ADC output count dispersion

Thread Starter

Challenger

Joined Jan 17, 2017
11
Hi everyone
I would need to get the ENOB for an ADC I am testing.
I know I would need to measure a constant voltage many times to get an histogram for ADC output and get ENOB with the following expression:

upload_2017-3-22_19-10-26.png

The ADC is a Delta Sigma and I need to use the input range Vref/2= 512 mv. The Verf is internal to the ADC = 1.024 mv
The ADC have other Input ranges and they goes from +/- 2.048 mv, +/- 1.024 mv, .........up to Vref/16.
The point of my question is about what how can I get a stable level of voltage to get that Histogram keeping in mind that the noise level of that voltage which drive the ADC should be low enough.
Sorry if it is a silly question, I am learning about mixed signals.
Thank you very much for your time.
 

Attachments

Thread Starter

Challenger

Joined Jan 17, 2017
11
Help us to help you: what is the part number of the ADC you are using?[/QUOTE


Hi OBW0549

I'm currently using the Cypress PSoC 5LP which is a Cortex M. It is system on chip with many analog and digital resources that you can enable according to the necessity.

I have attached for you guys the component 20 bits ADC Delta Sigma data sheet.
 

Attachments

OBW0549

Joined Mar 2, 2015
3,566
The point of my question is about what how can I get a stable level of voltage to get that Histogram keeping in mind that the noise level of that voltage which drive the ADC should be low enough.
I think if you're trying to measure the ADC code distribution to get the RMS noise, you don't need an external stable, low-noise voltage source; you can do that with zero volts differential input. Just connect the +Input and -Input pins together and connect them to the nVref pin, bypassed with a 10μF or larger capacitor as described on page 14 of the datasheet document you posted. I don't see any need to test with a non-zero differential voltage input.
 
Top