USB powered dac-amp musings
Sep 22, 2014 at 7:30 PM Post #31 of 43
I use OSHPark too, I find the quality to be quite good for the price.
From the time I place my order, until they show up in my mail box,
three weeks to a month. Probably a week longer for you.
You can PM me the details when they're ready.
 
Sep 24, 2014 at 8:37 PM Post #32 of 43
Subscribed!
 
I've been dreaming of a DIY USB-powered DAC/Amp combo for awhile and what do I find? Booyah! My personal ideal case size target would probably be the Hammond 1455C80x, but the 1455C120x is pretty damn close. And hey, I don't know a damn thing about circuit design, I just solder-by-number.
 
Maybe you already addressed this and I didn't see, but are you still thinking Blue Velvet for the pot? That won't fit in the 1455C120x will it?
 
Sep 25, 2014 at 7:15 AM Post #33 of 43
Some little things:
 
- No, the blue velvet wouldn't fit. The rk097 will have to do. 
- You could probably fit the amp in a 80mm deep 1455 case if you used a Grub DAC with it.
- I don't intend to distribute the boards myself. I'll however share the files on Oshpark once the pcb are built and tested, so that anyone could order those there. It will probably take about two months from now. I'll also post a Mouser BOM.
- The project is really geared towards high impedance headphones, to get more voltage out than the common 2Vrms (which is already a lot with most headphones). The extra voltage will come at the expense of current, so it's not good for orthos either. It would probably make more economic sense to buy a Dragonfly, geek out, dacport or the like if you don't need the extra voltage.
 
Sep 27, 2014 at 10:41 AM Post #34 of 43
For my Carrie Amp proto run, OSH Park was about the same price as Imagineering per board, but I got mine in a week. Min 5 boards though. 4 sq in, so $20 per board plus $24 for ground shipping to Canada.
 
Sep 27, 2014 at 12:56 PM Post #36 of 43
Oh snap. I gotta work on my reading comprehension. 
confused_face_2.gif

 
Sep 27, 2014 at 1:20 PM Post #37 of 43
00940, what's your take on the isolated grounds? When you connect signals L, R, and G from the DAC to the amp, doesn't that just end up circumventing the ground separation?
 
Sep 27, 2014 at 2:06 PM Post #38 of 43
  00940, what's your take on the isolated grounds? When you connect signals L, R, and G from the DAC to the amp, doesn't that just end up circumventing the ground separation?

 
 
Indeed. Still, when you consider the circulating currents, it shouldn't be a problem. There's always the possibility of a groundloop of course, as for all USB powered devices without isolation of the input lines.
 
jcx made me think a bit about the input/output filters for the dc-dc converters. 
 
Oct 23, 2014 at 11:33 AM Post #39 of 43
I received the PCBs from smart-prototyping  :)
 
They seem fine, the only complain is that the tenting of the smaller vias is not perfect. Otherwise, the drilling is very accurate and a quick check with a dmm doesn't show any problems. 
 
No reason anymore to procrastinate on the mouser BOM and order.
 
 
As a side note, I just got an ESI juli@ for testing. 
tongue_smile.gif

 
Nov 24, 2014 at 9:13 AM Post #40 of 43
How could you not love Mouser fast shipping ?  I received the parts. 
biggrin.gif
 It's nice to see how compact that Hammond case is in real life.
 
The good news is that my eagle home made libraries seem ok (especially for the headphones jack and that kind of things).
 
Let's heat the soldering iron !
 

 
Nov 24, 2014 at 12:10 PM Post #41 of 43
The DAC is alive  :)
 
No drama, everything works as planned. 
 
Here's a RMAA summary at 16bits/44.1KHz, with a PCM5101 rather than a PCM5102. The setup is pretty awful, there is some 50hz hash as the board sits near the computer and long unshielded cables go to the soundcard. The DAC is powered through a small hifimediy isolator and the recording soundcard is a ESI Juli@.
 
 
Frequency response (from 40 Hz to 15 kHz)dB +0.04, -0.02 Excellent
Noise level, dB (A)-96.0Excellent
Dynamic range, dB (A)96.0Excellent
THD, %
0.0019
Excellent
THD + Noise, dB (A)-86.8Good
IMD + Noise, %0.0053Excellent
Stereo crosstalk, dB-92.1Excellent
IMD at 10 kHz, %0.0062Excellent
General performance
 Excellent
 
Nov 25, 2014 at 8:43 AM Post #42 of 43
Allright, I built the amp part and cased the whole thing. I had no time for stability testing nor noise or thd measurement yet. But offset is at 3-4mV and the amp is silent into average headphones (I've got no sensitive IEM).
 
Building the amp was actually more difficult than the DAC. A first problem was that the inductor pads are a bit undersized and the inductors sink heat like crazy. Soldering isn't pretty around them... The second problem was the connection of the output. Wires need to be run from the output resistors to the output jack and I should have let a dedicated pad for those wires. The wires were meant to be soldered to the output resistors' legs but there's very little room under the pcb.
 
Rmaa testing of the whole thing this night I hope.
 
Nov 25, 2014 at 6:03 PM Post #43 of 43
And the final result, DAC+amp: 
 
Frequency response (from 40 Hz to 15 kHz), dB+0.50, +0.45Excellent
Noise level, dB (A)
-94.1Very good
Dynamic range, dB (A)94.1Very good
THD, %0.0018Excellent
THD + Noise, dB (A)-85.7Good
IMD + Noise, %0.0063Excellent
Stereo crosstalk, dB-73.9Good
IMD at 10 kHz, %0.0072Excellent
General performance Very good
 
 
Measurements are made with the same DAC, at almost 2Vrms into 100r (resistive). Playing with the scope for a while doesn't reveal anything bad either.
 
The summary doesn't tell the whole story though, so I upload the whole report as pictures  below:
 

 

 
 
 
 

 

Users who are viewing this thread

Back
Top