ADC Sampling rate for DC measurement

[hioyo]

Do we need to consider ADC sampling rate when we need to measure a DC value.

Let us say I want to measure 0 to 5V using an ADC

 

[barry]

if you want to do averaging, then yes.

we need more information.

 

[crutschow]

The sampling rate will, at a minimum, depend upon how fast you want an new reading when the 0-5V DC voltage changes value.
Digital voltmeters usually take a few readings a second.

 

[hioyo]

if you want to do averaging, then yes.

we need more information.

General question came to my mind when earning about data converters.
Regarding averaging you mean if sampling rate is high we can do better averaging.Please correct me if I am wrong

 

[KlausST]

Hi,

I´m a friend of developing: requirement --> solution
(not: first looking for a solution without having a requirement or problem. )

so if you have a requirement, please let us know:
* like: range, accuracy, precision, timing, .... whatever you can give us in values with units.

Klaus

 

[barry]

General question came to my mind when earning about data converters.
Regarding averaging you mean if sampling rate is high we can do better averaging.Please correct me if I am wrong

Totally agreeing with Klaus; without any requirements from you, we can throw out lots information about ADCs, sampling, etc. all day long.

How do you define "better" averaging?

Just understand this fundamental: if you increase the sampling rate and average, you'll get better noise performance and higher resolution (effective-number-of-bits). There's lots and lots of information about this if you search the net.

 

[andre_luis]

It sounds like you don't even know how to define your requirements, so here are some general tips:
1 ) Whenever possible, consider performing an oversampling taking the moving average of values, since even a variable signal has a 'median' DC value.
2 ) Choose a sampling rate that is not an exact multiple of the public network frequency or any other one whose presence is most predominant in your circuit, otherwise you will be running the risk of contaminating your sample with values synchronous to these fundamental frequencies.

 

[danadakk]

If you want to measure a DC signal, one that is not changing, sample rate
is of no consequence, sample once a day or once every big bang of the universe
recreation, does not matter. Unless you want it read with a specific periodicity
even though, by your definition its unchanging, so read once and never again.

AC another matter, so its changing. Question becomes what are you looking
for, peak, average, minimum, rise and fall time, harmonic content eg. you are
doing signal processing looking for other waveform characteristics......

# samples versus averaging :

Sample Size - Measuring a Continuous Variable

Sample Size - Measuring a Continuous Variable, Planning, Analysis, Continuous, Variable

accendoreliability.com

Regards, Dana.

 

[danadakk]

Apnotes on embedded averaging, attached.


Regards, Dana.