Arduino Uno Boucing Pot

I have an Arduino UNO board that im just using as a development board (im programmign directly to the Mega328 and not using the Arduino GUI)

I have setup 2 pots to the ADC0 and ADC1 that connect to a lcd... the values seem to be a bit bouncy.. ie. I will move the pot to a position and it will go 49,48,49,48,49,48
until finally it setlls settles on 49..

in the datasheet it says "AVCC with external capacitor at AREF pin"

Om tje Arduino specs it doesnt show (or im reading it wrong) where the AREF pins connects too.

so my questions.

1# is this what would stop the "bounce".. b# is that what is even happening?
2# how do i access this pin
3# what size capacitor can u suggest?

The AREF pin is clearly marked on the datasheet. On the Atmega328 in PDIP format it is pin 21.

I'm not sure what they mean by external capacitor, as far as I have seen it isn't usually used.

The analog read is likely to slightly differ from one time to the next. I suggest you average them.

Thanks for your resposne..
I can see where the pin is on the ATMEGA328 but I cant see how to access that pin from the Arduino UNO..

I wrote that quote about the capicator from the atmega 328 datasheet...

Im not really sure what to expect with POT's.. should it bounce a little around after just moving? and I may use code to average out as you said, or can i smooth the volatages so it gives me a more definite value?

Actually most designs do wire a .1ufd cap from the Aref pin 21 to ground, the Uno rev3 does. It gives a little extra noise filtering directly to whatever internal or external voltage is being selected as the voltage reference for the internal AVR A/D hardware. Don't recall if the AVR datasheet recommends that also or not, but it is generally practiced.

Lefty

stowaway:
Im not really sure what to expect with POT's.. should it bounce a little around after just moving? and I may use code to average out as you said, or can i smooth the volatages so it gives me a more definite value?

It's a mechanical device. Also the method of getting its value involves charging a capacitor (from memory) and seeing how long it takes to discharge. Minor variations between samples is to be expected a bit.

Pots come in different quality and use different materials in both their sliding contact and the resistive material. A 20 turn wirewound laboratory style 'Instrumentation pot' is a whole lot more accuracy, quite, and repeatable then a $1 pot. It's kind of 'you get what you pay for' in the pot world. That probably applies to the other pot world also? :wink: