Top-notch DAC.

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Peter Daniel said:
I'm glad you like it, it is also my favourite DAC. I was only experimenting with TDA1543 so far, but the results are so good, that I don't feel any need to explore TDA1541 teritory. What you see there, consists a year of research and experimentation. I have plans to release PCB version of that DAC (AC powered) and the protoype can be seen in this thread: http://www.diyaudio.com/forums/showthread.php?s=&threadid=31002&perpage=10&pagenumber=5
Looking forward to your dac project. ;)
 
Hajime said:
I believe DIY designs are a great value as you can use top-quality components and tailor things to your liking.

DIY is great value but I don’t equate great value with great performance. You specifically said you wanted to MATCH the PERFORMANCE of the BEST. I doubt a $700 DIY will match the best Wadia or Audio Note DAC and here’s why.

To match the Wadia sound you will need to implement and program a sophisticated DSP. Unless you are already an expert DSP programmer and have access to a DSP development system and ICE, a Wadia clone is not feasible. As far as I know, no one here has done a Wadia clone and I have not seen one elsewhere on the web.

Duplicating the sound of an AN DAC is more feasible. After all, they are little more than a DAC chip with a vacuum tube buffer and the schematics for all the models are readily available on the web. The hard part will be duplicating the unique sound of the DAC-5 without using the same silver-wire transformers ($$$).

I think your expectations are unreasonable, but have fun.
 
jbokelman said:


Hajime, what do you consider to be the best commercial DACs and what makes you think they can be equaled with a DIY project found on the internet?

jbokelman-

I would be interested to know what you consider CAN NOT be accomplished as a DIY project, especially in terms of DACs. To my ammature mind, it seems like what a DIYer lacks in expensive monitoring/test equipment, he gains through the absence of design constraints such as cost, manufacturing capability, size, looks, etc etc etc! And what about all the EEs who use their employer's labs to work on these wonderfull creations, and share the findings with the rest of us?
 
Till,

I like the looks of that DAC. Have you designed a PCB for your implimenation with the Pass Labs I/V stage? I read about the I/V stage on the linked Pass Labs site and it looks pretty neat. Has there been any comparison of this DAC with others?

I did no PCB, at the moment, i and some friends, compare different prototypes listening to them. I build different variations and test if i like the sound of them. Lets see if there will be a final implementation.
 
jbokelman said:


DIY is great value but I don’t equate great value with great performance. You specifically said you wanted to MATCH the PERFORMANCE of the BEST. I doubt a $700 DIY will match the best Wadia or Audio Note DAC and here’s why.

To match the Wadia sound you will need to implement and program a sophisticated DSP. Unless you are already an expert DSP programmer and have access to a DSP development system and ICE, a Wadia clone is not feasible. As far as I know, no one here has done a Wadia clone and I have not seen one elsewhere on the web.

Duplicating the sound of an AN DAC is more feasible. After all, they are little more than a DAC chip with a vacuum tube buffer and the schematics for all the models are readily available on the web. The hard part will be duplicating the unique sound of the DAC-5 without using the same silver-wire transformers ($$$).

I think your expectations are unreasonable, but have fun.

I find this kind of thinking very odd. The requisite skills to build a top notch dac do not depend on whether or not one is paid. Though duplicating the Wadia is difficult, it is not because of the DSP skills required. The hardest part would replicating Wadia's proprietary filters, though I don't doubt one could come close. Given the impossibility of knowing the choices Wadia made in arriving at their filter parameters, one might as well make one's own choices.
BTW, high performance dacs do not begin and end with AN and Wadia and there are many other routes one can take, The PassLabs D1 being one possibility.
 
amo said:
I would be interested to know what you consider CAN NOT be accomplished as a DIY project, especially in terms of DACs.

Amo, There are no limits on what an amateur DAC designer can create if he is willing to invest the time and energy required to investigate, learn, and experiment. I was questioning Hajime’s assumption that by copying a DIY project from an audio forum he would get performance that equaled the best commercial DACs.

Which DIY DAC projects, discussed in this forum, do you think MATCH the performance of the best Wadia or Audio Note DAC-5? Which MATCH the performance of the Zanden, Lavry, dCS, or Meitner?
 
rfbrw said:
Though duplicating the Wadia is difficult, it is not because of the DSP skills required. The hardest part would replicating Wadia's proprietary filters, though I don't doubt one could come close.

rfbrw, I think the functions used by in the Wadia filters are well known and the details could be easily discovered by examination of their DSP code; providing one had the knowledge and tools to do it.
 
jbokelman said:


rfbrw, I think the functions used by in the Wadia filters are well known and the details could be easily discovered by examination of their DSP code; providing one had the knowledge and tools to do it.

The most you can get hold of, without access to Wadia, is the hex code in the rom and thats assuming it is not a OTP device. Disassembling from hex to assembly code listing is not always an exact science.
 
Originally posted by jbokelman

DIY is great value but I don’t equate great value with great performance. You specifically said you wanted to MATCH the PERFORMANCE of the BEST. I doubt a $700 DIY will match the best Wadia or Audio Note DAC and here’s why...etc.
I have seen the inside of a Wadia and I agree it'd be almost impossible to clone. The Wadias are built with amazing engineering. I only listed them because I consider them a benchmark performance-wise.

If you could point me to the location of the AN DAC-5 schematic (perhaps in email) I would greatly appreciate it. I actually may be able to get ahold of some silver transformers.

Originally posted by till


I did no PCB, at the moment, i and some friends, compare different prototypes listening to them. I build different variations and test if i like the sound of them. Lets see if there will be a final implementation.
Neat, I'm interested in what your results will be.
 
Hajime said:
If you could point me to the location of the AN DAC-5 schematic (perhaps in email) I would greatly appreciate it. I actually may be able to get ahold of some silver transformers.

Hajime, if you really want to make a DIY DAC-5 clone why don’t you Do It Yourself? Do you expect me to search the web to find the schematics for you? I think this forum should be renamed cseAudio; for Copy Someone Else Audio. Except for a few contributors, there is scant DIY going on here.

You could start by studying Thorsten’s Adagio DAC. Its output stage is based on the AN DAC-4. Then visit Audio Note’s UK web site and learn what goes into the making of their silver transformers (hint: it’s more than silver wire) and learn the details of the different DAC models. Then, if you still think you can duplicate a DAC-5 for <$1K, Do It Yourself.

I assume your desire to clone a DAC-5 comes from you experience listening to one and not due to reading a hyped-up review on the Internet. If that’s not the case, you ought to audition one of the higher numbered AN DACs to see if it suits your fancy.
 
Someone mentioned the Scott Nixon dac.

I have one of these built from a PCB. It's good. It a no nonsense rhythmnic and dynamic presentation, a bit like Naim with a fuller tonal balance.

The loopfilter on mine is 499R/4.7uf Y5V 1206 SMD, and I changed the decoupling caps for quadruplets. The supplied OSCON types have sounded artificially fat and sweet when I've used them before. Sanyo 220uf 10v OSCON G-type do make excellent coupling caps, though.

1uh inductors on the power supplies also help, and the board sits on four thru-hole plastic pillars attached to small pine chopping board, which seems a good universal support.

I must dig it out again, and compare it to the 3D dac.
 
Hajime, if you really want to make a DIY DAC-5 clone why don’t you Do It Yourself?
Good point. But easier said than done. I am still suprised my semi-diy DDDAC1543 works.

I don't really like the idea behind the DDDAC. Creating a tower out of TDA1543 DACs doesn't seem like the way to get the
best sound quality out of a DAC.
There is theory and then there is real life. You would be suprised.

Here is another suggestion. It has been hyped to bits. But then again I trust Dave Davenport and Kevin Carter. It is a DAC set to come out in the weeks to come based on the (relatively new) PCM1794.

http://www.audioasylum.com/forums/KandK/messages/2063.html

I have posted this suggestion to just about any request for a new DAC project...so I'll forgive some people for thinking that I profit in any shape or form from bringing the RAKK DAC to your attention. (Well I do profit from one point of view...It should become easier for me to share note wrt to building and implementing the DAC....but that's it :angel:
I have no commercial links to K&K...or Raleigh Audio. Bla..bla..bla etc., etc, etc.
 
jbokelman said:


Hajime, if you really want to make a DIY DAC-5 clone why don’t you Do It Yourself? Do you expect me to search the web to find the schematics for you? I think this forum should be renamed cseAudio; for Copy Someone Else Audio. Except for a few contributors, there is scant DIY going on here.
I tried to find the schematic and I couldn't. If I had already found it, do you think I would have asked for help? I don't expect you to do anything, but some help from someone with experience is always welcome. :)

I would use the schematic as a base, and change the design to suit my desires. However, without a schematic I can't really do anything. I'm willing to admit I'm not talented enough to come up with my own schematic.

I don't know why you're so on the edge. Asking for help and suggestions on where to start is not a crime. I, again, admit that I am not talented enough to fully design a DAC. That doesn't mean I can't make changes to an already existing design and tweak it until I'm satisfied. I would consider that DIY as it's not the same as simply building a replica of someone else's design.

Edit: Bas Horneman,

Thanks for the recommendation. :)
 
Hajime said:
Asking for help and suggestions on where to start is not a crime. I, again, admit that I am not talented enough to fully design a DAC. That doesn't mean I can't make changes to an already existing design and tweak it until I'm satisfied. I would consider that DIY as it's not the same as simply building a replica of someone else's design.

Hajime, Didn’t I just give you some suggestions on where to start? Did you read Thorsten’s article? It seems he was able to get the DAC-4 schematics by just asking Audio Note. Have you tried that? Have you searched the archives and asked questions in other forums where the heavy-hitter audiophiles hang out?

No? I didn’t think so. You, like all the rest, want everything handed to you. You want the schematics, a ready-made PCB, a parts list and place to buy everything you need. Then, you think if you make minor part substitutions you have Designed It Yourself.

If you can’t design a circuit, why don’t you learn how or at least start with a less ambitious project? Here’s another suggestion: Buy a Dackit from Scott Nixon, add a couple silver-wire transformers and tell yourself that your DIY DAC is better than a DAC-5.

Sorry, I must be in the wrong place. I thought this forum was for DIY, not just tweaking existing designs. Isn’t anyone interested in DSP ala Wadia or Meitner? Isn’t anyone interested in designing converters with discrete components ala dCS or Lavry? Or, are the now discontinued TDA154x DACs the last word in digital audio that can never be surpassed or even equaled?
 
indeed it looks this board is going more and more twoards tweaking voodoo and selling kits, diskussing special sonic qualities like using another diode in the power supply of a kit amp, or DAC.


so far i build half a dozend different DACs on veroboard, and don´t belive anymore in kitsellers wisdom on this board... I would bhe interested to hear something about dsp usage in a DAC.

Hajime, what about starting with a very simple DAC for lets say 10 or 20€ parts cost, get it working, and have a look on a more difficult after? it will need one or two evenings to make a working prototype.
 
jbokelman said:

Sorry, I must be in the wrong place. I thought this forum was for DIY, not just tweaking existing designs. Isn’t anyone interested in DSP ala Wadia or Meitner? Isn’t anyone interested in designing converters with discrete components ala dCS or Lavry?

On the subject of DSP, in around 10 years of using the internet I have only ever seen two complete DSP based Audio projects including code and that was for the long gone DSP56001. I suspect that has alot to with commercial reasons. It is highly unlikely that anyone would give away the DSP code for something along the lines of the Meitner IDAT.


Or, are the now discontinued TDA154x DACs the last word in digital audio that can never be surpassed or even equaled?

There seems to be a level of effort, electronically speaking, beyond which very few are prepared to go and the TDA154x dacs complement this. Consider the ease with which one can create a working dac with the CS8412 and the TDA1543 compared to creating the balanced I2S stage shown in the AD1852 datasheet or replicating the AD1865N based dac stage of the Wadia 23,even if one replaces the proprietary filter with an off-the-shelf one. Wadia 23 pic here.
http://www.diyaudio.com/forums/showthread.php?s=&threadid=17277&perpage=10&highlight=&pagenumber=3
Still, jbokelman, this may be an opportunity for you to break the mould.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.