Today was a crazy day... We just finish to move to the new apartment and I just read the post. It is really long time that I don't fell so tired
Anyway, thanks Doede 🙂
The courtesy is mine because I am the one that I send you the board but all the credits should go to Frankie (palmito) and Ross (calsors). We did a great Team Job and I really enjoy this experience.
I want to say Thanks! to Frankie for the amazing support in the first design phase and Thanks! to Ross for continuous support during the design of the buffer circuit and during the first tests.
I hope you guys will enjoy your time with all the DACs.
All the Best,
Enrico
Stefan, with all the respect for the Koreans but tomorrow you guys will have an Italian/American dac at hands 😀
Ciao
Enrico
It’s only been yesterday, so this isa pretty quick follow up, but I woke up this morning with these words in my head so I better put everything on paper now.
As Supersurfer and Doede already commented on over the last couple of days. Yesterday, James, Doede, Stefan, Cees Pel & myselve, met up for a DDDAC bake-off/comparison/listening session.Especially James and Doede had travelled a considerable distance to make it to Stefan’s house in Utrecht, so a spot of lunch came first, but obviously the subject turned pretty quickly to all things dac related.
Between the five of us, we had brought along a lot of gear and options to discuss and listen to..
As soon as lunch was cleared the various dac’swhere spread out onto Stefan’s dinner table and we all got the first hand explanations of the individual builds.
For starters, in a nutshell, the three none stock tweaked dacs on offer were James his freshlyrebuild single boarder & Stefan’s and my four board DDDAC’s. There are a lot of small difference among then, but main setups involved:unregulated main supplies, different cap schemes mainly consisting of Silmic’s, CCS’s Tenslab shunt’s, and embedded source players, isolated, clocked/reclocked by Acko’s S03(BBB’s/synchronisefor James and me and RPi/A-synchronise for Stefan.)
Besides these three dac’s, Doede had brought along a trunk load full of dac’s and peripherals. He had brought a box standard one board DDDAC, a 8-board stack (all stock), Doede’s supplies but with a custom made Sowther output transformers (pictures of this build are on the first page of this thread). But besides these familiar devices was an intriguing looking breadboard filled with “stuff”… and last an DDDAC dac board made by Enrico with an integrated buffer stage (amongst other things).
We all gave a brief explanation and had a brief discussion about all these, during which Doede explained to us the basic concept behind his new discrete DSD dac prototype. The concept sounded very simple and elegant, my initial thought was that of slight disbelieve, a breadboard full of antennes is hardly ever going to “look” like it will sound good….(or maybe that’s just me) ;-)
From the outset it was obvious it was going to be hard to listen to all the options we had at hand and being able to hear & pinpoint what the effects of these build choice differences made. But since Doede had brought along a single “stock” build, this was a good place to lay the base line.
Stefan is the person to explain about his homegrown amps and speakers below, all I can say is that they make an amazing set and are a tribute to his knowledge/perseverance and ears.
Most of the subjective impressions below are mine, so I won’t elaborate to much , but I’ll describe the steps we went through.
Throughout the afternoon it became quite apparent that different people liked different thing in the Sound signature, and we all agreed that that one of the beauties of audio DIY’ing in that sound could be tailored to individual preferences. And besides that the “breed” stayed the same, so the differences were relatively small.
I’m afraid I didn’t keep track of all the numbers that were played from Stefan’s vast music collection, most of which I don’t know. Needless to say most where quality contemporary recordings, mainly acoustic with an emphasis on vocals.
The Single board DDDAC, sourced by a J-River/Wave-IO board (powered via USB) gave us that typical DDDAC NOS PCM1794 sound signature we’ve have all come to love and had hooked us on this dac, easy/dynamic/full but detailed, analogue sounding. Next Doede added a separate power supply to the Wave-IO, with an immediate effect of a fresher sound. We could have easily left it there, wasn’t it for all the other dac’s begging to be listened to.
Next up we made the transition with the none stock dac’s. Since mine had both a Wave-IO as well as the BBB/S03 installed this made a nice progression. Switching to my Wave-IO with tweaked dac downstream, was quite a big difference from the previous stock board, most striking for me was definition/colour/soundstage, just a lot more coming out, but then a lot of changes had been made to achive this and I won’t go into the detail of all of those, as I would have to repeat what has been discussed on this thread for the last year of so. (sorry) J
Next we switched to the BBB/S03 on my dac, which again made the SQ, cleaner/better defined. Not all steps are a step forward though, but in my ears this switch made quite a big step in the right direction and one that is hard to achieve otherwise.
If I remember correctly, I think we switched over to Squeezelite images at this point on the BBB’s and RPi.
Unfortunately James his dac had not totally survived the journey as some soldering was required and also, he had developed a problem in recent days with his output buffer, so this was left out as well as the Cinemags which would have needed them. So James his dac was left out for a minute.
Then it became interesting because Stefan switched dac’s, which can be seamlessly done with Squeezelite, not telling us which was which, between mine and his.. There were some obvious differences and I was quite taken by how big these were. However I made up my mind, on what sound I liked best and when Stefan revealed the order I was relieved (;-)) to hear I’d preferred mine. However I would love to incorporate some of Stefan’s sound signature and Stefan was of the opposite opinion, so there you go.. that’s Audio DIY for you. However later on in the evening I came to admit that my preference between these dac’s depended very much on the source material. (these are solely my observations🙂 Mine was a bit cleaner/slightly more focussed/ but thinner sounding with a bit more bass definition and a bit more base extension. Where as, Stefan’s dac sounded fuller with a bigger soundstage projection. Trying to identify what was making these differences Stefan explained it was his impression that it was caused by his different power supply scheme to the dac boards. If people are interested what these are, they’ll need to dig deep into the past posts in this thread to discover the sometimes “minute” differences, that made quite a considerable overall difference in the end. And give them a try..
One thing that struck us is how well the Rpi with A-synchronised clocking, compared to the BBB with synchronised clocking. They seemed a near match.
Fortunately at this point James his dac was up and running. His being a single board dac, but with similar unregulated main supplies to mine and Stefan’s and also with the CCS mod,shunts & BBB/S03. But lacking the buffer and Cinemag output which he had been planning to use, which is something we’ll have to save for next time. When listening to James his dac it was obvious it’s energy/punchiness was lagging behind a bit, but overall the SQ was very nice and very close to Stefan’s (closer then to mine). Overall I was very impressed and so was James I think.
Later in the evening when we started blind testing/switching via Squeezelite between James/Stefan’s and my dac’s, it became quite confusing at times as we were getting into a (friendly) disagreement on what dac we where listing at any given time. Illustrating just how close things were.
Not in chronological order & saving the best for last. After a wonderful dinner Stefan’s wife prepared for us, lasagne with tiramisu finished off with some excellent coffee. Doede first rigged up Wave-IO -> Enrico single board, which didn’t have the output buffer enabled do to some issues with it. We all agree it sounded very good and was a great effort but it was quite obvious the power supply was not there and we were listening to the Wave-IO again.. We then moved to the big 8-stack (stock) with the Sowther output traffo’s. It sounded like the bigger brother 😉 , of the single board dac we started off with, but with a lot more oempf.., to the point where I though the SQ became somewhat woolly. Doede also explained that with the 8 board the PSU was providing a solid 1amp and it was quite a lot to ask of the little 15w transformer (if I’m not mistaken).
So then the discrete DSD prototype/breadboard, with it’s “antenna’s”, was up. Doede had already warned us that there were still some noise issues. After he’d switched it on, this were instantly noticeable but not that intrusive, something like minus (+/-) 70dB high feq. white noise, nothing to nasty. I will not try to explain why what and how, as Doede will undoubtable do this when the time comes, if this dac makes it into the real world (I do hope so) . However when the music started, everybody in the room looked at each other surprised and in awe. By this time we were pretty tuned in to Stefan’s set-up, but this sounded very different & very special. There was an instant revelation of dynamics and space and the bass was so heavy and defined, almost as if the linearity wasn’t there anymore. Was this what it was supposed to sound like.., ok, right!? Quite the eye-opener and a great prospect of things to come.
But here is the catch (in my mind).., how much music we know has been made available, or will ever be available in “true” DSD format? Mmm..
Thanks guys, for a great day!
As Supersurfer and Doede already commented on over the last couple of days. Yesterday, James, Doede, Stefan, Cees Pel & myselve, met up for a DDDAC bake-off/comparison/listening session.Especially James and Doede had travelled a considerable distance to make it to Stefan’s house in Utrecht, so a spot of lunch came first, but obviously the subject turned pretty quickly to all things dac related.
Between the five of us, we had brought along a lot of gear and options to discuss and listen to..
As soon as lunch was cleared the various dac’swhere spread out onto Stefan’s dinner table and we all got the first hand explanations of the individual builds.
An externally hosted image should be here but it was not working when we last tested it.
For starters, in a nutshell, the three none stock tweaked dacs on offer were James his freshlyrebuild single boarder & Stefan’s and my four board DDDAC’s. There are a lot of small difference among then, but main setups involved:unregulated main supplies, different cap schemes mainly consisting of Silmic’s, CCS’s Tenslab shunt’s, and embedded source players, isolated, clocked/reclocked by Acko’s S03(BBB’s/synchronisefor James and me and RPi/A-synchronise for Stefan.)
Besides these three dac’s, Doede had brought along a trunk load full of dac’s and peripherals. He had brought a box standard one board DDDAC, a 8-board stack (all stock), Doede’s supplies but with a custom made Sowther output transformers (pictures of this build are on the first page of this thread). But besides these familiar devices was an intriguing looking breadboard filled with “stuff”… and last an DDDAC dac board made by Enrico with an integrated buffer stage (amongst other things).
We all gave a brief explanation and had a brief discussion about all these, during which Doede explained to us the basic concept behind his new discrete DSD dac prototype. The concept sounded very simple and elegant, my initial thought was that of slight disbelieve, a breadboard full of antennes is hardly ever going to “look” like it will sound good….(or maybe that’s just me) ;-)
An externally hosted image should be here but it was not working when we last tested it.
From the outset it was obvious it was going to be hard to listen to all the options we had at hand and being able to hear & pinpoint what the effects of these build choice differences made. But since Doede had brought along a single “stock” build, this was a good place to lay the base line.
Stefan is the person to explain about his homegrown amps and speakers below, all I can say is that they make an amazing set and are a tribute to his knowledge/perseverance and ears.
An externally hosted image should be here but it was not working when we last tested it.
Most of the subjective impressions below are mine, so I won’t elaborate to much , but I’ll describe the steps we went through.
Throughout the afternoon it became quite apparent that different people liked different thing in the Sound signature, and we all agreed that that one of the beauties of audio DIY’ing in that sound could be tailored to individual preferences. And besides that the “breed” stayed the same, so the differences were relatively small.
I’m afraid I didn’t keep track of all the numbers that were played from Stefan’s vast music collection, most of which I don’t know. Needless to say most where quality contemporary recordings, mainly acoustic with an emphasis on vocals.
The Single board DDDAC, sourced by a J-River/Wave-IO board (powered via USB) gave us that typical DDDAC NOS PCM1794 sound signature we’ve have all come to love and had hooked us on this dac, easy/dynamic/full but detailed, analogue sounding. Next Doede added a separate power supply to the Wave-IO, with an immediate effect of a fresher sound. We could have easily left it there, wasn’t it for all the other dac’s begging to be listened to.
Next up we made the transition with the none stock dac’s. Since mine had both a Wave-IO as well as the BBB/S03 installed this made a nice progression. Switching to my Wave-IO with tweaked dac downstream, was quite a big difference from the previous stock board, most striking for me was definition/colour/soundstage, just a lot more coming out, but then a lot of changes had been made to achive this and I won’t go into the detail of all of those, as I would have to repeat what has been discussed on this thread for the last year of so. (sorry) J
Next we switched to the BBB/S03 on my dac, which again made the SQ, cleaner/better defined. Not all steps are a step forward though, but in my ears this switch made quite a big step in the right direction and one that is hard to achieve otherwise.
If I remember correctly, I think we switched over to Squeezelite images at this point on the BBB’s and RPi.
Unfortunately James his dac had not totally survived the journey as some soldering was required and also, he had developed a problem in recent days with his output buffer, so this was left out as well as the Cinemags which would have needed them. So James his dac was left out for a minute.
Then it became interesting because Stefan switched dac’s, which can be seamlessly done with Squeezelite, not telling us which was which, between mine and his.. There were some obvious differences and I was quite taken by how big these were. However I made up my mind, on what sound I liked best and when Stefan revealed the order I was relieved (;-)) to hear I’d preferred mine. However I would love to incorporate some of Stefan’s sound signature and Stefan was of the opposite opinion, so there you go.. that’s Audio DIY for you. However later on in the evening I came to admit that my preference between these dac’s depended very much on the source material. (these are solely my observations🙂 Mine was a bit cleaner/slightly more focussed/ but thinner sounding with a bit more bass definition and a bit more base extension. Where as, Stefan’s dac sounded fuller with a bigger soundstage projection. Trying to identify what was making these differences Stefan explained it was his impression that it was caused by his different power supply scheme to the dac boards. If people are interested what these are, they’ll need to dig deep into the past posts in this thread to discover the sometimes “minute” differences, that made quite a considerable overall difference in the end. And give them a try..
One thing that struck us is how well the Rpi with A-synchronised clocking, compared to the BBB with synchronised clocking. They seemed a near match.
Fortunately at this point James his dac was up and running. His being a single board dac, but with similar unregulated main supplies to mine and Stefan’s and also with the CCS mod,shunts & BBB/S03. But lacking the buffer and Cinemag output which he had been planning to use, which is something we’ll have to save for next time. When listening to James his dac it was obvious it’s energy/punchiness was lagging behind a bit, but overall the SQ was very nice and very close to Stefan’s (closer then to mine). Overall I was very impressed and so was James I think.
Later in the evening when we started blind testing/switching via Squeezelite between James/Stefan’s and my dac’s, it became quite confusing at times as we were getting into a (friendly) disagreement on what dac we where listing at any given time. Illustrating just how close things were.
Not in chronological order & saving the best for last. After a wonderful dinner Stefan’s wife prepared for us, lasagne with tiramisu finished off with some excellent coffee. Doede first rigged up Wave-IO -> Enrico single board, which didn’t have the output buffer enabled do to some issues with it. We all agree it sounded very good and was a great effort but it was quite obvious the power supply was not there and we were listening to the Wave-IO again.. We then moved to the big 8-stack (stock) with the Sowther output traffo’s. It sounded like the bigger brother 😉 , of the single board dac we started off with, but with a lot more oempf.., to the point where I though the SQ became somewhat woolly. Doede also explained that with the 8 board the PSU was providing a solid 1amp and it was quite a lot to ask of the little 15w transformer (if I’m not mistaken).
An externally hosted image should be here but it was not working when we last tested it.
So then the discrete DSD prototype/breadboard, with it’s “antenna’s”, was up. Doede had already warned us that there were still some noise issues. After he’d switched it on, this were instantly noticeable but not that intrusive, something like minus (+/-) 70dB high feq. white noise, nothing to nasty. I will not try to explain why what and how, as Doede will undoubtable do this when the time comes, if this dac makes it into the real world (I do hope so) . However when the music started, everybody in the room looked at each other surprised and in awe. By this time we were pretty tuned in to Stefan’s set-up, but this sounded very different & very special. There was an instant revelation of dynamics and space and the bass was so heavy and defined, almost as if the linearity wasn’t there anymore. Was this what it was supposed to sound like.., ok, right!? Quite the eye-opener and a great prospect of things to come.
But here is the catch (in my mind).., how much music we know has been made available, or will ever be available in “true” DSD format? Mmm..
Thanks guys, for a great day!
Last edited:
thanks
Hello Stijn,
It looks like it has been a nice day. But having such a big load of possibilties that made it a bit hard i think.
I know next to nothing about computer based audio so my first thing to connect to my future dac will be a cd transport.
I will go for choke input in common mode connecting and a BIG r core. I will ujse a bigger choke LL2733 more mH but also higher serie resistance. The second choke will be the famous ll1694 also in common mode.
And the special sowther output transformers.
Probably 4 boards with shunts. If there is enough space i could add some more later.
If you are on a budget which way to go 4 boards with shunts or skip the shunts and get extra boards? That would have been a nice test on the meeting.
Anyhow i will wait for the new boards to arrive.
I will finish the power supply in a few weeks. Using different chokes i will need a power transformer with higher voltage. Actually have a 300va 15volts R core but will get a 21 volt R core soon. Expect to get output voltage around 11,5 volts with 4 boards shunt dac set up using the 21 volt R core
Will offer Stefan to compare it to his supply. Find a way to easily switch between the 2 supplies using the same terminal.
We will see and hear.
Greetings, edward
Hello Stijn,
It looks like it has been a nice day. But having such a big load of possibilties that made it a bit hard i think.
I know next to nothing about computer based audio so my first thing to connect to my future dac will be a cd transport.
I will go for choke input in common mode connecting and a BIG r core. I will ujse a bigger choke LL2733 more mH but also higher serie resistance. The second choke will be the famous ll1694 also in common mode.
And the special sowther output transformers.
Probably 4 boards with shunts. If there is enough space i could add some more later.
If you are on a budget which way to go 4 boards with shunts or skip the shunts and get extra boards? That would have been a nice test on the meeting.
Anyhow i will wait for the new boards to arrive.
I will finish the power supply in a few weeks. Using different chokes i will need a power transformer with higher voltage. Actually have a 300va 15volts R core but will get a 21 volt R core soon. Expect to get output voltage around 11,5 volts with 4 boards shunt dac set up using the 21 volt R core
Will offer Stefan to compare it to his supply. Find a way to easily switch between the 2 supplies using the same terminal.
We will see and hear.
Greetings, edward
Thx stijn, for pics and report. Seems like you had a good time together.
eduard@
I changed from series to common mode on the second choke in my lclc psu supply. If I remember right this was something "rickminnis" recommended.
This was not a succes. The sound lost details and clarity. Bass not so defined. Had no choice than to switch back to series. I am using LL1694.
Maybe things would have been different with both chokes in common mode, but since the sound is great as it is, I will not change from series to common mode. (I think )
eduard@
I changed from series to common mode on the second choke in my lclc psu supply. If I remember right this was something "rickminnis" recommended.
This was not a succes. The sound lost details and clarity. Bass not so defined. Had no choice than to switch back to series. I am using LL1694.
Maybe things would have been different with both chokes in common mode, but since the sound is great as it is, I will not change from series to common mode. (I think )
to choke or not to choke
Hello,
I am not sure who did say it. BUT i know that Lundahl gives the advice to do it. Are you sure you did connect it right?
The voltage drop will be the same so it is an easy modification to try.
Greetings, edward food is ready
Hello,
I am not sure who did say it. BUT i know that Lundahl gives the advice to do it. Are you sure you did connect it right?
The voltage drop will be the same so it is an easy modification to try.
Greetings, edward food is ready
DSD dac explanation?
"Doede explained to us the basic concept behind his new discrete DSD dac prototype."
I hope an explanation of this DSD dac is put on the forum for all of us to understand and look forward to seeing some day.
"Doede explained to us the basic concept behind his new discrete DSD dac prototype."
I hope an explanation of this DSD dac is put on the forum for all of us to understand and look forward to seeing some day.
DDDAC comparison Utrecht 28-2-2015
Hi Stijn,
Wow! you are fast! James and I were still waking up when you posted it (maybe the wine had something to do with it 😱) You did a very nice job with the report.
James and I have done some more listening and comparisons on Sunday;
-We had a set of cool live sunday morning Jazz, playing the stars from heaven: Miles Davis at the plugged nickel, Modern Jazz Quartet Live at The Music Inn, Stanley Clarke Live At The Greek, Bola Sete at The Monterey Jazz Festival. Nothing beats a Sunday morning with these giants and some good Cappuncino 😛
-We compared the Squeezelite to Volumio, we both find that Volumio is a tad more detailed, dynamic and colored in sound.
-We also listened to the 4 Pin20 mods James had prepared. After soldering in an extra connector on the power entrance on the dac board we were able to de-power the dac easily, leaving the BBB running, and swap the pin20 boards. It was very cool to hear the differences. It seems that James did not have the best pin20 mods in yesterday so he has left you guys out on some extra quality with his DAC 😉.
We have found the 2SK208 J-fet the best sounding of the four (luckily these are also occupying my DAC).
I think James will explain the differences of the four mods when he has recovered from his jet lag 😀.
About all the dacs we have heard; before the comparison I had my thoughts on the sound balance of the different dacs and I was happy to experience that these were quite in line with what I have heard. I cannot add more to the description of Stijn.
The big surprise for me was the DSD dac proto. This complete discrete design was intriguing, first thing we all were looking for was the DAC chip; but there was no DAC ship 😱😱😱
This is a very cool and novel design!
I would like to note at this stage that Doede (sort of 😉) promised that the first batch of the pcb's are reserved for us
This DAC sounded like no other source I have ever heard; so extremely open, transparant and detailed! This is VERY promising for things to come!
The only down side is that I only have a limited DSD catalogue in comparison to my Red Book collection. But that will not keep me from going the discrete DSD route. Lets only hope that avalability of good quality DSD files will expand in the future...........
I would like to thank Doede for the nice design; next to a good sound, it keeps the inspration going and makes for nice meetings with other enthusiasts.
About the attached pictures:
-The group picture shows (from left to right): Stefan, Doede, Stijn, Cees, James.
-Doede connecting the DSD DAC
-a collection of the finest DACs around
-different pin20 mods
Hi Stijn,
Wow! you are fast! James and I were still waking up when you posted it (maybe the wine had something to do with it 😱) You did a very nice job with the report.
James and I have done some more listening and comparisons on Sunday;
-We had a set of cool live sunday morning Jazz, playing the stars from heaven: Miles Davis at the plugged nickel, Modern Jazz Quartet Live at The Music Inn, Stanley Clarke Live At The Greek, Bola Sete at The Monterey Jazz Festival. Nothing beats a Sunday morning with these giants and some good Cappuncino 😛
-We compared the Squeezelite to Volumio, we both find that Volumio is a tad more detailed, dynamic and colored in sound.
-We also listened to the 4 Pin20 mods James had prepared. After soldering in an extra connector on the power entrance on the dac board we were able to de-power the dac easily, leaving the BBB running, and swap the pin20 boards. It was very cool to hear the differences. It seems that James did not have the best pin20 mods in yesterday so he has left you guys out on some extra quality with his DAC 😉.
We have found the 2SK208 J-fet the best sounding of the four (luckily these are also occupying my DAC).
I think James will explain the differences of the four mods when he has recovered from his jet lag 😀.
About all the dacs we have heard; before the comparison I had my thoughts on the sound balance of the different dacs and I was happy to experience that these were quite in line with what I have heard. I cannot add more to the description of Stijn.
The big surprise for me was the DSD dac proto. This complete discrete design was intriguing, first thing we all were looking for was the DAC chip; but there was no DAC ship 😱😱😱
This is a very cool and novel design!
I would like to note at this stage that Doede (sort of 😉) promised that the first batch of the pcb's are reserved for us
This DAC sounded like no other source I have ever heard; so extremely open, transparant and detailed! This is VERY promising for things to come!
The only down side is that I only have a limited DSD catalogue in comparison to my Red Book collection. But that will not keep me from going the discrete DSD route. Lets only hope that avalability of good quality DSD files will expand in the future...........
I would like to thank Doede for the nice design; next to a good sound, it keeps the inspration going and makes for nice meetings with other enthusiasts.
About the attached pictures:
-The group picture shows (from left to right): Stefan, Doede, Stijn, Cees, James.
-Doede connecting the DSD DAC
-a collection of the finest DACs around
-different pin20 mods
Attachments
Last edited:
Thanks again Stefan for hosting and everyone for their contributions. A very nice little DDDAC adventure with some new DSD thrown in 🙂 The quality of amps and speakers really helped to identify changes in the dacs and players very nicely indeed. I'm the first to admit that I'm fairly crap at listening tests generally, but it was very satisfying to have some nice clear and consistent results.
I will keep this brief-ish for now, but wanted to just highlight some conclusions that I have come to following our experiences and discussions this weekend.
Just a little caveat first that I am no electronics engineering expert, far from it.... Just a keen hobbyist who's relatively fresh to this still, but I like to understand, test and work things out and I like to always ask why 🙂
So this list is more of a brain-dump of my notes from this weekend that I'm sharing in-case it helps anyone else and not neccessarily the gospel as spoken by some great audio designer....
1) First, and quite importantly, a standard DDDAC with single deck and waveio sounds very good if the WaveIO has a good power supply rather than using the USB host for power. We heard this very early in the proceedings and for anyone who isn't keen on extensive modifications and who doesn't have the terrible and infectious brain disease of always needing to tweak something with their audio, then rest assured even the basic standard version of the DDDAC/WaveIO is capable of delivering some very good quality, enjoyable music 🙂
2) a DDDAC with shunt regulators, at least on the Analogue side sounds much more detailed and dynamic than one without. I would not swap my single deck with shunt regulators for an 8 deck without shunts.
3) Of the various CCS options for pin20 we heard today, there was a fairly obvious order for us. (worst to best)
LM334
E-301 current regulator diode
2SK246-Y
2SK208-o
Even the LM334 is not exactly bad, but as you work your way down the list, you get more air, spacial ambience and energy to the sound. The 2ks208-o really made us smile with how good it sounded. If anyone wants to borrow the other options and make their own testing let me know, but for us, the 2sk208-o was the clear winner here and that's what's staying in my dac now 🙂
4) When using shunt regulators, the analogue side of the dac really benefits from a good, uncomplicated, unregulated raw power supply to make it sing. As low impedance in the v+ as you can make it and as little as possible in the ground line. If using chokes like the Lundahl where you have 2 halves to each choke, using both halves in series in the v+ and not putting anything between the dac gnd and the transformer gnd/centre tap seems to work out best here.
I'm going to follow Stefan's example and change the wiring slightly to my dac deck so that the analogue regulators get a feed direct from the last cap of the LCLC choke supply with no series resistors in line and to separate this from the digital side slightly. Should be a very quick and simple change, I'll post some pictures later this week when I've done it.
5) The digital side of the dac deck and the mainboard isn't as concerned with low impedance and can be considered differently to the analogue in it's requirements. The digital side is kind of like a computing engine crunching away at the data at a very consistent rate, rather than the analogue side, which has a very varying demand for power depending on the music signal it's generating. A level of separation of this supply from the analogue makes sense, but using the same supply with a simple RC filter is perfectly effective and adequate here and has the benefit of lowering the supply voltages for the digital regulators, which are only trying to output to 5v or 3.3v anyway. Doing this and lowering the voltage to the mainboard only works if you are providing a separate supply to the analog regs on the dac deck(s).
6) Having separate power supplies for rpi/BBB/WaveIO etc etc where you have an isolator between the device and the DDDAC is highly recommended. But after an isolator, for example in the Acko S03 reclocker where you have the signal grounds joined to the dddac ground in the section for the clock and flipflops, you may as well power these parts from the same supply as the digital side of the dddac and mainboard, using an RC filter to get the voltage where you want it.
I had used a separate power supply for my s03 reclocker, but now I've compared with Stefan's setup, I get the impression my extra supply is only adding complication and the potential for grounding issues and is very unlikely to bring any improvements over using the dddac supply with an RC filter, so I'll chop this extra supply out of the loop this week and see. Less is more sometimes 🙂
6) Great to see the potential for DSD to really offer something special, and really cool to see the innovative solution planned. I hadn't paid DSD much attention before, but now I'm interested 🙂
7) X-ray is the new Chryo treatment! So now I'm back from my second international DDDAC adventure, my DDDAC has been through the x-ray security scanners at the airport a total of 4 times, and they seemed to make sure it got a real good dose each time, backwards and forwards while they checked out the weird shapes and suspicious wiring on their computer screens.... It's sounding better than ever these days, make of that what you will 😉 😛
thanks again guys for the adventure, the music, the company, the inspiration and all that jazz
🙂
James
I will keep this brief-ish for now, but wanted to just highlight some conclusions that I have come to following our experiences and discussions this weekend.
Just a little caveat first that I am no electronics engineering expert, far from it.... Just a keen hobbyist who's relatively fresh to this still, but I like to understand, test and work things out and I like to always ask why 🙂
So this list is more of a brain-dump of my notes from this weekend that I'm sharing in-case it helps anyone else and not neccessarily the gospel as spoken by some great audio designer....
1) First, and quite importantly, a standard DDDAC with single deck and waveio sounds very good if the WaveIO has a good power supply rather than using the USB host for power. We heard this very early in the proceedings and for anyone who isn't keen on extensive modifications and who doesn't have the terrible and infectious brain disease of always needing to tweak something with their audio, then rest assured even the basic standard version of the DDDAC/WaveIO is capable of delivering some very good quality, enjoyable music 🙂
2) a DDDAC with shunt regulators, at least on the Analogue side sounds much more detailed and dynamic than one without. I would not swap my single deck with shunt regulators for an 8 deck without shunts.
3) Of the various CCS options for pin20 we heard today, there was a fairly obvious order for us. (worst to best)
LM334
E-301 current regulator diode
2SK246-Y
2SK208-o
Even the LM334 is not exactly bad, but as you work your way down the list, you get more air, spacial ambience and energy to the sound. The 2ks208-o really made us smile with how good it sounded. If anyone wants to borrow the other options and make their own testing let me know, but for us, the 2sk208-o was the clear winner here and that's what's staying in my dac now 🙂
4) When using shunt regulators, the analogue side of the dac really benefits from a good, uncomplicated, unregulated raw power supply to make it sing. As low impedance in the v+ as you can make it and as little as possible in the ground line. If using chokes like the Lundahl where you have 2 halves to each choke, using both halves in series in the v+ and not putting anything between the dac gnd and the transformer gnd/centre tap seems to work out best here.
I'm going to follow Stefan's example and change the wiring slightly to my dac deck so that the analogue regulators get a feed direct from the last cap of the LCLC choke supply with no series resistors in line and to separate this from the digital side slightly. Should be a very quick and simple change, I'll post some pictures later this week when I've done it.
5) The digital side of the dac deck and the mainboard isn't as concerned with low impedance and can be considered differently to the analogue in it's requirements. The digital side is kind of like a computing engine crunching away at the data at a very consistent rate, rather than the analogue side, which has a very varying demand for power depending on the music signal it's generating. A level of separation of this supply from the analogue makes sense, but using the same supply with a simple RC filter is perfectly effective and adequate here and has the benefit of lowering the supply voltages for the digital regulators, which are only trying to output to 5v or 3.3v anyway. Doing this and lowering the voltage to the mainboard only works if you are providing a separate supply to the analog regs on the dac deck(s).
6) Having separate power supplies for rpi/BBB/WaveIO etc etc where you have an isolator between the device and the DDDAC is highly recommended. But after an isolator, for example in the Acko S03 reclocker where you have the signal grounds joined to the dddac ground in the section for the clock and flipflops, you may as well power these parts from the same supply as the digital side of the dddac and mainboard, using an RC filter to get the voltage where you want it.
I had used a separate power supply for my s03 reclocker, but now I've compared with Stefan's setup, I get the impression my extra supply is only adding complication and the potential for grounding issues and is very unlikely to bring any improvements over using the dddac supply with an RC filter, so I'll chop this extra supply out of the loop this week and see. Less is more sometimes 🙂
6) Great to see the potential for DSD to really offer something special, and really cool to see the innovative solution planned. I hadn't paid DSD much attention before, but now I'm interested 🙂
7) X-ray is the new Chryo treatment! So now I'm back from my second international DDDAC adventure, my DDDAC has been through the x-ray security scanners at the airport a total of 4 times, and they seemed to make sure it got a real good dose each time, backwards and forwards while they checked out the weird shapes and suspicious wiring on their computer screens.... It's sounding better than ever these days, make of that what you will 😉 😛
thanks again guys for the adventure, the music, the company, the inspiration and all that jazz
🙂James
Many thanks for the update Guys! Really appreciate the effort, you all have put in to keep us all inform. 😉
Cheers.
Sent from my iPhone using Tapatalk
Cheers.
Sent from my iPhone using Tapatalk
Where do they?
All I see is in the data sheet for the LL1694 where it states that using 1 half for the v+ and the other for the ground can be used "for improved common mode rejection"
But if you don't need improved common mode noise rejection, then I think it's better to have no unneccessary components in the dac gnd to make for a free and easy supply which the analogue side can use to really sing with no restrictions.
As you say, it's easy to try both ways and see what you prefer 🙂
common mode or not?
Hello,
I did ask Per Lundahl by email some time ago that if using the choke input choke ( ll2733) in common mode one could/should also use the second choke (ll1694) in common mode. He said there is no reason not to do so.
Here in my country there is a website who also did use common mode choke for heater and anode supply in tube amps and they stated it was an improvement.
I never did so i will try once the new boards are available.
However i did try and still use choke input for anode supply. And yes not every choke sounds the same.
Thank you James the information you did give is very nice.
Maybe when using the choke power supply like Stefan does and extra choke ( common mode?) could be used to '' isolate '' the digital from the analoque part
Could be higher induction because current is less especially when not using the shunts on the lower voltage?
Greetings, Edward
Hello,
I did ask Per Lundahl by email some time ago that if using the choke input choke ( ll2733) in common mode one could/should also use the second choke (ll1694) in common mode. He said there is no reason not to do so.
Here in my country there is a website who also did use common mode choke for heater and anode supply in tube amps and they stated it was an improvement.
I never did so i will try once the new boards are available.
However i did try and still use choke input for anode supply. And yes not every choke sounds the same.
Thank you James the information you did give is very nice.
Maybe when using the choke power supply like Stefan does and extra choke ( common mode?) could be used to '' isolate '' the digital from the analoque part
Could be higher induction because current is less especially when not using the shunts on the lower voltage?
Greetings, Edward
Thank to all of you for making it so much open to everyone here. You wrote great descriptions of the event, and very reviling for many. Especialy for new guys who want to find their way trough your examples (like me 🙂. Huge apreciacion to you for your work and your huge input!
Last edited:
I'm starting to be intrigued with this pin20 2SK208-o. James would you be so kind to give some explanation what is this actually? Where do you mount it, or instead of what? How many do you need? Maybe you could attach some scratch diagram with part list? And more datailed photo of that piece?
Regards
Please look at this thread
http://www.diyaudio.com/forums/digi...-upgraded-single-board-pcm1794-nos-dddac.html
Ed
Attachments
Last edited:
Hi stijn001,
Great review indeed! Thanks and sorry for the late reply but in addition to the 8 hours time difference between Korea and Europe I have some issue with the internet connection in the new apartment.
Anyway, I see that you guys enjoyed also the Italian cuisine... lasagne, tiramisu' and coffee... true connoisseurs 🙂
I am very glad that our small board was part of your review. The board is actually nothing special, we just try to condensate all the various mods well discussed in the thread in one more "flexible" configurable board.
The main goals was:
- maintain the connections with the motherboard as per Doede design
- connect the Tent's shunts near to the Vcc pins of the PCM1794.
- possibility to install the Belleson's regulator in the digital side that require a different configuration of input and output caps.
- use of SMD parts to optimize the layout and the performance.
- separate ground plane between digital and analogic side with jumper to reconnect in one ground plane.
- possibility to have a dedicated external PSU for digital and analogic sides
- option for a dedicates CCS circuit
- a buffer circuit integrated in the board for the "one" board layout with dedicated PSU +/-12V connection.
- maintain the possibility to stack more than one dack. In this case the parts for the buffer circuit are not to be installed.
I think that we reach our targets 🙂
From the first time we power up the board we make many improvement. The Waveio was already substituted with the Rpi and the reclocker board S03 from Acko.
The first improvement was to build the raw supply as per Stefan schematic. Thanks Stefan!!! Was quite hard to found the room for the big trafo plus the 2 chocke but finally i can get the PSU in place exploiting the possibility of the separated analogic and digital connections.
Additional improvement was the CCS circuit but as of now I try only the 2SK246-Y.
My final improvement is the Edel_NMR renderer that I love but this is another story...nothing to do with the board 🙂
I attached some pictures of my dac 4 boards configuration and the single board with buffer.
I would like to explain the reason why in past months we never show up our work and results in the forum.. was a choice dictated in respect of Doede thread and design.
Best Regards,
Enrico
PS: In one of your pictures I see the buffer circuit in the James's DAC. This is the same circuit that is in our board... just re-designed to be connected to the doede dac board. 😉
Attachments
Hi Enrico,
Your board is a very nice build. I like the possebility the layout gives for tweaking, also nice the convecrion holes for cooling the shunts, and the way you can adjust the CCS from the side of the boards.
Unfortunately we were not able to audition your buffer board, James his board was not completely ready and we listened to your board with the unbuffered output because Doede did not have the +\- 12v supply at hand.
So we will need you guys to inform us of the sound signature.
Regards,
Your board is a very nice build. I like the possebility the layout gives for tweaking, also nice the convecrion holes for cooling the shunts, and the way you can adjust the CCS from the side of the boards.
Unfortunately we were not able to audition your buffer board, James his board was not completely ready and we listened to your board with the unbuffered output because Doede did not have the +\- 12v supply at hand.
So we will need you guys to inform us of the sound signature.
Regards,