I am building a DAC with DIR9001 and PCM1794.
So far getting a reasonable signal out of DIR9001. It is configured with FMT0 and FMT1 high, i.e. 24-bit, MSB-first, I2S. PSCK0 and PSCK1 low, i.e. SCK0 128fs. Verified on scope that SCK is a bit over 6 MHz with 48kHz fs. Source is an old SB Live.
PCM1794 I have tried FMT0 and FMT1 low, CHSL high for Standard 24-bit I2S. And inverted, for Slow rolloff.
Current stage is similar as on JEDAC - yet another take on a DIY high-end USB Audio DAC, I have also reused the same layout (I made JEDAC).
Still I am not getting any signal out. Before soldering the 22R load resistors I had 5V out.
How does PCM1794 behave is format is wrong, does it still output something?
So far getting a reasonable signal out of DIR9001. It is configured with FMT0 and FMT1 high, i.e. 24-bit, MSB-first, I2S. PSCK0 and PSCK1 low, i.e. SCK0 128fs. Verified on scope that SCK is a bit over 6 MHz with 48kHz fs. Source is an old SB Live.
PCM1794 I have tried FMT0 and FMT1 low, CHSL high for Standard 24-bit I2S. And inverted, for Slow rolloff.
Current stage is similar as on JEDAC - yet another take on a DIY high-end USB Audio DAC, I have also reused the same layout (I made JEDAC).
Still I am not getting any signal out. Before soldering the 22R load resistors I had 5V out.
How does PCM1794 behave is format is wrong, does it still output something?
I am not any sort of expert. This is what I know,
1)You have deviated from the referenced design and changed it to use a DIR9001 instead of a PCM2707.
2) When I embark on a new to me design I try to follow the TI app notes and eval designs to a tee unless, I really know my stuff.
3) The other control signals like RST,mute on the PCM1794 are correct?
4) the zero signal out of the pcm1794 is in what state?
5) what is generating the ref signal for XTI on the DIR9001? I do not understand this part very well, so you are trying to recover the system clock from the received data?
I think I need I and probably you need to read the DIR9001 datasheet and understand it better.
1)You have deviated from the referenced design and changed it to use a DIR9001 instead of a PCM2707.
2) When I embark on a new to me design I try to follow the TI app notes and eval designs to a tee unless, I really know my stuff.
3) The other control signals like RST,mute on the PCM1794 are correct?
4) the zero signal out of the pcm1794 is in what state?
5) what is generating the ref signal for XTI on the DIR9001? I do not understand this part very well, so you are trying to recover the system clock from the received data?
I think I need I and probably you need to read the DIR9001 datasheet and understand it better.
Yes, the reset is pulled high through a resistor. Mute tied to ground. Came to think of measuring the zero pin after I left the lab, will check next time.
On DIR9001, Error pin is connected to a LED and that goes out. Also here, I should measure the ^Audio pin.
DIR9001 uses the PLL clock recovery.
Largely based on this: http://electronics-diy.com/electronic_schematic.php?id=806
On DIR9001, Error pin is connected to a LED and that goes out. Also here, I should measure the ^Audio pin.
DIR9001 uses the PLL clock recovery.
Largely based on this: http://electronics-diy.com/electronic_schematic.php?id=806
Further trouble... I have now perfected the input stage and soldered the opamp, but the output is not correct.
With a 1 kHz signal ( https://www.youtube.com/watch?v=3FBijeNg_Gs ) I get this output:
It does go to zero when video is stopped and it does change correspondingly when selecting another frequency.
Does this now mean that the format selection of the PCM1794 is not correct?
I also tried changing sampling frequency in asound.conf and DIR9001 correctly detects it and sets LRCK and SCK correctly.
The output looks idential regardless of sampling frequency.
With a 1 kHz signal ( https://www.youtube.com/watch?v=3FBijeNg_Gs ) I get this output:
It does go to zero when video is stopped and it does change correspondingly when selecting another frequency.
Does this now mean that the format selection of the PCM1794 is not correct?
I also tried changing sampling frequency in asound.conf and DIR9001 correctly detects it and sets LRCK and SCK correctly.
The output looks idential regardless of sampling frequency.
Last edited:
I guess you can tell by capturing the data length on the scope. It would be nice if the scope could decode I2S 
It sounds like using the PC as your test source has its problems. I would be in the same position.
I made a USB audio decoder once using a PCM2902B, I know it has SPDIF o/p as I used it to loop back to itself and send the audio stream back to the PC.
Good Luck
Rick
It sounds like using the PC as your test source has its problems. I would be in the same position.
I made a USB audio decoder once using a PCM2902B, I know it has SPDIF o/p as I used it to loop back to itself and send the audio stream back to the PC.
Good Luck
Rick
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.