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

Yes I know that, think also Rick McInnis suggested that. It's also on my list for testing. But because of the caps it's not easy to place the shunts there.

Sorry, I have no board with the 8V shunts soldered to the pads of the LF80. They have been from the beginning with the output at the pads of C18/C20.

From what I have read for years it is always best to have ANY reg as close as you can get it to the circuit. I was going to use the bypass caps and use their leads for an attachment point but agree it would be best to just do away with the cap entirely and use the cap's holes for mounting. After all, that is what Mr. Tent recommended. I thought it would add convenience and ease. Over the weekend I found it did not add any ease or convenience - it would be superior to just use the bypass caps "holes" and no caps.

The drawing of all traces is of immense help. I figured it was best to use wire for the other entry points but did not realize how the board traces ramble around. I was hoping that maybe the wire would not be necessary but now there is no question the wires would have to be better than the traces.

Reading a little more about the Vcom pins and their caps I see these are used for coupling and not storage - whether this will have an effect on the regulator is a mystery to me. I have asked the question of Mr. Tent who knows how to use DACs and I look forward to hearing what he has to say.

Seems like the traces to the Vcom caps are not unreasonably long but the fact they are daisy chained to the Vcc2L and R is of concern. We can connect with short wires to their respective bypass caps but those traces are still going to be in the circuit. Looks like another pair of spots for PCB surgery.

I thought that maybe a smaller value film might suffice but from what little there is to read about this on the internet I get the feeling this would result in a severe limiting of low frequency response. One way or another we are stuck with them. As SUPERSURFER mentioned he is going to try different caps in the circuit. I find very little to choose from that will fit without adding wires (antennas, for sure) for bypass caps.

There was talk that the Vcom caps can make a difference in the sound. I got the feeling the SILMICs might be fine here. A few cautioned that super low ESR caps are not a good idea in this position. There were a few mentions that 47uF was too small.

One can wonder if the MUSE might still be of use in front of the 8V regulator?

As far as a new board - that would be nice but I figure even with our modifications we are better off with this board until someone decides to offer something else instead of trying to breadboard a circuit ourselves. The ground place aspect, alone, is a great advantage. If we keep the extra wires in contact with the board we should be OK even if it will not be very attractive.

I like my circuits to be unenclosed and unshielded. Maybe there will be a need for the modified DDDAC to be enclosed? I hope not. All shielding has its price but as with all in audio compromise is required.

Long ago came to the conclusion that more space will be needed between the decks and with that we might as well use individual leads to each deck from the mainboard. My plan is to (eventually) have two decks above and two decks below the mainboard so these lengths will not get too long.

I will be using IE transformers. Though I considered for a moment using toroids but in combination with an IE isolation transformer between the "wall" and the toroids. But then I start worrying that this would lead to too much interaction between the two supplies and the whole point is to separate the digital aspect from the analog side.

Whoever makes their DDDAC the most unrecognizable from the original wins this contest! Is this another chapter in the old story of the silk purse from a sow's ear? Well, if it was a sow's ear we would not have had any interest in it to begin with.

The key to the design, the genius was how Doede implemented NOS with the chip and that is why it is not only worth the efforts but why the effort is rewarded.
 
Nice update Rick. Interested to hear ideas and recommendations for the vcom caps.

Still waiting.

Mr. Tent must get deluged with emails. Hoping by Thursday I hear from him. Especially since my question was included with an order.

By the way did any of you who bought his regulators take advantage of his "five pack" offer? I am wondering if this is really still in effect since his site looks the same as it did with I got some filament shunt regs ten years ago! Another question I cannot get him to answer and the option is not listed on his webstore. I figure as many of these things I will need eventually might as well take advantage.
 
Still waiting.

Mr. Tent must get deluged with emails. Hoping by Thursday I hear from him. Especially since my question was included with an order.

By the way did any of you who bought his regulators take advantage of his "five pack" offer? I am wondering if this is really still in effect since his site looks the same as it did with I got some filament shunt regs ten years ago! Another question I cannot get him to answer and the option is not listed on his webstore. I figure as many of these things I will need eventually might as well take advantage.

Waking him with an extra Email mostly works.
His is extremely busy with preparing shunts for us people!
Ed
 
After seeing all theses mods I wonder if it would not be better to just redesign the boards all smt affairs, much smaller in size and with all the components we don't plan to use left out? Doede maybe you can sell either fully populated or completely empty all smt boards?

This would get rid of empty component locations, unused vias, jumper wires and tall things acting as rf antennas, completely separate power supplies for analog/digital and maybe even left/right channels and keep the ground planes for digital and analog separated.

From this Analog Devices article: Staying Well Grounded :
Nevertheless, if a high-accuracy mixed-signal system is to deliver full performance, it is essential to have separate analog and digital grounds and separate power supplies.

on the first point, yes, this is on the short list for 2015. BUT before any design will be done, it is good, to go through all this testing and trying. that will help making a version were the shunts are on the board, still keeping things compatible with the current versions of course.... of course this will turn out to be a luxury higher budget version :rolleyes:

on the second point.... well, here the opinions are VERY different.... please do not start a discussion, as we will never come to a conclusion.

at least... look at the board. I have been trying to have return paths for analogue and digital, which is mostly below the trace, not to mix and cross..... I believe that is the best solution.
 
Why do you guys keep on modding an already well designed, well tested product? Doede is a guy that works almost like a scientist. He calculates and builds and listens as much as possible, trying out as much components as possible. When he comes up with a solution, I'm sure it's the best one and the best sounding he could get. Well, okay, hell, perhaps he knows there are better solutions, but perhaps they cost really a lot more !!
Anyway, I've had a little luck this month, got some money for free (always nice!) and I think I'm gonna spend it on his DDDAC, the full monty, with the Cinemagtransfo's. And I won't tweak anything because I believe in his skills. Thanks for all your time and research, Doede !
 
on the first point, yes, this is on the short list for 2015.
Thank you! Looking forward to the boards.
of course this will turn out to be a luxury higher budget version :rolleyes:
That is completely understandable!
on the second point.... well, here the opinions are VERY different.... please do not start a discussion, as we will never come to a conclusion.
No discussion, I already forgot I mentioned it... :eek:

Thanks for the answers Doede, much appreciated.
 
Why do you guys keep on modding an already well designed, well tested product? Doede is a guy that works almost like a scientist. He calculates and builds and listens as much as possible, trying out as much components as possible. When he comes up with a solution, I'm sure it's the best one and the best sounding he could get. Well, okay, hell, perhaps he knows there are better solutions, but perhaps they cost really a lot more !!
Anyway, I've had a little luck this month, got some money for free (always nice!) and I think I'm gonna spend it on his DDDAC, the full monty, with the Cinemagtransfo's. And I won't tweak anything because I believe in his skills. Thanks for all your time and research, Doede !

I agree with all what you are saying, but for me it means nothing negative that people discussing modifications here.

I got my DDDAC more than a year ago and a very long time I used it without any modifications. It's an excellent DAC. The best I have ever listened to. Before I also owned another DAC with the same PCM1794 chip which almost did cost twice as much, but wasn't half as good. So I know it's the good NOS design from Doede that is making the difference.

Now I am also doing some modifications (maybe it would be better to say upgrades), but it's not because I think something is wrong with this DAC. It's more that it is so good that it is worth to spend more effort and money in it.
The mods which are discussed here are on a very high level and some really cost a lot of money by using the best parts you can get. This parts make this DAC sound even better, but I am sure 90% or more of all audio equipment would sound better if it would be upgraded with such high quality parts. Other mods are also a matter of taste e.g. which caps to use, or made to fit the device better to the existing equipment.

Just wanted to say this because people which maybe only have read part of this thread might think that all this mods are necessary to get a good DAC. Thats not true, it is a very good DAC just without any mods.

Muug, I think it's a good choice if you buy this DAC. At least for me at was.
 
Member
Joined 2006
Paid Member
I agree with all what you are saying, but for me it means nothing negative that people discussing modifications here.

I got my DDDAC more than a year ago and a very long time I used it without any modifications. It's an excellent DAC. The best I have ever listened to. Before I also owned another DAC with the same PCM1794 chip which almost did cost twice as much, but wasn't half as good. So I know it's the good NOS design from Doede that is making the difference.

Now I am also doing some modifications (maybe it would be better to say upgrades), but it's not because I think something is wrong with this DAC. It's more that it is so good that it is worth to spend more effort and money in it.
The mods which are discussed here are on a very high level and some really cost a lot of money by using the best parts you can get. This parts make this DAC sound even better, but I am sure 90% or more of all audio equipment would sound better if it would be upgraded with such high quality parts. Other mods are also a matter of taste e.g. which caps to use, or made to fit the device better to the existing equipment.

Just wanted to say this because people which maybe only have read part of this thread might think that all this mods are necessary to get a good DAC. Thats not true, it is a very good DAC just without any mods.

Muug, I think it's a good choice if you buy this DAC. At least for me at was.

Thats my plan, replace a few caps and IV resistors. My DAC boards were missing the inductors, and I am going to get a few shielded inductors from mouser when I order some stuff. Some might come from partsconnexion.

As to the embedded streamer, I have currently a rPi, and BBB, and watching the thread about the Botic with interest, it sounds close to fruition. I am considering mounting the PSU units in a separate box, and mount the DAC, and Wave I/O and one of the streamers in the other box. That way I could open it up and swap back and forth if the streamer goes south. Or switch to an internal streamer that hooks up USB to the Wave I/O. All sort of options.

I am also trying to figure a way to get some sort of indicator from the computer to the front panel. With the Wave I/O board out of the loop, the BBB or rPi would have to output a signal for some LEDs or a display, perhaps with an arduino in the middle of that. Another cape for the BBB to run a display perhaps? I don't know enough about that to hazard a guess, and then as to whether it would coexist with the Botic.

Since Doede did the hard work of the DDDac, I can cogitate on other parts of the system.
 
I am also trying to figure a way to get some sort of indicator from the computer to the front panel. With the Wave I/O board out of the loop, the BBB or rPi would have to output a signal for some LEDs or a display, perhaps with an arduino in the middle of that. Another cape for the BBB to run a display perhaps?

I don't know much about the BBB. If it has a serial port you could use a display with a serial input. I am using this: GU140X16G-7003 Noritake | Mouser. It looks very nice.

You also will need a little program to output your messages. Think the BBB is based on Linux and it won't be difficult to write.
 

Attachments

  • 2014-06-11_18-50-12.jpg
    2014-06-11_18-50-12.jpg
    835.1 KB · Views: 594
  • 2014-06-11_18-50-30.jpg
    2014-06-11_18-50-30.jpg
    773.5 KB · Views: 561
I agree with all what you are saying, but for me it means nothing negative that people discussing modifications here.............The mods which are discussed here are on a very high level and some really cost a lot of money ..............This parts make this DAC sound even better..............Just wanted to say this because people which maybe only have read part of this thread might think that all this mods are necessary to get a good DAC. Thats not true, it is a very good DAC just without any mods.
I was thinking exactly the same ! , and that's why I posted my reaction. Thanks for understanding me. :D
 
Last time I checked this thread is contained within the DIY Audio forums and that implies experimentation by individuals on audio gear, not waiting for others to do what we want.

There are plenty of places where one can read about plug and play stuff

As I said earlier if we thought this device was ordinary who would bother exploring ways to make it better?

I will look forward to the revised boards but in the interim I will enjoy learning about how to make this device work better within my system. It will be interesting to hear how far apart a modified first generation will be from the second. And I will hope it is a significant step forward from what those of us of the obsessed DIY'ers come up with.

I recognize the value of SMD components but will miss the ease of trying out other components without having to use a magnifier!

All want this project to be a success in every way possible.

If it would be preferred should we start a new thread concerned with modifications?
 
Last edited:
My budget-modded single deck DDDAC is now complete:

Holco non-magnetic I/V resistors. 2x 250R in parallel
Tekdevices series regulators
12v Sjostrom super regulator / r-core trafo
Digital noise filtering caps

Total cost of all mods approx £120

Have to say it sounds considerably better than in standard form with almost no detectable background noise at full gain so very excited to hear what the other one is going to turn out like.