Hallo,

we want to commissioning our 4Mux Board and select via Microcontroller the Input Channels.

First we start the board stand alone. The power supply cames from a Labor power supply.

1. I saw the DATA at the output when I connect a spdif Signal.
2. I connect the I2S Signal via Minishrac -> Buffalo DAC to a IV Converter.
3. The signal what I saw was very bad.

Through the same line I sent a USB Signal via Amanero -> Minisharc -> Buffalo DAC -> IV Converter.
The Signal was perfect.

So it comes from the Receiver. What I measured with my oscilloscop when I changed the channel a pulsing on the I2S Line although the Input Signal was not connected. This pulsing produced a DC voltage and Interferences on the output.

I tried different startups, changing SW1, but nothing happends.

Anybody have had the same problem. Also I attached the oscillocop views.

Thanks

Greetings
Henning

Output_After_Switch.jpg (97kb) downloaded 6 time(s). Data_after_Switch.jpg (158kb) downloaded 8 time(s). Output_Signal_after_DAC.jpg (155kb) downloaded 9 time(s).

A picture of the test rig might help spot an issue.

Hallo Russ,

I tested the board stand alone. The digital signal what you see is the Output from the board DOUT when I switched the channel via jumper. No Input Signal. This produced the board at the output when no Input is selected.
First I thought also an EMI problem or a problem in the signal routing through the minisharc board.
Through the same way I send an I2S Signal from the USB receiver -> The signal was perfect at the output from the bufallo dac.
It is possible that could be a problem with the HW Mode of this chip ?
How must be the signal at the DOUT, BCK and LRCK without Input Signal ? Continues low ?


Thanks

Greetings
Henning

Hallo Russ,

I found that in an other forum. That describes the problem what I see.

Other forum.
I had to scrap a dozen boards with the CS8420 chip on it because of the "funny state" bug - disconnecting/reconnecting an input or otherwise causing a PLL unlock/relock would make the chip spit out garbage. Their recommendation was to reset the chip after a reconnected input, but even this would cause screwups on rare occasions. I sent them scope plots, and they were stumped... eventually they asked me to send them the physical chips I had to see if there was a problem with the particular batch!



Greetings
Henning

Hello,

ok. what do you think about the DOUT Signal when I changed the channel via Jumper ?

/Henning

Tell me about the signals - are they both consumer level?

I see only one properly connected source in your picture.

You could try simply terminating unused inputs to GND.

Try terminating unused inputs to GND with something like 470R-1K - or directly with a shunt wire if they will never be used.

Another very important note. For most I2S targets (all of our DACs to be sure) unless there is a bitclock and wordclock signal the state of the DATA signal is completely irrelevant. So you may be worrying over something not at all important.