F5X Pre My Way

It was the beginning od 2012. I was still leaving in Japan and in the process of building the F5X Amp. EUVL designed a matched Pre-Amp and I started to collect parts for it. This time I decided not to go for the full F5X Pre assembly, as designed by EUVL, but to integrate various things.
10 years passed, the unit is not 100% completed, front panels still need to be designed and milled, but it's fully functional. Here are the highlights:
1) Two chassis design (PSU and Audio sections)
2) Dual mono
3) Fully balanced
4) DCB1 input buffer
5) Muses 72323 volume control
6) F5X Pre gain stage
7) CLC Psu with common mode inductors
8) Plugs to power monoblocks and control them via relays
9) Controlled with Arduino (Display, Endoder and Remote)

Everything that could be matched in this unit is matched.

It was a long journey, but I am very satisfied with the results. I learned a loads of things in the meantime. I developed a new library for the 72323 (72320 libraries are not compatible)
Things to do:
1) Front panels. The front panel where the display is located should completely enclose and shield the microprocessor.
2) Galvanic isolators between Arduino and Muses chips.
3) Listen
4) Listen
5) Listen

IMG_8981.jpg



IMG_8983.jpg
IMG_8974.jpg
IMG_8979.jpg
IMG_8972.jpg
IMG_8980.jpg
IMG_8981.jpg
 

Attachments

  • IMG_8981.jpg
    IMG_8981.jpg
    76.8 KB · Views: 131
Last edited:
  • Like
Reactions: 1 user
Hi Nikon,

Nice! Quite a long journey.
As I’m building something similar, I’m interested in the reason why you opted for a buffer in front of the Muses, as the input resistance is arround 20k. Have you tried without the dcb1?
Because I never understood if the advertised input impedance of the digital potentiometer is constant on all volume levels. I tried both configurations and, with the limited measurements capabilities I have (Picoscope 4262), I could not see any difference with or without. So I left the buffer in. Like this I have 100k input impedance, rain or shine.

I also don't expect much differences using the galvanic isolator for the Arduino, but the idea was to go for the overkill.

D.
 
> why you opted for a buffer in front of the Muses..
Because I never understood if the advertised input impedance of the digital potentiometer is constant on all volume levels ..

Which is why we built our own switched-resistor, fully balanced attenuator (and not 2 SE in series).
And no bipolar-signal digital attenuator can have low enough distortion compared to switched bulk-foils.

If you MUST use a buffer, then use 2x 2SK209GR in parallel, instead of 2SK170.
This will give you a lower H3, and then you can cancel the H2 in the balanced F5XP downstream.
https://www.diyaudio.com/community/threads/njfets-for-source-follower-applications.329131/

Patrick