0
\$\begingroup\$

I am planning to build a PCB for a Eurorack Modular Synthesizer, incorporating an STM32 for DSP and a bidirectional Audio Codec like the TLV320AIC3104 to handle the ADC/DAC.

Eurorack audio signals typically range from -5V to +5V, so I need to attenuate this range and apply a DC offset to make it suitable for the ADC. The Output stage should take the DC offset DAC signal, and amplifies, as well as locut filters the signal, so it is in suitable Eurorack range.

As I'm relatively new to electronics design, I would appreciate your patience.

For the input and output stages, I plan to use summing opamps (Maybe NE5532 ?) in an inverting feedback configuration. I have been simulating these stages in ngspice within KiCAD , and the results seem satisfactory. Could you review my design to see if there are any issues?

Some notes regarding my design:

  • ngspice in KICAD does not support potentiometers, so i put in a manual resistor voltage divider.
  • This schematic does not include the coupling/decoupling required by the Audio Codec. I want to add Hardware specific circuitry, when the general idea is right.
  • I used generic components for the simulation, since i am pretty new to all of this, would this make a huge difference?
  • i am using a signal of 20Hz, since it is the lower border of the hearing spectrum, to verify the output filter is working correctly
  • the pot in the input stage is at a 0.2 ratio at the simulation screenshot, so there is still headroom to boost the signal

I added my simulation schematics, as well as screenshots of the input and output stage simulation to the post. schematic

Input Stage simulation Red: input from Eurorack; Green: input of Codec ADC input stage simulation

Output Stage simulation Blue: output from DAC; Purple: output to Eurorack output stage simulation

asked Jul 18, 2024 at 9:28
\$\endgroup\$

1 Answer 1

1
\$\begingroup\$

Input channel(s):

The negative DC offset (R2/7/4) is dispensable. Use instead a DC blocking capacitor at the output of the op amp. You are attenuating twice by 5:1. The input resistive divider (R8/2) can be removed. The op amp already has a gain of 0.2 which translates 10Vpp to 2Vpp (single ended) for the TLV codec. You could also reduce the resistors R1/2 to 1/10 of the original value (+ increasing C2 by a factor of 10) to get somewhat less noise.

Output channel(s):

Consider simple differential amps for the output channels to take advantage of the 6dB higher SNR and dynamic range. It takes only 2 more resistors and 1 capacitor per channel. Don't overcomplicate things with DC coupling, use AC coupling like for the input channels. To make 10Vpp from the 4Vpp differential output voltage, the ratio of the resistor pairs in the differential amplifiers needs to be 2.5.

----- EDIT -----

Example for single ended input stage and differential output stage with 1st order high-/lowpass response.

enter image description here

enter image description here

answered Jul 18, 2024 at 19:13
\$\endgroup\$
5
  • \$\begingroup\$ Thanks for the reply! Doesn't the TLV require a purely positive input signal for the ADC? This would not be given when dispensing the DC offset, right? Am i missing something, could you specify why leaving out the DC offset would work? \$\endgroup\$ Commented Jul 19, 2024 at 10:18
  • \$\begingroup\$ The analog input stages of the TLV codec are already biased internally with a suitable common mode voltage, you don't have to take care of that. My suggestion was to keep it simple by AC coupling your input (and output) amplifiers to the codec. \$\endgroup\$ Commented Jul 19, 2024 at 19:49
  • \$\begingroup\$ I have found the section in the datasheet. Thanks a lot for pointing me into the right direction! I will try to simplify according to the datasheet. \$\endgroup\$ Commented Jul 20, 2024 at 9:46
  • \$\begingroup\$ A question regarding your statement about using differential outputs: You mean i create a differential output out of the single ended one from the TLV? This would mean my output hardware would have to support differential inputs (which it doesn't)? Thanks for clearing things up for me! \$\endgroup\$ Commented Jul 21, 2024 at 12:04
  • \$\begingroup\$ The TLV already has differential outputs. You can use them as such and gain a few dB in SNR and dynamic range or just use them single ended. I'll add a schematic for the input/output stages to make that more clear. \$\endgroup\$ Commented Jul 21, 2024 at 18:52

Your Answer

Draft saved
Draft discarded

Sign up or log in

Sign up using Google
Sign up using Email and Password

Post as a guest

Required, but never shown

Post as a guest

Required, but never shown

By clicking "Post Your Answer", you agree to our terms of service and acknowledge you have read our privacy policy.

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.