Amanero Combo 384 digital distortion/Ciúnas USB converter

Hi after using the Ciúnas USB converter for about 5 hours the output becomes very distorted after turning the devise off/on it works fine but then the same can happen after an hour,ive replaced the Lifpo4 battery and the psu still the same with a steady voltage 3.45v. Inside is a modified amanero combo 384 my question is, is it ok to upgrade the firmware and will this resolve the issue or should i be looking else where ?
 
Did it work okay before it was modified? Did it always have this problem? If not, when did the problem start happening? Are you using it with a dac? If so, did you try with different dacs?

Is it one of these: https://www.tnt-audio.com/sorgenti/ciunas_iso_spdif_e.html

The first problem for us reading in the forum is that we don't have enough information to have a good idea of what you should do. We would probably take a look inside before doing anything else. Can you post high res pics? If so, that can sometimes be very helpful.
 
Fair point ,The modification was part of the original build Jkenny improved on the clocks and jitter in other area's ,working perfectly for 4 years ,similar model to your post ,never used to do it, fairly predictable occurrence now after 8hours on/off then after an hour ,feeding a Metrum Dac ,i will try swapping the DACs out,i didn't think of that thanks, there is a firmware upgrade for the amanero combo 384 https://www.amanero.com/ mainly for DSD which i dont use, Q could the firmware upgrade interfere with the clocks ?
 

Attachments

  • IMG_20230906_104147.jpg
    IMG_20230906_104147.jpg
    469.3 KB · Views: 118
  • IMG_20210627_114851.jpg
    IMG_20210627_114851.jpg
    407.5 KB · Views: 103
  • IMG_20210627_114724.jpg
    IMG_20210627_114724.jpg
    372.9 KB · Views: 86
  • IMG_20210627_114649.jpg
    IMG_20210627_114649.jpg
    382.8 KB · Views: 124
A couple of thoughts:
1. Depending on what the modifications are doing, its possible that changing the firmware could break it completely.
2. Can't tell much from the pics. Need to know part numbers of the devices on the added circuit boards. Pics aren't good enough to read the numbers. Also need to know where the added circuitry connects to the Amanero board. To the user interface pin header area at the end of the Amanero opposite from the USB connector? or does the added circuitry connect to Amanero somewhere else?
 
Is the battery directly connected to the output of the LM2596 buck converter? If yes it’s not the proper way.
The chip is a WM8804, a DIR. How do you choose input signal of your DAC between the USB and the SPDIF ?
It looks like the battery is shorted (!)...but it‘s likely a visual effect :
1695035053420.png


As said by Markw4, a schematic is required to start troubleshooting.
 
If a WM8804, apparently its a SPDIF transceiver, not just a receiver? Also, looks like a 24.576MHz clock too. Workmanship looks poor. Could be a bad connection somewhere. Is there an MCU, or is WM8804 in hardware mode? Maybe WM8804 is using the wrong clock source and eventually there is enough drift between clock domains that it stops working properly?
 
Good point, According to WM8804 DS, in hardware mode; clock source of the SPDIF TX Module could be MCK from the I2S (amanero) or the external clock. It depends on the status of CSB/GPO1 pin at reset. If the external clock is used, spdif signal will be generated with clock and data not in the same clock domain, which can explain the issue.
 
A couple of thoughts:
1. Depending on what the modifications are doing, its possible that changing the firmware could break it completely.
2. Can't tell much from the pics. Need to know part numbers of the devices on the added circuit boards. Pics aren't good enough to read the numbers. Also need to know where the added circuitry connects to the Amanero board. To the user interface pin header area at the end of the Amanero opposite from the USB connector? or does the added circuitry connect to Amanero somewhere else?
1.good point, ive made contact with Amanero and his thoughts are aline with both of yours .ive managed to get some more photo's, under the CN224 board is another chip upside down unless i start desoldering i just cant see it.but maybe these photos will help.if you need a specific area photographed let me know.
 

Attachments

  • IMG_20230920_203357.jpg
    IMG_20230920_203357.jpg
    561.9 KB · Views: 61
  • IMG_20230920_203126.jpg
    IMG_20230920_203126.jpg
    438.7 KB · Views: 62
  • IMG_20230920_203021.jpg
    IMG_20230920_203021.jpg
    449.6 KB · Views: 62
  • IMG_20230920_202921.jpg
    IMG_20230920_202921.jpg
    476.5 KB · Views: 53
  • IMG_20230920_210019.jpg
    IMG_20230920_210019.jpg
    344.3 KB · Views: 49
  • IMG_20230920_204008.jpg
    IMG_20230920_204008.jpg
    262.3 KB · Views: 57
From looking at the pics I can't tell exactly what is connected to what else, and or if there are any bad solder joints and or broken connections. If I had it here I would look at it under a 10x microscope and examine all the connections visually, and or wiggle them a tiny bit to see if they are loose. Also I would look for cracked solder joints.

If no explanation found, then the next step would probably be to sketch up a schematic. The way to do that is get the datasheet for each chip, for a standard clock module, and for the Amanero board pinout. Datasheets contain all that information.

Then get a big piece of paper and draw the Amanero board, the WM8804, etc. Then using start drawing in the wires to show what pins of one device are wired to what pin or pins of any other devices. Also, with power removed from the unit and from the Amanero USB connection, use a DVM on ohms range as needed to verify what is connected to what else.

Once you have all the wires and components drawn then we can use the schematic you just made to analyze the circuit to see how it is designed to work. Then we check to see if it is working as designed. For that you might need more than a DVM, we would have see when we get that far.