Asynchronous I2S FIFO project, an ultimate weapon to fight the jitter

[coolhead]

Originally Posted by iancanada

Hi coolhead,
The total power consumtion is up to what oscillators you are using. For the default configuration, the total current is around 110mA for all the 3 boards.
Have a nice weekend.

Got it.
Thanks.

[iancanada]

Originally Posted by UnixMan

you may consider integrating the USB interface from this project: "Open-source USB interface: Audio Widget"

Sounds good. Maybe I need buy one of that kit and trying to figure out how to integrate with my fifo.
Thanks,

[zinsula]

As the async USB is kind of "decoupled" from the source, there should be no need to pass the data though anoter FIFO.
But with the dual clock board, some nice thing could be done perhaps.

[glt]

Hi Ian,

In post 54 you said:

"6. Output the MCLK and I2S signals directly without passing through FIFO;"

What does this mean?

[glt]

Also, no Potato Semi flip-flop in the shipping versions of the board?

[iancanada]

Originally Posted by glt

Hi Ian,

In post 54 you said:

"6. Output the MCLK and I2S signals directly without passing through FIFO;"

What does this mean?

That means MCLK and I2S signals are output from clock board finially, not directly output form FPGA/CPLD which usually comes with much more jitter.

[iancanada]

Originally Posted by glt

Also, no Potato Semi flip-flop in the shipping versions of the board?

Hi glt,

I gave Ghz flip-flop a try on my dual clock boare prototype version 2.0. However, since prototype V3.0, I switched to 3 NXP PicoGate flip-flops. You may find them from pictures on my previous post since
Asynchronous I2S FIFO project, an ultimate weapon to fight the jitter P17, as well as from the pictures on the GB thread

Both 600Mhz Potato flip-flop and 550Mhz PicoGate flip-flops sounds great. They did perfect job on re-clocking. The only reason was Potato is a very small company, sourcing from a big company make me feel better on quality control because of my long time medical grade design experience.

I still have a couple of prototype 2.0 empty pcb. I can give you one of them for free if you want to try it again by yourself.

Have a good night.

[glt]

Hi Ian, thanks for the clarification. The only reason I ask is because I started reading the thread with more attention and then went over the literature over at Potato Semi and got all excited about the technology. Then I went to look for the chip on the clock board and I didn't find it. :-)

I don't think I'll take your offer for a PCB because I can't solder at that fine scale. I'll ruin the board.

In any case, I put up some pictures over at my blog. Let me know if what I said is accurate.

[mfr]

Hello,

I was not following the thread for a while. If for some reason there would be more kits, I'd have one of each boards.

Thanks,

MF

[gvelim]

FYI

Interesting startup in the oscillator market Accusilicon, we make ultra-low phase noise XO/VCXO/TCXO ICs
their XO clocks product reported as

12KHz-20MHz RMS jitter: <50fs (122.88MHz)