Non-oversampling DAC (TDA1543/CS8412)

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Here's my attempt on building a simple non-oversampling dac. I'd like comments on the lay-out.

Filtercomponents and the dac are placed on the downside. There are two smoothing regulators in the picture but I will only use one. Green is earth.
 

Attachments

  • dacpcbcomponent1klein.jpg
    dacpcbcomponent1klein.jpg
    89 KB · Views: 1,097
my friend - TDA1543 is VERY sensitive on the quality of the power supply.... no good quality power - no good sound.... i am affraid that using only one three terminal power supply regulator without anything that is better quality than that would not be enough to achive very good sounding dac.... well - it will be easier to make - but i am affraid that it would not be good quality...... than again - it is only my insignificant opinion about that.....
hope this helps
best regards
 
Francis_Vaughan said:



Groundplane? Control of RF?

The green wire is ground. I decided not to use a groundPLANE since I heared another TDA1543 dac that sounded extremly good...without using a groundPLANE...so it is possible I guess?

What else can I do to control RF?

sparkle said:
my friend - TDA1543 is VERY sensitive on the quality of the power supply.... no good quality power - no good sound.... i am affraid that using only one three terminal power supply regulator without anything that is better quality than that would not be enough to achive very good sounding dac.... well - it will be easier to make - but i am affraid that it would not be good quality...... than again - it is only my insignificant opinion about that.....
hope this helps
best regards

Thanks! I didn't include the power supply in the picture. Again, this choice was made after I heared several dac's using only one regulator for the dac. My power supply is based on a transformer and a few rectifiers, like 47labs builts it.
 
So how do you know that the lack of groundplane contributes to the good sound? More to the point, how do you know that it would not sound even better with a groundplane?

Your design regime seems to be one of picking odd ideas from other designs without really understanding why they were used. (Something sometimes called a cargo cult design.) If you do this fine, but then why ask for advice about the design? You have clearly already decided what to use.

Same goes for power supply, and layout. There are very good reasons for great care being taken.
 
Francis_Vaughan said:
So how do you know that the lack of groundplane contributes to the good sound? More to the point, how do you know that it would not sound even better with a groundplane?

I do know a lay-out without groundplane CAN sound awesome. Off course it's possible it sounds even better with groundplane. I guess I'll have an answer when I start experimenting, but I should begin with SOMETHING.

PS. I'm making a version WITH groundplane right know, so I can built that and simply compare with my ears ;)
 
Hi reMC;
I can give you an advice based only on my listening experiences regarding the sound of TDA1543/CS8412 NOS DAC.
I have attached the schematic and a photo of my last version of the DAC. My DAC is literally build by this schematic. This is the fifth version. The first version had just one LM7805 on power supply and a caps on power pins of TDA1543 (47uF) and CS8412 (47uF).
It is the holy rule on this forum to place the cap on power pin of TDA1543 (as close as possible , usually 47uF) and also a cap on both power supply pins of CS8412 (as close as possible , usually 22uF). When I removed those caps the sound improved. I was asking the other members of this forum if someone has the same experiences but all I got was answers about noise and filtering (mostly theoretical explanations from books). I guess nobody seems to try it. My experience is that the more you filter the power supply (big caps, chokes,...) the more the sound will be suppressed.
So the best you could do is to build the DAC on protoboard first and try to change components and just listen to the sound what sounds best to you.
Regards;
 

Attachments

  • TDA1543 NOS DAC_sch.JPG
    TDA1543 NOS DAC_sch.JPG
    77.4 KB · Views: 391
  • Picture 015.jpg
    Picture 015.jpg
    70.1 KB · Views: 276
For good implementation of this chip, read the "Pushing the limit of TDA1543 NOS DAC". Yours is a simpler approach, but it is easy to scale up the rail voltage close to chip's limit to lower distortion, e.g. using 7808.

It is normal and expected if someone is "impressed" by the sound of NOS TDA1543 with V/I using resistor. It is so natural, and the bass is so punchy. But the main limitation is the sonic/dynamic of it, of which sonic is known to be the main factor (at least according to me) determining enjoyment or boredom in listening to music.

You can do many things to improve upon your implementation, but imo the dynamic is the only issue of such design. The difficulty will be in choosing the right amplification stage after the DAC. My "best" result was using R V/I converter followed by tube amplification with complex power supply (inductor based). Later I will try again with 1000V/us op-amp.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.