Well,
I got the revision D from newark, I'm quite happy. I faxed the order to olimex for two boards and did a digikey order for the rest of the stuff. I bought 5 scientific conversions SC982-04 pulse transformers, so I have 4 left. I came across one problem when ordering the epson clock module--digikey is out of stock. In replacement, I found one from Citizen America Corperation. It's datasheet can be found here. It's the same package size and pinout, but will it work? It might not be as reliable as the one from epson, but I'm sure it will do for now. Also, instead of the amber LEDs, I ordered green, because the amber was out of stock. I like the color green better anyway.
The electrical characteristics are thje exact same, so they will work.
I hope now that since I have the right soldering tools I can get this DAC working. I plan on building two. It's one of my favorite DAC designs, because i wanted to build one with a D/A converter from cirrus logic.
I got the revision D from newark, I'm quite happy. I faxed the order to olimex for two boards and did a digikey order for the rest of the stuff. I bought 5 scientific conversions SC982-04 pulse transformers, so I have 4 left. I came across one problem when ordering the epson clock module--digikey is out of stock. In replacement, I found one from Citizen America Corperation. It's datasheet can be found here. It's the same package size and pinout, but will it work? It might not be as reliable as the one from epson, but I'm sure it will do for now. Also, instead of the amber LEDs, I ordered green, because the amber was out of stock. I like the color green better anyway.
The electrical characteristics are thje exact same, so they will work. I hope now that since I have the right soldering tools I can get this DAC working. I plan on building two. It's one of my favorite DAC designs, because i wanted to build one with a D/A converter from cirrus logic.
Hey jwb, is it possible to use multiple CS43122 in parallel as done with some other DACs to improve performance?
emuman, where'd you get the pulse transformers?
If I was to do this over, I'd go to the extreme and use a teflon/silver PCB
But for now I'm satisfied, and will be a couple of years until I build a new DAC...
emuman, where'd you get the pulse transformers?
If I was to do this over, I'd go to the extreme and use a teflon/silver PCB
But for now I'm satisfied, and will be a couple of years until I build a new DAC...
There was a minimum order of $25. They were 6.50 each, I ordered 5. Total came to around $35. Thing is, he doesn't seem to be selling them anymore.
Btw, would that clock module work inplace of the Epson? It has the same package size, pinout, and is the same frequency. I suppose it'll work.
Thanks
Btw, would that clock module work inplace of the Epson? It has the same package size, pinout, and is the same frequency. I suppose it'll work.
Thanks
Proper XLR connectors to use:
http://www.heavymoon.co.jp/hifi/weiss/medea/images/medea-rear_l.png
In this photo, on the left are the two otputs, and on the right, the three inputs (the balanced inputs are AES3). Here's an amplifier XLR input:
http://www.av-outlet.com/media/amp52R.jpg
So interconnects use female on one side and male on the other.
You can get good XLR jacks and connectors by Neutrik for a great price. For cable, I recommend Venhaus' silver/teflon recipe (I'm using it myself). (only pins 2 and 3 are connected)
http://www.heavymoon.co.jp/hifi/weiss/medea/images/medea-rear_l.png
In this photo, on the left are the two otputs, and on the right, the three inputs (the balanced inputs are AES3). Here's an amplifier XLR input:
http://www.av-outlet.com/media/amp52R.jpg
So interconnects use female on one side and male on the other.
You can get good XLR jacks and connectors by Neutrik for a great price. For cable, I recommend Venhaus' silver/teflon recipe (I'm using it myself). (only pins 2 and 3 are connected)
Well, after 7 hours of assembly and $300 or so more dollars spent I connected everything up, put my 1kHz sine wave cd in my dvd player and fired it up. To my amazment I heard a 1kHz sine out of my small bench amplifier. I was exstactic! I assembled it at work, so I packed it up in anti-static bubble wrap and put it in a box with the rest of my stuff. I take it home, set it back up on my bench and test it out with a somewhat better amplifier. I was shocked to hear silence. So after poking around a bit I saw 9V at the analog supply of the CS43122. I got quite angry that the regulator failed. It's quite odd that it just shorted like that. I removed it and it seems like my soldering lifted some of the traces of it's footprint and might have shorted some pins, but not from input to output. It's still weird that it did that. So I go to bed and work on it today. I remove the regulator and install the new one, test for 5.5V and bingo, 5.5V at the output. So I reflow the new CS43122. I connect everything up, turn on the dvd player and power up the DAC. IT WORKS!!!! IT WORKS!!!! IT WORKS!!!! IT WORKS!!!! IT WORKS!!!! IT WORKS!!!! IT WORKS!!!! IT WORKS!!!! IT WORKS!!!! IT WORKS!!!! IT WORKS!!!! IT WORKS!!!! IT WORKS!!!! IT WORKS!!!! Neil Young never sounded so good. I love this DAC. Now I have to put it in it's case, get some XLR and RCA panel mount connectors, install the toslink/coax/RS422/485 to coax circuit and I'm good to go. I love the sound, the design, the PCB, everything. It finally works now that I used the right soldering equipment.
The problem with the last one was the 8.2nF and the 1.8k were in the wrong places of the PLL filter. I changed them on the old one but I still couldn't get it to work so I won't bother with it anymore.
I used two new parts on the second build. the first was the different clock module because digikey was out of stock on the Epson one. I must have ordered the wrong part number because I got a low profile ceramic and metal case. It had the same pinout, and the length between the pins was the same as the footprint, but the width was about 1.5mm off. The leads were pads on the ceramic case, and most of the pad was on the bottom of the case, but there was a little bit on the side of it. I used wire to make contact with the small bit of pad on the side of the case and the clock seems to work quite well. Although I haven't measured anything yet, it could be less of a radiator than the Epson clock module because it's smaller and low profile, and has metal on the top of the case which could act as sheilding.
The second new part I used were green Panasonic LEDs rather than the amber LEDs. Digikey had the amber on back order, so I looked at it's datasheet and saw the green had the same forward current and voltage rating so I ordered those. They work quite well too. I have yet to try the 8.2uF Solen capacitors that I ordered, the 2.2uF metallized polyester caps I ordered from digikey sound just fine, but I'll give them a try.
Anyway, Thank you so much Prune and jwb for putting up with me and helping me out, it's greatly appreciated.
The problem with the last one was the 8.2nF and the 1.8k were in the wrong places of the PLL filter. I changed them on the old one but I still couldn't get it to work so I won't bother with it anymore.
I used two new parts on the second build. the first was the different clock module because digikey was out of stock on the Epson one. I must have ordered the wrong part number because I got a low profile ceramic and metal case. It had the same pinout, and the length between the pins was the same as the footprint, but the width was about 1.5mm off. The leads were pads on the ceramic case, and most of the pad was on the bottom of the case, but there was a little bit on the side of it. I used wire to make contact with the small bit of pad on the side of the case and the clock seems to work quite well. Although I haven't measured anything yet, it could be less of a radiator than the Epson clock module because it's smaller and low profile, and has metal on the top of the case which could act as sheilding.
The second new part I used were green Panasonic LEDs rather than the amber LEDs. Digikey had the amber on back order, so I looked at it's datasheet and saw the green had the same forward current and voltage rating so I ordered those. They work quite well too. I have yet to try the 8.2uF Solen capacitors that I ordered, the 2.2uF metallized polyester caps I ordered from digikey sound just fine, but I'll give them a try.
Anyway, Thank you so much Prune and jwb for putting up with me and helping me out, it's greatly appreciated.
Prune, Thanks.
But my luck as turned again, and the DAC is silent. I simply put it in a case, and when I powered it up, it took a second to hear the audio. After about a second, I heard a pop and then I heard audio. It kept behaving like this, and when I took it back out, I hear nothing. I'm going to reflow the solder, but other than that, I have no idea as to why this does this. I ordered another board from olimex so I can always rebuild it, but that was quite an undertaking.
But my luck as turned again, and the DAC is silent. I simply put it in a case, and when I powered it up, it took a second to hear the audio. After about a second, I heard a pop and then I heard audio. It kept behaving like this, and when I took it back out, I hear nothing. I'm going to reflow the solder, but other than that, I have no idea as to why this does this. I ordered another board from olimex so I can always rebuild it, but that was quite an undertaking.
As my sound card outputs TTL (5 V) digital audio, I decided to build this converter so as not to fry the DAC.
http://www.andrewkilpatrick.org/mind/spdif/spdif_circuits.gif (top)
I put it right in the computer, powered from the computer's 5 V supply.
Now I can use the computer as a source for the DAC, however the following strange problem occurs: crackling noises start and the ASRC keeps loosing lock. I was wondering if there was a grounding problem, and found that the problem disappears if I parallel a resistor (I used 1 K) by the 100 pF cap in front of the pulse transformer. So although I have a solution, I don't have an explanation of what the cause is or why the fix worked. Could there be a digital ground loop problem? The DAC chassis is earthed, as is the computer (so the digital output ground is too).
http://www.andrewkilpatrick.org/mind/spdif/spdif_circuits.gif (top)
I put it right in the computer, powered from the computer's 5 V supply.
Now I can use the computer as a source for the DAC, however the following strange problem occurs: crackling noises start and the ASRC keeps loosing lock. I was wondering if there was a grounding problem, and found that the problem disappears if I parallel a resistor (I used 1 K) by the 100 pF cap in front of the pulse transformer. So although I have a solution, I don't have an explanation of what the cause is or why the fix worked. Could there be a digital ground loop problem? The DAC chassis is earthed, as is the computer (so the digital output ground is too).
Finally found a digital camera so here are photos:
http://www.diyaudio.com/forums/showthread.php?s=&postid=474528
http://www.diyaudio.com/forums/showthread.php?s=&postid=474528
----------------------------------------------------------------------------------jwb said:
Not really, Switchcraft and Vampire are better.
Neutrik uses a material for their male XLR that has a high dielectric loss (I measured them) in comparison with others. Female ones don't seem to have same problem.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.