The draw of the reclocker/dac is less than 130ma, so its Sala is tuned for 150/250. The USB section draws less than 400ma, so it's Salas is tuned for 400/500ma. I usually go with more headroom, but I'm trying hard to control my thermals given my case choice. The ~100ma headroom is proving more than adequate.
Re: tips, you'll definitely want a lytic in C102.
Re: tips, you'll definitely want a lytic in C102.
Hi,
Thank you for the test. I'm glad that XMOS board works with RaspyFi.
Thank you for sharing your efforts with us and for squeezing everything from XMOS/AK combo
Regards,
Joro
Thank you for the test. I'm glad that XMOS board works with RaspyFi.
Just to share in my efforts; I've got an XMOS/AK stack powered by duel Salas shunts. Both with SILMIC II 220uf lytic, tuning for digital use. Pardon the 'low profile' build pattern; I plan to mount all this in a low profile case.
To the sound - very very nice for the price
Thank you for sharing your efforts with us and for squeezing everything from XMOS/AK combo
Regards,
Joro
Hi..
Thank you very much joro, the package arrived yesterday, beautifully packed.
Omne question please, the i2s output, which one is which? Can you give me a image showing me how to connectthe board to buffalo3? Also the spdif out, which one should be connected to which .
Sorry for i am a newbie on dac wiring.
Thanks.
Thank you very much joro, the package arrived yesterday, beautifully packed.
Omne question please, the i2s output, which one is which? Can you give me a image showing me how to connectthe board to buffalo3? Also the spdif out, which one should be connected to which .
Sorry for i am a newbie on dac wiring.
Thanks.
Hi,
You can find the datasheet of the XMOS board at Specifications
As for the connecting, you need to connect the outputs of the XMOS board (pin 13, pin14, pin15 and GND) to the corresponding inputs of Buffalo III board.
It is not necessarily to connect S/PDIF if you already connected XMOS board to Buffalo board through I2S. You can use S/PDIF input of Buffalo for connecting to other sources.
Regards,
Joro
You can find the datasheet of the XMOS board at Specifications
As for the connecting, you need to connect the outputs of the XMOS board (pin 13, pin14, pin15 and GND) to the corresponding inputs of Buffalo III board.
It is not necessarily to connect S/PDIF if you already connected XMOS board to Buffalo board through I2S. You can use S/PDIF input of Buffalo for connecting to other sources.
Regards,
Joro
Hi,
S/PDIF and I2S are working simultaneously up to 192kHz.
As for S/PDIF connection you can choose the pin above the label "out S/PDIF" for signal.
Regards,
Joro
S/PDIF and I2S are working simultaneously up to 192kHz.
As for S/PDIF connection you can choose the pin above the label "out S/PDIF" for signal.
Regards,
Joro
Hi Joro,
I now have the usb / 4396 combo running on a separated PS based on LM 317. The usb is now running on 4.3 V and the reclock / 4396 board is running on 5.2 V.
The sound is very good.
The question i have is: Is it usefull to try to improve the PS for the reclock / 4396 board anyfuther?
The clocks are locally regulated (LP59000?) how is the regulation around the 4396?
I think EUR 63 for the usb xmos board with isolation and reclocking is a real bargain!
Gr GJ
I now have the usb / 4396 combo running on a separated PS based on LM 317. The usb is now running on 4.3 V and the reclock / 4396 board is running on 5.2 V.
The sound is very good.
The question i have is: Is it usefull to try to improve the PS for the reclock / 4396 board anyfuther?
The clocks are locally regulated (LP59000?) how is the regulation around the 4396?
I think EUR 63 for the usb xmos board with isolation and reclocking is a real bargain!
Gr GJ
I'll share my experience gugi; I started out powering USB/AK each with 317's on the bench. It sounded pretty good. I had a couple of extra BIB's, so I built a couple of shunts to replace them. A definite improvement. More slam, more dynamic, better pace.
Just be sure to use lytics in your shunts if you go that way. I've found the Elna Silmic II's to be stellar with the BIB's built to power digital circuits.
Just be sure to use lytics in your shunts if you go that way. I've found the Elna Silmic II's to be stellar with the BIB's built to power digital circuits.
Hi,
Thank you for your words.
There is one LP5900, which is feeding only the working oscillator, and another LP5900 which feeds the reclock. So in this case LM317 is a pre-regulator. AK4396 entirely rely on external power regulator. For the other, you can trust to Mister Rogers.
Regards,
Joro
Thank you for your words.
There is one LP5900, which is feeding only the working oscillator, and another LP5900 which feeds the reclock. So in this case LM317 is a pre-regulator. AK4396 entirely rely on external power regulator. For the other, you can trust to Mister Rogers.
Regards,
Joro
I will try TPS7A4700
That looks quite good! Much simpler than a salas shunt: for ex. 4.17uV Ultralow noise DAC power supply regulator 3.3V/5V 1Ax2 - DIYINHK.
It would be interesting to try a linear power supply board based on Belleson or NewclassD regs, they seem to offer equal or better performance (the first above all: Superpower Super Regulator by Belleson) than a shunt, don't they?
With the TPS7A4700 it works...
Gr GJ
An externally hosted image should be here but it was not working when we last tested it.
Gr GJ
Last edited:
OK got it working now. How does it sound? Stil good, very good.
An externally hosted image should be here but it was not working when we last tested it.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.