Return-to-zero shift register FIRDAC

Jump to Latest
#4,002
Well definitely keep us updated , even though my system isn't a high end commercial based I do have friends that do use high end . I'm lucky I get chance to listen to their set ups . Marcels dac even how it is now would certainly not be embarrassed by those high end units .Once its running as best as I can get it I'll take it along in a biscuit tin 😆

I've not tried output transformers with this dac , I did years ago with previous other dacs but they must have been poor quality . The timing with complex music suffered , always gave the impression swimming in treacle . It seems to be very hard finding suitable good quality transformers so it would be interesting if those Chinese ones turn out to be good
 
#4,003
Today I found the reason for these tiny "idle tones" below -150dBFS at the signal level of -60dBFS.
In my case it was non gold plated plug for 3.5mm APU input port.
After cleaning:
 

Attachments

  • dsd_-60.png
    dsd_-60.png
    217.3 KB · Views: 55
#4,006
It refers to one of the products made by @IVX. IIUC its an amplifier and notch filter box that works at 1kHz and 10kHz. It can be used with the Cosmos ADC or some other ADC for FFT measurements. IVX is good at making low cost measurement devices. IMHO they can't sound that great, but for taking standard FFTs they actually fit the bill pretty well. And they are popular. https://e1dashz.wixsite.com/index/cosmos-apu
https://e1dashz.wixsite.com/index/cosmos
 
Last edited:
#4,007
So it's part of the audio measuring equipment. It's strange that the out-of-band signals at the audio output were still large enough to make a connector produce measurable intermodulation products, although those products were extremely small. Some crosstalk due to a poor connection maybe, causing more than the usual amount of stuff around fs/2 to reach the APU?
 
#4,009
Unless you really have a rotten contact, you normally need fairly large signals to get measurable passive intermodulation distortion. The peaks around fs/2 and its odd multiples (5.6448 MHz and its odd multiples for DSD256) that cause second-order intermodulation products in the audio band get suppressed by the FIRDAC and the analogue reconstruction filter. I'm surprised there is enough left of them to cause measurable passive intermodulation distortion, but then again, the levels were extremely small.

Anyway, good reason to use gold-plated headers between the DAC and the filter board of the DAC of this thread (like we already did). The peaks around fs/2 are only suppressed by a short FIRDAC and one capacitor there.
 
#4,010
Hi all ...

This is just a brief post to wish all of you a Merry Christmas - I hope the days to come may somehow smile to you in ways that each of you appreciate 🙂

And then I would also like to thank Marcel for having made & shared this RTZ FIRDAC design which to me has been a source of inspiration and not least musical joy for quite some time now. IMHO it is quite unique in its ability to convey unusual levels of nuance, detail and musicality - to my ears very truthfully - and also being surprisingly responsive to its "ancillaries", i.e. the circuitry, components, etc. that surrounds the DAC. A joy indeed ...

And with these words - best wishes for your holidays!

Jesper
 
  • Like
Reactions: tamra and Raj1
#4,011
Merry Christmas to you too!

To answer the question you asked on the PCM2DSD thread: with my configuration files, the delay will be around 65 samples of the input sample rate. That's mostly group delay of the interpolation filter. Switching to apodizing will not help, because the apodizing impulse response is shifted to have the same delay as the steep filter (otherwise the interpolation filter could not handle it).
 
  • Thank You
Reactions: diyiggy
#4,012
TNT said:
Maybe time to open an own commercial thread in the bazar section? Here we dwell in DIY for private use...

//
Click to expand...
Why, there is nothing to sell yet ? As far I can see it is staying a DIY exchange, whatever you agree or not. Make you own constructive ads on in spite of always just critisizing as if you were a wize man : input technical !
 
Last edited:
#4,020
idea for dsc2 power dac, the same concept using 74ahct595 but instead of driving resistors driving 32 primary windongs with center tap connected to HV, 64 iso gan fets in push pull configuration driving windings, and one only winding on secondary to drive speaker, might this work? Volume control by variable power supply. Thanks
 
Last edited:
You must log in or register to reply here.
Top Bottom