• These commercial threads are for private transactions. diyAudio.com provides these forums for the convenience of our members, but makes no warranty nor assumes any responsibility. We do not vet any members, use of this facility is at your own risk. Customers can post any issues in those threads as long as it is done in a civil manner. All diyAudio rules about conduct apply and will be enforced.

Amanero Isolator/Reclocker GB

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
How many of the multiple input voltages are required and/or highly recommended?

Similar to the SO3 board.
Isolated side power supplies are all separate but you can source them from a single 5V master supply if you wish to. There are on-board filters and decouplers for each power input. XO power is best from its own power source. There is an on-board 3.3V JFET regulator for the XO module. Option for external 3.3V XO power.
 
I just got my S03b and now comes the full version :rofl:

Anyway I got a question before my first 9018 build

In Sync operation, is 49.152MHz/45.1584MHz still within spec for 9018 to handle up to 384k material or must I use 98.304MHz/90.3168MHz?

Sorry if this question seems noob :p
 
In Sync operation, is 49.152MHz/45.1584MHz still within spec for 9018 to handle up to 384k material or must I use 98.304MHz/90.3168MHz?

49/45M is more than adequate for 384k and Crystek's CCHD957 is most recommended
The 98/90M option is there to test out if the ESS DACs perform better (jitter reduction) at higher clock frequencies.
 
49/45M is more than adequate for 384k and Crystek's CCHD957 is most recommended
The 98/90M option is there to test out if the ESS DACs perform better (jitter reduction) at higher clock frequencies.

Thanks Acko since the price of CCHD957 at my place are almost identical or even a little bit cheaper than other low jitter parts from Fox and Silicon Lab.

Another question, so will 9018 performs better under 98/90M? Or will it already do exceptionally well under 49/45M?

I have looked at other threads but seems I can't get an definite answer on clock freq selection.
 
Choice of XOs

After going through a number devices, I have gathered some information that will helpful in selection of oscillators. Every other manufacture now claims 'ultra-low jitter' specs so closer look at the characteristics is needed. The attached pics show the main contenders:

Crystek's CCHD 957 has the lowest phase noise, both close-in and floor

Top-end Fox and Silabs 5xx series are tad poorer in comparison even if they are of the high speed PECL types. Silabs types have 'spurs' (the effect of frequency synthesis) that could be bothersome

So CCH957 is the favourable choice.

As for 9018 DACs, these were originally promoted for Async operation where MCLK comes from a standalone XO, the frequencies must be greater than some multiple of Fs to operate correctly and for better jitter reduction:
>192 Fs for I2S
>386 Fs for Spdif

So to cover up to 384KHz I2S and 192KHz Spdif Async mode we are looking at MCLK >75MHz, so the choice of 100MHz types such as CCHD950/575 that have excellent jitter performance at this frequency.

Now we have seen that 9018 DACs can be operated in Sync mode with even better performance. In sync mode the MCLK frequencies can be much lower to operate correctly and as I have said before the 49.152MHz/45.1584MHz are more than adequate. (In fact without re-clocking the MCLK can be as low as 24.576MHz/22.5792MHz in sync mode to cover up to 384KHz)

So yes, 9018 in combination with CCHD957 (49/45M) will give excellent performance used in I2S Sync mode. Now whether it would work better at 98/90M is still yet to be determined. There is some mention that jitter reduction in ESS DACs will work better at higher frequencies but then again this could be negated by poor jitter performance XOs. I haven't seen any 100M XOs that could match CCHD957. The closest is Crystek's CCHD575 but these do not even come in sync frequency types. So we are left with Fox and Silabs ones and you may need to load these and see for yourselves. The S03 boards are meant exactly for these experiments. If anyone has done comparison tests please post your impressions. I will be doing my own tests in future (when I get some free time :))

The other thing that we need to be aware of:
if using 45/49M XOs, the I2S operation is covered nicely but if you are also using the 9018 for direct spdif inputs that are effectively Async, the highest spdif Fs will be limited to 96KHz! (386Fs).-just can't have the cake and eat it too :bawling:
 

Attachments

  • CCHD957.png
    CCHD957.png
    221 KB · Views: 654
  • Fox Expresso Ultra.png
    Fox Expresso Ultra.png
    179.5 KB · Views: 651
  • Silabs Si5xxx.png
    Silabs Si5xxx.png
    352.9 KB · Views: 646
  • CCHD 575 100MHz.png
    CCHD 575 100MHz.png
    326.7 KB · Views: 640
Last edited:
Hi Chris and thanks for the quick warning... I've had a good read on the miniSharc and realise adding I2S USB input isn't going to be straight forward. I'll also need inline Sample Rate Conversion ... that or re-sample all my digital music from 44.1 to 48MHz :eek:

My initial thought is Amanero > acko Iso/reclocker > TPA Metronome > miniSharc ... or something like that.

Paul.
 
Because the minisharc must be i2s master the mclk signal becomes and input rather than output and the isol chips are directional so the si/nve chip data sheets should show you the part number to allow that.... Would need to bypass reclocking I think too. I would potentially use this on the output only of the minisharc rather than input ... That would keep computer and DSP noise separate from DAC
 
Got you now :) ... Want to try using the TPA Metronome in full slave mode [input & output sides as slave] between the USB module & miniSharc. This should allow both USB and miniSharc to operate as master, the Metronome takes BCLK, LRCK & SD from USB [input side] and converts SR based on MCLK, LRCK, BCLK from miniSharc [output side]

USB needs to be 24 bit I2S 44.1 to 192MHz and SR output would be set by the minisharc plug-in [48 or 96MHz]. Metronome would be set-up as in/out slave, 24 bit I2S input and 24bit I2S output. SRC & output is based upon comparing input side LRCK to output side LRCK.

Will be a couple of weeks before I have all the bits and can physically try this out...


Was also thinking about using the iso/re-clk on the output side, between miniSharc & 2 x Curryman DACs... Long run, I'd like to go with a multichannel TPA Buffalo III but would require a multichannel version of Ian's FIFO Kit!?

Paul
 

Attachments

  • Amanero_miniSharc_set-up.pdf
    28.5 KB · Views: 164
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.