# AD7634 - Histograms near the extremity of the scale

Hi,

I'm currently working with the AD7634 and I have a question about the thickness of the histogram when working at the extremity of the scale.

Looking at the figure 7 of the datasheet (page 12), attached to this message, it is possible to see that the histogram shows a thickness of 7 (min: 0x20000 and max: 0x20006) and a standard deviation of 0.80. If we look at the figure 10 we can also see a thickness of 7 (min: 0x1FFFE and max: 0x20004), but a standard deviation of 0.75.

But for both histograms the measure was acquired at the center of the scale, i.e., near the code 0x20000 and I didn't find in the datasheet further details about the results when working outside of the center of the scale.

This way, I'd like to know what I should expect in terms of thickness of the histogram and standard deviation, when working near the codes 0x00000 and 0x3FFFF.

Another doubt: the description of both figures 7 and 10 says "Histogram of 261,120 Conversions of a DC Input...", but when I sum all of the counts of the histogram the result is 131,072. This way, I would like to know what this "261,120 Convertions" means.

Best regards,

Rafael Ito

Parents
• Hi Rafael Ito,

Each unique digital code corresponds to a small range of analog input voltages. This range is 1 LSB wide (the “code width”) and is centered around the “code center.” All input voltages resolve to the digital code of the nearest code center. The difference between the analog input voltage and the corresponding voltage of the nearest code center is the ADC quantization error.

Any spread in the histogram around the center code bin is caused by noise in the ADC. This test can also be used to determine the offset of the ADC as in the histogram test.

As the name suggests, histogram of a DC input at code transition refers to the uncertainty in the analog input voltage at which the ADC output code makes its transition. It’s also indication of transition noise.

Refer to these articles that you  should help you better understand this.

http://www.analog.com/media/en/training-seminars/tutorials/MT-004.pdf

If you are designing-in the AD7634 for your application, I’d recommend that you consider the new AD4003, which combines the ease of use features with high performance and reduces overall system complexity, lowers power and shrinks size.

regards,

Lloben

• Hi Rafael Ito,

Each unique digital code corresponds to a small range of analog input voltages. This range is 1 LSB wide (the “code width”) and is centered around the “code center.” All input voltages resolve to the digital code of the nearest code center. The difference between the analog input voltage and the corresponding voltage of the nearest code center is the ADC quantization error.

Any spread in the histogram around the center code bin is caused by noise in the ADC. This test can also be used to determine the offset of the ADC as in the histogram test.

As the name suggests, histogram of a DC input at code transition refers to the uncertainty in the analog input voltage at which the ADC output code makes its transition. It’s also indication of transition noise.

Refer to these articles that you  should help you better understand this.

http://www.analog.com/media/en/training-seminars/tutorials/MT-004.pdf