[fordgtlover]

Thanks for the tip. I'll work through the 3.3V side tomorrow.

Cheers

Quote:

Originally Posted by Calroth /img/forum/go_quote.gif Keep at it! The 5V side of the chip is the "analog" side; the 3.3V side is the "digital" side. So if there's a problem with the computer recognising the chip, then I'd check the 3.3V side.

 

[btrancho]

I finished my Alien DAC a few days ago and after a bit a work repairing solder bridges it tested out fine (3.3V & 4.6V), was recognized by my PC and worked fine. I ran it for a few days, on and off, between my Pimeta and the PC to break it in, waiting for the case to arrive.

Yesterday the case arrived and I disconnected it and gave the board a good cleaning with 91% alcohol and a toothbrush. This morning I cased it up and got no audio. Testing now shows I'm now getting 1.75V on the 5V test point. Looks like my cleanup has caused a problem.

I've gone back over my PCM2702 and it looks okay (I have a serious camera with macro lenses which helps a lot). Any hints as to where I should be looking?

 

[btrancho]

I think I've got the answer to my own question. R32 got pretty badly mangled after I had to de-solder it when I realized that I had laid it down overlapping the IC3 contacts. Once re-soldered it worked okay but I think that the toothbrush may have knocked off part of the cover that had been damaged.

 

[FallenAngel]

Quote:

Originally Posted by btrancho /img/forum/go_quote.gif I think I've got the answer to my own question. R32 got pretty badly mangled after I had to de-solder it when I realized that I had laid it down overlapping the IC3 contacts. Once re-soldered it worked okay but I think that the toothbrush may have knocked off part of the cover that had been damaged.

Congratulations on successfully photographing what an bomb shelter looks like after a nuclear blast, it looks brutal.

Now onto your AlienDAC - R32 is supposed to be connected across pins 3 and 4 of IC3. You might have disconnected the pads from the board. Try measuring resistance from the top of the resistor connected to pin 4 to ground to make sure there is none and again from the other side of the resistor (connected to pin 3), it should be the exact value of that resistor.

 

[btrancho]

Yeah, it certainly is ugly... but I swallowed my pride in order to get this DAC going again.

Pin 4 to ground is zero, Pin3 to ground settles at 51.3K (that was an 11K for regulated 4.75v out).

So I guess I my options are:
1. Get another R32 and try not to butcher it
2. Go unregulated, de-solder IC3, R32, R31 & C32

Correct?

 

[error401]

Quote:

Originally Posted by btrancho /img/forum/go_quote.gif I think I've got the answer to my own question. R32 got pretty badly mangled after I had to de-solder it when I realized that I had laid it down overlapping the IC3 contacts. Once re-soldered it worked okay but I think that the toothbrush may have knocked off part of the cover that had been damaged.

If you haven't already, you probably want to replace that resistor with a new one. The resistive element in SMD metal film resistors is actually a very thin layer underneath the black insulator with the marking on it. It sits between this black enamel and the white insulation that's showing in the photo. Since the element is obviously damaged, I'd expect that the resistance has probably increased substantially. At least meter it and check...

 

[amb]

More pics here.

 

[colonelkernel8]

Quote:

Originally Posted by btrancho /img/forum/go_quote.gif I think I've got the answer to my own question. R32 got pretty badly mangled after I had to de-solder it when I realized that I had laid it down overlapping the IC3 contacts. Once re-soldered it worked okay but I think that the toothbrush may have knocked off part of the cover that had been damaged.

That is a stunningly detailed and clear photo. You must have an amazing camera and macro lens. Sorry for the off-topic.

 

[btrancho]

Wouldn't it be nicer if were of one of amb's soldering jobs? Actually, I am fairly proud of the rest of the SMD sodlering I did on the DAC, but I really mangled this resistor during the de-solder process.

I shoot macro photos regularly. You can see them at a closeup view - my photoblog.

Once the DAC is repaired I'll post some photos. My Hammond case is all done - just waiting for an occupant.

 

[btrancho]

My Mouser order arrived today and I replaced the damaged 11.5k with a 9.5k. Now I only get 3.96V at the 5V post and the resistor only reads 8.03k when measured from Pin 3 to ground. Something else is going on. Could the pads be disconnected? They looked okay before I soldered the new R32.

 

[splaz]

That doesn't make much sense, if the pad's weren't connected properly the resistant should be higher, not lower.

Also if it is 8k, assuming your R31 is 31.2k output voltage should be around 6.15V which I'm not even sure the reg101 or 102 is specced for.

 

[btrancho]

One more shot at this before I scrap this attempt and go for another.

I removed IC3, R31, R32 and C32, and jumpered AV+ to the 5v pad, making it unregulated. I quadruple checked the PCM2702 - I even removed C13 temporarily to get a better view of the 5V side and ran a resistor lead behind the pads - it looks fine even under my 3X macro lens.

Yet when I plug it in to my PC I only get 3.66V at the 5V pad. My USB is putting out 5.08V. Any suggestions will be greatly appreciated.

 

[error401]

Quote:

Originally Posted by splaz /img/forum/go_quote.gif That doesn't make much sense, if the pad's weren't connected properly the resistant should be higher, not lower.

Under normal operation, a current flows into Iadj through R31 to set the voltage. This means that there must be a path to ground through this pin. Measuring resistance in-circuit isn't always foolproof as there can be other paths for the measurement current to take. I can't tell for sure if that's what's happening, but it's definitely plausible. Also possible of course is that the IC is fried and you're measuring some effects of its failure.

If your voltage rails are sagging that badly when you plug it in it's almost certainly a short somewhere. This doesn't necessarily mean that two pins on the PCM2702 are shorted, but that's the most obvious conclusion to come to. I'd try to measure it out - use a DMM to test for continuity between all adjacent pins and compare to the datasheet (there are a few pins on the IC that are internally connected). It could also be away from the PCM2702, say a ferrite accidentally placed where a capacitor should be, or a short to the ground plane where the solder mask has been scraped off or something like that. That said, I had the same problem with my HPDAC build and wasn't able to track it down without first destroying the solder pads. Good luck

.

 

[btrancho]

I'll check the PCM2702 continuity. Since the DAC worked well for several days prior to cleaning the board, I'm sure that there are no misplaced components. I think that there may be a short to the ground plane at the mangled R32 pad. If I eliminate the PCM2702 as a possibility, I'll probably start over from scratch. Not a big deal - I have confidence that my next attempt will be error free. My SMD soldering skills have gotten a fairly good workout.
Thanks for the advice.

 

[achina]

Try pulling the other electrolytic capacitors. If they were stressed they could be shorting internally. Maybe they weren't sealed good and some alcohol got in there. The output capacitors probably wouldn't have any effect though, if they shorted it wouldn't affect the power supply.

http://www.elna-america.com/tech_al_reliability.php