BALANCED! With tda1543

[Elso Kwak]

Inverting two's complement

Are you using some form of unsigned data? I mean... you can't just throw an inverter on two's-compliment...

(Invert... and add one)

Hi Gridstop,
yes you can, it works except for the lowest bit.
Philips does it this way in the CD850 too!
See Horowitz table 8.1.
Besides that I constructed a WORKING balanced DAC this way.
A greater problem is the non-similarity of the positive and negative DAC. (Unequal output)😎 😎

 

[rfbrw]

Are you using some form of unsigned data? I mean... you can't just throw an inverter on two's-compliment...

(Invert... and add one)

It would appear nobody, commercial outfits included, gives a gnats gonads about the last bit. The "add one" hassle probably outweighs any possible gain.

 

[rfbrw]

Re: Inverting two's complement

A greater problem is the non-similarity of the positive and negative DAC. (Unequal output)😎 😎

This problem is lessened with each 2ch dac converting a L/R pair.

 

[Gridstop]

I guess at higher bitdepths it's not such a problem, I'm not sure I'd want to screw with the LSB w/ a 16-bit DAC though. Don't have too many bits to begin with!

 

[alchemy]

hi, who are build a dac with tda 1543 in balanced mode and what the result?

it's a real didderence with the the small sine wave?

i haven't found a shematics for look this type of arangement, i haven't realy understand how this works. if you have a links for a shematics i'm interesting


thanks

 

[alchemy]

my english is so bad, nobody can understand that i have write😀


😉

 

[rfbrw]

There is a schematic in a thread about using the TDA1541 in offset binary mode that can be modified for use with the TDA1543. Additional logic will be needed to carry out 2's comp inversion correctly but frankly I wouldn't bother adding the 1 if I was using the TDA1543. Search through the threads by a poster called rah or search the digital forum using the term 'offset binary'