enter image description here

I need some help understanding this discriminator circuit for FM demodulation.

(L1, C1) is the primary LC tank which resonates with the carrier frequency, whose FM signal is applied at V12. As the FM signal oscillates according to the signal, the instantaneous voltage and current amplitudes of the LC tank vary across the Bode plot for the circuit.

The RC filter then picks out the AF signal from the varying amplitude due to the signal's skipping around the amplitude. In particular it does so according to the voltage relations between that across L3, and the two halves of the centre tapped secondary.

Warning: I don't understand transformers comfortably yet.

Questions: Most important:

  1. When the LC tank is above or below resonance, how do you find the phase of the current and voltage through and across L1?
  2. When the LC tank is above or below resonance, what is the phase of the voltage across L3?

Just to verify:

  1. Are the voltages and currents of L1 and L2 in phase / phase difference of 180 degrees?

  2. Do the capacitors C and C4 cause a double 90 degree phase shift (180 degrees) on the voltage across L3?

I care about 1) because I don't know how to currently explain to myself why the phase of voltage across L1 and the current through L1 can be anything else than 90 degrees. Why? Because the voltage across L1 is di/dt which is a 90 degree phase shift off of the current through L1.

I care about 2) because in combination with 1), knowing the phase of the voltage across L3 would allow me to understand the voltage relations for the RC filter circuit. 3) Would allow me to understand the same. 4) Is supplementary.

  • 1
    \$\begingroup\$ Bode plots don't deal with instantaneous voltages or currents. \$\endgroup\$
    – Andy aka
    Jan 11, 2021 at 16:27
  • \$\begingroup\$ @Andyaka aside from that? \$\endgroup\$ Jan 11, 2021 at 16:45
  • 1
    \$\begingroup\$ I stopped reading after that. \$\endgroup\$
    – Andy aka
    Jan 11, 2021 at 18:13
  • \$\begingroup\$ @Andyaka Gee thanks. \$\endgroup\$ Jan 11, 2021 at 18:53

1 Answer 1

  1. Above f, C conducts more; below L conducts more phase leans from 180 away from the one with less conduction only when R is in loop= DCR or ESR or R
  2. Reactive elements always have 90 across them but attenuated off resonance. Only R shifts the phase in a loop.
  3. Always when inside the BW of the transformer.
  4. C3 and C4 are Notch filters with shared L3 to suppress carrier after using slope of BPF to convert FM to AM and thus Diode carrier detect of resulting AM to decode FM
  • \$\begingroup\$ For 1): When (generally) is there a phase difference between the current through L1 and the voltage across L1? \$\endgroup\$ Jan 11, 2021 at 17:04
  • \$\begingroup\$ Always 90 deg lag in current, but the voltage depends on the impedance ratio of load R to sqrt(L/C) and they XL=XC are equal at resonance and V drops on either f side due to L or C or f shift. So a pure // LC with no load is very high Impedance, whereas the notch series CLC filter is very low impedance 0 in theory ( when tuned) - Research Z, Q for LC filters and learn to use http//www.Falstad.com/afilter with simple filter menu and add phase \$\endgroup\$ Jan 11, 2021 at 17:14
  • \$\begingroup\$ Off // resonance impedance drops with amplitude \$\endgroup\$ Jan 11, 2021 at 17:15
  • \$\begingroup\$ Thanks, this is some great help, and I will definitely use the applet in future. What still bothers me is that my source claims that the voltage across L3 is in phase with the current through the transformer at resonance. Then above or below resonance the voltage across L3 deviates from phase with the current through L1 and L2. My (hopefully mistaken) belief is that the voltage across L3 is in phase with the voltage across L1 and therefore at 90 deg w.r.t. the current through the transformer. Why is this claimed to be otherwise? \$\endgroup\$ Jan 11, 2021 at 17:31
  • \$\begingroup\$ I didn’t address that. You may be right due to the centre tap \$\endgroup\$ Jan 11, 2021 at 17:35

Your Answer

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

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