ADC Resolution & Accuracy

[Majid_Vatan_Parast]

Hello everybody
I have read a sentence but can not understand it . Sentence is this " A 2-bit ADC with 10 bit accuracy". I do not understand , what does it mean ?
Please help me :sad: :sad:
Thanks

 

[dreamlover]

At first i would like to define you what is accuracy in ADC?
accuracy of the ADC determines how close the actual digital output is to the theoretically expected digital output.
Sentence "A 2-bit ADC with 10 bit accuracy" means an 2-bit ADC produces 2*2, or 4 output codes where 10 bit accuracy means 1024 code close to theoretical value.
but i think the sentence might be "24-bit ADC with 10 bit accuracy" because accuracy must be always lower then resolution.
lets take an example: An ADC can provide resolutions as high as 24 bits. A given 24-bit ADC converter may only provide 10 bits of accuracy. In this case, the 14 LSB’s represent random noise produced in the converter. However, these noise bits are used with digital filter algorithms to increase the useful measurement resolution at the expense of a lower sampling bandwidth.

 

[KlausST]

Hi,

I have read a sentence

It's always a good idea to give a link to this source of information, or upload the document...because sometimes the meaning of the sentence becomes obvious if one knows the context.

Maybe "A 2-bit ADC with 10 bit accuracy" means that the threshold from one step to the other is very precise. In this case 10 bits may mean 0.1%.

But I agree that it is likely that there is a typo in the sentence. Maybe 12bit ADC with 10 bits accuracy, maybe 24 bit ADC with 10 bits accuracy...

Klaus

 

[schmitt trigger]

Hi,


Maybe 12bit ADC with 10 bits accuracy,...

Klaus

This scenario is the one that makes the most sense. Having the least LSB 2 bits in an ADC wobble with noise is fairly common, but as others have noted, can be overcome by simple algorithms.
The simplest is multiple sampling and averaging.

 

[KlausST]

Hi,

Even with noise-free digital signal the phrase "12bit ADC with 10 bits accuracy,..." can be true....with an INL of +/-2LSB.

Klaus

 

[erikl]

"A 2-bit ADC with 10 bit accuracy"

maybe 24 bit ADC with 10 bits accuracy

There may be more possibilities, depending on their architecture and its nomenclature: some ADCs with high resolution (16 ... 24bit) are called 1-bit ADC, sometimes also 2-bit ADC.

Also, sometimes redundant signed digit (RSD) ADCs are designated as 1.5-bit or 2-bit ADC, because of their 2-bit output per resolution bit.

 

[FvM]

I won't suspect a typo in the first order. A 2-bit ADC with this huge accuracy overhead can make sense as building block of a cascaded high speed design.

Reading the respective quotation thoroughly will clarify...

 

[Majid_Vatan_Parast]

Re: ADC Resolution & Accuracy

Hello

Please have a look on this page ,this part " Component Accuracy "

https://www.maximintegrated.com/en/app-notes/index.mvp/id/1023#collapse3

Component Accuracy
Digital error correction does not correct gain or linearity errors in the individual DAC and gain amplifiers. The front-end S&Hs and DAC actually need about 12-bit accuracy, whereas the components in subsequent stages require less accuracy (for example, 10-bit accuracy for Stage 2, 8-bit for Stage 3, and so forth).

Yes this ADC is a building block of a pipeline ADC .

- - - Updated - - -

In this schematic, the analog input, VIN, is first sampled and held steady by a sample-and-hold (S&H), while the flash ADC in stage one quantizes it to three bits. The 3-bit output is then fed to a 3-bit DAC (accurate to about 12 bits), and the analog output is subtracted from the input.

Digital error correction does not correct gain or linearity errors in the individual DAC and gain amplifiers. The front-end S&Hs and DAC actually need about 12-bit accuracy, whereas the components in subsequent stages require less accuracy (for example, 10-bit accuracy for Stage 2, 8-bit for Stage 3, and so forth).