[nikongod]

 
Quote:

"Best" how?  Whenever someone ask "What is the BEST of this or that", 99% of the time there is no right answer.
Given that there are not many fully differential opamps out there, it does narrow the playing field to just a handful.  I would look at the OPA1632, but there are others which are also interesting.  Note that using a fully-differential opamp is quite different than a "normal" opamp, and puts some constraints on input impedance and choice of feedback resistor values, how you implement the volume control, and your choice of volume pot value.

This post gets my vote for the best answer.

 

[lampee]

Sorry, I didn't specify what I really wanted. Of course I am speaking about the sound quality, the THD or the naturalness of the sound, which comes out. But now I don't think I will go fully differential. The standard M3 with OPA627 and do it balanced will do the job for me pretty well I think.

 

[lampee]

Here is the next step in the design and manufacture period: Here you can see the Build of Material (BoM) list. Please make your comments. The prices are in USD and in HUF, because I live in Hungary.
http://mtws01.rmki.kfki.hu/~lampee/doksik/M3Balanced.xls

 

[tranhieu]

not new but yet very cool idea. subscribe.

 

[lampee]

I didn't post anything lately, but here is my design of the sigma11 power suppy. The current regulator diodes has been changed to a properly set variable current source (LM334). Here is the shematic:

and here is the separated top and bottom layer:
Top:

Bottom:

 
Of course 4 drill holes will be added to the 4 corners, but till then I would like to hear your comments especially amb's, if it is not too much to ask. Thank you guys!

 

[lampee]

The current sources are not properly placed, all three has to be pin swapped pin1 <-> pin3 but anyway tell me if you find any design faults.

 

[nikongod]

Why not just buy 2 M3 boards and populate 2 channels on each? You will save yourself the small-qty costs of getting yur boards made, and are more likley to get something that works.
 
I wonder if this is an effort to build an amp &psu that you dont have to call an M3 or σ11. It seems like way too much work to DIY something that is available but gives the designer(s) credit as due.

 

[lampee]

I just want to make everything by myself. This way I can use exactly the parts I want, and only those will fit. And of course this is 50% cheaper than buying, and this way I can use the other current source, and won't suffer any parts shortage or if it goes out of production. My other aim is to learn how to diy pcb's professionally. I read lot of things about it, and it is managable to do everything at home, including solder mask, chemical tinning and silkscreen too. If I have the result I will publish it, and I hope everybody will be amazed.
Again please look at it, and find any mistakes, please.

 

[lampee]

and it was not too much work. The M3 caused a lot of headache, the sigma11 was ready in 5-6 hours

 

[MASantos]

I'm sure you know this, but these power supplies have been tested using specific parts and circuit and altering the CCS scheme might introduce oscillation or other problems.
 
What is the problem with the CRD method used by AMB?

 

[johnwmclean]

 
Quote:

 
What is the problem with the CRD method used by AMB?

 
 
I get the impression lampee wants to learn how to DIY his own pcbs, he’s never said there was a problem with AMB’s method. IMO I think this would be a good experience and cannot see any harm in this project as it all looks to be a personal learning. Good Luck lampee!   

 

[MASantos]

My question wasn't  regarding the design and build of the PCB,
 
it is regarding the schematic and theory of operation of the power supply, which could lead to instability and oscillation in the circuit.
 
He changed the constant current source from a CDR to a variable current source (LM334) and this could be harmful if not properly tested.
I'm not saying it will, just giving lampee a heads up!

 

[johnwmclean]

 
Quote:

My question wasn't  regarding the design and build of the PCB,
 
it is regarding the schematic and theory of operation of the power supply, which could lead to instability and oscillation in the circuit.
 
He changed the constant current source from a CDR to a variable current source (LM334) and this could be harmful if not properly tested.
I'm not saying it will, just giving lampee a heads up!

My mis interpretation, sorry MASantos.

 

[lampee]

On amb's forum there has been a discussion about this kind of substitution, and the sigma11 was working well with it. Even amb gave his blessing, but he said, he will now post it as an official substitute, because it is only capable of handling a maximum of 40V and it would get the general usability away from the PSU.

 

[MASantos]

No need to apologize!  
Quote:

 

My mis interpretation, sorry MASantos.