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

FifoPi Q3 - HDMI - Topping D90

FifoPi Q3 has an isolated DSD_EN signal. So I confirmed there is no problem working with Topping D90 or other exteranl DACs that need this DSD_EN signal to switch between PCM and DSD.

Both HdmiPi and TransportPi can be used for this application through a HDMI cable. A small two pin cable is also needed to connect the DSD_EN signal from FifoPi.

The only issue is that D90 has click noise when change between DSD and PCM. Just hope they can fix this issue by upgrade the FW.

Good weekend.
Ian
 

Attachments

  • FifoPiQ3_DSD_D90.jpg
    FifoPiQ3_DSD_D90.jpg
    675.5 KB · Views: 487
  • FifoPiQ3_PCM_D90.jpg
    FifoPiQ3_PCM_D90.jpg
    776 KB · Views: 501
Disabled Account
Joined 2020
I was one of those who mentioned it. I had been looking at it ~2years ago..
Just for curiosity. Yes, these are identical.

Also the phase noise parameters are equal to those cited for Pulsar..
I think the Pulsar project really was a group buy adventure.. But for some reason did not hold..

Ciao, George

Good stuff.

I'm allocating this year as my learning curve for Q3 / Accusilicon.

If all goes well, I estimate beginning of Summer 2021 I may start hunting for these clocks. Maybe earlier, but conservative estimate. Will follow a GB if one shows up.

I'm looking to build the most optimal small footprint battery power transportable Music streamer possible for personal use. Don't think that goal can be obtained without these clocks.
 
Disabled Account
Joined 2020
MX037/14D, MX037/14 ?

MX037/8P looks better.

Thanks for sharing. Saves me from having to contact about obtaining more plots. I cannot believe these alternatives are available.

Please share which product you would consider most optimal with the Q3? The one closest to plug and play. I just assumed the consensus was the 14D?

I might contact Russia next month and try to get info on pricing, payment, shipping, etc..
 
Last edited:
Good point.. ]

Actually I mistook 1Hz for 0.1Hz, 14D looks better to me and could be the sweet-spot in their lineup for power consumption, size and phase noise.

@ A123, yes 14D. Of interest is that initial samples measured better than what became end Pulsar product (I refer to 9x.xxxx MHz). I did also find reference to Pulsar using selected 'low phase noise' OCXO from MagicXtal. The single sample which I have suggests this may be true (which runs counter to the discrepancy mentioned previously), and I'm sure there's a back-story there, and quite possibly an intermediary as well, reading between lines makes picture clearer but also introduces risk of error :)

Regardless, a shame that Pulsar is no longer making these available in QTY 1+ to end user.

HK
 
Last edited:
Actually I mistook 1Hz for 0.1Hz, 14D looks better to me and could be the sweet-spot in their lineup for power consumption, size and phase noise.

HK

Yes, I do understand what You say..
Though would take that "sweet" rather 'bitter-sweet'..
The MXOCI line has 10dB lower phase noise at 1; 10Hz..
Not small difference. Actually, with 1Hz /-120dB, 10Hz /-145dB , at 10MHz, it starts to be tough competition..
 

Attachments

  • 10MHz 14 16-629-102.png
    10MHz 14 16-629-102.png
    73.8 KB · Views: 365
  • 10MHz C ref4_4 Heat insulation .png
    10MHz C ref4_4 Heat insulation .png
    69.7 KB · Views: 358
Thanks guys for your great suggestions.

Here is the StationPi V2.0 PCB layout.

Ian

Ian can you make simple like your Rpi i2s adapter board but in oposite configuration so it can be mounted on male gpio and can go on fifo? then using ufl cables you can mout pi and fifo in any direction also can incapsulate pi in to matal chassis. I will give endless options in not expensive way.
 
Ian can you make simple like your Rpi i2s adapter board but in oposite configuration so it can be mounted on male gpio and can go on fifo? then using ufl cables you can mout pi and fifo in any direction also can incapsulate pi in to matal chassis. I will give endless options in not expensive way.

That's a really good idea. I'll take it into consideration.

Good weekend.
Ian
 
Disabled Account
Joined 2020
I got a quote from magic xtal last year: MXO37/14P-A17D3T-22.5792 MHz and MXO37/14P-A17D3T-24.576 MHz at USD 350/ea., 15 weeks lead time, moq 5 pieces. They said these models had the best phase noise at these frequencies.

Awesome. No need for I to contact Magic Xtal at this time.

Looks like 14P > 14D.

MOQ down to 5 from 10. Tears down the barriers to entry. A GB of ten a bit difficult. GB of five not so much.

14P:
22x15x10 mm

1 Hz, -105/----, -90/----
10 Hz, -135/-100, -120/-90
100 Hz, -155/-130, -145/-120
1 kHz, -165/-155, -155/-150
10 kHz, -170/-170, -165/-165
100 kHz, -172/-172, -165/-165

Phase-noise level: -173 dBc/Hz, floor

14D:
22x15x11 mm

1 Hz, -105/----, -85/----
10 Hz, -135/-100, -115/-85
100 Hz, -155/-130, -143/-115
1 kHz, -165/-155, -150/-148
10 kHz, -170/-170, -165/-165
100 kHz, -172/-172, -165/-165

Phase-noise level: -165 dBc/Hz floor

-----

Not sure if lower height footprint available to distance from HAT/DAC on top:

8 mm height is available. Please consult the factory.
 
Last edited:
Disabled Account
Joined 2020
Regardless, a shame that Pulsar is no longer making these available in QTY 1+ to end user.

HK

Yes, it is a shame. I just time-traveled the Pulsar Clock thread and glad to read about the wonderful experiences like yours. I don't like sinking costs and time on incremental upgrades (Accusilicon) and wish Pulsar Clock was a current option before entering this project. Well, I'll just look at it as a educational sunk expense on how to clock-roll.

I hope I can experiment this when ready to upgrade to "Pulsar" Clock:

Pulsar Clock said:
A Digital to Analog Converter is like an orchestra that plays music following the exact time stated by the Director, the Clock.

Each hesitation of the Director makes the orchestra less clean, more cloudy. At the same way a jittered clock makes even the best DAC sound more blurred loosing the details that makes music alive.
Due to its exceptionally low phase noise and a consequent negligible jitter the Pulsar Clock pushes at the top the performance of any DAC, to test this performance and compare different clocks we made two different setup for synchronous and asynchronous DAC convertion.
 
Last edited:
Disabled Account
Joined 2020
I already sunk a few hundred on the Accusilicon + stuff I don't need but since I'm ordering from Audiophonics anyways order. So I need to put some use to the Accusilicons before I can justify an upgrade. I'll re-visit this 2nd Quarter 2021.

Otherwise, I would snap call a "Pulsar" Clock GB.

Russian customs is a pain so I don't know if you want to purchase outside of Russia. I've sold audio gear to Russia and it was always held up.

There's also a complication in "Dual Use" (Customs), so will need to be addressed before attempting an order.

Original Pulsar Clock Datasheet:
Available in a wide selection of audio frequencies Designed for High-End Audio DAC
Makes digital music a unique listening experience Ultra Low Phase Noise
Femtosecond Jitter
Very Low Power Consumption
Fits into DIL14 Clock Sockets HCMOS Output
Easy Installation
Laser Etched Serial Number

I'm guessing "HCMOS Output" is the option to choose for this potential buy since there's a few output options.
 

Attachments

  • S1 - Pulsar Clock Test - Synchronous DAC Block Diagram.jpg
    S1 - Pulsar Clock Test - Synchronous DAC Block Diagram.jpg
    97 KB · Views: 392
Last edited: