[amb]

digger945, the passive filter found in the Wolfson PDF is for a DAC with unbalanced outputs. For WM874x you need to use the same filter on both the hot and cold outputs. To make it even simpler, you could connect each of the hot and cold outputs to the 470 ohm resistor as shown, but instead of running a 4.7nF capacitor to ground after that, just run a single 2.2nF cap across. The rest is the same -- coupling caps and drain resistors to ground.

Note that this passive filter design is 1st order (6dB/octave slope), whereas the active one implemented with the opamp is 2nd order (12dB/octave).

 

[cerbie]

Is the OPA2365 all that cranky of an opamp? The AD8656, FI, warns of impending doom on every other page. If it's not, the datasheet has a nice Butterworth that would just need values tuned for 100kHz [size=xx-small](I would also assume that the phase problems Butterworths are known for are not a big deal at <1/5Fc)[/size]. A shallower passive one would be simpler, though.

 

[MisterX]

Compared to the AD8397 I would consider them both to be non cranky op-amps.

 

[jar]

Quote:

Originally Posted by digger945 /img/forum/go_quote.gif An evolutionary step towards the full featured desktop dac?

Quote:

Originally Posted by amb /img/forum/go_quote.gif glt, thanks for your comments. There is no room on the γ2 board for a microcontroller, nor is there the power budget for it (the whole thing, γ1 and γ2 combined, must run off USB power). We're trying to keep things simple (i.e., no programming, no firmware, etc), so we run the chip in hardware mode. Also, I don't see a need to upsample to any frequency below 96KHz.

I'll definitely build a gamma2, but I second digger's and others' question if the full featured dac is still on the horizon. Microcontroller power consumption seems like it would be a non-issue, but I understand the desire to avoid the complexity. There are plenty of processors that can do what you'd need to configure the DAC for tens of microamps while configuring and tenths of microamps afterward.

What I'd really like to see eventually is a DAC with the same cost and complexity goals as I perceive in the B22. Make it as good as it can be, but don't throw money away. If you need any help on the micro-controller side, let me know, as it's what I do at my day job.

 

[hopeless]

Quote:

Originally Posted by cerbie /img/forum/go_quote.gif Is the OPA2365 all that cranky of an opamp? The AD8656, FI, warns of impending doom on every other page. If it's not, the datasheet has a nice Butterworth that would just need values tuned for 100kHz [size=xx-small](I would also assume that the phase problems Butterworths are known for are not a big deal at <1/5Fc)[/size]. A shallower passive one would be simpler, though.

I've used the OPA2365 in my C&C BOX... no it didn't look crankier than anything else

Except that it made a big (steep and hard) power on and off pop, so be prepared to that.

That is why I suggested the OPA2376 as a real alternative to it, but no one would listen.

Pity because the OPA2376 sounded quite good too, just as the OPA2727 (another good choice).

 

[amb]

Quote:

Originally Posted by jar /img/forum/go_quote.gif What I'd really like to see eventually is a DAC with the same cost and complexity goals as I perceive in the B22. Make it as good as it can be, but don't throw money away.

That's still down the road, but isn't what the γ2 is.

 

[jar]

Quote:

Originally Posted by amb /img/forum/go_quote.gif That's still down the road, but isn't what the γ2 is.

That's all I wanted to hear. I'm sure what you learn from the γ2 will make the eventual no holds barred DAC even better. I won't have the time/money for it for a while anyway, which is the only reason I haven't joined the γ2 prototype team.

 

[MrSlim]

I think the concept is brilliant, and for the majority of the people out there, this will be as high end as it needs to get, particularly with the capability to use the WM8741/2. '

Sure, having access to the software selectable filters, balanced outputs and perhaps a discrete output would be nice, but then it would be a different animal, at a different price point.

And what does it all come down to?
What has to be paid for the extra 3-5% improvement in sound quality? (OK, Im just throwing a number out there, but we all know about the audio laws of diminishing returns, and I'm a big proponent of "bang for the buck". And the y2 is going to have some serious bang.)

 

[DoYouRight]

buffalo goes with b22 I would imagine

 

[amb]

I am still looking to add a few more people to join the γ2 prototyping team. See post #37.
This is not for the total novice, but you don't need to be an advanced builder either. It will be a bit more challenging to solder than a Mini³. If you're interested, please PM me.

 

[digger945]

I would like to add to what amb said above that there are some extremely helpful online tutorials on smd soldering that made building the Y1 a breeze for me. I will try to find the links and post them.

 

[MisterX]

Curious Inventor - Guides : Surface Mount Soldering

 

[Lifthanger]

and a special hint from someone, that butchered his 100% diy opus and got perfect smd joints on the gamma1: get a pen with liquid no clean flux.. its worth at least three times the price.

 

[MrSlim]

Man, I joined the proto team and how I gotta do homework..

Good thing I've finished real school for now (3 years of Info Sec at night)...

Is there a thread that has some good info on SMD work and what the best accessories are?

How about toaster oven or electric fry pan SMD work?

 

[digger945]

Watch the video MisterX linked, get a flux pen like Lifthanger mentioned(I got Kester #186 and #951). No special accessories required, the correct placement of the devices themselves is the hardest part. Most valuable tool in completing the Y1 for me was a pair of straight tweezers.
I still remember plugging it into the computer and seeing the balloon pop up stating "Windows is now installing drivers for your device". Toooo cooool. I wanna learn more.