Help required for practical tank circuit

[gary36]

I have a voltage which is of the form Asinwt*cos(w1t-phi), where w= 50hz and w1= 1khz. I need to extract the carrier in a simple way using parallel tank circuit tuned at 1KHz.

Suggest me a circuit that works.

 

[danadakk]

Parallel Tank response :

You have to account for input and output Z the tank has to respond to. Al;so G needed in circuit and
its BW.


Regards, Dana.

 

[gary36]

input to tank circuit is opamp and output from tank circuit goes to zero crossing detector. I used the link https://www.redcrab-software.com/en/Calculator/Electrics/RCL-Parallel-Resonance-Circuit and got in values R=1k, C=202uf and L=125uH. All connected in parallel, but output is same as input.

 

[FvM]

Asinwt*cos(w1t-phi) is suppressed carrier AM, means that carrier phase is inverted every 10 ms.

Tank Q must low enough to allow output to follow input signal. With infinite Q you get zero output.

 

[gary36]

so what do you recommend?

 

[FvM]

Depends on what you want to achieve...

 

[D.A.(Tony)Stewart]

What phase error , amplitudes and distortion?
Given phase inversion, which phase do you want?

 

[gary36]

i want to recover carrier as it is

 

[D.A.(Tony)Stewart]

which carrier phase? both? How do you define the transition? in amplitudes and phase?

To remove all modulation requires just a limiter but phase inversion distorts zero signal = 100% modulation.

 

[gary36]

since carrier is suppressed, is there a way to recover 1 khz carrier cos(w1t- phi)? dropping the first term (btw, i do not wish to use costas loop)

 

[D.A.(Tony)Stewart]

that sounds like a contradiction.

 

[danadakk]

Costas loop .....attached.

Oops I see no Costas, this approach, also discussed in attached -

\

Is this phase synchronous, I did not read the entire doc ....? Does not look like it is....

Note your attempt at just a BPF implies, because of depth of carrier suppression, a
lot of G is needed to recover, hence would have to be low noise, and one would
have to consider jitter (FM in this case) resulting from noise in recovered signal path.


Regards, Dana.

 

[D.A.(Tony)Stewart]

A PLL can achieve more than an ILL, Injection Locked loop.

Here using a Phase Shift Osc as a BPF and limiting the AM-SC injected as ground ref. noise causes FM noise at subthreshold at low carrier levels

Top signals are AM input, Carrier out and Limiter out (B)
Bottom plots show more signals and different time scales.

Falstad Site if Offline, but I have the standalone version.

 

[FvM]

A possible solution is double carrier frequency by a full wave rectifier. This eliminates 180 degree phase jump on modulation polarity change. Then apply bandpass filter.

 

[D.A.(Tony)Stewart]

A possible solution is double carrier frequency by a full wave rectifier. This eliminates 180 degree phase jump on modulation polarity change. Then apply bandpass filter.

Yes I thought of that but then dividing by 2 to recover the carrier frequency has phase ambiguity.

 

[FvM]

but then dividing by 2 to recover the carrier frequency has phase ambiguity.

The carrier extracted from input signal has same ambiguity unless you combine it with modulation signal, as it's e.g. achieved in Costas loop.

 

[D.A.(Tony)Stewart]

Hence the contradiction in this question.