[tomb]

  A "Bottlehead Crack" and a profile pic of Rob Ford? Beautiful.
 
Jokes aside, the 4.75V may not be problem but my DC offset is way off.

I agree that offset is way too high.  I'm at a loss to really pick out anything in your pics, though - except the resistors.  Cobaltmute spec'd the resistors around the opamp at 1% tolerance.  Those aren't V-D RN50's that you've used, so I'm not sure if they're in the recommended tolerance.  Just an FYI, but even Tangent used to (still does?) have articles online about matching resistors around an opamp for ensuring the minimum in offset with his amps designs.  Matching resistors used to be de riguer when the CMoy, PIMETA, and PPA were first in vogue (not that they still aren't).  Cobaltmute even designed the very critical resistors to be installed on the PCB bottom - right next to the opamp - in an effort to reduce lead lengths and potential oscillation issues.  The pupDAC will definitely result in opamp oscillation (especially with the OPA2836) if everything is not just so-so with the resistors and lead lengths.
 
I'm not saying for sure that's what's going on - it's still only a guess.  Perhaps the resistors could be causing the issue and maybe the opamp is oscillating.  Strange things can go on when that happens, including ticking sounds when the oscillating opamp causes regular/intermittent capacitor discharges.

 

[COOKIESnPIE]

Post deleted by user.
 
 

 

[lordearl]

Is anyone using the pupDAC with JRIVER and Windows 8?
What is the optimal setting?  I'm finding WASAPI EVENT STYLE to be a little unreliable, so currently use the DIRECT SOUND with 0.10s buffer and everything upsampled to 96khz - any others using similar settings?

 

[tomb]

  They're 1% KOA metal films, not "military grade" though. I might be able to get some more accurate resistors near the opamp for a spin, but besides from that I don't know where to look either. I spoke to someone in the industry before building this and they said that rarely you need mil grade resistors unless you need high reliability - just the right tolerances are fine. He could be wrong though...

What's with the lead lengths?

Well, "someone in the industry" may not know about mismatched resistors on a sensitive opamp.  It's not necessarily the precision of the resistors, although that buys you some consistency from resistor to resistor with the same rating.  It's the differences between two resistors of the same rating that can cause offset when applied to the two different Left and Right channels.  For instance, a 100R resistor at 1% precision may be 99 or 101 ohms, a 1K resistor might be 990 vs. 1010 at 1% and so forth ...  Differences of that magnitude can definitely cause offset.
 
You might try measuring the resistors and see if you get a difference in the same rating resistor from one channel to the other channel.
 
Actually, this can apply to solder joints, too.  A few sub-standard joints that may predominate in one channel over the other can combine to increase the resistance in that channel.  That might result in the offset, too.  It kind of explains why two people can build the same thing with the same parts and yet get two different performance results.  It's also why re-flowing solder joints over an entire PCB can sometimes result in fixing a problem. 

 
Again, these are guesses ...  

 

[jboehle]

Hooked it up to my 17" MacBook Pro tonight and got some new measurements.
 
5V = 4.989V
-5V = -4.9702V
-2V5 = -2.4890V
2V5 = 2.4897V
4V5 = 4.7235V
3V3 = 3.308V
R = 1.04mV offset
L = 0.1mV offset
 
All looking great!
 
Hooked it up to my work 15" MacBook Pro which had a charger hooked up.  Fed it through my KSA5 clone amp.  With Monoprice 8323 'phones I can hear ticking in both channels from 50%+ volume.  With AT ESW9 'phones it starts about 65%+ volume.
 
Looking forward to trying the Doodlebug!

 

[lordearl]

  Is anyone using the pupDAC with JRIVER and Windows 8?
What is the optimal setting?  I'm finding WASAPI EVENT STYLE to be a little unreliable, so currently use the DIRECT SOUND with 0.10s buffer and everything upsampled to 96khz - any others using similar settings?

 
Actually I don't know that the upsampling is working correctly, as (according to the datasheet) the PCM2707 only takes up to 48kHz....anyone else tried this?

 

[COOKIESnPIE]

Post deleted by user.
 
 

 

[tomb]

  I removed and checked the resistors near the opamp... while the 1.5k ones were exact, the 10 ohm ones were at 10.8 - I'll probably replace them later. Anyways, I also resoldered everything else on the board (including the chips) and the ticking seems to have disappeared. However, the high DC offset and the noise at high volumes still remained....

Strange.

Well, take satisfaction that something improved.

 
 
As for the resistors - maybe I wasn't clear before or perhaps I'm not understanding now: absolute values matter little.  What matters is whether you have 10.8 on one channel of the opamp and 9.2 on the other channel.  If they're both 10.8 even though they're supposed to be 10, it's OK - it's still a match.

 

[tomb]

BTW, I don't think I ever answered the question about lead length ...
 
Leads are extraneous parts to a circuit.  However, being metal and being conductive, they have an inductance per unit length.  On very small circuits that operate at high frequencies, that little bit of inductance may be enough to upset the stability in the circuit.  That's why the critical resistors for the opamp on the pupDAC are designed to be installed on the bottom of the PCB - closest to the opamp.  It's to reduce any instability caused by the lead length.

 

[COOKIESnPIE]

Post deleted by user.
 
 

 

[tomb]

 

  If they're both 10.8 even though they're supposed to be 10, it's OK - it's still a match.

 
Hmm... that means that the noise must be coming from somewhere else, not opamp oscillation. Since jboehle also has the noise too, it might not be related to the DC offset problem I have. I'm running it on a desktop, so there's plenty of power coming in.

Well, after all of this - I have to come down on the side of Avro: it's probably the power supply in one way or another.  Your offset is still high, but that could be power related as well.*  I can't think of anything else it might be and your board pics look fine.  Some of the regulators may be a bit dodgy from their appearance, but if you're measuring all the proper voltages (you are), it should be OK.
 
A couple of the chips that appear like regulators are actually not.  For instance, U6 controls the reset on the DAC chip.  I'm guessing and out of my element here, but without this, the DAC may fall completely offline if the voltage drops and then you would have to disconnect and re-connect to allow the PC to bring it back online.  Or, it  could be that very bad dropouts occur during the middle of the music stream - I'm not quite sure.  Regardless of which, this is a terrible nuisance, so U6 controls the reset to the DAC chip if the voltage drops below acceptable limits.  It could very well be that your voltage from the PC is marginal and the chip is continuously re-setting the DAC output.  That might cause the ticking.  Again, just a Wild Ass Guess (WAG).
 
There's also U7.  This is the charge pump controller.  It essentially creates a zero-voltage reference, allowing a + and - analog output, which obviates the need for output coupling capacitors.  Again, if this chip is malfunctioning, it may seriously affect the signal.  It's quite possible that it could cause offset in one channel, too - especially since it's the thing that setting the zero voltage point of the output.  There again, a WAG.
 
Maybe some others more knowledgeable than me can chime in and correct my guesses or elaborate.

 
 
 
* Plenty of power doesn't necessarily mean it's at the right voltage.

 

[COOKIESnPIE]

Post deleted by user.
 
 

 

[COOKIESnPIE]

Post deleted by user.
 
 

 

[KimLaroux]

Measuring an unloaded power rail is not a reliable way of learning what voltage a load is fed. The only way to know for sure is to connect the load, make sure its powered, and then measuring the voltage at the load.

 

[mcandmar]

USB can be anything from 4.5v to 5.2v with a nominal of ~5v.  In reality it fluctuates depending on supply load, for example plug in a USB memory key and copy some data to it while monitoring the voltage and it will jump all over the place.
 
One thing i have found the PupDAC susceptible to is dropouts due to spikes in the output ground. I can fairly consistently make it drop off by power up my valve amp connected to it, or even turning on my cheap Chinese eBay soldering station can cause it to drop off, though interestingly the USB connection to my UPS drops off too when i use it.
 
It doesn't seem to make any difference if i power it from the computers USB output, Doodlbug or a completely overkill power supply / Isolator box i previously built so it has to be small voltage spikes coming back into the RCA output grounds that is causing it, or a voltage differential between the input and output grounds.
 
Whatever amp you are connecting to the PupDAC, make sure its powered from the same power point as the computer, other than that i haven't really got any solutions to avoid it.