• These commercial threads are for private transactions. diyAudio.com provides these forums for the convenience of our members, but makes no warranty nor assumes any responsibility. We do not vet any members, use of this facility is at your own risk. Customers can post any issues in those threads as long as it is done in a civil manner. All diyAudio rules about conduct apply and will be enforced.

Input and switch boards for Soekris DAM1021 DAC

Jump to Latest
meldano

meldano

Member
Joined 2005
My solution
 

Attachments

  • Soekris0.jpg
    Soekris0.jpg
    220 KB · Views: 795
  • Soekris1.jpg
    Soekris1.jpg
    394.3 KB · Views: 816
  • Soekris2.jpg
    Soekris2.jpg
    171.5 KB · Views: 788
  • Soekris3.jpg
    Soekris3.jpg
    189.1 KB · Views: 734
  • Like
Reactions: 1 users
carsten.witt

carsten.witt

Member
Joined 2006
@Søren, all,

found the solution for my dam problem with normundss help, thanks again! :eek:

On my other boards i solder a connection for the isolator (from J2 => J3) for fast testing the boards and on my "problem board" i didn't make these connection. :confused:

Now it workss as expected without issues, tested only I2S with my waveio! I will take a look on waveio board where to get 3.3V for the isolator. :)
 
carsten.witt

carsten.witt

Member
Joined 2006
Hi together,

@normundss,
question to the input board, at the connector (input board) for the front panel pin 1+3 how much voltage should there (i have only 0.88V)? On the front panel all led are out.
Measure the 3.3V at the front panel and they are missing.

I powered the input board with 7V, the dam1021 ist powered with +-9V

And i had a higher sample rates distortions (>= 192k), what could it be with the amanero Combo384?
When i remove the input board and "replace" it with my waveio there are no problems.

An externally hosted image should be here but it was not working when we last tested it.
 
carsten.witt

carsten.witt

Member
Joined 2006
Some more measure and testing ... the flat cable was it not made a new one ... the "Power LED" pulses with the same interval as the the sync led from the soekris (on a not uses input), a used input the voltage of the Power LED and the Input LED is around 0V.
Also "Auto" does not work.

is there any test i can make?
 
oneoclock

oneoclock

Member
Joined 2010
V2.3 board fails with 352.8 and 384.
I tried to put EXT-POWER to Amanero board.
I tried another Amanero board, if mine was damaged.

Finally the problem was not Amanero board, is Murata NFL21SP106X1C3D. Eliminating this 3 filters, V2.3 board works perfectly at 352.8 and 384.

attachment.php
 

Attachments

  • V2.3.jpg
    V2.3.jpg
    137.8 KB · Views: 753
N

normundss

Member
Joined 2009
oneoclock said:
V2.3 board fails with 352.8 and 384.
I tried to put EXT-POWER to Amanero board.
I tried another Amanero board, if mine was damaged.

Finally the problem was not Amanero board, is Murata NFL21SP106X1C3D. Eliminating this 3 filters, V2.3 board works perfectly at 352.8 and 384.

attachment.php
Click to expand...

That is true, the data line filtering is too aggressive for 3xx kHz and DSD256 sample rates to work reliably.
There are two ways to fix it:
1) Replace the bottom two CF filters with NFL21SP206X1C7D. The middle filter already is NFL21SP206X1C7D. I am still working on verifying this solution for all cases that I can think of, but I am fairly confident it will work for all builds.

2) Replace the bottom three CF filters with 0805 size 47R resistors. While this will not attenuate other high frequency noise, it will at least reduce the ringing of I2S signal transitions.

If you do not need the highest sample rates, and use sources up to 192k and DSD128, there is no need to change anything.

The new batch currently on order will have this fixed.

If you really need to run the highest sample rates and are not able to replace the SMD parts yourself, you can mail the board to me and I will replace the filters. Just notify me by PM before sending anything.
 
carsten.witt

carsten.witt

Member
Joined 2006
@oneoclock,
thanks for the hint, i tryed this but it is the same problem. I'll remember when the input board works if the problem are still there.

@normundss , all,
it seems to be the problem is on the input board. Without the input board (amanero I2S direct to the dam) all works as expected (all sample rates and all dsd files).

At input board i removed the R3 and power the raspberry pi with the "+5V OUT", at the input board "+7-9V IN" is powered with 8V, the jumpers "EXT POWER" and "LED ENABLE" are set.

Any other hints?

An externally hosted image should be here but it was not working when we last tested it.
 
N

normundss

Member
Joined 2009
It looks like your Amanero is unmodified and thus outputs 3.3V in parallel with the input board 3.3V supply. That is a no-no. For a quick test, open the EXT_POWER jumper to supply 3.3V from Amanero.
A better solution would be to modify the Amanero not to use power from USB bus. Details on that are linked in post #325.

The other problem with missing 3.3V on the switch board is unrelated, please check the voltages at the regulator there as I previously described.
 
oneoclock

oneoclock

Member
Joined 2010
normundss said:
1) Replace the bottom two CF filters with NFL21SP206X1C7D. The middle filter already is NFL21SP206X1C7D. I am still working on verifying this solution for all cases that I can think of, but I am fairly confident it will work for all builds.

2) Replace the bottom three CF filters with 0805 size 47R resistors. While this will not attenuate other high frequency noise, it will at least reduce the ringing of I2S signal transitions.
Click to expand...

Thanks Normundss
 
carsten.witt

carsten.witt

Member
Joined 2006
@oneoclock,
thanks again for the hint, see below. :)

@normundss,
thanks for your advices
- i removed the U2 from amanero and powered the input board
plays as before, distortions >192k
- i replaced the three CF filter with 3 x 2x100R :cool:
no distortions anymore at all sample rates, great :D
=>i'll order end of next week the NFL21SP206X1C7D and test again

The 3.3V problem was very simple to fix, my own stupidity was to blame. :down: I forgot to solder the connector from the input board to the dam (J2). Stupid, it cost me so much time, without words.

A last question (hopefully), i think i read here in the thread about this but on a quick look i didn't find it.
Works "auto" withe the RaspberryPi and the amanero together?
My build doesn't switch between Pi and amanero automatically only manuel input select works.

Many thanks to all helping hand! :worship:

An externally hosted image should be here but it was not working when we last tested it.




@awietek,
looks nice, what for a enclosure it is?
 
N

normundss

Member
Joined 2009
carsten.witt said:
- i replaced the three CF filter with 3 x 2x100R :cool:
no distortions anymore at all sample rates, great :D
=>i'll order end of next week the NFL21SP206X1C7D and test again
Click to expand...
Don't worry too much about the filters. They are a bit of an experiment, and I do not claim they will make a noticeable difference vs resistors. Except when they don't work :blush:

A last question (hopefully), i think i read here in the thread about this but on a quick look i didn't find it.
Works "auto" withe the RaspberryPi and the amanero together?
My build doesn't switch between Pi and amanero automatically only manuel input select works.
Click to expand...
Auto selection is done in DAM1021, and since it has only one I2S input, it can not auto select between Raspberry and Amanero. In Auto mode the input board feeds I2S-1 (Amanero/DIYINHK) to DAM, so I2S-2 will be ignored.

With a small hardware modification it would be possible to change the Auto selection to I2S-2 (Raspberry), but there is no logic on the input board that will auto select among several I2S inputs, and DAM can not do that either.

To be honest, I find the Auto select feature somewhat useless, since DIYINHK USB interface will always output bit clock signal when powered on. That results in it always being auto selected, even if USB cable is not plugged in. Soeren has even said he considers the behaviour of turning off the clocks to be a problem with Amanero.

If you are really desperate for auto selection of all inputs and are using Raspberry or another microcontroller, you could write an externally controlled auto selection routine by cycling through various inputs and monitoring LXXX output on DAM serial line...
 
N

normundss

Member
Joined 2009
I just received a shipment of connectors from China for the new batch of boards. The supplier has sent the wrong parts again, for the second time ! :mad: :headbash::headbash:

I have now ordered them from another source, hopefully this works out better. This supplier is more expensive, so I will probably need to raise the input board price by 1 EUR.

Sorry to keep everyone waiting. The boards will be available as soon as I get the connectors.
 
You must log in or register to reply here.
Top Bottom