PCM2706 not working

[Eddy Dean]

Hello everyone,

I have been trying to build a PCM2706 soundcard with I2S output to an TDA1543, and this morning I etched the PCB and built it. It however does not work, when I plug it in Windows sees it as an unknown device.

schematic (PDF)

I measured all connections and they seem to be fine, so the problem must be in the schematics or a broken component.

The first time I hooked it up there was a small solder bridge from XTO to ground, could that have killed the PCM? Of course I removed that bridge, but it still didn't work.

[Guido Tent]

Originally posted by Eddy Dean
Hello everyone,

I have been trying to build a PCM2706 soundcard with I2S output to an TDA1543, and this morning I etched the PCB and built it. It however does not work, when I plug it in Windows sees it as an unknown device.

schematic (PDF)

I measured all connections and they seem to be fine, so the problem must be in the schematics or a broken component.

The first time I hooked it up there was a small solder bridge from XTO to ground, could that have killed the PCM? Of course I removed that bridge, but it still didn't work.

Eddy

check if the oscillator runs, you can measure the dc voltage at xto, this normally is half of the power supply voltage. If it is zero, the oscillator doesn't run at all.

succes

Guido

[Eddy Dean]

Thanks for your reply

XTO is 3,23V, and XTI is 2,94V (to GND). I unfortunately do not have an oscilloscope, so there is not much I can measure.

There is no AC voltage at all on the crystal.

[dsavitsk]

Schematic is a little messy, but is HOST connected to anything? It needs to be pulled up or down.

[error401]

I would clean up the schematic so that it's readable and then verify the PCB layout. There are a number of spots where I'm uncertain there's a junction, for example XTO connecting to the crystal and DO connecting to U1.

If there is no AC voltage on the XTI or XTO pins, the oscillator is not running. I suspect the above mentioned missing junction is the problem.

[Eddy Dean]

Thank you for your comments. I cleaned up the schematics a bit (but honestly, I'm terrible at drawing schematics...)

All the connections that you were not sure of are made, so these are (unfortunately) not the problem.

I attached the new schematics and the PCB (both in the same PDF).

When I look at the schematics of the AW-D3 I see that he connected everything that needs to be pulled up to Vdd, while I connected most of it to Vccp, because I thought that the block diagram (pag 8) in the datasheet showed that all these pins are essentially the same. Is that a mistake?

If that is actually a mistake, could someone give me tips on how to desolder a TQFP32? The previous time I did that I used a scalpel to carefully bend all legs up, but the IC never really looked good afterwards.

pdf

[Guido Tent]

Originally posted by Eddy Dean
Thank you for your comments. I cleaned up the schematics a bit (but honestly, I'm terrible at drawing schematics...)

All the connections that you were not sure of are made, so these are (unfortunately) not the problem.

I attached the new schematics and the PCB (both in the same PDF).

When I look at the schematics of the AW-D3 I see that he connected everything that needs to be pulled up to Vdd, while I connected most of it to Vccp, because I thought that the block diagram (pag 8) in the datasheet showed that all these pins are essentially the same. Is that a mistake?

If that is actually a mistake, could someone give me tips on how to desolder a TQFP32? The previous time I did that I used a scalpel to carefully bend all legs up, but the IC never really looked good afterwards.

pdf

Hi Eddy

I suspect your oscillator doesn't run as both input and output are near power supply voltage, whilst the output should be low when the input is high and vice versa

succes

Guido

[Eddy Dean]

I'm afraid so, but what to do about it? Is it possible that my crystal is broken, or is something about my PCB/schematics wrong?

I'm terribly sorry if these are "stupid" questions. If needed I'll test the crystal by hooking it up to an microcontroller and see if it works. I don't know an other way to test it...

[Guido Tent]

Originally posted by Eddy Dean
I'm afraid so, but what to do about it? Is it possible that my crystal is broken, or is something about my PCB/schematics wrong?

I'm terribly sorry if these are "stupid" questions. If needed I'll test the crystal by hooking it up to an microcontroller and see if it works. I don't know an other way to test it...

what you can do to check the oscillator in the 2706 is:

remove crystal

connect multimeter to crystal output

take resistor (10k-ohm) and connect crystal input to ground resp power supply (using this resistor in series)

The output should go from high to low voltage

best

Guido

[Eddy Dean]

Pulling XTI low makes XTO low aswell, and when XTI is not pulled low they're both very close to Vcc (perhaps because XTI is floating?).

I'll connect the crystal to an uC now to see if the crystal actually works. I got it all on the breadboard already anyway.

edit: The crystal works... I'm not sure if I should be happy about that...