Connecting audio output to an amp with shared power supply (same Ground)

#1

Hi,
I plan on connecting one audio output channel to an amp with a shared power supply. The amp itself has a direct connection from the power supply ground to the Ground of the audio output ("sleeve" of the plug, I guess). Is it safe?

In axoloti, there is a resistor from the "Ground" of the audio output to the real power supply ground... If I connect to the amp's inputs, I will effectively shortcut the "audio ground" to the real power ground. Any problem? Any pointer to something I should look into/study?

To be precise, I plan on using this adafruit amp built around a MAX9744: https://www.adafruit.com/product/1752

By the way, I read this interesting thread: https://sebiik.github.io/community.axoloti.com.backup/t/might-have-fried-the-output-of-axo/573
Thanks!

#2

are you sure? i have not checked, but it would be strange. about every amplifier has ground connected to audio-connectors. so if there is indeed a resistor in axoloti it has to be expected to be shorted. or are you talking about the headphones output? this is biased and has a "virtual" ground. (and should therefore not be connected to line in of an amp)

#3

Thanks, I was talking about the axoloti normal 1/4 output, not the headphone out - I think I saw that on the schematics, will double check. If you say most amps have this direct ground connection in the audio jack, then it should be safe anyway, indeed.

#4

Just getting back to confirm: I just tested a shared 12V power supply, and it looks like everything is working fine.

#5

i measured with a multimeter, and indeed there seems to be a 22 ohm resistor. maybe some pop protection?

#6

It's what I guess as well, pop protection for the moment when you connect a 1/4" plug, when the tip touches the "sleeve" connector (i.e. "GND").

#7

The line output ground terminal has indeed a 22 ohms resistor to real ground.
The reason is to avoid ground loop noises.

Without any countermeasures in the design, here is a common scenario that would cause significant ground loop noises:

  • Axoloti Core is connected to a computer's USB port

and

  • Axoloti line out is connected to an unbalanced line input of a device that is also connected (perhaps indirectly) to another USB port of the same PC.

The current consumption of Axoloti Core fluctuates with processor activity, sdcard activity, and perhaps a midi controller with some blinking leds.
Imagine that the USB cable to Axoloti has a resistance of 0.2 Ohms, and the amplitude of current consumption fluctuation is 100mA, we'd find a 20mV signal between PC ground and Axoloti Core ground. That signal will likely be annoying in the audible spectrum.

Unless it is compensated for, and that is what the line output circuit does. It uses the TPA6132A2 to convert the audio DAC balanced (and biased) output to an unbalanced output that is not ground referenced but "soft"-ground referenced. And it does this DC-coupled as a bonus.

The 22Ohms resistance between line out ground and Axoloti ground prevents supply current (and fluctuations therein) to be carried over the audio ground wire, as there will be a different path with a lower resistance for supply ground current.

This application note from Texas Instruments probably explains it better.

#8

Hi,

I've a similar problem, so I add here my request without opening a new topic.

I also need to connect Axoloti to an amp board that has grounded input, but since I'm going to put both boards in the same enclosure, I would like to use the rear audio out pins that should be biased like the headphones output (I would leave the output jack for external connection to a mixer).

I'm trying to find a solution, I was thinking about an op-amp with unitary gain in the middle, that should not stress Axoloti hardware but I'm not sure.

Have you got a better solution?

Thanks in advance

#9

I believe that the RTSRTSTSR pins just at the back of the jacks are really connected directly to the Tip Ring Sleeve of the Jacks, it's equivalent. If you're talking about that, it's going to be safe to use the RTS pins corresponding to the main OUT 1/8 jack (but NOT the TRS pins corresponding to the headphones out). It's what I've done on my project, and it works well.

#10

Excellent!, I was so focused on the rear audio pins near the GPIO pins line that I forgot the RTSRTSTSR connections. They make the job

Thanks

Max

#11

Hi, I converted my Axoloti to a module for my 5U Modular Synth, so the inputs and outputs will be connected with other modules that share ground with Axoloti.
When I patched the axoloti I had a 3KHz pitched noise, very faint, but audible, specially in a PA.
To fix that, I connected a 20 ohm resistor in series between the ground of the outputs of the Axoloti and the sleeves of the jacks outputs, and another 20 ohm resistor between the ground of the inputs of Axoloti and the sleeves of the jacks inputs.

@johannes In the TI article you linked the resistor used to avoid ground loops is in series with the ground (as I connected) not in parallel (as in Axoloti). I don't know if both methods works, but using the output/input ground of the Axoloti (with the resistor in parallel) with shared grond didn't worked for me...