DIY AD1853 and AP measurement

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Share with you guys my diy AD1853 DAC.

Simple design, no ASRC or reclock.
DIR1701 > AD1853 > OPA2604 I/V > OPA2604 LPF

There is an Elma selector and a DACT volume pot at the output.
 

Attachments

  • 1853dac-pre-toptview.jpg
    1853dac-pre-toptview.jpg
    77.3 KB · Views: 1,115
ezkcdude said:
Banana, you have some nice DIY speakers on your website. I built speakers with the SS8554 as well. Are you going for a complete DIY stereo system? That's my goal.

Ezkcdude, that 8554 was made long time age and the foam surround is deteriorating. I would use 1 inch dome to mate with 8554 instead if i was to do it again.

There's more if you can read Chinese. My next target is UCD400 and Manger active 2-way, but it already took me to long to build.

12M4631 3-way
Manger never completed
 
banana said:
Bernhard, thanks for the file. Which kind of dither is it?

Lower noise floor might be good thing, but I'm affraid there's no chance for us to enjoy 24bit audio source in foreseeable future.....

No idea what kind of dither. I did not make it, look at the link address.
Please measure :)

16bit data + os filter + >16bit DAC also gives lower noise floor.
 
banana said:
To have better view of harmonics, I applied 32 times averaging and do the -60dB FFT again.

K3 now appears, and also a tiny bit of K5.

Hi Banana,

Your results appear very good, even considering use of the
OPA2604 which isn't the best of opamps.

I recommend you try some OPA627's. The measurements
-may- improve but the sound definately will.

Is this with 1853 in mono or stereo mode?

Cheers

Terry
 
Bernhard said:
Hi Banana,

thanks for posting the graphs, looks good which is expected from a quality delta sigma DAC.

Bernard,

These are delta sigma but they are, from memory, 5 bit unity weighted.
So out of band noise is much better than the old 1 bit designs.

They also have proper ciurrent sink OP as opposed to SCF that
Cirrus and AKM uses. I like the AD approach better, although there
isn't that much in the measurements these days.

Even compared to the much praised PCM1704, measurement wise,
the 1853 gives nothing away to the BB chip and betters it in some
areas.

So I have to ask the question, are the old R2R dacs still better?
Or is it really a question of implementation.

Not many people are aware, but the AD1853 has a current set
resistor, (R28 p 14 data sheet) to pin 10 which can be used as a
true analog volume control.

I know of no one who has used this feature but it is
potentially a very good idea, making totally pre amp free signal
chain possible.

Cheers,

Terry
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.