An idea! Fully balanced TDA1541A DAC idea. Please comment!

[cartman]

Hi Guys!

I have an idea to make a fully balanced TDA1541A DAC.


It could works?

Thx!

 

[Tazzz]

Yes it would work but you would not get the really cool benefits of running it balanced.

So use this instead:

http://www.pedjarogic.com/1541a/I2Sbal_split.htm

 

[cartman]

Thanks Tazz for the comment...

...but I really don't see why better the Pedja's method?

Maybe the channel separation is better, but at my idea, the RCA and XLR outputs are simply to operate.

What do you think?
Why isn't good this method than Pedja's?

 

[cartman]

Perhaps the phase inverter would be more simple than i2s splitter...

 

[Tazzz]

You could read the threads about how to decouple the tda1541a lots of pointers there.

For one thing running one dac differential would make the complimentary audiocurrents flowing trough Va+ constant, wich would be most desireable.

 

[Bricolo]

the way the 1541 is built, the AC current going to the supply pin is the same as the current in both outputs

so if you have a classical L/R implementation, it's just 2 separate channels

on the other hand, if you have L+/L-, the AC supply current becomes (L-)+(L+) so L-L=0
you get only (or almost) a DC supply current to the chip, making the regulation, decoupling, PSSR, problems almost disappear

 

[X.G.]

the best way is to make a glue logic for working in sim' mode of TDA1541, like this by CPLD:

http://www.cdream5.com/htm/blance_dac1541A.htm
http://www.cdream5.com/htm/I2S_MK.htm

 

[abzug]

I have always wondered, why not use sign magnitude loading with dual dac like TDA?

 

[Bricolo]

the best way is to make a glue logic for working in sim' mode of TDA1541, like this by CPLD:

http://www.cdream5.com/htm/blance_dac1541A.htm
http://www.cdream5.com/htm/I2S_MK.htm

why is it better?