I Have not seen the schematics for VDA2/Channel Islands DAC but it is only single ended output, which is a shame. I use Pass labs gear and balanced output is a must for me.
Once you go balanced out you will never go back! Any distortion and noise cancels itself out and the back ground is just dead silent.
THe PCM 179x is a beautiful balanced current output DAC and I don't understand why any designers would sum the + and - phases together through an op amp.
Also discrete class A I/V is much better than op amps in general, but harder to design properly.
The Pass Labs I/V circuit from the D1 is just magic as one example.
-David
Once you go balanced out you will never go back! Any distortion and noise cancels itself out and the back ground is just dead silent.
THe PCM 179x is a beautiful balanced current output DAC and I don't understand why any designers would sum the + and - phases together through an op amp.
Also discrete class A I/V is much better than op amps in general, but harder to design properly.
The Pass Labs I/V circuit from the D1 is just magic as one example.
-David
dw8083 said:
WRT balanced, agreed but there are few things to consider.
If the dac is fully balanced as in D1 I-V any common mode noise
will not get cancelled -before- coming out of the dac box. So the
noise will still be there on each phase and will be nulled at the
next port of call, be it pre amp or amp.
I'm not sure whether this is a good thing or not.
So the ideal I-V is balanced in and balanced out but -also-
rejects common mode signals before the OP.
With opamps this is easy to acheive but with discrete open loop
D1 type circuits it is not so easy. I have done it but it took some
thinking.
And at the end of the day I'm not even sure which sounds better.
cheers
T
As this thread has escalated from just "which DAC chip is best", I'll go ahead with som application questions for the 1794.
Input:
Running I2S directly from a CD drive will result in original 44.1 kHz sampling frequency through the DAC, right? So I'll need an upsampler like the AD1896 or whatever to run 96 kHz or 192 kHz. Also, there's the question of using an input transformer or not, specially if I'm splitting up the signal to two DACs.
Output:
I'll either use the active I/V suggested in the TI datasheet, using LM4562 opamps (possibly class A-biased), or the discrete class A ZAPfilter mk2 from LC Audio. An immediate advantage with the opamp setup is that I can run true dual mono with two 1794 in mono mode, whereas the ZAPfilter is a stereo output stage connecting to a single 1794 in stereo mode.
Tips and comments, anyone?
Input:
Running I2S directly from a CD drive will result in original 44.1 kHz sampling frequency through the DAC, right? So I'll need an upsampler like the AD1896 or whatever to run 96 kHz or 192 kHz. Also, there's the question of using an input transformer or not, specially if I'm splitting up the signal to two DACs.
Output:
I'll either use the active I/V suggested in the TI datasheet, using LM4562 opamps (possibly class A-biased), or the discrete class A ZAPfilter mk2 from LC Audio. An immediate advantage with the opamp setup is that I can run true dual mono with two 1794 in mono mode, whereas the ZAPfilter is a stereo output stage connecting to a single 1794 in stereo mode.
Tips and comments, anyone?
Hi Lars,
The PCM1794 will work fine with 44.1K I2S source. If you really need 96K/192K up-sampling than the AD1896 / SRC4192 ASRCs are required. Note the AD1896 output master mode only up to 96K but the SRC4192 will work at 192K master mode.
Running the PCM1794 in mono will be excellence idea, (132 dB dynamic Range) and let us know the project outcome as I am very interested on this DAC chip.
The very new LM4562 appears to be an awesome dual op-amp from the specs, and set a new benchmark for the dual chip, just like the venerable NE5532 when it debut 30 yrs ago. Pls report your opinion on the LM4562. Thanks.
Regards
Max
The PCM1794 will work fine with 44.1K I2S source. If you really need 96K/192K up-sampling than the AD1896 / SRC4192 ASRCs are required. Note the AD1896 output master mode only up to 96K but the SRC4192 will work at 192K master mode.
Running the PCM1794 in mono will be excellence idea, (132 dB dynamic Range) and let us know the project outcome as I am very interested on this DAC chip.
The very new LM4562 appears to be an awesome dual op-amp from the specs, and set a new benchmark for the dual chip, just like the venerable NE5532 when it debut 30 yrs ago. Pls report your opinion on the LM4562. Thanks.
Regards
Max
I know the 1794 works fine with 44.1 kHz, but isn't upsampling a major advantage? I'm no expert (yet), but from everything I've read and heard, properly applied 192 up-sampling beats "pure" 44.1 hands down. It does complicate things though, for example by requiring more clocks.
Yes, the LM4562 looks good on paper, and from numerous reports, it sounds incredibly good as well. Some still prefer the OPA627, but that seems to be mostly due to personal preference - it beats everything else. That's why I'm tempted to go for the simplicity of single chip opamps.
If you want some LM4562, be sure to check out the group buy, it's just about to close.
Kittikun: Unless you need the TDMCA support and/or SPI control interface, I think you'll be better off with the 1794 over the 1796. It's better on just about every measurement; noise, dynamic range, distortion etc.
Yes, the LM4562 looks good on paper, and from numerous reports, it sounds incredibly good as well. Some still prefer the OPA627, but that seems to be mostly due to personal preference - it beats everything else. That's why I'm tempted to go for the simplicity of single chip opamps.
If you want some LM4562, be sure to check out the group buy, it's just about to close.
Kittikun: Unless you need the TDMCA support and/or SPI control interface, I think you'll be better off with the 1794 over the 1796. It's better on just about every measurement; noise, dynamic range, distortion etc.
I've ordered one of www.ezdiyaudio.com pcb's to get an idea how the PCM1794 sounds, mainly being a fan of the TDA1541S2 I'll look forward to trying this one out to see how it compares
As soon as the pcb arrives and I can source all the parts I'll get it knocked up
As soon as the pcb arrives and I can source all the parts I'll get it knocked up
I too have always preferred the TDA1541A sound; I have a couple of modified Arcam Alpha 5+ CDPs.
However, I have just received a Rockna RD-2 DAC (PCM1796 based machine) and I am beginning to waver. It seems to offer the PRaT features of the 1541 with the detail and space of bitstream.
I would be interested in hearing the ezdac once you have built it.
However, I have just received a Rockna RD-2 DAC (PCM1796 based machine) and I am beginning to waver. It seems to offer the PRaT features of the 1541 with the detail and space of bitstream.
I would be interested in hearing the ezdac once you have built it.
Hi Lars,
I have no experience on up-sampling with the ASRC chips, but if you want to go 192K up straight then the TI/BB SRC4192 will work with 192K output master mode for minimum jitter.
Yes, the LM4562 did generate lots of good buzz recently, I may try them out later this year. Just for your information, Nat Semi did not stop at the Dual LM4562 and just released the Single chip counterpart, the LME49710 !!!
http://www.national.com/pf/LM/LME49710.html
With these kind of performance and pricing, it is going to be a killer. Since the LME49710 is only released two weeks ago , I've failed to locate any distributor stocking it. Have Fun !
Regards
Max
I have no experience on up-sampling with the ASRC chips, but if you want to go 192K up straight then the TI/BB SRC4192 will work with 192K output master mode for minimum jitter.
Yes, the LM4562 did generate lots of good buzz recently, I may try them out later this year. Just for your information, Nat Semi did not stop at the Dual LM4562 and just released the Single chip counterpart, the LME49710 !!!
http://www.national.com/pf/LM/LME49710.html
With these kind of performance and pricing, it is going to be a killer. Since the LME49710 is only released two weeks ago , I've failed to locate any distributor stocking it. Have Fun !
Regards
Max
D'oh! Maybe I shouldn't have bought all those 4562, then... But I think I'll survive.
Thanks for mentioning the SRC4192, it fits my application like a glove. And I do like TI.
There is the question of reference clock, though. Like you say, the SRC4192 will work in master mode, but the DAC (PCM1794) still needs a system clock at a multiple of 192 kHz, so I can't just run the 16.9344 MHz clock from the CD unit (probably a CD-Pro2) as total system clock. Any tips?
Thanks for mentioning the SRC4192, it fits my application like a glove. And I do like TI.
There is the question of reference clock, though. Like you say, the SRC4192 will work in master mode, but the DAC (PCM1794) still needs a system clock at a multiple of 192 kHz, so I can't just run the 16.9344 MHz clock from the CD unit (probably a CD-Pro2) as total system clock. Any tips?
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.