diyAudio (
-   Digital Line Level (
-   -   PCM2707, I2S out (

00940 18th September 2004 03:48 PM

PCM2707, I2S out
2 Attachment(s)
Is this implementation correct if I simply want to use the PCM2707 as I2S interface ? The decoupling caps are obviously missing on the schematic. ;)

link to the PCM2707 datasheet :

Kochkurov Maxim 19th September 2004 07:17 AM

You must use 1.5k pullup resistor for D+ pin. If this resistor is absent, USB device cannot be found by system. SUSPEND pin 11 is recommended to connect with pin 10.

00940 19th September 2004 11:14 AM

2 Attachment(s)
thank you :) Pretty obvious mistakes :headbash:

I've updated the schematic below (I can't see how to edit the first post :xeye: ).

Do you have some experience with that chip and external clocks ? The datasheet says that :


An external clock can be supplied from XTI (pin 28 for PCM2704/5, pin 12 for PCM2706/7). If an external clock is supplied, XTO (pin 1 for PCM2704/5, pin 13 for PCM2706/7) must be left open. Because no clock disabling pin is provided, it is not recommended to use the external clock supply. SSPND (pin 27 for PCM2704/5, pin 11 for PCM2706/7) is unable to use clock disabling.
Basically, the question is : is the SPACT system really good enough to provide a decent clock or should I try to implement a XO clock on it ?

Kochkurov Maxim 19th September 2004 03:45 PM

No point to do it. You 'll not improve nothing. SPACT system has own VCO, which determine phase noise level of the output clock. XTAL clock only provide functionality of SPACT system, but it is not reference signal of SPACT. Phase random walks, which depend by jitter of XTAL clock, has enough high frequency spectrum to be eliminated by analog PLL, which followed after SPACT.

00940 19th September 2004 03:47 PM

Thank you :)

niles 20th September 2004 06:08 PM

00940, are you designing a higher end USB dac? I have built your 2702 USB dac and have been impressed with it, but I am interested in pushing the envelope to see what can be done with a 2707> I2S >pcm1798. I am still wating on my 2707 sample to ship. Is it time to start a new thread for a USB super DAC?


00940 20th September 2004 06:28 PM

We're discussing here a DAC based on the PCM2707, followed by the PCM1794. The idea would be to have a dac based on a few boards : a rectifier/preregulator board with LM317-337 (providing +5V, +7V and +/-15V), a digital board with the chips and the final rectification and finally an output board with the output stage. I found an old post in this board, showing how BelCanto used a THS4130 differential opamp for I/V conversion in the DAC2. That could be the basis of the output stage, with an opa627 for balanced to unbalanced conversion and LPF.

niles 20th September 2004 09:41 PM

thanks for the link, I havent checked in on that thread in awhile. It looks like you and the others are already working out some of the same questions that I had independently arrived at. I am not registered at head-fi, so forgive me for crossing the forums for the same topic, but I still have one unanswered question. The 2707 is independently clocked from within using the SPact circuitry. However, the 1794 requires its own master clock, preferably a discrete design such as Guido's. I always was poor at timing diagrams, but it seems to me that the Data clock out of the 2707 on the I2S line needs to be synchronous with the master clock of the 1794. Is this correct, and is some glue logic going to be necessary? Anyone?

BTW, I really think that a discrete open loop I/V with a minimal gain stage is the way to go with these converters. In my mind it makes no sense to use an op-amp when it isnt a lot more difficult to use a discrete circuit. This is especially the case if your design calls for discrete boards for each functional block.


00940 20th September 2004 11:40 PM

have you something in mind for the I/V stage ?

niles 21st September 2004 01:54 AM

yes. Jocko's simplified circuit or a minor modification.

There are some follow up posts that include schematics for some improved versions. I am a fan of minimal circuits. Some of the other participants may wish for the convenience of op-amp implemenations, and that may be the best part of your proposal to break up the function blocks. That way, we can roll our own output stage based on own needs and design philosophies.


All times are GMT. The time now is 08:46 AM.

Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 18.75%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Copyright ©1999-2018 diyAudio