0
\$\begingroup\$

Does the output of a DAC have quantization noise when outputting a DC voltage value? It seems to me that it will not as long as the DC value does not fall between two output levels of the DAC.

All the information I have been able to find regarding quantization noise of a DAC discusses it in the context of outputting a sawtooth wave or sine wave, and then it ends up being proportional to 1 LSB spread uniformly over the bandwidth of the DAC.

\$\endgroup\$
2
\$\begingroup\$

For a DC output, a DAC will not have quantization noise, but it will have quantization error.

EDIT - While noise is technically any unwanted or erroneous component of a signal, in practice it is used to refer to the AC component of such error. For instance, noise is normally characterized as having a spectral density measured in $$\frac{quantity}{sqrt{Hz}}$$ or volts/amps/power/etc divided by the square root of the measurement bandwidth. DC, however, has zero bandwidth, so any such measurement would give an infinite value for DC "noise". As a result, static errors are usually referred simply as errors, or DC errors if context requires. In the case of a DAC with a DC output, the error caused by quantization is called quantization error.

\$\endgroup\$
  • \$\begingroup\$ I am not sure I understand the distinction between the two in this context. Could you please elaborate on your response a bit? \$\endgroup\$ – cstraats Jul 30 '15 at 19:36
  • \$\begingroup\$ @cstraats - OK, I've edited. \$\endgroup\$ – WhatRoughBeast Jul 30 '15 at 19:58
  • \$\begingroup\$ Ok, I think I understand now. My interpretation is that the quantization error really only translates into a noise with some spectral density if the output of the DAC is changing (i.e. a voltage ramp or sine wave output), otherwise it is just a DC error. Is that about right? \$\endgroup\$ – cstraats Jul 30 '15 at 20:30
  • \$\begingroup\$ @cstraats -Exactly. \$\endgroup\$ – WhatRoughBeast Jul 30 '15 at 20:54
  • \$\begingroup\$ @cstraats - Well, let me backtrack and explain a little. I was referring to broadband noise as having a spectral density. This does not apply to single-frequency noise. For instance, if a circuit is picking up a pure 60 Hz tone, we would talk about "x volts of 60 Hz noise", rather than referring to spectral density, since a pure tone also has zero bandwidth. But DC is still referred to as error or sometimes offset. \$\endgroup\$ – WhatRoughBeast Jul 30 '15 at 21:32

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.