my DAC design - pup1 DAC
Dec 8, 2012 at 2:31 PM Post #571 of 675
It's alive! 
etysmile.gif

 
This has to be a record for me, package came in the mail yesterday, started it and stayed up late last night, got all the SM parts on front before I went to bed.
This morning, did the SM on the back side and all the TH parts.
 
Did have a few problems, all my fault.
I had a solder bridge on the dac chip, found it because I decided to ohm out between each pin to make sure things were OK  Could see it, even under magnification, but a couple pins ohmed to 0, so I used some wick to remove and then it was fine.  FYI, if you ohm out some pins end up getting tied together that are not obvious so you need to check against the schematic.
 
When I plugged it in, USB device was not recognized.  I had a few of the voltage regs that I didn't solder correctly (it was late last night).  Once I fixed those, it was recognized, device installed, and it made music
L3000.gif

 
I did hear some "static" at first.  I dried it out a little with a heat gun at low, carefull not to heat too hot, and put it in the case, then static went away.
1mv offset on left, 0mv offset on right.  Used the "default" opamp.
 
It has plenty of drive for an iem, and for no breakin with cheap buds it sounds really good. 
 
Will let it run for a while, and then try with better iems.
 
I did make a few changes, just because :wink:.  I used some different caps around the dac, and replaced some of the lytics with other ones I had, silmics and an organic polymer.
I also lost one cap in my carpet :frowning2:, so it was replaced with the closest I had to it.  That was a 2.2uf, it is now a 4.7uf.  Had to improvise that one becaue it was a bigger package.
Of course mine does not look nice like tomb's example, but it works.
 
Kit was very nicely put together, made it easy to build, everything was well labeled and packaged.  Really worth it to buy the kit instead of trying to buy the stuff yourself from Mouser.
My one suggestion would be to add reference designations to the labels, especially for the IC's, where you can.  I know that is one IC used in multiple places, but all the rest are unique so they could be labeled like this, if you don't use the same IC's for other kits.  Would make it easier.
 
Randy
 
Dec 8, 2012 at 2:54 PM Post #572 of 675
Glad to hear it, Randy!!
 
About the parts labels - it was one of those things that happen when working remotely with people on a project.  I had many conversations with the salesperson at Mouser and we were both emphasizing that each series of parts would be individually-bagged and labeled with "part numbers."  It wasn't until I received the entire volume of kits that I realized their "part numbers" were different than my idea of "part numbers."  That's why I emphasized referencing the BOM and include one in every kit.
 
This was the very first time I ever had Mouser completely make up the parts kits.  I think they did a great job.  We just blew it on communication about the part numbers.  It'll be fixed on the next round. 
 
 
P.S. Told you guys about that moisture thing.  It's gratifying to hear that you didn't lose your cool and got out the heat gun.
wink.gif
wink.gif
  
 
Dec 10, 2012 at 10:04 AM Post #573 of 675
So when I listened right after I built them I was a little disappointed with the sound, but after letting it run in a little, I think it sounds really nice.
 
I mostly connected it to my O2, driving some IEM's.  I can drive IEM's straight out of this, but it has too much gain, so I have to be at a low volume in windows.
 
I looked at the artwork, and it looks easy to cut the traces to the headphone jack, and connect a small resistor to the jack instead.  I'll probably start with 50ohms, and see how that goes.
 
Randy
 
Dec 10, 2012 at 6:05 PM Post #575 of 675
I've been told by another person besides Randy that a few of the capacitors' part #'s are different than on the BOM.  When in doubt, please check the part number's rating at Mouser and reference that with the rating on the BOM.
 
I will have this documented later tonight, will post it, and include corrections on the BOM since I'm including a printout in the kits.
 
Dec 11, 2012 at 2:02 AM Post #576 of 675
So I cut the traces to the headphone jack tonight and added a 50 ohm resistor in series to the jack.
 
First resistor was too big (a resista) so it didn't fit :frowning2:
The board wouldn't slide back into the case.
 
Found a smaller dale which is the same type in the kit I think, and it just fits :)
 
It helps, before when I plugged directly into the jack before I was almost at min volume, now I am maybe at a 1/3.
 
But I am worried if I make the resister bigger it may degrade the sound, so I might try lowering the opamp gain now.
 
I will probably try using a 1k for the feedback resistor, and see how that works.
 
Randy
 
Dec 11, 2012 at 6:36 PM Post #577 of 675
No offense, but that seems like a lot of cutting on a DAC that was primarily designed as a source.
wink.gif
  I mean, we've said it should power efficient IEM's  or the like, but when I've tried it - I just used the volume control on Foobar.  I know that's heresy, but it's not like it was a permanent scenario - just something fun to try outside of the primary purpose as a source.
smily_headphones1.gif

 
Dec 11, 2012 at 9:23 PM Post #578 of 675
OK - here is the story on the kit parts vs. the BOM.  I am sorry for the confusion, but there are only two parts bags that have different part #'s than the BOM and that are not totally obvious:
 
C1, C23, C30, C37 - the kit part # is GCM2195C1H103JA16D
C7, C31 - the kit part # is GRM2185C1H153JA01D
 
Even so, the parts counts of the two items match exactly and there is no other possibility of confusing these parts with others.  Plus, one could always look the parts # up on Mouser to be certain.  Nevertheless, the intent was to make it simpler than that.
frown.gif

 
BTW, I've made this BOM different than all the other BOM's I made in that if you actually click on the link for a part, you should get a multi-part search result on Mouser that lists all of the parts available with the required ratings.  The part #'s in question, along with the part #'s that were on the original BOM, are all in that list when you click the link.  The reason the mixup happened is that while negotiating with Mouser over making up the kits, several of the parts lines went temporarily out of stock.  To simplify things and get the kits going, I told Mouser to use the most inexpensive, available part that had the qualifying ratings.  So, while I standardized the small 805 caps on the BOM with Kemet "C0805" part numbers, Mouser substituted Murata caps for the two items above.  They were still in the list if you clicked on the link in the BOM.
 
Regardless, I have changed those two part items to match the bags in the kit.  I've also reduced the part # text string for the other 805 caps, even though 75% of the string matched the kit parts bags exactly - only some qualifying suffix characters are different (such as "ACTU" or "ACAUTO").  These really have no bearing on the ratings of the part, which comes in all the numbers/characters before that.  That may not have been totally obvious, but hopefully reducing those parts' text strings will remove confusion.  Just as a for instance, clicking on the BOM Mouser link for "C6, C8, C11, C12, C17" results in 9(!) pages of parts, all of which meet the required ratings.  The important part of the text string is simply, "C0805C104K."  Even then, one of the other mfrs with totally different part #'s can be used - if it's in the list for that rating.
 
The following are also different part #'s on the kit bags than on the BOM, but surely these are totally obvious:
X1 - the clock oscillator (the kit uses a series C3392 instead of C3391)
LED1 - the LED (it's 3mm and it's red ...)
C3, C18 - the Mica caps (there's nothing else that remotely looks like purple hearts with leads sticking out)
wink.gif

I'm not going to worry about changing the part numbers on these items.
 
All the other changes have been made and uploaded to the pupDAC website.  Future kits will include printouts of the revised BOM.
 
Dec 12, 2012 at 8:27 AM Post #580 of 675
My understanding is nothing is "wrong" per se.

The issue, again to my understanding, is that some of the part numbers in the kits do not match the part numbers on "the BOM". This will mean that you as builder may need to take a step or two more to ensure that you put the right part in the right place.

The issue is that there are many, many parts that could fit a position on the board. For example, a 1uF SMT 0805 X7R capacitor. You could also use X5R or X8R. For stocked parts, Mouser has 137 different parts that fit that bill. All of them will work, but based on past history/bias/phase of the moon, we choose a part for a position. And then when tomb makes a kit, for some reason, another part may be put in the kit instead due to stocking issues. An example of this is X1 - Mouser changed the minimum qty on the part we were using, so the part had to be changed.

As a result the BOM is a living thing, that changes depending upon what is in stock when parts are ordered.

So, to my understanding, there is nothing wrong with the BOM or the kit - the issue is that some parts are not labelled in a way that is immediately clear where that part goes on the PCB. You may have to check the part value, to the BOM value to be clear where it goes.
 
Dec 12, 2012 at 11:59 AM Post #581 of 675
Quote:
No offense, but that seems like a lot of cutting on a DAC that was primarily designed as a source.
wink.gif
  I mean, we've said it should power efficient IEM's  or the like, but when I've tried it - I just used the volume control on Foobar.  I know that's heresy, but it's not like it was a permanent scenario - just something fun to try outside of the primary purpose as a source.
smily_headphones1.gif

 
[size=medium]Hi[/size]
 
[size=medium]First, let me say this is not intended to be argumentative, just want to present my point of view on my “cuts”.[/size]
 
[size=medium]It is “interesting” that the PupDAC was intended to be a source, but I think that is selling it short.  I started out using it as a source, with my laptop and an O2 as the amp.  But since the PutDAC has an opamp, it seems to be quite capable of driving IEM’s.[/size]
 
[size=medium]It makes a nice, easy to use, portable setup, using my laptop, PupDAC and a set of IEM’s.[/size]
 
[size=medium]My only “complaint” is that the PupDAC has too much gain for my IEM’s, using mostly TDK BA200, RE-262 and Sony MH1-C at this time.  The Sony is the hardest to drive, but still with a stock PupDAC, I am very low on the Windows volume scale for my listening levels.[/size]
 
[size=medium]So I am making some mods to my PupDAC to make try to get the windows volume at least above 50% for my use.  I am also worried about data loss using digital volume control so I like to be high up on the scale when I do use it.[/size]
 
[size=medium]As for the cuts, there are two required, and they are easy.  I made the cuts on either side of the headphone connector, top side of board.  The traces go from the RCA jacks output to the headphone out jack, one coming from each side.  These traces are all by themselves, with NOTHING around them, so it really would be hard to mess it up.  When cutting, I make sure the trace ohms out before cutting, then cut and make sure that it no longer ohms out.[/size]
[size=medium]Before I added the 50 ohm resistor, I did notice some hiss on my IEM’s when I would pause the music.  That hiss is no longer audible after this mod.[/size]
[size=medium]I am also considering other “mods” to my PupDAC, just because I’m an engineer and I can’t leave well enough alone.[/size]
 
[size=medium]As for the parts “substitutions”, I wanted to add that I don’t consider the parts to be mistakes.  My company does it all the time, depending on what is available for purchase.  When you use a standard size and value surface mount cap, there are probably at 5 different companies at Mouser that make basically the same part.  If the size, value, voltage rating and dielectric are the same, we would consider them to be equivalent and use them interchangeably.[/size]
[size=medium]It just may be a little confusing if you don’t understand how capacitors are numbered.[/size]
 
[size=medium]Randy[/size]
 
Dec 12, 2012 at 2:06 PM Post #582 of 675
I guess if you're already at the low point on the Windows volume scale, then all that makes sense.

And yes, there is nothing "wrong" with the BOM.

Thanks cobaltmute and good post, Randy!
 
Dec 12, 2012 at 2:21 PM Post #583 of 675
It is “interesting” that the PupDAC was intended to be a source, but I think that is selling it short.  I started out using it as a source, with my laptop and an O2 as the amp.  But since the PutDAC has an opamp, it seems to be quite capable of driving IEM’s.


I agree with your point, but one thing to be aware of is that the negative rail generator can only create 60mA before it limits.

So, yes it can drive headphones. But at loud volumes or larger current requirements, the negative rail to the opamp could collapse in all sorts of interesting ways.
 
Dec 15, 2012 at 12:51 AM Post #584 of 675
So I played a little with gain, and it is better (for me) now.
 
You need to change R6, 8, 17 and 19.  What matters is the ratio between R5, 7, 16 and 18 and these other resistors, ratio is 15 to 1 using the default values.
I actually changed all of these resistors, and changed the ratio to 7 to 1, and am around 1/2 on the volume scale now.
 
I may drop it a little more, to around 5 to 1, and will probably settle with that.
 
Randy
 
Dec 18, 2012 at 1:31 AM Post #585 of 675
Got a box with a picture of a puppy on it today.
Checked all the items off the BOM during my lunch break and everything is accounted for. The packaging for the OPA was a bit ridiculous, a silica gel pack and moisture indicator card, it took up more space than the rest of the parts combined. 
beyersmile.png

Hopefully i can find some time to construct it tonight.
 

Users who are viewing this thread

Back
Top