My "audiophile" LM3886 approach

if someone could be patient enough to explain, then I would like to know the difference between the RevA and RevC and can we convert Rev A ( stereo Board ) to Rev C, I do understand we would have to manage to solder the components for the change. but is it possible?
 

Attachments

  • RevC Mods.pdf
    370.4 KB · Views: 731
Thank you, I m going with my build a little slow.

Bought the 0.1% resistors and ordered other components awaiting delivery...

also working on a PGA2311 with Relay for input selection.

Regular work is taking a lot of time from me and leaving a little while to relax but going on with the build ( this build is getting a bit expensive for me :D).
 
Finally assembled the board and now doing some testing.
My transformer is having a secondary 16-0-16 @4.5 amps. So will be ordering new one once I see things are working and not stuck anywhere with issues.
 

Attachments

  • 8FF864DC-6A01-4729-9D28-CD2FA87C2442.jpeg
    8FF864DC-6A01-4729-9D28-CD2FA87C2442.jpeg
    483.9 KB · Views: 425
I have been listening to the amp since couple of days. I like it.
Need to find a suitable cabinet in india and not too expensive.

cool amp. Doesn’t heat up more than being a bit warm at my listening levels on 8ohm load.

also need a input switching circuit with remote ( no tone control).
Help on the cabinet would be great
 
I'm going to start my own amplifier design, these modifications will move it even more into the High End class

  1. It will be implemented as a pure monoblock (moduschop)
  2. 200W toroid (talema) + 2x 22.000uF / for 1 channel
  3. I will replace the zener diodes for lm318 with stabilizers LM317/337
  4. the value of the input capacitor is increased to 6u8
  5. The original rev c scheme will be retained
  6. The 0.47ohm/7W output resistor will be replaced by a metalized 0.6w (10x 4,7ohm parallel)
That all :)
 
  • Like
Reactions: 1 user
Member
Joined 2019
Paid Member
Yes, I'll give a positive vote to UnixMan's suggestion. The Fremen edition, now in V1.8, is a sonically beautiful implementation of M Penasa's design. The BOM ranges from "frugal build" to "audiophile build." It's one of the best sounding LM3886-based amps out there.

Cheers
 
  • Like
Reactions: 1 user
After years I've decided to build the RevC. last year, thanks to recent mandates. I'm not a pro, just the basics. But I have a very uncommon issue. I have a 180 degree phase shift between amps. It's very obvious, no doubt. Over software I shift the phase of one channel and we're back in business. Interestingly this shift is not present with a 4ohm load, different speakers per se. But 8ohms, phase shift! It's almost perfect to where I can flip the polarities of one speaker. Other than this discreetly the amps work just fine; no hum no noise, just perfect. By Mauro's (RIP) pdf's I'm guesstimating it's one of the compensation networks. All resistors are .1, caps are far from perfect. I'd be glad if someone can point me into the right direction before I start flipping parts.
 
After years I've decided to build the RevC. last year, thanks to recent mandates. I'm not a pro, just the basics. But I have a very uncommon issue. I have a 180 degree phase shift between amps. It's very obvious, no doubt. Over software I shift the phase of one channel and we're back in business. Interestingly this shift is not present with a 4ohm load, different speakers per se. But 8ohms, phase shift! It's almost perfect to where I can flip the polarities of one speaker. Other than this discreetly the amps work just fine; no hum no noise, just perfect. By Mauro's (RIP) pdf's I'm guesstimating it's one of the compensation networks. All resistors are .1, caps are far from perfect. I'd be glad if someone can point me into the right direction before I start flipping parts.
Nevermind, thought I've checked the cables gazillion times from the board to the chassis, wondering how a perfect phase shift could even happen by compensation... but I had nothing left. One of them shrodinger moments.
 
I can confirm that it's more than 1,5V. The exact value I don't remember, too...
My assumption is that here Mauro had tried to optimize between a still practically acceptable level and a safe dynamic headroom so as not to intervene too fast with real music.
I can confirm that it works quite well in practice. it's rare that the error would manifest like a 4V DC stable level ..
Are there in circulation other, more precise circuits? Today I would say yes.
Other 'feature' of the MyRef protection circuit is the insensitivity for negative polarities..
 
  • Like
Reactions: 1 users