[time-nuts] Clocks for Audio gear

[Heinzmann, Stefan (ALC NetworX GmbH)]

Hal Murray wrote:

> As a practical matter, is clock jitter or phase noise from a typical
> low cost crystal and decent board layout a significant problem in audio
> gear?  How hard is it to measure?

The answer depends a lot on the circumstances (as usual).

If you refer to jitter effects on a conversion between analog and digital (either way), and you're interested in whether these can be audible, you will want to measure the effect on the audio signal, and not necessarily the jitter itself. For example, you might want to use a high frequency sine wave close to the upper bandwidth limit as the audio signal, and measure the jitter-related distortions using an FFT analyzer. It may be hard to distinguish jitter-related artifacts from other distortions, but as you are interested in the overall signal fidelity, that's probably what you want anyway. The good thing about that is that you need no extra gear that you don't already have as an audio developer. No expensive phase noise analyzer, for example.

If you refer to jitter effects on a digital transmission, you will be interested in what they do to the bit error probability, or whether you are still conformant to some standard that puts a limit to the allowable jitter. In that case you are more likely to find yourself looking at eye diagrams on an oscilloscope, or perhaps using a bit error analyzer.

There is little use for a proper phase noise analyzer in audio, and R&D labs of most manufacturers I know don't have one.

> Is clock accuracy a practical problem?  How good are people with
> perfect pitch?  It wouldn't surprise me if there are a few that are
> much much better than others, but how good is that relative to 50 PPM
> which I can get in a low cost crystal?

Clock accuracy can be a problem, but not because of pitch perception. Crystal oscillators are easily accurate enough for human perception, even the crappy ones.

Clock accuracy does matter in system applications, when pieces of gear need to lock to each other, or to a house clock. Here, the clock accuracy needs to match the lock range of downstream PLLs, or else you can't lock reliably. The AES11 standard has rules for that.

> Video geeks solved their clock distribution problems by using frame
> buffers.
> Is there a similar trick for audio?  Is there a need for it?

A sample buffer of at least one sample is contained in pretty much every S/P-DIF or AES/EBU receiver chip, so the answer to your question would be yes. Storing a sample is cheap, however, compared to storing a video frame.

Cheers
Stefan