I have two pieces of CS8412 and I believe one of them sounds better ( more open ) with 4 DACs per channel.
With one DAC per channel I'm not sure yet.
Also with 4 DACs per channel there is a click tone every second only with the better CS8412 in silent passages. That gets worse ( faster ) when I touch the data line.
With one DAC per channel there is no such problem with that chip.
For sound comparison I make a fast change of CS8412 while power is on and music playing...
With one DAC per channel I'm not sure yet.
Also with 4 DACs per channel there is a click tone every second only with the better CS8412 in silent passages. That gets worse ( faster ) when I touch the data line.
With one DAC per channel there is no such problem with that chip.
For sound comparison I make a fast change of CS8412 while power is on and music playing...
The better chip has date code 9924 and the worse one 9442.
So the newer chip sounds better but seems to have less drive capability ???
My non os setup is:
Word clock goes to one HC04 inverter input, output goes to 4 PCM56 pins of one channel & another HC04 input, output to 4 inputs of other channel.
Each other line goes to 8 PCM56 inputs directly.
Before I tried word clock to one HC04 inverter & 4 PCM56 pins of one channel, HC04 output to 4 inputs of other channel and it had more problems with clicking noise.
Do 8414/16 sound better ?
I found a smd date code 9529, is it a new one ?
So the newer chip sounds better but seems to have less drive capability ???
My non os setup is:
Word clock goes to one HC04 inverter input, output goes to 4 PCM56 pins of one channel & another HC04 input, output to 4 inputs of other channel.
Each other line goes to 8 PCM56 inputs directly.
Before I tried word clock to one HC04 inverter & 4 PCM56 pins of one channel, HC04 output to 4 inputs of other channel and it had more problems with clicking noise.
Do 8414/16 sound better ?
I found a smd date code 9529, is it a new one ?
I have no drawn schematic of my DAC, it is the standard receiver schematic with inverter for word clock on one channel. Passive I/V 500 ohm per one DAC chip. Transformer with small cap on secondary for filtering + standard op amp output buffer.
Now I'm fiddling with 4 parallel chips per channel to cancel DLE without using MSB adjustment pot.
Now I'm fiddling with 4 parallel chips per channel to cancel DLE without using MSB adjustment pot.
Bernhard said:
I suppose as you are not one of those who claim to be able to hear picoseconds of jitter it doesn't matter but I would still get rid of the delay. It is very easy with 16 bit dacs.
My remote viewing skills are not what they used to be when the spoonbender and I were in business but I'll give it a shot.
If it is just one CS8412, then perhaps it is out of spec, damaged or dying. Try using a buffer. If both are causing problems then perhaps one of your PCM56's is at fault. Another thing to do is look at the signal on the scope. Mind you, driving 8 data inputs without a buffer is asking for trouble.
My experience was that CS8414 sounds better than CS8412. I was using CS8414 on SOIC to DIP converter board so I could switch between them.
My CS8412 was made in Taiwan and I can't remember its date code. The CS 8414 was probably made in Korea.
To describe in subjective term, the CS8412 sounded a bit mechanical compared to the CS8414. DAC chip used was tda1541a.
No idea about 8416.
My CS8412 was made in Taiwan and I can't remember its date code. The CS 8414 was probably made in Korea.
To describe in subjective term, the CS8412 sounded a bit mechanical compared to the CS8414. DAC chip used was tda1541a.
No idea about 8416.
Hi guys
Where would one check this date code?
If the last 4 digits below CS8412 indicate date, then the spare I have is 0303, is that possible?
On the bottom it says Taiwan 106365-001
I am really temted to plug it in the old Theta DAC (an early 90s design Pro Basic II) and see what happens....
Where would one check this date code?
If the last 4 digits below CS8412 indicate date, then the spare I have is 0303, is that possible?
On the bottom it says Taiwan 106365-001
I am really temted to plug it in the old Theta DAC (an early 90s design Pro Basic II) and see what happens....
Right, done, listened, used wife who did not know what was what....
Anyhow, difference is marked, initially I preferred 0303 vs 9212, wife did not, felt the reverse. I recently changed the AD829 I/V to LM6171 but was never sure and felt uneasy.
Changed those back and listened again, the clarity returned and a fuzz was gone. I prefer AD829 in the Theta aplication. Decoupling is quite well done as standard btw so I doubt it is that.
Then we listened again to the two 8412 chips.
We are in agreement, the 0303 sounds more spacious and clearer then the 9212, by quite a margin but it is not raspy or bright.
Lovers of the Meridian sound (talking 602/203 DAC vintage stuff here, not sure how their new stuff is) may like the darker slightly warmer presentation of the old chip. imo, it is veiled in comparison.
This is a worthwhile upgrade imo (easy as big as interconnect cable) and I now wonder what an 8414 would do......
Anyhow, difference is marked, initially I preferred 0303 vs 9212, wife did not, felt the reverse. I recently changed the AD829 I/V to LM6171 but was never sure and felt uneasy.
Changed those back and listened again, the clarity returned and a fuzz was gone. I prefer AD829 in the Theta aplication. Decoupling is quite well done as standard btw so I doubt it is that.
Then we listened again to the two 8412 chips.
We are in agreement, the 0303 sounds more spacious and clearer then the 9212, by quite a margin but it is not raspy or bright.
Lovers of the Meridian sound (talking 602/203 DAC vintage stuff here, not sure how their new stuff is) may like the darker slightly warmer presentation of the old chip. imo, it is veiled in comparison.
This is a worthwhile upgrade imo (easy as big as interconnect cable) and I now wonder what an 8414 would do......
In general CS8414 ( SMD only ) sounds less good than recent DIP versions of CS8412...
Bernhard, if you need 2002 version of CS8412 please send me a PM. Revision G if I remember correctly. I have some new/unused chips left.
Guillaume, did you also change the loop filter to Monkeysects version ? Very good mod, costs near to nothing and really is better.
Wildmonkeysects loopfilter for CS8412: 3300pF between filt and Agnd, 220nF and 470R in parallel to the 3300pF cap.
Use good quality caps and build this as small as possible and/or with shortest possible lead wires. I use 3.3 nF SMD caps right between the pins so I can use the existing solder pads for the 220 nF and 470R.
Bernhard, if you need 2002 version of CS8412 please send me a PM. Revision G if I remember correctly. I have some new/unused chips left.
Guillaume, did you also change the loop filter to Monkeysects version ? Very good mod, costs near to nothing and really is better.
Wildmonkeysects loopfilter for CS8412: 3300pF between filt and Agnd, 220nF and 470R in parallel to the 3300pF cap.
Use good quality caps and build this as small as possible and/or with shortest possible lead wires. I use 3.3 nF SMD caps right between the pins so I can use the existing solder pads for the 220 nF and 470R.
Wildmonkeysects loopfilter for CS8412: 3300pF between filt and Agnd, 220nF and 470R in series together parallelled to the 3300pF cap.
So the original loopfilter with different values with a new cap parallel to it. I use 3.3 nF SMD ceramic caps soldered right between the pins and Wima MKS ( = MKT ) 5 mm for the 220 nF with good results. Better than without the mod anyway. For those who want to use big caps, beware: the bigger the worse here. Those pins are sensitive. See the datasheets of later Crystal CS chips.
This mod is very old. I am surprised you don't know it
So the original loopfilter with different values with a new cap parallel to it. I use 3.3 nF SMD ceramic caps soldered right between the pins and Wima MKS ( = MKT ) 5 mm for the 220 nF with good results. Better than without the mod anyway. For those who want to use big caps, beware: the bigger the worse here. Those pins are sensitive. See the datasheets of later Crystal CS chips.
This mod is very old. I am surprised you don't know it
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.