifier for a curing chamber. The DHT22 was working perfectly when I was testing it at 3.3V, but when I rewired it so it could reach the fridge about 8m away, the DHT22 stopped working. There were no other changes made other than replacing the wires with longer ones (5cm > 8m). On researching the issue, this seemed like an expected result and the solutions suggested include using shielded cable and running the DHT22 at 5V, rather than 3.3V, but I’m having little success.
I’m a total beginner and while there is a tonne of support and information available on how to:
Run the DHT at 3.3V from a Raspeberry Pi
Run the DHT at 5V from an Arduino
There is very little support for running the DHT22 at 5V from the Raspberry Pi where there is a need to reduce the signal voltage down to 3.3V. So, I’m reaching out and hoping someone here can help.
I’ve tried numerous things to try and get it working, but nothing seems to be giving reliably consistent results. The attempts I’ve made are:
Replace standard wires with shielded cable and keep hookup the same (VCC = 3.3V) = Inconsistent readings, hangups and odd results including temperatures <0 and humanities > 3000%.
Inconsistent readings, hangups and odd results including temperatures <0 and humanities > 3000%.
I really want to use the Rpi and had everything working really well when the DHT22 was close to the Rpi, so I suspect that the software is fine and that this is a hardware/hookup issue. To rule out a faulty DHT22, I have ordered another one this morning and keen to give that a shot. If anyone can advice the technically correct way to run the sensor 8m from a Rpi that would be enormously appreciated.
That’s a shame @Daniel67857 - the simplest solution would have been to move the Pi.
Another possibility might be to use a wireless option - I have seen DHT22s mounted on an ESP wireless module (I think it was an ESP8266) - maybe something like this could work, then you could use (presumably) WiFi to collect the reading from the DHT22.
Thanks again, Arb. Will have a look at that option. The other constraint is that the pi is already connected to the home Wifi network and I can’t easily run a network cable to it as it’s in a remote building. Really hadd my finger crossed there was a reliable way to get it to work at 8m.
@Daniel67857, my understanding is that the ESP will connect to your home WiFi, then your Pi can communicate with it wirelessly. Shouldn’t matter how far apart they are that way.
Hey Daniel,
I would add a capacitor from the supply pin to ground Long cables are very prone to noise. I would also measure the resistance of the full cable and see how much voltage you are loosing over the full length.
If you have a way to ground the sensor I would suggest a 100uF at each end of the cable, If not just the Pi side may do. The capacitors should be wired in so that the are connected to the supply line and ground(see picture below).
Another thing that might help would be to measure the voltage drop across the wires at each of the pins. Do you have an Oscilloscope that you can use to have a look at what the data signal is doing with?
Thanks Clinton, unfortunately I don’t have an oscilloscope, but can give the capacitors a shot. Still concerned about the risk of sending 5v to the GPIO though. Is that a valid concern? I did measure the voltage drop and it was minimal (<0.1V).
Yeah it is, did you measure the voltage drop with the sensor running or under no load conditions? Yeah, be very careful that the voltage back to the PI does not exceed 3.3V. The capacitor will look like an open circuit to the DC 5V which is why we connect it like this. Do you have a logic level shifter that you could place on the Pi side of the circuit? Step down modules are generally designed for power supply and can be too noisy to do logic shifting.
Thanks again Clinton. Unfortunately no, I don’t have a logic level shifter. Would be happy to give it a crack, but starting to think the path of least resistance may well be to relocate the PI. I will try and add a capacitor at the pi end and run at 3.3V, but failing that, will just shift the PI i think.
