diyAudio

diyAudio (http://www.diyaudio.com/forums/)
-   Digital Line Level (http://www.diyaudio.com/forums/digital-line-level/)
-   -   Hacked TDA1541 CD Player USB DAC Idea (http://www.diyaudio.com/forums/digital-line-level/228200-hacked-tda1541-cd-player-usb-dac-idea.html)

themartianboy 19th January 2013 08:29 PM

Hacked TDA1541 CD Player USB DAC Idea
 
So I've got this idea to use a USB to I2S board something like this http://www.ebay.ca/itm/110906343285?...84.m1423.l2649 and patch it in to pins 1,2, and 3 of the TDA1541A chip on my old CD player, completely abandoning the actual CD playing function and eliminating any oversampling/filtering (ie. SAA7220).

Would this work? If so, would I need any attenuation or buffering between the USB/I2S board and the TDA1541A chip?

josiphal 20th January 2013 11:02 AM

Quote:

Originally Posted by themartianboy (Post 3333394)
So I've got this idea to use a USB to I2S board something like this http://www.ebay.ca/itm/110906343285?...84.m1423.l2649 and patch it in to pins 1,2, and 3 of the TDA1541A chip on my old CD player, completely abandoning the actual CD playing function and eliminating any oversampling/filtering (ie. SAA7220).

Would this work? If so, would I need any attenuation or buffering between the USB/I2S board and the TDA1541A chip?

Board have I2S out, and this should work with TDA1541A.

It will be better option, to carefully take out TDA1541A from CD board without damaging chip/board, and make new DAC board based on Pedja Rogic DAC. Also, if 16/44.1 is enough, PCM2707 can be used for DIY PC USB/I2S bridge.

themartianboy 21st January 2013 06:31 PM

Thanks for your response! I am trying to do with this with very little money so I don't think I can afford to make a new DAC board. Plus, I am having difficulty finding a schematic of Pedja Rogic's design. He now sells complete DACs and doesn't sell DIY parts or kits.

I am curious about why you recommended the PCM2707. I thought I read that it can have jitter issues, but maybe I'm wrong?

josha 21st January 2013 08:32 PM

If you use as short a cable as possible then you shouldn't need any buffering / attentuators. However using a cat5 cable with one of each pair grounded at both ends helps as I understand it.

Your idea will work though, I've done exactly that myself.

There are better solutions than the 2707 like the one you linked but the 2706/2707 is easy to diy (especially if you can make your own pcb's) and they're cheap - I think you can still get free 'samples' if you know where to look(!). One option is this - just leave out all the coax/optical bits and you've got an easy to make (and cheap) usb to i2s converter.

themartianboy 22nd January 2013 03:01 AM

Thanks; that's more food for thought. Speaking of the 2707, this looks like a really nice USB kit: http://www.dddac.de/pics/dddac1543mk...iver32-003.jpg (receiver chip soldered on the underside) but at 39 Euros plus 29 for XO clock plus shipping it's getting a bit pricey...

josha 22nd January 2013 07:58 AM

For that money your first link would make more sense I think - or even something based on a CM6631. These would give you more options for trying different DACs in the future as well.

Adding low jitter / high quality clocks to the 2706/07 has always puzzled me a bit as that chip reclocks internally so I'm not sure how much of an impact the quality of the external clock makes - plenty of kit makers do it though so they probably know something I don't!

In my humble opinion the 2706/07 only really makes sense if you're scratch building these days, it is smd but I've managed to solder a couple of them using a standard iron, they're pretty robust.

themartianboy 22nd January 2013 06:45 PM

I wasted a bunch of time studying the datasheet and can't help but draw the same conclusion, since it recovers the clock from the incoming data. Maybe an upgraded clock makes that recovery process less prone to errors? Nevertheless the manufacturer recommends against adding an external clock anyway...

I forgot to mention that I also need compatibility with MacOS 10.4. TI officially confirms this for the 2706/7, which is nice.

themartianboy 23rd March 2013 05:14 AM

So I ended up buying the TE7022 board.

I hooked up the I2S signal to the I2S input of my TDA1541A CD player's filter chip (which happens to be a Yamaha YM3414 instead of the usual SAA7220).

The audio comes through but there is way too much gain in the signal and there is peak distortion even at the lowest source (computer) gain levels.

How do I fix this? Attenuation with resistors between the USB board and the filter chip? If that's all I need to do, what's a good resistor value to start with?

themartianboy 25th March 2013 09:15 PM

3 Attachment(s)
Can anyone help me with this? I've attached a picture of the USB converter board, a copy of the DAC/output stage of my CD player, and the datasheet for the YM3414 receiver chip for reference.

I disconnected the DAC/output stage from the Main PCB and left XTAL dangling since it's not needed by the USB board.

From the USB board I connected DATA to SDI (Pin 7) on the YM3414, LRCK to SDSYS (Pin 6), and SCLK to BCI (Pin 5). I connected GND to a ground point on the CD player.

I'm testing with AIFF files with 16/44.1 sampling.

Help?

AudioLapDance 25th March 2013 10:42 PM

Good job so far!

There is talk of attenuating the 1541 I2S levels but this has to do with cleaning up a digital signal and doesn't lead to 'overdriven' analog output. So don't worry about that right now.

It sound like you might be sending the 1541 24 bits worth of data and it can only accept 16 bits so the resulting waveform is "too loud"

Are you able to change the bit depth (16, 24 bit) and rate (44, 48, 96, 192 ...)?

Cheers,
Jeff


All times are GMT. The time now is 08:35 PM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio


Content Relevant URLs by vBSEO 3.3.2