I've taken a look at the AK4118 and I can't figure out how it can drive the AD1865 correctly, given that the DAC requires simultaneous (both L and R channels) input and the AK4118 has only one data output which outputs L and R sequentially in all possible output modes. Its possible the designer has done a fudge and updates the L and R channels alternately. In any case to hack it to accept I2S doesn't look trivial to me, just from the point of view of additional logic being required.
The AK feeds I2S to a 74HC7266A. I think you can hack the lines between these two? This is mine (small attached pic on the left) but it seems there is another version now with different routing, regulators closer to the DAC IC and bypass caps on the output electros (large attached pic on the right). But given mine has serious issues (probably something oscillating) I can say for sure, don't buy it.
Attachments
Last edited:
Ah yes, you're right of course. I didn't even bother to check since it's going back. Thanks for pointing that out. I can check again when I'm home if needs be, or the pics should show the hardware settings. the underside only has caps so everything relevant is in the pic I think - no hidden traces under the IC cos of that cluster of caps under it. So, one transistor (?) and one resistor and pins 1, 3 , 8 , 9 all tied together. the rest is spdif in and power in related I think. The aliexpress ad still has the first design so better pics are there AD1865 decoding finished board r2r decoding nos mode DAC hifi audio good-in Digital-to-Analog Converter from Consumer Electronics on Aliexpress.com | Alibaba Group
Last edited:
I looked up the logic gate and I think you're right. I guess he uses the LR CLK to toggle data output between L and R, but if that's the case, it would seem to me they would either be out of sync or out of phase?
Without any data storage being used (like some shift registers) I reckon the best that can be done is updating the two channels alternately within the sample period. So they'll have a time delay between them of about 11uS. That's not out of sync but it is out of phase, very much like the original Sony CD player which time-shared one DAC between L and R.
11uS doesn't matter at all. It's like sitting off centre or moving your head when listening to music. And the accoustics in your room probably aren't that great anyway. But if you're super fussy perhaps it matters?
I've got old CD players with mono DACs and NOS DACs that multiplex the output. I don't think it matters at all. But that's just me.
I've got old CD players with mono DACs and NOS DACs that multiplex the output. I don't think it matters at all. But that's just me.
The multiplexing only happens on each word, not each bit, and there is a gap between each word, so I don't think the LR switching affects jitter as much as you might expect.
It's just going L R L R. Which makes it the slowest signal on the board.
But perhaps I'm wrong?
I never personally heard anything I'd call a jitter problem from these circuits.
I think transports can do more damage.
Last edited:
I want to add that passive crossovers in speakers do more damage to sound than jitter in DACs to put some perspective.
Furthermore, amplifiers trying to drive loads with varying capacitance and inductance does terrible things to the sound.
Since I went to active, most sonic issues I previously attributed to the DAC actually turned out to be passive crossover and amp driving whacky load issues!
Furthermore, amplifiers trying to drive loads with varying capacitance and inductance does terrible things to the sound.
Since I went to active, most sonic issues I previously attributed to the DAC actually turned out to be passive crossover and amp driving whacky load issues!
Last edited:
I have a first order between the mid and tweeter using mundorf supreme silver/oil and Jantzen cross coil, and I bi-amp them with isobaric woofer pairs. My speaker issues largely relate to the room and listening position and like you, I'm well aware that speakers are the biggest issue. I am going to get a MiniDSP at some point and tri-amp. One of the reasons I've come back to listening to DACs again is to work out what to try with the MiniDSP.
USB is a pretty horrible source of jitter - what xo are you using for the CM108? These Taiten OX-U are very impressive with very low phase noise and very very cheap for that they are ?????? 12M ????Taitien-???
XMOS with a low phase noise clock works well, as does CM6631a. I just got an SA9227 to try and that will just leave me the Amanero to try.
No xo just a very cheap low profile 12MHz XTAL.
I have CM6631A too, that's not adaptive its async.
a pierce oscillator type thingamy? I have tried low phase noise xo's with PCM270X, CM108, CM102, pci-e cards like CMI8768, ASRC like SRC4392, AD1896, and many many more digital devices ....it's my go to standard improvement if the gear doesn't have one already, along with clean power dedicated to the xo ..... there's a benefit every time.
Is there a reason you can't use async? drivers?
Last edited:
- Status
- Not open for further replies.