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

To pick up on a few points made over the holiday period:

1. I think we all agree the standard DDDac gives great sound: that's why we are all here. The main thing for me is the tireless nature of the sound which I have only found previously with other NOS solutions, and never with any DAC containing digital filters and oversampling solutions.

But there are many ways to modify it to improve the sound, by adding more boards or improving the capacitors regulators etc. Some of these mods eg the clock delay on the blue board may have implementational subtleties that need to be taken into account like the change of resistors if this proves to be the case.

2. When going away from the standard build it makes sense to measure what we can, to spot any obvious flaws, before concluding one version is better or worse than another but I agree that measurement is not the end of the SQ story necessarily. Doede's Wave psu measures very well indeed but both a friend and I recorded decent increases in SQ by using a custom CRC supply with a 5v super reg to feed the Wave. But I don't see why we can't use our own measuring tools and publish the results for peer review and feedback rather than trying to standardise measurements.

I have a standard two decker which is a great reality check to gauge the progress of my single board version.

3. Personally I am not in a one board or multi board camp. IME 4 boards is better than two, but my upgraded single decker is better than my 4 board standard build by a good whack. Unfortunately, I cannot even say it is just the improved psus/caps/regs that are giving me those results as I use the blue board with my single decker and that in itself is an improvement on the red board. In an ideal world of course I would step test every single dimension, as was done originally but life's too short.

Here's to a happy and successful 2015 to all on this thread,
David
 
@ Supersurfer :

I have found that the Wolfson SPdiff reciever implemation on the blue boards is not so good sounding. It might be that you are better off with a separate SPdif-I2S board (if the correct reciever ships are used).
For instance the chrystal CS8416 reciever chip with an input pulse transformer will probably sound a lot better. The Tent labs clock is not the bottle neck on the blue board design. I have used the Tent clock and shunt supplies for the SPdif part on the blue board.

Clock delay circuit is better on the blue board though!

Tks for the info. My current problem is to get the best from the SPDIF input because I currently use the old fashioned 1543-mkII SPDIF->I2S conversion board, and the results are better from the USB input (which delivers directly delivers I2S flow to the mainboard) than from the SPDIF input (which ses this SPDIF->I2S conversion card).

I am OK to install another SPDIF->I2S conversion card, but what is the best one to use ? I know the Crystal 8416 (it is an update of the 8414), but do you know any available SPDIF->I2S cards using a CS8416 ? (in order to feed the 1794 or 1794S mainboard)

Thanks
 
I strongly recomend new dddac users to let this exellent sounding dac get some time to play in their setup before start different types of tweaking.

When I compare dddac against other dacs, there is one important difference, and that is the non fatigue sound signature. And this is a very important element in my world. It doesn't help how clean and detailed the sound is if you got tired of listening after a while.

System setup is also critical here. Since we all have different preferences there is no simple answer. A revealing setup will not enjoy a dac which is on the edge clear and to focused on details. A system not so revealing, will probably prefer it.

I am running dddac without caps and trafos on output and have changed the I/V resistors to shinkoh tantalum 2 watt and bigger transformers ( 250va ). That's all. And to be honest, the change in SQ going from stock transformers to ten times bigger, is very small, if any.
To verify this I must change back to stock trafos, but since the psu boards have been cut of this won't happen.

Hope all of you clever tweak guys not read this the wrong way. Some tweaking is ok, and often a smart thing to do, but it's important to reverse things after a while, and this is often hard to do after desoldering and cutting traces etc,,,, sometimes, when reverse tweaks, I have been surpriced as it turns out that it actually sounded better before.

Keep on tweak this great dac guys, but don't forget take a step back and reverse things when this is possible, just to be shure. And remember, it takes ofter longer time to evaluate SQ than you might think. Things that in first place sounds great, can over time tip over, and you are on the "tweak train" again.

Your right Mr Smooth
The only problem, is some of these chaps mod at the speed of light, if you dont keep up then you get left in the dust :)
I'm kind of in tht position, problem is I've got too many other builds on at the moment ;-)
 
I am running dddac without caps and trafos on output and have changed the I/V resistors to shinkoh tantalum 2 watt and bigger transformers ( 250va ). That's all. And to be honest, the change in SQ going from stock transformers to ten times bigger, is very small, if any.

If you simply increase the size of the psu transformers with no other changes, I'm not surprised you got this result. Benefits come from moving to an unregulated psu, and using super regs.
 
Not had any email contact as such but I did order a clock from him just before Christmas and it arrived within a few days.

Also keep in mind that we have just been through almost 2 weeks of holidays and hopefully Guido has had a nice break too. Perhaps he is trying to catch up with emails after the holidays.
 
Hi

Has anyone had any email contact with tentlabs over the last couple of weeks. I have sent numerous emails regarding some shunt regulators that aren't correct but i have had no response. Thanks, Si

Hi,

Guido is a very busy man, next to his Tent Labs he is the co-founder of Grimm Audio (professional recording and monitoring equipment).
I am afraid that he is getting a bit overwhelmed with all the shunt orders for this project.
Every shunt needs to be adjusted to the correct current and voltage for this dac, and that takes some time.

If you have a current setting issue on your shunts than I might be able to help you. Just send me PM with the details.

Kind regards,
 
Thanks for the replies. I have tried the ormada email as well as info and orders @tentlabs. The regulators I have should be 8v but I'm only getting 4.6v out when supplied with 12v. Under no load I get 8.3v with a 12v supply. I think they may be 15v regulators, and wanted to see if guido kept records to verify this.
 
Hi Ian

I tried the raspberry pi via i2s into the dddac a few weeks ago. The sound was ok. I wasn't blown away to be honest. It was with only a wall wart supply though. I have a linear supply to fit in the case to power it but the dac is out of action at the moment whilst I sort the regulators. Will certainly revisit the pi though and spend some more time on it.
 
This surprises me David, I always thought that a regulated supply is preferable. Can you elaborate?
The DAC wants a neatly regulated power supply, but it has local regulators to take care of that. If you're using a shunt regs or similar here, what they thrive on best is a good, low impedance power supply. This is why people are hearing improvements from removing the resistors and using a bigger transformer in the standard power supplies. The shunt regs handle the accurate regulation of voltage, so as long as the power supply feeding them has filtered out any large spikes, it's more important that it offers a nice, free, low impedance supply than it is that it's exactly bang on a perfect regulated 12v at all times. An unregulated supply using just rectifiers, capacitors and chokes provides a real low impedance supply which makes the shunts happy, so they sing with sweeter voices :)
At least, that's what listening tests have shown.
An externally hosted image should be here but it was not working when we last tested it.
 
Member
Joined 2010
Paid Member
An externally hosted image should be here but it was not working when we last tested it.


Hi enrico
Here's a photo of one of the regulators in-situ. Not sure if it's clear enough to see what you need to see. If not I'll try again tomorrow. Thanks for your help.

Hi Undarra,
The picture is clear enough... as you can see is missing one part (the only one missing in the board). In that position you should solder a 30 ohm resistor to make the shunt working properly.

Let me know...

Regards,
Enrico
 
Hi Ian

I tried the raspberry pi via i2s into the dddac a few weeks ago. The sound was ok. I wasn't blown away to be honest. It was with only a wall wart supply though. I have a linear supply to fit in the case to power it but the dac is out of action at the moment whilst I sort the regulators. Will certainly revisit the pi though and spend some more time on it.

Not just me then Si. To be honest, in my system it sounds terrible. I have been using the HiFiBerry Digi + with the Pi into the SPDIF on the new 1794s mainboard for the last few days and it's been really quite impressive for such a cheap streamer.

Connected up the I2S tonight and it just doesn't sound right at all. I know the clocks on the Pi aren't up to much but the timing is completely off, sounds quite flat and recessed too. I've got one of ACKO's S03 re-clockers waiting to for parts to be built, I'm hoping that will at least bring it up to the level of the SPDIF input. Saying that, it might also help to use shorter cables too but I wouldn't have thought 150mm was excessively long.