Optimum current for analog measurements?

Other Hardware Sensors
1

I want to achieve the best possible analog measurement with an Arduino Uno and a simple voltage divider (soft potentiometer). What resistance should the potentiometer have for best precision? The potentiometer is disconnected when not in use so a pullup or pulldown resistor is required to detect this condition.

10 Kiloohm potentiometers seem to work fine and are often used with Arduino projects. A 100 Kiloohm potentiometer might save energy, but there seems to be random noise sometimes. Is there a way to calculate the best resistance to match the Uno's AD input?

I could either activate the internal pullup resistor by software or use an external one. Which is better? Is pullup or pulldown better, internal or external?

I understand that the pullup/pulldown resistor will slightly distort the measurement. So would it be a good idea to disable it once I have detected a connection? (Speed should not be an issue, I need less than 100 measurements per second).

Thank you for any explanations or links. This subject should certainly have been discussed before, but I cannot find it anywhere (this is my first posting).

2

Datasheet says device should have 10K output impedance to drive the internal sampling & hold capacitor that makes up part of the ADC.

Internal pullup is fine for reading ~1023 and nothing connected externally.
External pullup/pulldown can not be disconnected - internal can:
pinMode (Dx, INPUT);
digitalWrite (Dx, LOW);
then perform analogReads.

When done reading, turn it back on if desired:
pinMode (Dx, INPUT);
digitalWrite (Dx, HIGH);

3

Thank you for the fast response, this helps a lot.

I had not found these data before. Arduino.cc does not give any detailed spec of the AD converters, but refers to the data sheet of the Atmel ATmega48PA/88PA/168PA/328P. On page 257, in the chapter "23.6.1 Analog Input Circuitry", this document states what you were probably referring to:

"The ADC is optimized for analog signals with an output impedance of approximately 10 k? or less. If such a source is used, the sampling time will be negligible. If a source with higher impedance is used, the sampling time will depend on how long time the source needs to charge the S/H capacitor, which can vary widely. The user is recommended to only use low impedance sources with slowly varying signals, since this minimizes the required charge transfer to the S/H capacitor."

"If conversion accuracy is critical,...: If any ADC [3..0] port pins are used as digital outputs, it is essential that these do not switch while a conversion is in progress."

Regarding the suggestion to turn off the pull-up resistor during measurement, I will probably have to wait a few microseconds after turning off and before measuring, in order not to spoil precision. What do you think?

4

Perhaps. Perhaps not. Try it without and see. If the 20K to 50K pullup affects your signal, turn it off.