A NOS 192/24 DAC with the PCM1794 (and WaveIO USB input)

Disabled Account
Joined 2002
Hello,
In the end someone will come with the right solution to your problem.
If nothing works why not ask Doede to take a look after all you are both in Europe.
I am happy i did decide to wait untill the big improvements are included into the DDDAC boards. Using a single dac board could be easy to troubleshoot but once more boards are interconnected and one board has an issue you will have a busy day finding it.
Soldering and desoldering in this set up will need serious skills and concentration all along the way. Replacing some caps will probably just take time and could be done done but adding extra circuits and erasing circuit traces gives me goosebumps in a negative way.
Soon i will just get 4 boards, maybe use some special caps, get Doedes output transformer and built an over the top power supply ( surely choke input)
Greetings, Eduard
 
yes, I am always happy to look at any DAC some one built and has issues. I get a few every quarter to check/repair....

just consider...

also, de-soldering decks is EASY when you have de-soldering station. A piece of equipment EVERY DIY should have I believe. I do this regularly...

poor man solution: cut the wires just under the PCB and remove the deck. you can solder back later the decks at those points easily as well
 
Dear UMarcus,
Hope I am not stepping on any toes.

Do you have a preference between the Wave IO, Amanero, JL Sounds and DIYINHK USB to I2S ?

The IAN FIFO is beyond by budget and abilities.

Thank you !

Hi,
i did not hear or test the Diyinhk USB.

My preference woulde be :

1. IAN FIFO
2. JL Sounds + Clock Board + Crystek
3. Wave I/O and JL Sounds
4. Amanero

As karahana already mention I could also confirm the JL Sounds clock board with crystek is the game changer here and make a clear and noticable improvment of the sound.
Beside the crystal as I correct construe an technical difference is that on the Wave I/O the galavanic isolation IC is after the clock generation against on the JL Sounds the isolation is before the clock generation / clock board. So generated clock has not traverse an additional stage.

I recommend the FIFO as best solution, improvment is again noticable ... on other hand cost is about 100 Euro higher than JS Sounds + Clock Board + Crystek.

Irrespective of your solution, please try to keep the I2S Cables as short as possible :)

Regards
Marcus
 
I just updated my website with a short reference to the new dac board with the Tentlabs Shunt and pin20 Current Source, so you can download the technical documents if you like.

or directly here ---> Click here for link to document

Audio creative is updating their web site as the kits are available now... Click to go to web shop

have fun :cool:
 

Attachments

  • dddac1794_pbt_4.jpg
    dddac1794_pbt_4.jpg
    526.6 KB · Views: 448
  • dddac1794_pbt_1.jpg
    dddac1794_pbt_1.jpg
    325.6 KB · Views: 436
Disabled Account
Joined 2002
triple wow

Hello,
Just did order a 4 board dddac kit to be used with my rega cdplayer. So oldfashioned i know!
Now i will need to find some time to construct it. Hope it will come with an updated manual but it surely will be easier than modifying the old boards with Tent shunts.
I honestly believe that the incorporated shunts will be more reliable than the diy solutions seen so far and adding the other shunts to these boards ( but there will be mods that will offer a better return on investment if we should believe the almighty Doede) will be easier because there is the right space available.
Will ask Doede soon for a pair of output transformers.
Greetings, Eduard
 
I just updated my website with a short reference to the new dac board with the Tentlabs Shunt and pin20 Current Source, so you can download the technical documents if you like.
Great!

I just have a question, the 7810 regulator is mandatory?
I'm powering my (2) dddac boards with low noise regulated 12V, one for left side, one for right.

Tent shunts can accept 12v directly?

Thanks for your work, ones again! :cheers:
 
Last edited:
Hi,
i did not hear or test the Diyinhk USB.

My preference woulde be :

1. IAN FIFO
2. JL Sounds + Clock Board + Crystek
3. Wave I/O and JL Sounds
4. Amanero

As karahana already mention I could also confirm the JL Sounds clock board with crystek is the game changer here and make a clear and noticable improvment of the sound.
Beside the crystal as I correct construe an technical difference is that on the Wave I/O the galavanic isolation IC is after the clock generation against on the JL Sounds the isolation is before the clock generation / clock board. So generated clock has not traverse an additional stage.

I recommend the FIFO as best solution, improvment is again noticable ... on other hand cost is about 100 Euro higher than JS Sounds + Clock Board + Crystek.


Regards
Marcus

Hmmm, I personally don't believe in synchronous reclocking - simply because it can not eliminate all jitter due to its design - it can only create a "fixed jitter". Maybe "fixed jitter" is preferrable over random jitter, I don't know, but to my mind "no jitter" is nevertheless better than "fixed jitter".

I did some measurements of the synchronous Acko S03 reclocker some time ago and was not really convinced:

http://www.diyaudio.com/forums/group-buys/227502-amanero-isolator-reclocker-gb-156.html#post4168978

It is, moreover, a real pity that all those jitter-killer solutions that can be found are not checked by means of measurements. To me, assuring that there is no jitter seems to be mandatory before trying to check for audible differences.

So, the real solution seems to be asynchronous reclocking by buffering the audio stream in a FIFO and clocking the I2S data out via a high precision clock (Crystek). This is quite complex as there needs to be some kind of logic that recognizes clock speed and handles buffer over/underruns, which can be done for instance in an FPGA.

The only solution I'm aware of is Ians reclocker, however, I could (not yet) find out how to get one - any clues in this forum?

Best Regards,
Hermann
 
Hmmm, I personally don't believe in synchronous reclocking - simply because it can not eliminate all jitter due to its design - it can only create a "fixed jitter". Maybe "fixed jitter" is preferrable over random jitter, I don't know, but to my mind "no jitter" is nevertheless better than "fixed jitter".

I did some measurements of the synchronous Acko S03 reclocker some time ago and was not really convinced:

http://www.diyaudio.com/forums/group-buys/227502-amanero-isolator-reclocker-gb-156.html#post4168978

It is, moreover, a real pity that all those jitter-killer solutions that can be found are not checked by means of measurements. To me, assuring that there is no jitter seems to be mandatory before trying to check for audible differences.

So, the real solution seems to be asynchronous reclocking by buffering the audio stream in a FIFO and clocking the I2S data out via a high precision clock (Crystek). This is quite complex as there needs to be some kind of logic that recognizes clock speed and handles buffer over/underruns, which can be done for instance in an FPGA.

The only solution I'm aware of is Ians reclocker, however, I could (not yet) find out how to get one - any clues in this forum?

Best Regards,
Hermann

Thank you for your feedback and measurement.

May be I get don't get your point, but ordere IAN's FIFO from here

Regards Marcus
 
I just updated my website with a short reference to the new dac board with the Tentlabs Shunt and pin20 Current Source, so you can download the technical documents if you like.

or directly here ---> Click here for link to document

Audio creative is updating their web site as the kits are available now... Click to go to web shop

have fun :cool:

Hi Doede,
that's fantastic.

Many thanks for your effort of support time and sharing that amazing DAC !

Regards
 
Great!

I just have a question, the 7810 regulator is mandatory?
I'm powering my (2) dddac boards with low noise regulated 12V, one for left side, one for right.

Tent shunts can accept 12v directly?

Thanks for your work, ones again! :cheers:

Of course that will work as well.... Guido Tent recommended the Pre-Regulation though... you can easily try it out for your self of course?

Can you reduce the 12 to 10 Volt? That would be even better for heat dissipation in the transistors
 
motherboard has 5 volt regulator. maybe the optional capacitors on the deck helps? i will try it and i wil check the soldering. when something would be wrong with the 1/2 clock, it would be on both channels, wouldn´t it? and why is it getting worse/for other volume-levels? thanks, mario

I had a very similar issue after upgrading my "red board" DDDAC with a 1/2 clock delay board. After reading the data sheets of the logic chips I decided the 3.3 Vdd supply should be fine, so I left it at 3.3 V. Then I got the crackle / hiss in one channel, but not the other. After I changed the voltage regulator to 5.0V, the crackle / hiss went away. Based on this experience I would not exclude the 1/2 clock delay as a possible source of your problem.

Did you swap the 100Ω / 1kΩ resistors on the DAC boards?
 
hi,
thanks for sharing your experiences. i have to desolder the decks (ordered a desolder-station yesterday :) and will check these things. the resistors are the right ones. hope to have good news in a few days...

While a desoldering station is probably nice to have, I just cut the wires to remove the DAC boards and swap the "delay" resistors on the DAC boards.
 
Of course that will work as well.... Guido Tent recommended the Pre-Regulation though... you can easily try it out for your self of course?

Can you reduce the 12 to 10 Volt? That would be even better for heat dissipation in the transistors
Thanks Doede,
I should be able to reduce to 10V, as I said, it is already regulated, but will power the right (or left) side of two boards at the same time, digital side included.

Well, I'll try!
Where is the fun if not experimenting :D
 
Doede, yet again great work and thank you!

I understand the 7810 is Reg the 12 to 10v to improve the dissipation in the tent shunt. I thought the whole idea of the shunt was to stop having regulation for improved low noose supply so by putting the 7810 there it is not helping with that.

What do I not understand please as I am missing something here? Thank you