The Best DAC is no DAC

Hey, guys thanks so much for the support, now I'm embarrassed by all your kind comments:blush:

I found some time this morning and have been assembling a board and taking some pictures so I can document the modification. I'll post it later. I've almost finished the board so will be able to test it a bit later - I'll let you know how it goes.

Thanks again for the positive comments.

Ray
 
OK, some good news:)

I've assembled a balanced board and have just been testing it; with the modification everything works fine, producing very nice music. The only thing I haven't tested is the mute function and will do that a little later.

I actually tested it in my Silk TVC project. I assembled the PCB with a LP filter at 40KHz then took the outputs from the board straight to the Silks, with no caps. It works well and all of the white noise/mush is gone, background is now really quiet. Music sounds wonderful.

So, if you make the mod to the PCB you should be fine.

Here's a walk through of the modification for anyone who isn't sure about doing it.

Step 1. Cut the PCB tracks like this;

An externally hosted image should be here but it was not working when we last tested it.


Step 2. Solder the header sockets to the bottom of the PCB (I install them onto the JLSounds board and then solder a couple of pins in situ to make sure everything is aligned) but DO NOT solder pin 1.

Step 2. Insert the 1N4148 to the PCB and jam a length of fine wire into the pad nearest the board edge and solder the pad. Slip some sleeving over the wire and bend it around pin 1 of the header socket. Solder pin 1 and the wire to the PCB. It should look something like this (n.b. I routed the wire to allow the screw terminal to be soldered to the board later);

An externally hosted image should be here but it was not working when we last tested it.



Step 4. Turn the PCB over and, using the wire from the 1N4148, make a connection in a similar manner to the screw terminal pad. Bend the end of the wire at 90deg and push it into pad. Like this;

An externally hosted image should be here but it was not working when we last tested it.


Step 5. Insert the screw terminal to the board and solder. It should end up looking something like this;

An externally hosted image should be here but it was not working when we last tested it.


And from the top;

An externally hosted image should be here but it was not working when we last tested it.


Hope that helps.

I feel a bit more chilled now.

Ray
 
BTW, the board I assembled uses the higher output current FlipFlop that I mentioned a few posts back. I can't honestly say if it made any difference as I also incorporated the LP filters at the same time but overall this build is a dramatic improvement over the previous attempt at using the Silks as the LP filters so the new FlipFlop doesn't seem to be doing any harm.

I'll measure the DC across the FlipFlop output later too.

Ray
 
Good stuff Ray :) Do you reckon your improvements are due to the flipflop giving approx double the output volume with better signal to noise and not requiring the dc blocking caps in the signal path?

That seems probable, though I can't demonstrate it as I haven't really progressed the projects in a controlled manner, making several changes at a time and only measuring the results with my ears. Makes me smile though.

Ray
 
I have also successfully tested the mute function although not with Fabio's delay circuit (because I forgot to order the preset potentiometer).

Anyway, having made the PCB modification I now have a correctly functioning balanced (FlipFlop) board that is producing great sounds.

I did check the DC offset across the outputs with music playing and, not surprisingly, it does fluctuate but I didn't see it go above 4mV.

Ray
 
Hi, friends, I checked on my flip flops card, modified as Ray said , and for the right channel I have 0v between + and - (2.5- and 2.5 +) but on the left the 5v (+ 5V and 0-) this without music coming out. Fabio.

How's this looking now Fabio?

Have you checked for solder bridges? Maybe start by checking continuity to see if the left positive line has a connection to the 5V line?

Ray