A NOS 192/24 DAC with the PCM1794 (and WaveIO USB input)

Jump to Latest
#4,921
dwjames said:
Evening all 🙂
Has anyone experimented with adding small ferrite beads ..
Click to expand...

I tried a few clip-on beads around the dac, at the PS input to the Wave IO, and the main feed to the "blue" board, but didn't notice any difference at the time. Mind you I don't have a dedicated supply to the digital side of the PCM1794's, so I wasn't able to test this.

Instresting video on the subject:
https://youtu.be/81C4IfONt3o
 
#4,922
Movember again !! Please donate (even a small amount is welcome !!)

OFF TOPIC.... it is Movember again 😎

Like last year, please have a look at the Fund Raising Site from my son Erik. Please donate a small amount if you like my DAC 😀 This is really a good goal (for us Men...)

https://de.movember.com/en/mospace/9946341/

Thanks!


The Movember Foundation
The leading global organisation committed to changing the face of men’s health.
 
#4,923
My 16 decker is working wonderfully up until now so I am starting on a dual 16 deck.

I have built a separate power supply which comprises of 2 x 225 Va 25v transformers into the modified DDDAC power supply which now has 2 x 10000 uf Mundorf caps instead of 4700 uf. The smoothing resistors have been replaced by Lundahl 2733 chokes and the TIP122 changed to a TIP142 to power the decks.

The Wave IO power supply also has the Mundorf caps, the 15v + 15 v transformer unused for the dacs fitted instead of the 9 + 9 v .

I am going to use the Nichicon FPCAP NU series 47uf 50v in place of the 47uf 25v Muse supplied. 710 ordered ....

I am going to use the S Series resistors from Charcroft Electronics for Iref 6k and for the IV conversion I will use the Charcroft VSR range.

I have also bought some Sowter 1 - 1 transformers as I do not need the high output level.
 
#4,924
The missing link

Maybe this is off topic but I have to recommend the following component: Intona High Speed USB Isolator. I'm using this little gem for one week with waveIO now and it seems that this is the missing link for USB transport between PC and USB DAC. It gives real galvanic isolation with 480 MBit speed and the sound is fantastic! There is a special version with "NF modification" (LED blinking modified) which is available instead of the standard configuration. Intona-Products

P.S.: I'm not afiliated with that company ...
 
#4,926
Galvonic isolation does not mean the device will isolate the down stream device from the computer noise, the layout is critical for any attenuation of PC noise, there is also some noise coupling through the Adum device that they use inside. So these devices do tend to let the HF noise pass, but they do protect against high voltage which is their primary function....
 
#4,929
marce said:
Galvonic isolation does not mean the device will isolate the down stream device from the computer noise, the layout is critical for any attenuation of PC noise, there is also some noise coupling through the Adum device that they use inside. So these devices do tend to let the HF noise pass, but they do protect against high voltage which is their primary function....
Click to expand...

You're surely right that all components and the layout of the pcb is critical. What I've read seems to be done right in every aspect. I'm not an expert in this field and tried a couple of other devices over the years. Nothing comes close to this isolator.
 
#4,930
Without looking at the design I couldn't comment... But as said galvanic isolation is for high voltage protection and possible ground loops. The main coupling method for the high frequency noise from a PC is capacitive, this is why when you look at certain DAC/ADC mixed signal design guides where separate analogue and digital powers and grounds are recommended they go to great lengths to advise against any overlap of the separate plains. The Adum devices themselves have a certain amount of capacitive coupling between the input and output (listed on the data sheet) which will also couple some of the HF noise so extra measures and filtering do have to be employed to minimise the noise coupling. The example shown will work quite well at USB 2 full speed as it is only 12MHz, but does have some detrimental effect on the signal integrity, this is very small and would not effect USB 2 full speed but for the full blown USB in the 100s of MHz a bespoke USB hub device or FPGA solution is required. The Adums only do up to 12MHz anyway.
I do this as my day job and the design posted is from some work I have done previously where noise isolation was required as well... 🙂
 
#4,940
Ok modification done.

Acutally did not affected the balanced output due to my wiring on the preamp, however for the unbalanced one, this is better, thank you 🙂.

But the "hiss" is still here.
Actually on a 24 bit file, there is no "hiss", but on a 16, it is there.
I use a raspberry pi with PI core player.
 
You must log in or register to reply here.
Top Bottom