• State Resolved
  • Locked Locked
  • Replies 1 reply
  • Subscribers 45 subscribers
  • Views 1948 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

LM358: Isolating two op amp amplifiers from each other

jing tan
Prodigy 60 points
Part Number: LM358

My circuit consists of two oscillators: one VCO in audio range and an LFO (low frequency oscillator - below the audio range). The signals from the two oscillators are amplified with two non-inverting op amp amplifiers. The two amps (LM358) share the same split voltage supply/virtual ground source. Because of that the LFO effects the frequency of the VCO. The two oscillators should be entirely independent. How can I avoid the two oscillators effecting each other?

  • 0
    TI__Mastermind 22825 points

    Hello,

    The LM358 crosstalk rejection in the frequency range you are considering is very high (typically 120dB). So, if one OPA output section is swinging 1Vpp, this could cause the offset voltage of the other OPA section to change by 1uVpp (or 3.7uVpp of output disturbance at the noise gain that you are operating). I don't think this amount of crosstalk is discernible unless you are seeing the disturbance on a spectrum analyzer or other tuned instrumentation?

    I'm curious why your split supply (R13, R14 junction) is not adequately decoupled / filtered using shunt cap of appropriate value? Even then, due to the finite impedance of this node, I can see one section creating crosstalk in the other. Have you tried separating / bypassing the split supplies for the two sections to see the effect? Better yet, it is best to create this supply using an LDO or other "stiff" source with adequate decoupling.

    Also, have you provided physical separation between the traces / components of the two oscillator sections? Using ground plane and isolation traces between sensitive nodes is also another method.

    You've not shown any power supply filtering or bypassing. Some of the crosstalk could be getting in through power (or ground) connections. A ground plane will reduce ground impedance / inductance and helps. LC or RC low pass filtering is another method used to decouple signals / power.

    Let me know whether you have any comments  or clarifications.

    Regards,

    Hooman