Did you get your DAC from Wei Zhou (snow46_8)? It's actually a 'she' but definitely get in touch with her regarding the inputs and jumper positions etc. She always replies.
From Snow
I bought from Snow. I sent her an email yesterday to Snow468 at gmail but haven't heard back yet. So I might have to look at the circuit and install a jumper wire to get the coax working with the 9001.
I bought from Snow. I sent her an email yesterday to Snow468 at gmail but haven't heard back yet. So I might have to look at the circuit and install a jumper wire to get the coax working with the 9001.
thx for the quick how-to, Rich, but i'm gonna go for the trafos. but being in china, i have no access to premium products unless i pay premium in shipping also. so i'm gonna go for several cheap alternatives and treat them like lottery tickets.
on this page, several seem like an o.k. candidate- A, B, C, D, E i would appreciate further inputs.
as for the listening fatigue, it's turned out to be a rather serious deal. i think i induced myself a temporary hearing loss from listening to one song on the dac yesterday. one song! (9 min long though) this morning my ears feel numb and my hearing has definitely reduced a notch.
i'll be ordering 4397+8416 pair since on the earlier part of the thread their measurements showed to have a rolled off highs. and also so that i can try the trafo mod right away. i was not fond of the cleaned and waxed sound of the dac- it was way too artificial for my liking. whether it was due to the op-amp or the 1798, we'll see/hear.
i've only found one account of listening fatigue on this thread and the poster's comment was regarding the 5532, saying that there 'seemed' to be a listening fatigue. i just found out yesterday that living a quiet life full of Chinese herb consumption (ginseng, reishi, ect.- daily) had restored my hearing greatly. i was able to hear clearly the 16khz tone on the test which was the highest freq. offered. i could only hear up to 12khz back at the states two years ago. hooray!
. so my guess is that whatever the listening fatigue is, it's a problem in the high frequency region, and would look much more serious when a detailed measurement is done.
on this page, several seem like an o.k. candidate- A, B, C, D, E i would appreciate further inputs.
as for the listening fatigue, it's turned out to be a rather serious deal. i think i induced myself a temporary hearing loss from listening to one song on the dac yesterday. one song! (9 min long though) this morning my ears feel numb and my hearing has definitely reduced a notch.
i'll be ordering 4397+8416 pair since on the earlier part of the thread their measurements showed to have a rolled off highs. and also so that i can try the trafo mod right away. i was not fond of the cleaned and waxed sound of the dac- it was way too artificial for my liking. whether it was due to the op-amp or the 1798, we'll see/hear.
i've only found one account of listening fatigue on this thread and the poster's comment was regarding the 5532, saying that there 'seemed' to be a listening fatigue. i just found out yesterday that living a quiet life full of Chinese herb consumption (ginseng, reishi, ect.- daily) had restored my hearing greatly. i was able to hear clearly the 16khz tone on the test which was the highest freq. offered. i could only hear up to 12khz back at the states two years ago. hooray!
. so my guess is that whatever the listening fatigue is, it's a problem in the high frequency region, and would look much more serious when a detailed measurement is done.PreSapian - when you say 'trafos' do you mean power supply transformers. or output transformers?
If you want to try a good power transformer get this one. Highly recommended:
50W R Core Transformer,115V//230V, for Preamp,DAC/R26-x on eBay (end time 02-Sep-10 07:57:32 BST)
The model you need is R26-34 which has both 9V (for digital) and 15V (for analogue) windings. It would be better than the ones on the page you linked to. Also it's in China.
If you meant to try output transformers then you need 1:1 line level output transformers for the CS4397.
If you want to try a good power transformer get this one. Highly recommended:
50W R Core Transformer,115V//230V, for Preamp,DAC/R26-x on eBay (end time 02-Sep-10 07:57:32 BST)
The model you need is R26-34 which has both 9V (for digital) and 15V (for analogue) windings. It would be better than the ones on the page you linked to. Also it's in China.
If you meant to try output transformers then you need 1:1 line level output transformers for the CS4397.
maybe this is the answer to my listening fatigue...
though i'm too ignorant to even take a guess at why xtr3me's measurements differed so much from the others posted here, if this is by any chance the same symptom with my unit, then i think it certainly explains why my ears were hurting after a while.
the designer, mr.liu sheng, says that the op-amps are at fault but xtr3me's measurements were made with output trafos. and since the curve in his measurement aims straight up beyond 20khz, it can't be pleasant for my dog either... poor kid
i should also note that he vacated the living room when i started running the dac yesterday. if this is an innate character of the dac that's gone unnoticed to those who hasn't used it on ribbon tweeters, then i might have to give away this dac and restart my hunt for another giant killer
AK4395 vs DIR9001/ CS43122
So with a brief comparison, my first impression is that the DIR9001/ CS43122 sounds better than the AK4395 in my DEQ2496 even though there is a slight hiss issue right now with the big dac board and I can't use the excellent sounding upsampler board due to it causing an even louder whistle. Is this LM3-V10 still the best deal as a platform for a direct out CS4398 or are there any other boards available that may be a better starting point for mods?
So with a brief comparison, my first impression is that the DIR9001/ CS43122 sounds better than the AK4395 in my DEQ2496 even though there is a slight hiss issue right now with the big dac board and I can't use the excellent sounding upsampler board due to it causing an even louder whistle. Is this LM3-V10 still the best deal as a platform for a direct out CS4398 or are there any other boards available that may be a better starting point for mods?
There is a daughter board with PCM1798. It's excellent starting point. Last weekend I compared my PCM1798 DAC (tube output stage) with dCS Elgar, dCS Puccini and McIntosh MCD500. And PCM1798 DAC sounds best.
The rest of the system was Sonus Faber Guarnieri Hommage, BAT VK60, my DIY tube preamp, dCS Verdi SACD/CD transport.
And the second system in far bigger room was Sonus Faber Amati Hommage Anniversario, McIntosh MC1000 monoblocks, McIntosh C1000 preamplifier, McIntosh MCD500 as a transport.
Last edited:
Hey Bill,
did you read my last post about filtering? Per Lundhal did not suggest an RC filter but just a 7K resistor on his LL1690. I wonder which is the best way to go and why so many point of view.
Best
P
1798
I was going to try the PCM1798 next. What is a good AND convenient I/V conversion circuit.
Different trafos. You can't compare apples to oranges. LL1690 was designed as input trafo. Monacor is an output trafo, completely different. I've repeated this so many times I can't even guess how many.
Here's another mod I tried which works very well. It's complicated but I'll try and explain:
I installed a dedicated 5V supply for the DAC VREF pin. On our standard DAC, the VREF shares it's input with the 5V analogue input on pin 22. Now having a high quality dedicated supply for VREF along with decent decoupling to REF_GND is a well known trick and pays dividends.
Using the space vacated by the +12V LM317 regulator, I dropped in another 5V reg booster. Removed the trimmer and fitted setting resistors to get 5v. There is a circuit trace that runs from the new regulator output cap to pin 17 of the adapter, I cut this and connected the positive pin of the output cap to pin 15 of the CS4398 adapter with wire under the board.
On the CS4398 adapter the VREF of the SMD chip (pin 17) is connected to pin 15 of the adapter. It then runs a trace to pin 28 and from pin 28 to pin 22 under the main board (VA +5V input).
I cut the track from adapter pins 15 to pin 28 and fitted decoupling capacitors from 15 to 16 (VREF to REF_GND).
It definitely works and has improved depth and possibly more bass and midrange clarity.
Pics:
New 5V regulator booster in +12V position:
Wire connecting output cap positive pin to adapter pin 15:
Decoupling on adapter board, note cut in track near edge:
I installed a dedicated 5V supply for the DAC VREF pin. On our standard DAC, the VREF shares it's input with the 5V analogue input on pin 22. Now having a high quality dedicated supply for VREF along with decent decoupling to REF_GND is a well known trick and pays dividends.
Using the space vacated by the +12V LM317 regulator, I dropped in another 5V reg booster. Removed the trimmer and fitted setting resistors to get 5v. There is a circuit trace that runs from the new regulator output cap to pin 17 of the adapter, I cut this and connected the positive pin of the output cap to pin 15 of the CS4398 adapter with wire under the board.
On the CS4398 adapter the VREF of the SMD chip (pin 17) is connected to pin 15 of the adapter. It then runs a trace to pin 28 and from pin 28 to pin 22 under the main board (VA +5V input).
I cut the track from adapter pins 15 to pin 28 and fitted decoupling capacitors from 15 to 16 (VREF to REF_GND).
It definitely works and has improved depth and possibly more bass and midrange clarity.
Pics:
New 5V regulator booster in +12V position:
Wire connecting output cap positive pin to adapter pin 15:
Decoupling on adapter board, note cut in track near edge:
Hi Ryssen
With your shunt regs the DAC is obviously getting a very high quality supply as it is. However a separate supply to the VREF is well worth a go. The higher quality (low noise) the reference voltage is, the better.
You could try running a connection from an existing shunt reg initially to pin 15 if you cut the track on the adapter board. But for a quick and cheap solution, do try the regulator boosters with a dedicated rectifier / cap / filter supply board.
All the digital voltage rails in my DAC have their own upgraded supplies. It really works very well.
Here is a layout for a low current regulator booster:
The booster is a VBE based on the PFM flea circuit. It works.
With your shunt regs the DAC is obviously getting a very high quality supply as it is. However a separate supply to the VREF is well worth a go. The higher quality (low noise) the reference voltage is, the better.
You could try running a connection from an existing shunt reg initially to pin 15 if you cut the track on the adapter board. But for a quick and cheap solution, do try the regulator boosters with a dedicated rectifier / cap / filter supply board.
All the digital voltage rails in my DAC have their own upgraded supplies. It really works very well.
Here is a layout for a low current regulator booster:
The booster is a VBE based on the PFM flea circuit. It works.
DIR9001 input is switched
There is an electronic switch under the bottom side of the DIR9001 board so I assume it is set up to duplicate the selection based on the SA and SB jumper settings the same as the CS8416. I don't know why mine is not switching to coax with SA and SB set to 0. I may just end up lifting the switch and soldering a jumper wire to select.
It is also weird that the DIR9001 requires an opamp in the input path.
.
.
There is an electronic switch under the bottom side of the DIR9001 board so I assume it is set up to duplicate the selection based on the SA and SB jumper settings the same as the CS8416. I don't know why mine is not switching to coax with SA and SB set to 0. I may just end up lifting the switch and soldering a jumper wire to select.
It is also weird that the DIR9001 requires an opamp in the input path.
.
.