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



Thanks Ed and Chanh. it's more or less what I was suspicious of. There seems to be a fault occurring, in/on the I2S interface either due to the WaveIO not starting correctly or to slow.
Doede, do you have a suggestion on what might be going on? You wrote before that the PCM1794 don’t like it when they don’t see a I2S signal, but I didn’t think they would spit out white noise. Or could it be a case of the Wave IO not powering up correctly?
 
Too many things can cause this, I had instances where it was the pc, the power supply or indeed the board.

Just for the sake of excluding causes, can you run the waveIO on usb power for a while? See what happens. Also do you have an opportunity to use another pc system? May be you have one staying around or use a note book pc or what ever. I had some one with problems, who just re installed his operating system and all was fine.

It is all frustrating, but there are many variables and hence hard to point in one direction.

I had one issue, some one had a WaveIO working on PC and not on a Mac. Very strange... He sent it to Dick and later to me and you guess right, at our place all worked fine.

One question, if the noise burst comes, does it go away or you need to do some thing? Stop music and start again, or do you have to reset the WaveIO?

I hope you get the point on how to approach things like this. Good luck!
 
dwjames,

First replace the 1000 ohm resistors for R7 and R9 with 100 ohms to get rid of the slight delay caused by the 1000 ohm resistor. I think that NOT doing this with my initial test with 3.3Volts is what caused a negative first impression.

Try the 1/2 chip delay first with the 3.3V Vreg setting. After the circuit WORKS and you listen to some familiar recordings then switch your Vreg to 5.0 volts. The 2 chip circuit adds almost no extra load to the Vreg. According to the data sheets the 74VHCxxxx chips switch state much more quickly when operating at 5.0 volts.

I am eager to read your impression of any difference in sq you can hear from adding the circuit and does 5.0 volts give any more improvement?

My engineering mind cannot figure out why new components, circuits, PS, etc in the digital domain of 1's and 0's have to go through a burnin period, but my ears tell me this is necessary. Anyone want to pontificate on this?
That all makes sense, thanks 🙂

Does anyone know how critical the 100ohm value is for R7, R9 on the dac deck and R13 (in the BCK line to the 2 new chips)?
Could I use 120R or 133R here?

thanks,
James
 
Cheers guys. I just cracked on as I'm really impatient 😛
I figured if Ross is using a 121r resistor on the bck feed on his clock mod board, and if Nige has his resistors replaced with wire jumpers on the dac board, then it's not too critical. So I built up using 120r for both.

An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.


Works first time! 🙂 very happy with that and very impressed with the improvement in clarity! Highly recommended! 😀
 
Cheers guys. I just cracked on as I'm really impatient 😛
I figured if Ross is using a 121r resistor on the bck feed on his clock mod board, and if Nige has his resistors replaced with wire jumpers on the dac board, then it's not too critical. So I built up using 120r for both.

3d8a066cbc9798c1cdc81a3a61a5a963.jpg
77c536df10ece1c38efb762b5cc2ec87.jpg
f86a0b91853f93056fd43fd01e594aa1.jpg
6a6db34eab6c93a576cbef37130a26a2.jpg
191dd1df5e58872659c97d357d19a866.jpg
51943f303e908b37938912bfe2a533b8.jpg
918231cef63fcadfb820bdbc4d300411.jpg
28a96f0e9bb86909e21ab88939e9a481.jpg
53af2a8ef7bc1cfa0c816481b2e3a208.jpg


Works first time! 🙂 very happy with that and very impressed with the improvement in clarity! Highly recommended! 😀

great stuff
well worth the cost 🙂
 
given the Dac module-digital and analogue- consumes around 100ma I am wondering why you guys are using the filament chokes with high amperage and low inductance rather than lower amperage higher inductance?
David

I think it is because folks were wanting to use the same transformers they had.

If you went with high resistance you would lose voltage and have to get another transformer.

Nonetheless, low resistance chokes usually sound better but who knows in this application if it would make a difference.

As G600 says if you use more than one card you could draw some real current but with separate supplies for digital and analog you would have to have lots of cards to approach 1A for either "side".
 
dwjames:

Welcome to the club of ugly boards making better music.
Now to hear about 3.3V versus 5.0V..........
It's an ugly creature with a beautiful voice 😀

Great timing too as I have a couple of music loving friends coming round this evening for a musical appreciation session 😎

I'm going to quit changing stuff for a little while as I've been doing far too much soldering and not enough enjoying the music, but I'll check out the 3.3v vs 5v for the mainboard soon.

Last couple of mods I've done just to finish off are to change the 220pf polystyrene caps across the RCAs to 150pf MKP caps
IMG_20141007_110718.jpg


Then 1 final mod I've just made was to add a wire to join the ground from my Raspberry Pi power supply to the star ground for my dddac power supply and this has killed off the last bit of hum and buzz I had, so I'm pretty pleased with the improvement there, certainly for the cost and time involved 🙂
I'd definitely urge others who have a couple of different power supplies to give that a try running extra wires joining all the grounds together to a point in the case before they get to the dac. It's probably common practice, but hadn't occurred to me until just now and it's just taken the noise floor right away so the silences are truly silent.
Man it's sounding good now :sax::sax::sax::sax::sax::sax::sax:

Has anyone who's changed their dac regulators to shunt regs or similar experimented with running wires and star grounding the grounds from the regs straight back to the power supply ground rather than going via the boards? Nige mentioned it a while back and I've been wondering....
 
Just ordered the "new" mainboard from AUDIO CREATIVE.

Hoping I get credited for the VAT and the clock! If one waits long enough to post a note they can update it - well, they credited the clock but not the VAT - I thought this was settled long ago. I have never had this problem with any other European seller, Why the problem with AUDIO CREATIVE? I hate having to deal with this kind of thing. Only word for it is stupid.

Is anyone else as annoyed as I about "handling charges"? I thought these were things one would build into their selling price. Next thing they will add "energy costs" to the invoice.

Not the easiest webstore I have dealt with.

Glad the stuff is available just wish it did not take three attempts to get the order to take.

For those who decide to make an order - make sure you first "unclick" the box for a different shipping address. If you go too far with this "unclicked" you will have to start over completely.

A cautionary tale ...
 
I think something that needs to be added to this project is a polarity switch.

Getting the polarity right is important for the best possible sound.

I will be placing a switch at the output of the DAC board - assuming a balanced output, if you use the single ended route I do not think the switch will work. But, I am not sure.

John Broskie at TUBECAD offers a good rotary switch for this purpose or one can easily come up with one of their own. I used to use a pair of toggles for this. The switch allows changing both channels at once.

cPLAY offered a polarity switch within the player. Unfortunately, MQNplayer does not and one can understand the minimalist approach of sbgk so a switch is needed. Not sure about other players others are using. If you have built in polarity switching, good for you.

Digital recordings do not have a protocol for signal polarity. LPs are bad enough but digital recordings can change polarity with different tracks.

Once you become attuned to the difference this makes you will not mind having to make adjustments.

The difference can be, on many recordings, substantial. Incorrect polarity tends to bunch the sound between the speakers and when the change is made there is a broadening of the sound. Solo instruments and voices are another quick indicator of correct polarity. Everyone who does this seems to have their own first indicator. This is a skill worth developing.
 
I think it is because folks were wanting to use the same transformers they had.

If you went with high resistance you would lose voltage and have to get another transformer.

Nonetheless, low resistance chokes usually sound better but who knows in this application if it would make a difference.

As G600 says if you use more than one card you could draw some real current but with separate supplies for digital and analog you would have to have lots of cards to approach 1A for either "side".

My thinking was I would rather shed some volts in a choke than add another resistor. My brief try of the 15h choke instead of the 10uh one was very promising sound-wise but I did lose some output used that way without adjusting the voltage
 
Has anyone who's changed their dac regulators to shunt regs or similar experimented with running wires and star grounding the grounds from the regs straight back to the power supply ground rather than going via the boards? Nige mentioned it a while back and I've been wondering....

Just to be clear: do you mean simply commoning all the negs from the regs and psus etc to a central point or doing that and then connecting that star point to mains earth?
 
Just ordered the "new" mainboard from AUDIO CREATIVE.

Hoping I get credited for the VAT and the clock! If one waits long enough to post a note they can update it - well, they credited the clock but not the VAT - I thought this was settled long ago. I have never had this problem with any other European seller, Why the problem with AUDIO CREATIVE? I hate having to deal with this kind of thing. Only word for it is stupid.

CORRECTION - Turns out I am the STUPID one - the VAT was deducted at checkout and I missed noticing this. My dollar/euro calculator was not working very well this morning; the one in my head.

I tried to edit my post but the time had run out.

Mea culpa and my apologies to AUDIO CREATIVE.
 
My thinking was I would rather shed some volts in a choke than add another resistor. My brief try of the 15h choke instead of the 10uh one was very promising sound-wise but I did lose some output used that way without adjusting the voltage

You ABSOLUTELY should not use any resistors. If you need to shed volts you are correct - use the choke's resistance.

Brian Lowe HATES the thought of resistors in the path leading to his regulators!

You are on the right path. Chokes store energy, resistors simply waste it.