Effective number of bits of 14-bit ADC

[newbie_hs]

I have a 14-bit ADC. However, looking at the datasheet (see table 2 on page 5), the effective number of bits (ENOB) is always less than 12 bits.

Why is my the DAC claiming to be a 14-bit ADC when it only has 12-bit accuracy? What is the point of having two extra bits if they are meaningless?

 

[danadakk]

I always tried to stay with the OP´s question, his ADC and it´s specifications.
Are you still discussing about the thread topic .. or is this some off topic discussion now?
Or in other words: Is this something you would recommend the OP to do?

I wanted to make sure OP understood inaccuracy of dither input to signal path just analog,
in earlier post, also has solution both at analog and digital interface, not just analog :

Dither discussion one more possibility to improves OPs concern over ENOB.

On topic.....

 

[barry]

Hi,

The document says:
...the theoretical SNR of an ideal N-bit ADC: SNR = 6.02N + 1.76 dB....

I fully agree.

But we don´t talk about ideal ADC. It´s a real ADC with a real datasheet, with real specifications.

All the above formulae are correct.

Klaus

Klaus, I see N (number of bits) as a factor in SNR in your above equation, which is exactly the point I was trying to get across, that you seem unwilling to accept, that number of bits affects SINAD. As I said, you can make an ADC of any number of bits worse, but you can't make it better than its theoretical limit.

Look at any table of ADCs from any manufacturer; the higher resolution devices have higher ENOB. That's all I'm trying to say.

 

[KlausST]

Hi,

in SNR in your above equation

It wasn´t me who brought up this formula.
It is the "theoretical limit" of an ideal ADC, caused by the quantisation.
But the OP talks about a real ADC. And here the SNR is different. The SNR is a real value, not a theoretical one.

https://www.analog.com/media/en/training-seminars/tutorials/MT-001.pdf

it´s Introduction says:

You don't have to deal with ADCs or DACs for long before running across this often quoted
formula for the theoretical signal-to-noise ratio (SNR) of a converter. Rather than blindly
accepting it on face value, a fundamental knowledge of its origin is important, because the
formula encompasses some subtleties which if not understood can lead to significant
misinterpretation of both data sheet specifications and converter performance. Remember that
this formula represents the theoretical performance of a perfect N-bit ADC. You can compare the
actual ADC SNR with the theoretical SNR and get an idea of how the ADC stacks up.

This absolutely supports what I´m saying. Don´t misinterprete the formula!
Noboby denies that an ideal 14 bit ADC has a better SNR than an ideal 12 bit ADC. But that´s simply not the OP´s question here.

****

you seem unwilling to accept, that number of bits affects SINAD

Oh, no. I surely did not. Instead I clearly showed it has an influence!
Apparently you didn´t read my last post.
I not only showed the math that the number of bits influence the SINAD .. but also how much.
And the point is "how much".
From my previous post: (3 different values for 3 different bit numbers, but you say I refuse to accept the influence...Why? How?)
--> for 14 bits: 1468 uV pp
--> for 13 bits: 1493 uV pp
--> for 12 bits: 1589 uV pp
Now move back all numbers to RMS:
Signal = 2V pp = 0.707V RMS (pure, single sine, full scale)
Noise14 = 245uV RMS (noise, a bunch of frequencies, used same factor of 6 as above. depends on frequency distribution)
Noise13 = 249uV RMS
Noise12 = 265uV RMS

.. and calculate SINAD and ENOB:
14 bit: SINAD 69.20dB = ENOB 11.20 bits
13 bit: SINAD 69.06dB = ENOB 11.18 bits
12 bit: SINAD 68.52dB = ENOB 11.09 bits

So using the OP´s ADC
* omitting 2 LSBs (using 12 bits only) gives an ENOB of 11.09
* omitting 1 LSBs (using 13 bits only) gives an ENOB of 11.18
* omitting 0 LSBs (using all 14 bits) gives an ENOB of 11.20
It shows there is a (mathematical) benefit, but no practical benefit, since it most probably is below the production variance.

And all math above is already done in your favour and to lower my argument. How more can I express that I understand your argument? But sadly you did not gave any math or numbers.
* Calculating with peak-to-peak values would improve my argument
* if we take external noise (which surely exists: in signal, in power supply, in VRef..) into account would improve my argument.

Sorry, but if you don´t accept math, if you can´t say where my math is wrong (it´s surely possible), just bring up blury words, wrong accusations ... it´s impossible to discuss a technical problem.

Klaus

--- Updated Jul 20, 2023 ---

Look at any table of ADCs from any manufacturer; the higher resolution devices have higher ENOB. That's all I'm trying to say.

It´s simply not true: both ADCs from AD
* ADS8380: 18 bit, SINAD typ 91dB, ENOB 14.82 bits
* AD677: 16 bit, SINAD typ 92 dB, ENOB 14.99 bits
Also this time I´ve done the research in your favour and compared only from same vendor.
It becomes even worse if you compare different vendors, different speed, different technologies...

If you say "most, mostly" ... then I fully agree.
But as soon as you say "any, any, higher" then it is simply wrong.

Klaus

 

[barry]

Klaus, my friend, the OP’s original question was ”Why is my the DAC claiming to be a 14-bit ADC when it only has 12-bit accuracy? What is the point of having two extra bits”. (I think OP meant ADC, but, whatever).

I was attempting to explain why those are not “extra bits”. And the basic point is this: “The more bits you have, the better the ENOB will USUALLY be”. And, “You cant have a better ENOB than your number of bits; if you need an ENOB of N, you’re going to need an ADC with more than N bits.”

Somehow this discussion veered off (as they usually do) into things like dither and what-not.

 

[danadakk]

Somehow this discussion veered off (as they usually do) into things like dither and what-not.

Oh my feelings are so hurt now. And I thought object of discussion was to show
OP methods to improve what he has or may get/wind up with. PM me and I will
send you my address where you can ship tear wipes.

Regards, Dana.