• State
  • Locked Locked
  • Replies 2 replies
  • Subscribers 31 subscribers
  • Views 1560 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

PCM1792A volume control - does it affect THD+N performance? Better use PGA2320?

Sebastian Engl
Prodigy 10 points
Other Parts Discussed in Thread: PCM1792A, PGA2310, PGA2320

Hello,

Please help me to get this right: I have to decide whether to use the PCM1792A's integrated volume control / attenuator or to use the 1792 together with a PGA2310/2320 volume control. Goal is the highest possible sound quality, even with low volume levels (attenuation, say -40dB). The circuit will feed right into a power amplifier.

As the PCM1792A's volume control works in the digital domain, I'm not sure if it will affect or even degrade THD performance. If I set the attenuator to -40dB for example, the wanted signal will be attenuated this amout. But the distortions won't be attenuated, as they originate in the DAC stage (right?). Therefore, the output signal's THD performance degrades from the specified -108dB (0,0004% @Fs 44,1kHz) to -68dB.

Opposed to that, when using a external attenuator like the PGA2320 set to -40dB, it will attenuate not only the wanted signal, but also the distortions, so the THD performance will stay unaffected. Of course, the PGA2320's own THD will dominate in this configuration. Unfortunately, it is only specified at 0dB Gain (with -108dB), there is no data for THD vs. Gain or THD vs. output amplitude. But I expect THD to decrease with higher attenuation.

So the PGA2320 solution will be far superior to the PCM1792A's integrated attenuator, won't it? Am I missing something here?

Thank you very much for any suggestions.

finalstage

  • 0
    Prodigy 70 points

    I have a non-audio application for this part and desperately need noise and bandwidth vs. attenuation. Since it is a nonaudio application I would like to see the noise in either volts, spectral density, or a plot but without any weighting. Also, what do the offset and gain errors do as the attenuator is traversed?

  • 0
    Prodigy 70 points

    Hello finalstage,

    You are describing a scenario which is actively debated amongst the audio gurus. Either one is an acceptable solution for high fidelity. The end result hinges strongly on design implementation. Let me elaborate a little.

    If you use the PCM1792A's digital volume control, the DAC simply multiplies the data stream by a coefficient of less than one, i.e. you lose bits. With the PGA2320 working in the analog domain, you don't lose bits but you add distortion and noise not just from the PGA2320 but from the input buffer and output buffer (a must when using the PGA2320). Your board is now more complicated and you now have at another mixed signal device that you'll have to carefully plan your ground planes and power delivery/regulation for.

    My personal opinion (I like KISS) is to utilize the PCM1792A volume control with a 24-bit I2S signal. This way, you can drop bits but still maintain CD-quality resolution.This is how Bel Canto did their DAC3.5VB and it is a highly regarded product (info obtained from reviews). Benchmark Media and Centrance went the other way and use a plain potentiometer in the analog domain. These are all highly regarded products so you can see that it all boils down to implementation.

    Good luck.