[cheapskateaudio]

 
Quote:

Built 3 of these DAC's over some period of time last year.  I must say, they are great sounding for the price.  Fantastically fun and simple build process too!  All thanks to its creators!!

Good job man, I think I will build another as well, but .... I need to get a magnifier first, lol. I felt like I was going blind building this thing. I agree though it was a fun build, definitely a different process with the surface mount stuff, each piece is it's own tiny adventure. 

 

[akgfan]

Quote:

...each piece is it's own tiny adventure. 

Yeah, especially when you try to find that tiny SMD in carpet.

 

 

[tomb]

Quote:

Yeah, especially when you try to find that tiny SMD in carpet.

 

Isn't that the truth?  My worst episodes are when my tweezers try to play tiddly-winks with the TPS regulator.  For some reason, that's the one I have the most bad luck in handling.
 

 
 
Quote:

 


Woot, it works now! Sounds awesome...
 
Initial impression of the sound is that there is solid and abundant bass and excellent detail throughout the sound spectrum. Very revealing and transparent in combination with the O2, almost disturbingly so. 
 
The O2 + Grubdac + HD448's is a pretty nice sounding combo, very airy and extended sounding, I've never heard the HD448's sound this good, I'm impressed. Still don't like the O2 driving the HD650's even with grubdac, but it is an improvement over the uDac 1/2 in pretty much every way. 
 
My one complaint is that it occasionally sounds a bit "digital", I don't think it's the grubs fault necessarily, just certain tracks.
 
BTW, thanks for your help Tomb!

So glad you got it working!

 
 

 

[Ikarios]

A tip for those who tend to lose SMD chips in carpet: get a vacuum cleaner with a hose attachment and cover it with some light cloth (pantyhose probably works best, cheesecloth may also work) and work over the area where you might have lost it. The chip should get sucked in by the suction but will be trapped by the cloth.
 
Also a 35W iron shouldn't be too bad, if you have a steady hand and have worked with SMD pieces before. I've done two  GrubDACs on my 25w Hakko Dash with no trouble at all, but for a first-time SMD solderer I would not really recommend it. Desoldering braid and lots of flux is your friend.

 

[Handy Ray]

Definitely a mag is useful.  After building these I felt like I became tunnel visioned and near sighted for quite a while.
 

 
Quote:

 


Good job man, I think I will build another as well, but .... I need to get a magnifier first, lol. I felt like I was going blind building this thing. I agree though it was a fun build, definitely a different process with the surface mount stuff, each piece is it's own tiny adventure. 

 

 

[cheapskateaudio]

 
Quote:

Yeah, especially when you try to find that tiny SMD in carpet.

 

Wooot! I'm not the only one who did that!
 
 
 

 

[zeffa]

Built the grubDAC last night. I was pretty happy with the results for my first time doing SMD soldering. Anyway, initial power on seemed to work properly - all the voltages were correct, and the LED turned on. When wired up though, I'm getting sound, but only out of the left side. I tried on Mac and PC and it was the same result. I've attached pics of the PCB. Any thoughts?
 
 

 

[tomb]

Two possibilities:
1. You've mis-wired the output connector.
2. The WolfsonDAC chip is messed-up/soldered bad.  Check the furthest pin opposite the dot.  That would be the pin that's closest to the Right channel V-D resistor (blue one) and the C9 capacitor.  That's the LINEVOUTR pin.

 

[zeffa]

Hi Tom, thanks for the advice! On closer inspection of the Wolfson chip, I noticed that there wasn't any solder at the LINEVOUTR pin you mentioned. Now that its been on soldered properly, everything seems to be working! Thanks again!

 

[tomb]

Quote:

Hi Tom, thanks for the advice! On closer inspection of the Wolfson chip, I noticed that there wasn't any solder at the LINEVOUTR pin you mentioned. Now that its been on soldered properly, everything seems to be working! Thanks again!

That's great!  Glad you were able to fix it!

 
 

 

[hitman47]

Hi everyone, I finished my second grubDAC about a week ago (after messing up the first one already) and I'm having trouble again... Something's causing a short and I'm not sure how to check what it could be, Tomb advised me to ask my questions here as more people are probably looking at this thread. Details are in my thread at http://www.head-fi.org/t/573197/mosfet-max-tube-biasing-issues-solved-now-about-grubdac-issues/45, thanks in advance for the help!

 

[tomb]

 
Quote:

OK, yeah I thought that because of the way caps work, there should only be a resistance showing up at the beginning as it's being charged (whichis what happens with C5). I'm just not sure how to check what's causing the short, as the VRM is turned off by overcurrent protection (I'm assuming) so measuring currents/voltages won't do the trick I think and checking for resistance across caps (they're usually what's connected to ground) doesn't work as you told me... I'm also gonna leave a link to this thread in the grubDAC thread, thanks for the tip.

The difference in the caps is probably which ones are charged (low resistance) and which ones aren't (high resistance initially).  C5 is on the same circuit as the LED, so it probably drains every time, whereas the other caps could get a partial draining, only.
 
Can you be more specific as to the problem besides just "the short?"  What is telling you there's a short ...
 
 
P.S. Measuring resistances is fine for troubleshooting.  Just don't do it when the DAC is plugged in and don't try it again on capacitors.

 

 

[hitman47]

I figured it was a short based on what cobaltmute told me in my thread - the LED lights up, but it flickers and there's only a tiny voltage on the 3V3 point while the VRM gets hot as hell. And if I can't measure resistances in caps to determine if they're busted, I'm unsure how to find the short, since pretty much all points prone to causing a short are caps connected to ground as far as I can see.

 

[tomb]

 
 
Quote:

I figured it was a short based on what cobaltmute told me in my thread - the LED lights up, but it flickers and there's only a tiny voltage on the 3V3 point while the VRM gets hot as hell. And if I can't measure resistances in caps to determine if they're busted, I'm unsure how to find the short, since pretty much all points prone to causing a short are caps connected to ground as far as I can see.

Here's some principles to guide you in troubleshooting:
 
1. The ground plane is the ground (sounds redundant, I know).  In this case (sort of an unofficial standard), Ground is negative (-).
 
2. Traces that are not connected to the ground plane are of opposite polarity from the ground.  That means they're positive (+).
 
3. Placing DMM probes connecting any two points of 1) the ground plane, and 2) a trace or a part of component connected to a trace - should measure some definite resistance, as dictated by the components in the path between the two DMM probes.  I was going to say capacitors are exempt from this rule, but in actuality - you're trying to measure pads on the PCB vs. the ground plane.  It won't matter if you measure a capacitor for resistance to the ground plane in this instance, because even a capacitor will have "some" resistance other than zero.\
 
3.a. To make this easy, pick a Ground point for one of your probes and simply keep it there.  Anything connected to the ground plane should have zero resistance to this point.  "OG" should suffice, for instance.
 
4. If you measure anything close to -zero- between two points as described in #3), then you have a short.
 
5. Try to follow the schematic for the GrubDAC: http://www.diyforums.org/GrubDAC/GrubDACschematic.php.  If the voltage regulator is getting super-hot, look for the circuit of the voltage regulator from the power-in point (the USB connection) to the regulator itself.  That will be the circuit path for powering the voltage regulator.  You may need to follow the output path of the regulator - from its output to the component that its supplying, but worry about 2nd, not 1st.
 
6. Once you have the schematic path pinpointed for your part in question (the regulator), try to make sense of the schematic with the actual PCB.  Look at either the layout - http://www.diyforums.org/GrubDAC/GrubDAClayout.php, or the board photos - http://www.diyforums.org/GrubDAC/GrubDACboard.php, so that you understand how the circuit on the PCB goes exactly how the schematic shows it.  IOW, if you've identified the path on the schematic for the voltage regulator, then look for the components in that path on the PCB.  See if you can find the traces on the PCB that connect all of the components in that path on the schematic.
 
7. Measure the resistance of those component connections - with ground - all along that path that you've identified.  The values are not important when reading resistance.  What's important is whether you read any two points along that path, between a trace and the ground plane, that are -zero-.  THAT'S YOUR SHORT.
 
Give it a try and let us know.

 

 
 

 

[hitman47]

Well, it seems C14 is at fault. I get practically 0 resistance from both sides to ground (I get a value of 1.4 ohms, which is the same as when I just touch the leads together). I'm gonna replace it with the one from my old build and see if that fixes the issue. Thank you very much for your very helpful tips, I'll let you know how it goes!
 
Edit: I replaced it, and the same thing happened again (albeit only after a few seconds as far as I could see, at the beginning the light wasn't flickering). What could cause that cap to blow continuously (the resistance to ground is 0 again, and it wasn't on the first one I built, I measured it before taking the cap out)?
 
Edit2: After thinking about it (and having another look at the schematic), I think it's probably C15 after all since it's in parallel to C14.
Edit3: Replaced C15, no improvement. It could also be C6 or C10 but somehow I doubt that. I've noticed that the PCM chip gets hot as well, so currently my bet is on that being fried. I'm gonna try and see if I can get a replacement somewhere, but I'm not sure it's worth it. This is really sad since it's already my second failed attempt and I was planning on giving my dad a Mosfet-Max with a grubDAC for christmas which isn't gonna work out from how it looks now.
 
Well, I managed to replace the chip with the one from the very first board I tried to assemble and messed up because I had such crappy soldering equipment. Now the PC says the device malfunctioned, so I hope it's just a badly soldered pin that can be fixed. I'll go try again.