2
\$\begingroup\$

I am working on a noise simulation of a circuit with an RF amplifier, demodulator, and a baseband amplifier. I would like to accommodate the demoduluation in the noise simulation such that the noise near the carrier is shifted to baseband, where the noise models for the audio amplifier components sum with the RF noise to produce a realistic end-to-end result.

In other LTSpice simulations I've simulated the demodulation accurately in the time domain. What are suggestions for a way to perform this frequency shift operation within the noise simulation which is based in in the frequency domain?

\$\endgroup\$
1

1 Answer 1

1
\$\begingroup\$

In other LTSpice simulations I've simulated the demodulation accurately in the time domain. What are suggestions for a way to perform this frequency shift operation within the noise simulation which is based in in the frequency domain?

This is a common problem, the noise needs to be modeled but its dependent on other things in the circuit.

If preforming a .noise simulation, then frequency effects don't mean much because only the Vin vs Vout the gains and noise added from resistors is calculated. The gains are similar to an .ac analysis. So a .noise analysis is more like a .ac analysis but with noise sources added from components and amplifiers.

If the noise simulation you want to preform depends on time then it may be best to simulate noise in a transient simulation, this can be difficult with complex circuits as you need to add a noise source to the noise sources that matter (usually the largest ones, or noise sources before amplifiers, noise adds by the sum of the squares so very small noise sources do not contribute and need not be modeled)

\$\endgroup\$
1
  • \$\begingroup\$ Thanks. Thinking of .noise as .ac with resistor noise included is helpful. However, in the case of a bandpass RF amplifier, the noise at audio frequency is significantly lower compared to the noise at the carrier. That is why I want to shift the profile at the carrier to audio so it will experience the remaining audio noise contribution. Finding an equivalent noise resistance of the RF output to control a new voltage source for the audio section might be worthy solution. This would approximate the RF noise voltage density as being constant, which may work as an input to the audio stage. \$\endgroup\$
    – Pietro
    Dec 14, 2020 at 17:23

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.