0
\$\begingroup\$

I am working on a project which involves 8 sets of LEDs being controlled by a micro-controller. The micro-controller is also communicating with an IC which is sampling an audio signal. The audio signal comes from either a 3.5mm jack or an onboard bluetooth audio board. In my first version all the components were connected to the same ground seeing as the micro-controller doesn't have an AGND for me to use. Here is a basic schematic of it:

Basic circuit diagram

This caused two problems:

  1. The Bluetooth audio board caused a ton of awful noise on the audio signal when a speaker was plugged in to the project. I learned this was caused by the bluetooth sharing a ground with the audio so I planned to use a B0505S DC-DC isolator to fix this. However, this didn't solve the second problem.

  2. Even when the Bluetooth audio wasn't installed i heard noise on the audio line which corresponded to what the LEDs were doing. I was thinking of using an DC-DC isolator again for the Micro-controller and audio IC, but then I realized that if the Micro-controller was isolated from the ground of the buffer and the LED's i don't think the Micro-controller would still be able to control the LEDs.

My question is how can I isolate the audio from the rest of the circuit while still letting the Micro-controller sample the music and also control the LEDs? Would the DC-DC converter still let the micro-controller communicate with the LEDs even though they are on different GND networks? IS there an LC circuit I should use instead to stop any noise on the audio line?

Improve this question
\$\endgroup\$
11
  • \$\begingroup\$ can you detail how the LED's are driven, and what the "buffer chip" is doing? \$\endgroup\$ – Pete W Jan 13 at 20:21
  • \$\begingroup\$ It looks like the audio circuit does not need the full 5V supply (??), so you could filter BOTH its VDD and GND. This might change how you send the signals to it, somewhat. Knowing some details about the noise spectrum would be helpful to do this intelligently. (and if you do that, it's possible you end up figuring out how to reduce the noise at the source instead, which would in principle be better) ... also why is CMN disconnected on the BT chip \$\endgroup\$ – Pete W Jan 13 at 20:24
  • \$\begingroup\$ The leds are ws2812s so they are driven by pwm, the buffer takes the signal from 3.3v to 5V \$\endgroup\$ – Spencer Brennessel Jan 13 at 21:47
  • 1
    \$\begingroup\$ The ws2812 are pwm internally. They will generate some noise. These need to be adequately bypassed. \$\endgroup\$ – Kartman Jan 13 at 22:09
  • 1
    \$\begingroup\$ Is there feedback from LED current return into analog gnd? ADC or is it aliasing noise? Are all wires twisted pairs? \$\endgroup\$ – Tony Stewart Sunnyskyguy EE75 Jan 13 at 22:15

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.