DIYHiFi.org

Got an s/pdif signol from it

still don't know why it won't output audio...

now i'm trying to hook it u to an AD1843..

Are you sure that you want use AD1843 is Serial-Port 16-Bit SoundComm Codec? 80PQFP or 100TQFP?

Ah damn, typo!

1853 (actually, i incidentally ordered 1852, which i try to connect now, but i will use the 1853)

Is there a post for ugliest DAC? sure win here;)

As i understand, i can directly connect these chips? no need for reciever chip.

Bas

Yes, you can directly connect them.

Couple pictures of my dac
1) USB to I2S converter

2) AD1852 DAC board
[/img]

now this is really DIY stuff :)
are you happy with the sound? how does it compare to what you had previously?

Looks great! How'd you do the PCB? It must've been the toner/transfer method, right?

Looks fantastic.


How does it sound? :)

Yes, this is pure DIY project. Everything I did myself - putting datasheets together, PCB layout, made PCB using toner/transfer method (laser-iron ), etching and soldering.
Its more stable than my previous project with PCM2902, no sudden disconnect from USB port, no RFI interferencies, very stable.
Sound is better than PCM2902 with internal DAC, but because I dont have made proper PSU jet (waiting for transformers), I cant give final conclusion about sound.

Next project will be PCM2707 with AD1851.
Have anybody used I2S to serial converter (Analog Devices Application Note 207)?
Are 74HC00 nand gate and 74HC74 flip-flop suitable for this?

Hello All,

When in self-powered mode, the USB 5V must be connected to HOST and the USB Ground must connect to the external 3.3V PSU ground.

First, any idea what the HOST pin leads to inside the IC? In other words, would it be beneficial to remove noise on the USB 5V before feeding it to HOST?

Second, has anyone looked into removing noise on the USB Ground pin before connecting to the 3.3V ground? I've seen various suggestions to add a ferrite between USB ground and PSU/PCM2706 ground... is this advisable?

Thanks,
Bryan

Of course it would. This is where you want the ferrite bead. Use it to filter the 5V coming in off the bus before you feed it to the chip. From there, the internal linear regulators do the rest.

In a future design I plan to run the filtered 5V through a discrete linear regulator, a lower noise design than what is probably on chip. This can then feed the analog sections of the chip separately. Note, to make this work, you have to back-bias the internal regs by operating at higher than 3.3V, say 3.6V or so.

jh

_________________
Jim Hagerman

I think he is asking about putting a bead on the ground.

I wouldn't. We know what Charlie thinks about beads to begin with, so ignore him. [joke]

Jocko

My understanding is that when in self-powered mode, the IC runs off an external 3.3V referenced from the external ground (which is then connected to USB ground). The USB 5V is run to HOST to indicate that it is in self-powered mode (instead of being run to VDD when in bus-powered mode, where HOST would be held low). As such, the USB 5V would not be used for powering the chip in self-powered mode (if I understand correctly).

Anyway, as for filtering in general, my thinking is that it would only help (or at least not hurt) to remove HF garbage from the dirty USB 5V and ground lines. I'm mainly asking because there are members here that have delved significantly beyond the datasheet, both in discussions with TI and through experimenting with test equipment.

As for HOST, if someone can tell me that noise on the HOST pin will not propagate to other circuitry in the chip in a harmful way, then I will drop this. If not, then I may experiment in my first 2706 prototype here soon...

As for ground, this is more of an issue since the USB ground is directly connected to the external PSU ground. It seems that this would contaminate what an external regulator worked so hard to regulate...

Thanks,
Bryan

Hi all,

I've been lurking here for quite a while & based on the excellent info contained herein I have jumped into USB diy DAC.

I took the easy route & bought a kit- USB-I2S non-os DAC (won't mention name here). I will report on build if there is interest.

Anyway after a bit of investigation & study of PCM2706 datasheet, I have some comments - Bryman79 is almost correct - when IC is self-powered USB 5V is run to Host pin 3 to indicate attach/detach state of USB (self powered state is actually selected by pin 16 PSEL set low).

I wonder if anybody has tested the difference between the pseudo self powered mode mentioned by Hagerman (5V supplied from external reg thru Vbus, USB 5V to Vbus & full self powered external 3.3V to Vbus(regulation is done externally in this case))

I also wonder if anyone has tried to minimise power connections to IC - I am not interested in headphone output, Spdif output, so I was thinking of not powering the relevant pins (26-31) - does this have any significant sonic benefit?

Enough questions
John

My "portable" DAC with proper PSU. Sounds very nice!

pupsik46,

When you say "proper" PSU do you mean PSEL to ground & regulated 5V supply to Vbus & regulated 3.3V supply to Vccp, Vdd, Vccl, Vccr?

If so how much of an improvement over just supplying regulated 5V to Vbus & PSEL set high?

Also, what method did you use to ensure USB cable is connected before powering on because D+ needs to be connected before power (else chip is damaged)?

Just trying to evaluate payoff versus effort required before embarking on this course

John