next topic »

[on:]

Hi guys, I'm using a Pololu motor driver with the VNH2SP30 chip (http://www.pololu.com/catalog/product/706) to control voltage to a thermoelectric cooler and every time the driver comes on and sends current to the thermoelectric device my DS18B20 sensors start putting out incorrect readings. Notably, they drop by several degrees then fluctuate around. If I leave the driver operating and disconnect one of the leads to the cooler then the sensor readings begin coming back normal. I'm guessing this is some problem with the ultrasonic frequency making its way to the wires carrying the DS18B20 signal, so what can I do?

Thanks!

[Reply #1 on:]

I'm guessing this is some problem with the ultrasonic frequency making its way to the wires carrying the DS18B20 signal,

It could be in which case add decoupling to the supply:-
http://www.thebox.myzen.co.uk/Tutorial/De-coupling.html

On the other hand it could be how you have wired the grounds together. Make sure that the ground to your sensor is not caring heavy current from your cooling device.

[Reply #2 on:]

The sensors ground to the arduino, which grounds to the power supply.

[Reply #3 on:]

Yes but it is how it is physically laid out that matters. It all has to be electrically at the same potential but if the sensor return ground is going through a wire that is also carrying the ground from a lot of current you get a slight increase in the ground voltage and it upsets readings.

[Reply #4 on:]

The ground wire for the 2 DS18B20s only connect those sensors and go directly back to the Arduino and connect to nothing else.

I should note that this starts when the motor driver has a pwm input of only 2.5-3 out of 255... So it barely takes any input to start mucking the sensor readings up.

[Reply #5 on:]

So try adding the decoupling.

[Reply #6 on:]

OK, so I switched the DS1820B's from power mode to parasite mode and now no issues with the motor driver. I was powering the DS1820B's from the 5V on the Arduino and the only other thing sharing that power line is a DHT22 temp/humidity sensor. Any thoughts?

[Reply #7 on:]

I go back to decoupling. Add a 0.1uF ceramic cap across the power and ground as close to the DS1820B as possible.
You might also try a series inductor in the power line as well.

[Reply #8 on:]

Ok, will try that. I guess I'm just confused as to why there is this interference when the 5V comes off the Arduino and has no other connection than to the DS18B20's.

Could it be the proximity of the motor driver to other wires, such as signal/power wires?

[Reply #9 on:]

Could it be the proximity of the motor driver to other wires, such as signal/power wires?

Yes, direct pickup by wires is very possible. They act like antenna.