Is an envelope detector an RF mixer?

[neazoi]

Hello, this is a thing that is not clear to me. Is an envelope detector (https://en.wikipedia.org/wiki/Envelope_detector) actually an RF mixer?

I do not see any mixing in the process, it operates more by allowing the positive edges of the RF to pass through to a capacitor. Then you detect the capacitor charge (voltage level) variations as a sound wave.

but some others say that the envelope detector diode will mix the AM carrier with the sidebands to produce audio, so it will infact act as a mixer.

However wikipedia states that the carrier is filtered by the capacitor after the diode, but the diode still sees and mixes the RF signals components.

I am really confused, can anyone explain this in detail? What is actually happening there?

 

[D.A.(Tony)Stewart]

a balanced diode bridge can be used to modulate or demodulate AC.

Your understanding is correct for demodulation.

 

[neazoi]

a balanced diode bridge can be used to modulate or demodulate AC.

Your understanding is correct for demodulation.

So it is indeed NOT an RF mixer by any means?

 

[D.A.(Tony)Stewart]

By definition, a modulator is an RF mixer, which produces sum and difference frequencies.

 

[neazoi]

By definition, a modulator is an RF mixer, which produces sum and difference frequencies.

I am confused because this type of demodulator does not use a local oscillator to downconvert the RF into audio.
How can this be considered a mixer though?
The only reason for that would be to accept that it mixes the AM carrier with the sidebands in an AM modulated signal.
But the wikipedia article does not refer to that at all?

 

[D.A.(Tony)Stewart]

A full wave bridge is a frequency doubler and the difference frequency for CW is DC, thus you see the sum 2x and difference DC.
Adding a cap suppresses the 2x f and modulation appears as baseband above DC.

 

[jiripolivka]

A full wave bridge is a frequency doubler and the difference frequency for CW is DC, thus you see the sum 2x and difference DC.
Adding a cap suppresses the 2x f and modulation appears as baseband above DC.

In principle any non-linear element will generate linear combinations of two or more signal frequencies when applied to it.
So with one input signal at a certain frequency the non-linear element (a diode for instance) will function as a peak detector with a lower-frequency corresponding to input signal modulation.
Also, input signal harmonics will appear in the current flowing through the diode.

Different signals at different frequencies can be separated by filters. To improve diode operation, there are various methods how to reject or support odd or even frequencies by using parallel, antiparallel or bridge connections of two or more diodes.

For more details read some books on detectors, mixers and modulators. See www.minicircuits.com for various commercial mixer, detector and modulator versions and their parameters.

 

[D.A.(Tony)Stewart]

I just learned that "Rules of Thumb" came from archaic religous beliefs.

But here are some sage rules of thumb for RF Engineers.

https://www.microwaves101.com/encyclopedias/microwave-rules-of-thumb

 

[jiripolivka]

I just learned that "Rules of Thumb" came from archaic religous beliefs.

But here are some sage rules of thumb for RF Engineers.

https://www.microwaves101.com/encyclopedias/microwave-rules-of-thumb

Thanks Sunnyskyguy for mentioning Rules of Thumb. I can see they came from personal experience.

I could write mine but not sure someone is interested. I think edaboard is nice for anyone.

 

[neazoi]

In principle any non-linear element will generate linear combinations of two or more signal frequencies when applied to it.
So with one input signal at a certain frequency the non-linear element (a diode for instance) will function as a peak detector with a lower-frequency corresponding to input signal modulation.
Also, input signal harmonics will appear in the current flowing through the diode.

Different signals at different frequencies can be separated by filters. To improve diode operation, there are various methods how to reject or support odd or even frequencies by using parallel, antiparallel or bridge connections of two or more diodes.

For more details read some books on detectors, mixers and modulators. See www.minicircuits.com for various commercial mixer, detector and modulator versions and their parameters.

jiripolivka, I see your point,
So assuming that the input signal is a single carrier with varying level (not the real case), the envelope detector should just follow the envelope.
But because AM is the Carrier plus the sidebands, then the diode in the envelope detector ALSO produces signals at it's output because of the mixing of the AM carrier with the sidebands, due to the non-linearity of the diode.
These signals are filtered (sort of) with the shunt capacitor of the envelope detector. This capacitor is used both to smooth the rectified signal as well to filter out RF harmonics.

Have I understood the whole thing correctly?

 

[jiripolivka]

jiripolivka, I see your point,
So assuming that the input signal is a single carrier with varying level (not the real case), the envelope detector should just follow the envelope.
But because AM is the Carrier plus the sidebands, then the diode in the envelope detector ALSO produces signals at it's output because of the mixing of the AM carrier with the sidebands, due to the non-linearity of the diode.
These signals are filtered (sort of) with the shunt capacitor of the envelope detector. This capacitor is used both to smooth the rectified signal as well to filter out RF harmonics.

Have I understood the whole thing correctly?

I do not think the AM carrier is mixing with the side bands . With no modulation, a peak detector only generates a DC current. With AM modulation the low-frequency detected voltage follows the modulation pattern.
With SSB, there is no AM carrier and detector output does not offer any suitable audio signal. We must inject an artificial "carrier" and adjust its frequency to get an intelligible output.
You are right in that the capacitor at audio output from detector is a part of the low-pass filter which limits detector output to only audio response.
A detector or mixer is a more complex issue than it seems. Follow my advice and study it deeper to better understand the device.

 

[neazoi]

I do not think the AM carrier is mixing with the side bands . With no modulation, a peak detector only generates a DC current. With AM modulation the low-frequency detected voltage follows the modulation pattern.
With SSB, there is no AM carrier and detector output does not offer any suitable audio signal. We must inject an artificial "carrier" and adjust its frequency to get an intelligible output.
You are right in that the capacitor at audio output from detector is a part of the low-pass filter which limits detector output to only audio response.
A detector or mixer is a more complex issue than it seems. Follow my advice and study it deeper to better understand the device.

Thank you very much.
So when you say
"Also, input signal harmonics will appear in the current flowing through the diode." you refer to the harmonics of the RF frequency (2nd, 3rd etc.), not the sidebands of the AM signal?

If that is so, then in a clean, harmonic-free AM signal, there is no mixing occur in the envelope detector diode itself, but just rectification, as described in the wikipedia article, right?

 

[jiripolivka]

Thank you very much.
So when you say
"Also, input signal harmonics will appear in the current flowing through the diode." you refer to the harmonics of the RF frequency (2nd, 3rd etc.), not the sidebands of the AM signal?

If that is so, then in a clean, harmonic-free AM signal, there is no mixing occur in the envelope detector diode itself, but just rectification, as described in the wikipedia article, right?

Again no, you do not understand properly the non-linear device behavior.

A diode fed with a single frequency signal will recitify it and this will generate a DC current. At the same time it will generate harmonic multiples of that frequency.
If you apply more signals like sidebands, all will generate DC , all will generate their harmonics, AND the diode will generate harmonic combinations of each pair of frequencies, either applied or already generated.
All this requires filtering that mess to operate a diode (or more diodes) as detectors, mixers, modulators and the like.
More fun you can see if you add noise.

Go and study if you want to really understand a simple diode function.

 

[neazoi]

Again no, you do not understand properly the non-linear device behavior.

A diode fed with a single frequency signal will recitify it and this will generate a DC current. At the same time it will generate harmonic multiples of that frequency.
If you apply more signals like sidebands, all will generate DC , all will generate their harmonics, AND the diode will generate harmonic combinations of each pair of frequencies, either applied or already generated.
All this requires filtering that mess to operate a diode (or more diodes) as detectors, mixers, modulators and the like.
More fun you can see if you add noise.

Go and study if you want to really understand a simple diode function.

Thanks a lot, this is the answer I was looking for.
If a clean single frequency input signal passes through the diode it is rectified, but at the same time the diode generates harmonics, due to it's non-linear behaviour.
It is very clear to me now what is happening, thanks a lot.

 

[D.A.(Tony)Stewart]

Use mouse to draw signal or choose from menu and see spectrum or visa versa

https://www.falstad.com/fourier/

DC is the first bar.
then fundamental, then all harmonics up to N