Alien DAC v1.1 Construction Thread
Jun 10, 2008 at 4:38 AM Post #1,246 of 1,562
Quote:

Originally Posted by CingKrab /img/forum/go_quote.gif
Hmm, I hooked everything up to my amp and while the right channel sounds fine, the left channel has no sound. It shows 0V AC on the DAC output. I also checked upstream at the PCM side of CL, and it is also 0V AC. Is there anything else I should check? I hope the PCM isn't fried...


Maybe the offset meant something. Time for pics to see if we can find anything amiss.
 
Jun 10, 2008 at 5:23 AM Post #1,247 of 1,562
Quote:

Originally Posted by tomb /img/forum/go_quote.gif
Maybe the offset meant something. Time for pics to see if we can find anything amiss.


I've attached pics of the DAC. I should clean it, but I was hoping to leave that until after it worked.

I've reflowed pin 23 (VoutL), but no improvement. I left the solder bridge on pins 13 and 14 because they're both connected to ground anyway.



 
Jun 10, 2008 at 7:25 AM Post #1,248 of 1,562
Quote:

Originally Posted by CingKrab /img/forum/go_quote.gif
I've attached pics of the DAC. I should clean it, but I was hoping to leave that until after it worked.

I've reflowed pin 23 (VoutL), but no improvement. I left the solder bridge on pins 13 and 14 because they're both connected to ground anyway.






Clean up the flux and take another photo
smily_headphones1.gif
You also seem to have some questionable solder joints on the back.
 
Jun 10, 2008 at 11:00 AM Post #1,249 of 1,562
Quote:

Originally Posted by FallenAngel /img/forum/go_quote.gif
Clean up the flux and take another photo
smily_headphones1.gif
You also seem to have some questionable solder joints on the back.



Agreed. I'm not so sure about the PCM chip and the regulators, either.

I don't mean to give useless advice after the fact, but it helps to clean one up before soldering in the electrolytics and the USB connector.
smily_headphones1.gif
 
Jun 10, 2008 at 4:17 PM Post #1,250 of 1,562
I'm using a microscope to solder the main smd chip onto the Alien DAC board. I can very clearly see each pin and its connection to the board, etc. This is the first time I've worked with such small parts.

Most of the pins are securely soldered to the board. I can see the solder and I can use a sharp probe to push on the pin and see that it does not nudge off its pad.

A few pins refuse to solder to their pads. I have cleaned off the board, soaked it in too much flux, then tried resoldering. The main issue is that the solder just does not seem to want to stick to the pads.

I have ordered a few more boards and a new DAC chip -- I figure I may have ruined this pair and can start again with the AMB kit I have.

Things I will try next time I can get back to working on it:

* Trying a different solder. I've been working with Cardas Quad.

* Really scrubbing the board with solvent again. I've been using Electrosolve contact cleaner (what my local shop stocks) and a tooth brush. I might soak in pure alcohol, etc.

* I will try a much more precise application of rosin flux -- using a toothpick this time, instead of an earswab that tends to flood the whole area with flux.

But does anyone have any other suggestions for this situation?

Thanks,

__Roy
 
Jun 10, 2008 at 6:12 PM Post #1,252 of 1,562
Quote:

Originally Posted by royewest /img/forum/go_quote.gif
But does anyone have any other suggestions for this situation?


Yes, drop the microscope, flood the SOICs with solder and clean up with desoldering braid. Takes about 30 seconds to do the PCM, safest and easiest way I found. I don't care much for solder or chemical cleaners, but do use a toothbrush and 90%+ alcohol. Some compressed air afterwards leaves it nice and clean.
 
Jun 10, 2008 at 7:02 PM Post #1,253 of 1,562
Quote:

Originally Posted by royewest /img/forum/go_quote.gif
<snip>A few pins refuse to solder to their pads. I have cleaned off the board, soaked it in too much flux, then tried resoldering. The main issue is that the solder just does not seem to want to stick to the pads.


Two things cause this, but it can be a combination of both:
1. Not enough heat in the right place.
2. Not enough flux.

After soldering together several of these things using a small chisel point with a Hakko 936, I've come to the conclusion that ~375 deg.C. works a lot better than something lower. Everybody has their methods, but it seems to work best for me if I do a lot of the horizontal soldering in a sweeping rolling motion away from the pins - wiping, as it were. This is detailed in Tangent's vids as a "clean-up" procedure for the final touch in soldering chips like these. Seems to work for me, too.

Lower temperatures will simply not work with pads that are connected to the ground plane on the Alien DAC. Waiting for the pads to heat up in that instance causes more heat transfer to the chip. Whereas the high heat option allows a very short exposure - almost allowing the heat to dissipate before it reaches the chip because you are quickly removing the iron.

More flux will help, rather than less - as long as there's not too much solder. The flux will not allow the solder to stick to anything that's not solderable. If you have pins soldered together, then that's too much solder - not too much flux. As I said, everyone has their method - braid didn't work for me. However, a clean, hot horizontal iron works very good to wipe away excess solder on the pins.

Quote:

<snip>
* Trying a different solder. I've been working with Cardas Quad.


That should be good enough - if it's thin. Maybe ~0.025"?

Quote:

* Really scrubbing the board with solvent again. I've been using Electrosolve contact cleaner (what my local shop stocks) and a tooth brush. I might soak in pure alcohol, etc.


Agree with FallenAngel in this regard. 90% Walmart alcohol and a toothbrush are fine. It helps if you do all this before adding the electrolytics and the USB connector.

You can check for continuity of the pins on the PCM chip with a meter. Check for some sort of resistance between adjacent pins. You'll get different ohm readings between adjacent pins, but as long as it's not zero, then it's OK. Then place one probe on a pin, then the other probe on the pad's destination across the board. If zero, it's a good connection. Note that there are a couple that are shorted together by the board, anyway.

Quote:

* I will try a much more precise application of rosin flux -- using a toothpick this time, instead of an earswab that tends to flood the whole area with flux.


I use the flux pens. I don't use them for anything else, but for the Alien, they're great. The stuff is so sticky you can almost use it to glue the chips in place prior to soldering. Again, precise application of the solder is required, not the flux. The flux can't make the solder stick to an unsolderable surface.
 
Jun 10, 2008 at 8:21 PM Post #1,255 of 1,562
Quote:

Originally Posted by CingKrab /img/forum/go_quote.gif
New pics after I cleaned up the flux:






Well, you definitely have a few cold solder joints around, the ones that look dull, not shiny should be reflown. As for 0V off the other channel, you're measuring from the DAC pin to ground right? If this is the case, your PCM is likely fried, happened to me a couple of times before, usually when you apply power to it while it has a solder bridge. If that's not the way you measure, do so.
 
Jun 10, 2008 at 9:53 PM Post #1,257 of 1,562
Just double checked with the DMM, continuity test says no bridge with the bottom two pins and 3rd, as well as 4th and 5th. The 4th pin is bent for some reason, so it's pretty close to the 5th, but it's not touching. The white gunk left over from the alcohol I used from cleaning makes some of the areas look pretty dirty. Is there anything you guys do to avoid this?

I did measure from the DAC on the left channel, and it gave me 0V. If the PCM is indeed fried, would you say it's worth the effort to remove and replace it or to start over with a new board/parts?
 
Jun 10, 2008 at 11:37 PM Post #1,258 of 1,562
I would take your DMM and ohm out all the pins for the left channel to the board to make sure that they are all making contact. Ohm these from the chip itself to the board pad/trace and verify connection. I would also take some solder wick and clean things up some, many joints don't look too hot.
 
Jun 11, 2008 at 2:09 AM Post #1,259 of 1,562
Quote:

Originally Posted by Pars /img/forum/go_quote.gif
I would take your DMM and ohm out all the pins for the left channel to the board to make sure that they are all making contact. Ohm these from the chip itself to the board pad/trace and verify connection. I would also take some solder wick and clean things up some, many joints don't look too hot.


Just checked VCCL, AGNDL, and OUTL. All ~0 ohms (0.1 according to my meter) from top of the pin to the pad. VCCL is around 4.71V at the pin. AGNDL is connected to both the USB receptacle shield and OG according to the continuity test at the pin. I'll try reflowing some stuff related to those pins, but it looks like the left channel analog stuff is properly powered at least. I'll try reflowing R15 again.
 
Jun 11, 2008 at 3:14 AM Post #1,260 of 1,562
Good luck!

BTW, the white stuff is dried powdered flux. It means you need to rinse it more. I've found that toweling off the alcohol with a good absorbent paper towel works well. Otherwise, you're just spreading around the flux each time you rinse with the alcohol. You should start to notice your paper towel turn a bit brown as it soaks in.
smily_headphones1.gif
 

Users who are viewing this thread

Back
Top