power consumption of an ADC per sample

Status
Not open for further replies.
P

pankaj jha

Full Member level 3
Full Member level 3
Joined
Apr 16, 2010
Messages
170
Helped
1
Reputation
2
Reaction score
1
Trophy points
1,298
Location
jaipur
Activity points
2,457
Hello everyone !!!

How to calculate/estimate the power conversion of an ADC per sample in terms of the parameters like sampling frequency, etc. ???????????

Necessary papers/reference would be helpful.
 

In my understanding, the "power per sample" does not have meaning. I think that you are interested in the "energy per sample" of the converter. It is the energy consummed for obtaining a sample, and is simply the product of power consumption by the sampling period, i.e.:

Es [Joule/sample] = P [Watts] * Ts [seconds/sample] = P [Watts] / Fs [samples/seconds]

Less is better, but other parameters (like accuracy) are not taken into account.
In ordet to compare converters with different accuracies, the effective nomber of bits should be taken into account [bits/sample] giving "energy per bit" [Joule/bit].

Regards

Z
 

    V

    Points: 2
    Helpful Answer Positive Rating

Thanks a ton..ur reply was quite informative...
Can u give me the formula for calculating energy per bit...

- - - Updated - - -

erikl said:
It's called figure of merit (FOM). See this thesis paper: View attachment 103238 .
Click to expand...

Thanks erikl...
 

zorro said:
Es [Joule/sample] = P [Watts] * Ts [seconds/sample] = P [Watts] / Fs [samples/seconds]
Click to expand...

pankaj jha said:
Thanks a ton..ur reply was quite informative...
Can u give me the formula for calculating energy per bit...
Click to expand...

Quite obvious, after the above:

Eb [Joule/bit] = P [Watts] / ( Fs [samples/seconds] * Nb [bits/sample] )

Z
 

pankaj jha said:
Can u give me the formula for calculating energy per bit...
Click to expand...

The FOM definition - which is often used for ADC efficiency comparison - you can find on p. 16 (p. 24 of the PDF), equ. (1.9) of the above linked thesis paper. See here a snippet from an ADC paper:
 

Agree. It's fair to evaluate energy per conversion level instead of energy per conversion bit as a significant FOM.
From the point of view of accuracy, or signal to quantization noise ratio (after some processing in "normal" conditions), doubling the sampling frequency is nearly equivalent to double the number of significant levels (to increase one bit in accuracy).
It that way,

El [Joule/effective_level] = P [Watts] / (Fs [samples/seconds] * (2^ENOB) [effective_levels/sample]

There are other possible figures attending to different criteria (see for example https://converterpassion.wordpress.com/generic-adc-fom-classes/ )

Z
 

Status
Not open for further replies.

Similar threads