USB Audio Chips

[6AQ5]

Season's greetings!

I am in search for a USB chip for implementing an audio interface for a small amateur radio project. All I need is an good 24-bit ADC-DAC with sampling rates of 48 kHz and, if possible, 96 kHz.

Searching the net, I've found a few possibilities, but all found so far have some constraints.

- The TI PCM290x family is very nice and simple to use, but after building a prototype, I've found that these chips suffer from a sample delay problem between the L & R channels. The ADC passband is also not very clean.

- The Micronas UAC355xB chips are similar to the TI ones (without the sample delay problem). They are used in the iMic USB Audio interface and works reasonably well for my application. The problem is that Micronas does not want to talk to Individuals and doesn't reply to any email inquires. I have found no way to purchase the chip in small quantities.

- C-Media has a few USB Audio chips, but like Micronas, C-Media refuses to talk.

- TI TSUB3200A and TAS1020B. These are very interesting chips and are available in small quantities from multiple sources, but they would need some firmware written.

My question to the group:

1. Do you know of any other solutions?
2. Does anyone with experience in writing firmware to the TUSB3200A would like to share (or sell) code?

Regards,

-- Edson

[Fast1one]

This might be of interest: http://www.twistedpearaudio.com/opus/opus.aspx

[analog_sa]

Originally posted by Fast1one
This might be of interest: http://www.twistedpearaudio.com/opus/opus.aspx

Or maybe not. Did you understand the post above?

[analog_sa]

Originally posted by 6AQ5
TAS1020B. These are very interesting chips and are available in small quantities from multiple sources, but they would need some firmware written.

If you know someone with a M-Audio's Audiophile USB you might be able to use the drivers for the TAS1020B

[analog_sa]

Or shoot John Swenson an email. He has disassembled the drivers and may be willing to help. If you manage to get the TAS1020 working in async mode please let us know

[Fast1one]

Originally posted by analog_sa




Or maybe not. Did you understand the post above?

I guess not, it was pretty late

[00940]

well, one could argue that the pcm2707 used in the opus is not of the pcm290* familly Gordon Rankin once said that the pcm2702 and pcm2902 were not working in the same way as the pcm2707. Jitter measured at the output of the 2702 was ten time worse than the 2706-7 according to him.

[analog_sa]

Don't know if it's just me or the Opus spam is getting completely out of hand. Need new wheels? Why not have a look at Opus?

Jitter measurements? Gordon Rankin? How does this help if one is looking for an ADC?

[00940]

Memories messed up... I would have sworn the pcm2706 had an ADC

Apologies, it hasn't.

NB/ jitter was just an example to show how the TI parts have been redesigned.