[Alf]

Quote:

Originally Posted by t52 some more info on my problem: the dc offset on both channels is the same within 1mV, i also checked with a dmm in the 200kohms range from output to ground (before output caps) and measured 2x ~90kohms, also identical (dac off). i also looked at the output with my old scope (built 1969, all tubes) which is not very precise, but both outputs show a nice sine wave form when fed with a 1khz mono sine signal... it's just that one channel's amplitude is lower. could it have something to do with the digital attenuator in the chip (probably being programmed to some value on one channel)? i don't even know if the attenuator can be set explicitely for one channel only. nor do i know how to set it at all, can someone shed some light on this or should i contact ti about this? if it's not a "software" problem, i probably have a broken chip here...

My understanding is that the chip attenuator is controlled by the driver via USB. The volume is not hard coded in the chip. You move the volume slider in your OS/driver, the chip setting changes. I assume you have played with the slider already. What happens if you move the slider up and down? Do channels still sound different? Do you use Windows?

If you are confident about your soldering and your software settings, then I think the broken chip is the only possibility. Contacting TI would be an interesting option too. I would like to hear what they have to say on the subject.

 

[t52]

Quote:

Originally Posted by Alf My understanding is that the chip attenuator is controlled by the driver via USB. The volume is not hard coded in the chip. You move the volume slider in your OS/driver, the chip setting changes. I assume you have played with the slider already. What happens if you move the slider up and down? Do channels still sound different? Do you use Windows? If you are confident about your soldering and your software settings, then I think the broken chip is the only possibility. Contacting TI would be an interesting option too. I would like to hear what they have to say on the subject.

ok, i'll try some more things tomorrow, up to now i ran this particular board with linux only. i'll try windows too... maybe the initial setting for the attenuator is different for l/r and is kept for some reason while adjusting the slider in the mixer. possibly turning down volume to 0 will fix it... this really smells like a software problem, as the sound is very good on both channels, no distortion etc... will report back!

 

[jdimitri]

Hey all, just ordered 2 boards for me to experiment with..
I've soldered quite a bit before, mostly on guitars, but never these tiny tiny chips
Anyone got any guides that i can look to?
Thanks folks, looks like i'm going to come back here with problems soon enough

 

[Alf]

Quote:

Originally Posted by jdimitri Hey all, just ordered 2 boards for me to experiment with.. I've soldered quite a bit before, mostly on guitars, but never these tiny tiny chips Anyone got any guides that i can look to? Thanks folks, looks like i'm going to come back here with problems soon enough

http://www.myexposition.com/diy/usbdac/assembly.html

 

[t52]

just got back from experimenting with my "problem board"

seems like the chip is really broken. same symptoms under both windows and linux. i carefully checked with different mono sine waves and white noise, with white noise i also noticed the right (quieter) channel being slightly more bright - there's something really wrong here.
of course i'm not tooooo confident with my soldering, though everything looks and measures ok. i'd be willing to ship the problem board to somebody else for doublechecking, as this really bothers me... anybody?
will order some more pcm2702s soon as i have three boards left...

 

[n_maher]

I need some advice on checking continuity of the 2702 connections. I tried the "flood and suck" method of soldering my first one down tonight and while the results look good (no visible bridges under heavy magnification) my first test of pins one and two shows a short.

I've tried sucking, and re-sucking and nothing will keep my meter from telling there's a short. And yes, I'm quite confident that I'm holding the probes and different pins. In fact, as I started to go down the line I get continuity from pin one to a bunch of other pins. What!? Is this normal or have I already pooched my first Alien DAC before I basically got started. If so, you can expect a full kit of parts + a PCB for sale in the forums shortly. I've had way more than my fare share of troubles with SMD stuff lately and I'm nearly done for good with it.

 

[MisterX]

You are ok so far.

 

[Yen]

Quote:

Originally Posted by jdimitri Hey all, just ordered 2 boards for me to experiment with.. I've soldered quite a bit before, mostly on guitars, but never these tiny tiny chips Anyone got any guides that i can look to? Thanks folks, looks like i'm going to come back here with problems soon enough

Try YouTube videos, if you like visual examples.

 

[MASantos]

Quote:

Originally Posted by n_maher I need some advice on checking continuity of the 2702 connections. I tried the "flood and suck" method of soldering my first one down tonight and while the results look good (no visible bridges under heavy magnification) my first test of pins one and two shows a short. I've tried sucking, and re-sucking and nothing will keep my meter from telling there's a short. And yes, I'm quite confident that I'm holding the probes and different pins. In fact, as I started to go down the line I get continuity from pin one to a bunch of other pins. What!? Is this normal or have I already pooched my first Alien DAC before I basically got started. If so, you can expect a full kit of parts + a PCB for sale in the forums shortly. I've had way more than my fare share of troubles with SMD stuff lately and I'm nearly done for good with it.

If you have some liquid flux or a flux pen, try to flux the pins and heat them all over again. This has worked very well for me many times and now I do this in every board I solder. Liquid flux is the best since it actually gets under the IC.

 

[Sinbios]

Quote:

Originally Posted by n_maher I need some advice on checking continuity of the 2702 connections. I tried the "flood and suck" method of soldering my first one down tonight and while the results look good (no visible bridges under heavy magnification) my first test of pins one and two shows a short. I've tried sucking, and re-sucking and nothing will keep my meter from telling there's a short. And yes, I'm quite confident that I'm holding the probes and different pins. In fact, as I started to go down the line I get continuity from pin one to a bunch of other pins. What!? Is this normal or have I already pooched my first Alien DAC before I basically got started. If so, you can expect a full kit of parts + a PCB for sale in the forums shortly. I've had way more than my fare share of troubles with SMD stuff lately and I'm nearly done for good with it.

From the datasheet it doesn't look like pin 1 should be connected to anything other than ground and pin 28... are you sure it's pin 1? Pins 13 and 14, 15 and 16 are connected. If you have an unused chip try measuring between the pins on it to see if it's meant to be shorted. If not, try adding flux to the pins and wipe the iron across. This makes any stray bits of solder melt and flow to each individual pin.

 

[tomb]

Guys,
I thought MisterX was saying Nate was OK so far. So maybe those readings are correct?

 

[psilosome]

Quote:

Originally Posted by tomb Guys, I thought MisterX was saying Nate was OK so far. So maybe those readings are correct?

A short between pins 1 and 2? Hmmm - the schematic shows pin 1 going through a cap to ground, whereas pin 2 goes through an inductor to +3.3 volts and through a cap to ground. My guess is a meter shouldn't show a short despite the common ground path via the caps, but I could be wrong.

 

[~n00beR]

Wondering if y'all could help have read through all the thread, but as ever I'm still not 100% sure about my choice.

As I am ordering from farnel, I am mostly going to order the same parts as Alf has suggested in his BOM.

However for parts C2 470uF Sanyo OS-CON, and C13,C23,C33 47uF Sanyo OS-CON.

For C13, C23, and C33 the rating is for 20V, audio wise I presume this has no detremental effects?, but better asking before I shell out on those OS-CON's.

Cost is not too much of an issue I at least want to make 1 all out DAC.

Also how would this compare to an all blackgate dac, as the cost is not going to be much different.

Have not decided for CL, CR yet, I have some 50v 10uf SILMIC II's but may buy the 4.7uF HiQ blackgates.

Thanks in advance guys

 

[MASantos]

Quote:

Originally Posted by ~n00beR Wondering if y'all could help have read through all the thread, but as ever I'm still not 100% sure about my choice. As I am ordering from farnel, I am mostly going to order the same parts as Alf has suggested in his BOM. However for parts C2 470uF Sanyo OS-CON, and C13,C23,C33 47uF Sanyo OS-CON. For C13, C23, and C33 the rating is for 20V, audio wise I presume this has no detremental effects?, but better asking before I shell out on those OS-CON's. Cost is not too much of an issue I at least want to make 1 all out DAC. Also how would this compare to an all blackgate dac, as the cost is not going to be much different. Have not decided for CL, CR yet, I have some 50v 10uf SILMIC II's but may buy the 4.7uF HiQ blackgates. Thanks in advance guys

Why would you want all blackgate dac? Or OS-CON's in all those positions? There is only one place on the board where a blackgate capacitor makes sence and that is CL/CR.

Capacitors have different functions in diferent sections of the circuit and the characteristics of blackgates don't suit the need of all the positions. The same applies to the OS-CON's. Use them only in the C13 position, it is where it makes sence. On the other positions use panasonic FM or nichicon UPW or something similar. Blackgate capacitors in those positions won't improve the performance of the DAC.

A good DIY build is also one where the builder can make a good balance between price and performance.

Sorry if I seem a bit harsh with this post, but I see many people who think that blackgates or elnas in every position will give the best performance and that is not true. And I am not speaking without knowledge, I have done some extensive listening and part selection with the Alien DAC both during the first prototiping stage and with the final boards.

Good luck with your build.

 

[~n00beR]

Quote:

Originally Posted by MASantos ........ Sorry if I seem a bit harsh with this post, but I see many people who think that blackgates or elnas in every position will give the best performance and that is not true. And I am not speaking without knowledge, I have done some extensive listening and part selection with the Alien DAC both during the first prototiping stage and with the final boards. Good luck with your build.

No your words are not harsh at all. I have read this thread through and appreciate your comments as I see you have input alot into this thread, and others.

Thanks again for your advice, im off to ammend my list

Ps my wallet says a big thank you too