I have assembled a parallel DAC using TDA1541A. Wondering is it worth the effort to go fully balanced? My system is in this configuration.... WM8804 (Twisted Pear Audio) input receiver, 4X oversampling digital filter (not a crappy SAA7220), A pair of Red Baron TDA1541A boards (Version 5) and a 25 Ohm I/V with SRPP ECC86 with 2SK170 JFET (from a Swedish DIY forum). My DAC pair are S1 Crown's purchased in 1990. I have a schematic for an I2S splitter. My reading suggest that 16 bit dac's in parallel gives about 17 bits depth due to averaging. This clearly beats my other two TDA1541 (not A) dac builds. Open to comments. Thanks in advance....
1-bit would require 4 DAC chips.
This would likely most depend on whether your following gear (preamp, power amp) is balanced or not. If the succeeding gear is single ended, then there's probably little reason to go balanced. DAC outputs connected in a balanced configuration are effectively in series with each other. While the net improvement in SNR will be the same as for a parallel configuration, an balanced configuration will also exhibit twice the output signal amplitude but four times the output impedance versus an parallel configuration using the same number of chips.
Paralleling DAC chips results in an 3dB improvement in signal-to-noise ratio (SNR) for each doubling of the total number of chips. So, two chips = +3dB, four chips = +6dB. 1-bit of increased DAC resolution represents an 6.02dB improvement, so, to improve the system by the equivalent of 1-bit would require four times the number of DAC chips. There is no improvement to SNR simply from re-configuring the same number of DAC chips from parallel to balanced (series).
This would likely most depend on whether your following gear (preamp, power amp) is balanced or not. If the succeeding gear is single ended, then there's probably little reason to go balanced. DAC outputs connected in a balanced configuration are effectively in series with each other. While the net improvement in SNR will be the same as for a parallel configuration, an balanced configuration will also exhibit twice the output signal amplitude but four times the output impedance versus an parallel configuration using the same number of chips.
Paralleling DAC chips results in an 3dB improvement in signal-to-noise ratio (SNR) for each doubling of the total number of chips. So, two chips = +3dB, four chips = +6dB. 1-bit of increased DAC resolution represents an 6.02dB improvement, so, to improve the system by the equivalent of 1-bit would require four times the number of DAC chips. There is no improvement to SNR simply from re-configuring the same number of DAC chips from parallel to balanced (series).
Last edited:
Balanced
Hi Ken, Thank you for responding. I appreciate your view. I noted a few comments that it was a way to get the most quality out of the TDA1541A. Most of my downstream equipment is single ended. There are practical challenges with the glue logic in making the attempt to go balanced. Speed and propagation delays. Proper splitting of the I2S signals. Something I will need to consider. Thank you for clarifying the bit depth issue. I do hear more detail in the low bit levels with the Parallel DAC. Quite obvious with Roger Waters "Amused to Death" CD.
Balanced
Hi didiet78, Yes. The parallel DAC is quite brawny. It is interesting how detailed and musical it is as well. I may revisit the Balanced version in the future. Thank you again for the comment.Paralel brawn, balanced beauty.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.