Soekris' DAC implementations

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
It was said that using I2S source, ISO +3,3V pin should be fed with external power supply (ex. from USB-I2S interface). I noticed strange thing: after disconecting external +3,3V isolators still works fine and on ISO +3,3V pin I got 2,28V...
Is it normal?

I have noticed this, also.

I am using a battery to power the isolators.

A few days ago I was listening and thought it was a little hashy. Eventually noticed the battery was not connected. I did not expect it to work but I know nothing about those things.

I did not make any voltage measurements.
 
For maintaining integrity, I just had cut the USB 5V PS line, it's easy, a clearly exposed trace. This way the onboard 5V >> 3.3V local regulator is kept, and by a solder drop one can always restore for only usb operation.
External supply clearly helps, but keep grounding local: practically the external supply ground plain has to join directly that one of the amanero: best is a sandwich of the two with bottom sides together. I'm using a texas eval board.

Ciao, George
 
if both were in stock configuration single dac board
soekris dac is much better in every way but flow/fluidity

I'm still trying filters and modding to beat my version of the dddac
I think it will as it's far more detailed
definitely try it
good project :)

Hi Nige,

Stock single board dddac was also not my preference but my dddac has evolved to a fully tweaked 4 board design that is far better than the stock one. I am just wondering if this dac has the potential to be as good or (preferably) better than a tweaked out dddac.

What I read on the other soekris thread is that the sound is not so good, seems fairly harsh sounding and flat imaging. This maybe due to the used filters ( I understand that there is still improvement going on in filter programming).

On Hifiduino are some links to sound reports:
https://hifiduino.wordpress.com/2015/02/22/dam1021-r-2r-dac-mods/

I just want to prevent investing money and time in a project that eventually ends up at a dead end because of design flaws that are impossible to modify.

Regards,
 
Hi Nige,

Stock single board dddac was also not my preference but my dddac has evolved to a fully tweaked 4 board design that is far better than the stock one. I am just wondering if this dac has the potential to be as good or (preferably) better than a tweaked out dddac.

What I read on the other soekris thread is that the sound is not so good, seems fairly harsh sounding and flat imaging. This maybe due to the used filters ( I understand that there is still improvement going on in filter programming).

On Hifiduino are some links to sound reports:
https://hifiduino.wordpress.com/2015/02/22/dam1021-r-2r-dac-mods/

I just want to prevent investing money and time in a project that eventually ends up at a dead end because of design flaws that are impossible to modify.

Regards,

well your gonna have to decide for yourself
it's definitely more detailed
the filter's will play a huge part
I think it will surpass
waiting and letting the dust settle will answer more questions
 
I tried MixedPhase1, MinPhaseSlowRev1, BAM-COMB1, TotalCRaP_rev2, CRaPMagic.

Right now, I plan to stay on CRapMagic for a little bit. I have been using MinPhaseSlowRev1 for a while. Also tested the others. MinPhaseSlowRev1 is very smooth, mellow, easy to listen. CRapMagic seems to bring things to sharper focus. I can hear all the details with MinPhaseSlowRev1, but they seem much more obvious on CRaPMagic. I think CRaPMagic also has better imaging probably due to enhanced details. It also feels more dynamic compare to MinPhase. The cost is that the "sss" sound or sibiliance in some vocal tracks also become slightly more emphasized. But it is a tradeoff I am willing to make.
 
1021filt_35HR_LP.skr 1021filtNewNOS.skr 1021SA1.skr CRaPMagic_LP.skr
1021filt_MixPhase1.skr 1021filtNQR_NOS.skr 1021SA2F2v1.skr totalCRaP_rev2.skr

Tried all of these, the CRaPMagic appears the most transparent and "filling". All of the NOS filters just clip for me on certain louder tracks and I'm not using a volume pot.
 
Going back to the optical input I sadly need more than one. Currently searching for ideas but first I need to find useful optical receiver. And maybe add a transmitter so I can tap into the data stream for easy AB listening. I prefer looking for parts in current production as an update to TORX147’s max 96kHz.

The ADAT standard make use of the same optical cables used in commercial audio. Modern implementation are able to feed 192kHz/24bit stereo though this cable. But so far I’ve drawn blank trying to find info about optical devices these vendors use in current productions. Anyone had the lid of a multichannel sound cards and had a look inside?

From the TORX147 datasheet there is an interesting read:
“Noise resistance: It is believed that the use of optical transfer devices improve noise resistance. In theory, optical fiber is not affected by noise at all. However, receiving modules which handle signals whose level is extremely small, are susceptible to noise.”
And then they go on with tip to filter power supply as spzzzzkt already have shown us. But less known I think are the need to screen the device. Picture below show an example on how to screen the optical device. Sadly the screen also prevent me from prying on the optical R&T devices.
An externally hosted image should be here but it was not working when we last tested it.
 
Anyone believe the LVC595’s could benefit from being thermally connected?

I’ve read somewhere that early R-2R test done by others had problems with uneven on resistance in the 595 device used in their early implementation. Of course the LVC version used here is of much higher quality and If I remember correctly Søren measured switching speed at 70us.

What I’m thinking is a thin sheet of copper covering all the resistors and LVC595’s. With double sided thermal stickers on each LVC.
 
I’ve just wired up the DAM dac, fed by a Rpi running Volumio and fire it up for the first time. But I’m not yet getting any sound. When I play a track the green led on the DAM comes on solid which I suspect indicates it is sensing a signal. I’ve rechecked the I2S pin-out, the ISO side it being fed 3.3V. I have selected the generic I2s output within Volumio.

Some of you have got this to work. Is there anything else I need to configure on RPi?
 
Last edited:
I’ve just wired up the DAM dac, fed by a Rpi running Volumio and fire it up for the first time. But I’m not yet getting any sound. When I play a track the green led on the DAM comes on solid which I suspect indicates it is sensing a signal. I’ve rechecked the I2S pin-out, the ISO side it being fed 3.3V. I have selected the generic I2s output within Volumio.

Some of you have got this to work. Is there anything else I need to configure on RPi?

I spent quite a bit of time on this and have some info for you, PM me for a few suggestions.
In general the only super stable method I have found for Rpi i2s to DAM1021 is running Logitech Media Server and Squeezecenter. Easy and glitch free.
 
I spent quite a bit of time on this and have some info for you, PM me for a few suggestions.
In general the only super stable method I have found for Rpi i2s to DAM1021 is running Logitech Media Server and Squeezecenter. Easy and glitch free.

Thanks Robertrowett, I wasn't planning one using squeezelite/box/centre.

Do we know of anybody else who managed to get RPi/Volumio to work as a transport for the Dam1012?
 
Could some kind people quickly confirm that the serial - USB converter with output 5V is OK to connect with DAM?

I just bought Unitek Y-1050 from a local store, measure with multimeter I got GND - TX is -5.5V

Thanks a lot!
 

Attachments

  • FullSizeRender.jpg
    FullSizeRender.jpg
    116.5 KB · Views: 455
Last edited:
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.