[ezkcdude]

So, I'm getting ready to place an order for the DAC that I've designed (http://www.cellandtissue.com/diy/dac/proto_all.png).

I've gotten a lot of advice at various forums, including this one, and I think it's ready to go. The PCB is the mini-board size from ExpressPCB, so I will get 3 boards. I'd like to keep one for myself, but I'm looking to find one or two experienced DIYers (comfortable with SMD) who might want to build a prototype. I'd give you a board free of charge. Of course, I can't afford to give the components, but except for the chips and transformer, most of them are fairly standard and cheap. The board takes dual inputs (+ ground), and I've configured the BOM for +/- 15 V, although there is quite a lot of variability here. If you are truly interested in doing this, all I would ask is that you build it in a timely fashion (within a month or so), and let us know about your experiences. You could resurrect this thread when the time comes, and post pictures. Well, I've said my peace. I know it's a long shot, but if there are any brave and selfless DIYers willing to help a newbie out, send me a PM. Before placing the PCB order, I'd ask you to review the PCB as well, to check for mistakes (but not major design changes). Oh, I forgot to mention that obviously, you would be able to keep the DAC after you built it. For more info on the PCB, you can click on my link below.

 

[Garbz]

if you want more responses you best say more about the design. like schematics or something so people know what they can estimate for costs.

 

[dsavitsk]

Look at the link in his signature for more info. It looks like a pretty solid design to me.

 

[[AK]Zip]

I'm in on this. I will take a good look over it over the weekend.

-Alex-

 

[ezkcdude]

Sweet. That makes 2 testers, so we're all set now.

 

[ezkcdude]

I just wanted to provide a quick update, and see if there is anyone else who would like to prototype this PCB. It is up to version 0.95 now, so obviously, I'm considering the design virtually set. Even if you don't want to prototype, it always helps to have more sets of eyes going over the design to look for misconnections or shorts (although I don't think there are any at this point). You can click on my www, or use the links below to check out the latest update of the board. I can't wait to build this thing! Maybe I'm reinventing the wheel here, but hey, you gotta start somewhere

.

Top+Bottom+Silk
http://www.cellandtissue.com/DIY/DAC...b_all_v095.png

Top
http://www.cellandtissue.com/DIY/DAC...b_top_v095.png

Bottom
http://www.cellandtissue.com/DIY/DAC...b_bot_v095.png

Pads+Silk
http://www.cellandtissue.com/DIY/DAC..._pads_v095.png

Here's the BOM link:
http://www.cellandtissue.com/DIY/DAC/ezdac_bom.html

 

[abcheng]

sent you a pm.

 

[ezkcdude]

I finally ordered the boards today for my DAC! Keep your fingers crossed for me...

 

[t52]

one question: why did you put a cap from one of the spdif input lines to gnd? why not leave it completely floating?

and another one: could you up schematics with values and/or part numbers? that would make discussions much easier!

 

[Filburt]

Have you considered any other chips for I/V? Not saying the AD8610 is bad, but it wouldn't be my first choice I spose.

 

[ezkcdude]

Quote:

Originally Posted by t52 one question: why did you put a cap from one of the spdif input lines to gnd? why not leave it completely floating? and another one: could you up schematics with values and/or part numbers? that would make discussions much easier!

To answer your first question, I got the SPDIF input circuit from dsavitsk's TDA14541/5 designs, who has had success using that approach. Also, if you read the data sheet for CS8416 it recommends coupling the coax sleeve to ground of the chassis. I have been told the input circuit could be made even better by adding buffers, but since my design is using an ASRC for re-clocking/jitter reduction, it seems to me the potential gains from adding buffers don't outweigh the additional cost and complexity that would introduce. As much as possible, I want the "ez" in "ezdac" to mean something, besides my initials.

As for schematics, I will work on updating that now that the prototype is set. I think I will put the component labels on the schematics, and leave the values/part numbers for the BOM on my website.

 

[ezkcdude]

Quote:

Originally Posted by Filburt Have you considered any other chips for I/V? Not saying the AD8610 is bad, but it wouldn't be my first choice I spose.

You must have got that post in as I typed my last response. I've only heard good things about it until now, I guess. What would you suggest? I should add that I probably would build my first board using some cheaper chip, like AD8065 or OPA134.

 

[t52]

Quote:

Originally Posted by ezkcdude To answer your first question, I got the SPDIF input circuit from dsavitsk's TDA14541/5 designs, who has had success using that approach. Also, if you read the data sheet for CS8416 it recommends coupling the coax sleeve to ground of the chassis. I have been told the input circuit could be made even better by adding buffers, but since my design is using an ASRC for re-clocking/jitter reduction, it seems to me the potential gains from adding buffers don't outweigh the additional cost and complexity that would introduce. As much as possible, I want the "ez" in "ezdac" to mean something, besides my initials.

hehe, ok... esy is not bad! the buffered input circuit could be added outboard if one prefers, no problem. as well as a low noise clock instead of the integrated one.
one reason why i stepped back from my dac project was it's complexity - not good for it's my first project in this size... so let's get this one going

btw. regarding the grounding cap on the spdif input - i still don't see why it should be good - i thought a complete separation of potentials through the transformer would be a good idea... anyways we can try this easily and just leave it out

 

[ezkcdude]

Quote:

Originally Posted by t52 one reason why i stepped back from my dac project was it's complexity - not good for it's my first project in this size... so let's get this one going

Exactly. I wanted something that is technologically modern, but still easy enough to understand and explain. In other words, something good for a beginner. All the parts are readily available and still in production, and there's no microcontroller programming involved. I felt like this was a project I could really finish. Looking at some of the other designs out there, I think I would need another year or more just to understand exactly what was going on.

Quote:

btw. regarding the grounding cap on the spdif input - i still don't see why it should be good - i thought a complete separation of potentials through the transformer would be a good idea... anyways we can try this easily and just leave it out

Yeah, that makes some sense to me, too. Like you say, you can build the board without it first, and then add it if you want later. There's a lot of components on the board that can be left out, for example, decoupling caps, and the resistors on the clock lines can be substituted for zero-ohm values, as well. I may eventually build one "minimal" board like that, and then one that is "maxed" out, just to see if I can hear the difference. (I hope I do!)

 

[00940]

Quote:

Originally Posted by ezkcdude Yeah, that makes some sense to me, too. Like you say, you can build the board without it first, and then add it if you want later. There's a lot of components on the board that can be left out, for example, decoupling caps, and the resistors on the clock lines can be substituted for zero-ohm values, as well. I may eventually build one "minimal" board like that, and then one that is "maxed" out, just to see if I can hear the difference. (I hope I do!)

Certainly don't left out the decoupling caps. You already don't have that many onboard.