3
\$\begingroup\$

I'm building a simple audio mixer. It's going to be running from a USB power supply, so I'm using a TLE-2426 to derive a 2.5V reference voltage for the summing op-amp.

My question is, what do I connect to the sleeve pin of the TRS input jack? My first thought is to use the 2.5V reference as the ground, but then I thought that if the other component connects the cable's sleeve to its chassis ground, then suddenly there's a lot of current flowing through the cable. So, should I actually use the 2.5V reference as ground, or the actual 0V rail as ground? Should I include a decoupling capacitor, and if so, what size?

schematic

simulate this circuit – Schematic created using CircuitLab

Thanks to Transistor for pointing out the circuit schematic editor I completely missed, so I can illustrate what I'm trying to do. My original question was whether to connect Point #1 to Point A or Point B, but as I created the schematic, I realized that I don't know about Points #2 and #3, either. I think, based on Transistor's answer, that #2 should go to B and #3 should go to A, am I right?

\$\endgroup\$
  • \$\begingroup\$ Perhaps you can capacitively couple your inputs and outputs, keeping the ground a ground. \$\endgroup\$ – Richard the Spacecat Jul 19 at 15:17
5
\$\begingroup\$

What do I connect to the sleeve pin of the TRS input jack?

Always connect to chassis ground. Your thought process is correct. Any other connection between the two system grounds would short-circuit your 2.5 V reference to ground. The current would be limited but your system would cease to operate.

Should I include a decoupling capacitor, and if so, what size?

No decoupling capacitor is required on the ground lines. If you are asking about a decoupling capacitor on the signal (to block any DC voltage superimposed on the audio) then please edit your question and maybe add in your schematic as it stands.


There is a CircuitLab button on the editor toolbar. Double-click a component to edit its properties. 'R' = rotate, 'H' = horizontal flip. 'V' = vertical flip.

Note that when you use the CircuitLab button on the editor toolbar an editable schematic is saved in your post. That makes it easy for us to copy and edit in our answers. You don't need a CircuitLab account, no screengrabs, no image uploads, no background grid.


schematic

simulate this circuit – Schematic created using CircuitLab

Figure 1. Original and modified circuit.

The modified circuit should make everything clear. The main thing we've done is to ground the potentiometer bottom tag. It is generally safe to assume that there is no DC component from the source. We then move the DC blocking capacitor between the wiper and the op-amp input. This prevents upsetting of the DC setup for the circuit.

A couple of notes at this point:

Generally you want the load on the wiper of a pot to be at least ten times the value of the pot. With the components in Figure 1b there is a 10 kΩ load (R5) on a 10 kΩ pot. This will adversely affect the perceived linearity of the pot. To fix you can scale R5 and R6 up by a factor of 10. This will help us in your next question.

Should I include a decoupling capacitor, and if so, what size?

Because the inverting input is a "virtual ground" C4 and R5 act as a high-pass filter. You need to set the corner of the cut-off about 0.1 times the lowest frequency you want to hear. The cut-off frequency is given by \$ f_c = \frac {1}{2 \pi RC} \$ so increasing the value of R5 by a factor of ten means we can reduce the value of C4 by the same factor for a give frequency.

Further reading:

\$\endgroup\$
  • \$\begingroup\$ Thanks! I somehow completely missed the schematic editor button. I've added a schematic to my original question. I also realized as I was making it that I didn't know "which ground" some other points should go to, so I've added that to the question as well. \$\endgroup\$ – HiddenWindshield Jul 19 at 19:15
  • \$\begingroup\$ See the update. \$\endgroup\$ – Transistor Jul 19 at 19:35
  • \$\begingroup\$ Thanks again! I should point out that "R4" is actually a dedicated volume control chip (LM1972), which actually has an impedance of 40k. (I went back and added that to the schematic once I noticed the mistake, but I guess you'd already started typing your answer by then.) Which means that R5 should be a minimum of 400k, giving a cutoff frequency of about 40Hz, correct? \$\endgroup\$ – HiddenWindshield Jul 19 at 20:04
  • \$\begingroup\$ In that case I would consider using a non-inverting configuration. Going too high with the resistor values may cause some increase in noise susceptibility. By the way, I presume your choice of TL081 was the default in CircuitLab. It wouldn't be a good choice in practice. \$\endgroup\$ – Transistor Jul 19 at 20:07
  • \$\begingroup\$ You're correct, the op-amp I'm actually planning to use is an OPA1678. Thanks again for the advice! \$\endgroup\$ – HiddenWindshield Jul 19 at 20:23

Your Answer

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.