Envelope detector, how does it actually work?

neazoi · Nov 14, 2012

Hello,
I have found this https://michaelgellis.tripod.com/mixerscom.html
which says that the envelope detector mixes the carrier of an AM signal with the sidebands to produce audio, which makes sense.
From the other side, I thought that an envelope detector does not used as a mixer, but it justs rectifies RF and converts it into a variable voltage (AC audio waveform).
Which of the two are correct?

BradtheRad · Nov 14, 2012

neazoi said:
From the other side, I thought that an envelope detector does not used as a mixer, but it justs rectifies RF and converts it into a variable voltage (AC audio waveform).
Which of the two are correct?

This is how ordinary AM works. It can be picked up by the simplest AM detector. You can understand the audio that comes from it.

I have found this https://michaelgellis.tripod.com/mixerscom.html
which says that the envelope detector mixes the carrier of an AM signal with the sidebands to produce audio, which makes sense.

This is how single-sideband radio works. It is commonly used by ham radio operators. You need a special radio to pick it up.

The transmitter extracts the carrier, broadcasting only a modulated sideband. (I'm not sure how it can do this without a carrier.)

If you receive the signal on an ordinary AM radio, you will not be able to understand the speech.

Your radio needs to have a BFO (beat frequency oscillator), which adds the carrier to the incoming sideband signal. Then it can be treated the same way as an ordinary AM signal, and you can understand the speech.

The BFO also allows you to hear morse code broadcasts. Such transmissions consist only of a carrier which is simply turned on and off. Since it is unmodulated it does not produce audible sound through an ordinary AM receiver.

SSB transmission is efficient because a smaller percentage of power goes to the carrier, and a greater percentage of power to the signal which is carrying speech (information).

neazoi · Nov 14, 2012

BradtheRad said:
This is how ordinary AM works. It can be picked up by the simplest AM detector. You can understand the audio that comes from it.

This is how single-sideband radio works. It is commonly used by ham radio operators. You need a special radio to pick it up.

The transmitter extracts the carrier, broadcasting only a modulated sideband. (I'm not sure how it can do this without a carrier.)

If you receive the signal on an ordinary AM radio, you will not be able to understand the speech.

Your radio needs to have a BFO (beat frequency oscillator), which adds the carrier to the incoming sideband signal. Then it can be treated the same way as an ordinary AM signal, and you can understand the speech.

The BFO also allows you to hear morse code broadcasts. Such transmissions consist only of a carrier which is simply turned on and off. Since it is unmodulated it does not produce audible sound through an ordinary AM receiver.

SSB transmission is efficient because a smaller percentage of power goes to the carrier, and a greater percentage of power to the signal which is carrying speech (information).

Yes, I know all these stuff. But the website confused me by saying that the carrier is mixed with the sidebands in an envelope detector?

betwixt · Nov 14, 2012

The wording is ambiguous. There are things called synchronous detectors in which an oscillator is phase locked to the received signal's carrier then subtractively mixed with it. It's the opposite of the original modulation and recovers the original modulating signal. It works for AM only and is very good in situations where distortion from selective frequency fading is a problem. It also works on SSB and DSB although it does not work synchronously in these modes because there s no carrier to lock to.

Brian.

neazoi · Nov 14, 2012

betwixt said:
The wording is ambiguous. There are things called synchronous detectors in which an oscillator is phase locked to the received signal's carrier then subtractively mixed with it. It's the opposite of the original modulation and recovers the original modulating signal. It works for AM only and is very good in situations where distortion from selective frequency fading is a problem. It also works on SSB and DSB although it does not work synchronously in these modes because there s no carrier to lock to.

Brian.

What I mean is that in the first case, the diode rectifies the RF and charges the capacitor, which keeps the voltage and thus creates a copy of the audio waveform.
In the site explained case he claims that the carier is mixed with the sidebands, ie multiplied (assumming AM).

Which of the two happens? or both?

betwixt · Nov 14, 2012

It's in the wording, there isn't actually a mixing process because there is nothing to mix with. Your rectification explanation is correct, it is actually a rectifier and a simple peak holding circuit with a short time constant.
The AGC output is the same signal with a longer time constant so even the audio is filtered out and only the very low frequency of the signal level itself is seen.

Brian.

neazoi · Nov 14, 2012

betwixt said:
It's in the wording, there isn't actually a mixing process because there is nothing to mix with. Your rectification explanation is correct, it is actually a rectifier and a simple peak holding circuit with a short time constant.
The AGC output is the same signal with a longer time constant so even the audio is filtered out and only the very low frequency of the signal level itself is seen.

Brian.

Thank you Brian,
I understood that now.

Just for the purpose of wondering, Why a diode cannot mix the carrier with the sideband? There are two slightly different frequencies signals, that vary in frequency by the audio tone. The carrier could be thought as a LO whereas the sideband could be thought as the signal to be downconverted.

zorro · Nov 14, 2012

As the diode is a nonlinear device, it is right to say that there are mixing or intermodulation products of the different components.
This is not the most simple way to understand the process of envelope detection, but this is an important concept for undertstanding the behavior of the detector at very low SNR, i.e., below threshold.
Regards

Z

Syncopator · Nov 15, 2012

neazoi said:
I have found this ..... which says that the envelope detector mixes the carrier of an AM signal with the sidebands to produce audio, which makes sense.

From the other side, I thought that an envelope detector does not used as a mixer, but it justs rectifies RF and converts it into a variable voltage (AC audio waveform).

Which of the two are correct?

Both are correct. They are two ways of explaining the operation of the circuit - the simplified and the more rigorous. It is unfortunate that the author tries to mention both in the same explanation. It does nothing for clarity.

Read this, https://www.davidbridgen.com/mixers.htm, it should help.

betwixt said:
It's in the wording, there isn't actually a mixing process because there is nothing to mix with.

Of course there are things to mix, three of 'em. The carrier and both sidebands, which are going through a non-linear device.

Read my write-up again.

neazoi · Nov 15, 2012

Syncopator said:
Both are correct. They are two ways of explaining the operation of the circuit - the simplified and the more rigorous. It is unfortunate that the author tries to mention both in the same explanation. It does nothing for clarity.

Read this, https://www.davidbridgen.com/mixers.htm, it should help.

Of course there are things to mix, three of 'em. The carrier and both sidebands, which are going through a non-linear device.

Read my write-up again.

Ok this is a good explanation.

So basically one could think of the envelope detection as rectification, following the envelope of the AM sidebands, with the help of the capacitor.
But at the same time, intermodulation products occur in the diode mixer, because of the more-than-one signals (carrier and two sidebands) present at the input of the diode that mix together.

How does this sound?

The article says a sentence that confuses me: For a.m., ....demodulation (detection) and mixing are exactly the same process.
This does not agree with my conclusion above, they are different processes occurring simultaneously.
I think of envelope detection to be a different process than mixing.

Syncopator · Nov 16, 2012

neazoi said:
Ok this is a good explanation.

So basically one could think of the envelope detection as rectification ...

That's the normally encountered explanation.

... following the envelope of the AM sidebands ...

No, it follows the carrier's envelope.

But at the same time ...

... there is a more detailed, and accurate, description of what is happening.

...intermodulation products occur in the diode mixer, because ...

... it is non-linear.

The article says a sentence that confuses me: For a.m., ....demodulation (detection) and mixing are exactly the same process.

That's correct, they are.

Don't worry about being confused, just stay with it and think about it. There are sure to be other things which, during the learning process, will at first confuse.

This does not agree with my conclusion above, they are different processes occurring simultaneously.

They are the same process, described in different ways.

neazoi · Nov 16, 2012

Syncopator said:
That's the normally encountered explanation.

No, it follows the carrier's envelope.

I think that is not true. The carrier contains no information. What is modulated are the sidebands. Think about it on the amplitude/frequency domain.

I still haven't got a technical explanation that describes WHY envelope detection and mixing ARE the same thing...

FvM · Nov 16, 2012

I think that is not true. The carrier contains no information. What is modulated are the sidebands. Think about it on the amplitude/frequency domain.

I fear the problem is in your capability of imagining the duality of time and frequency domain.

The term "carriers envelope" is possibly a bit sloppy and doesn't refer to the distinction between different components of the modulated signal - carrier and sidebands, as they clearly appear in frequency domain. In my understanding, the term rather means the envelope of the whole signal, the modulated carrier, because the components don't appear isolated in time domain.

I still haven't got a technical explanation that describes WHY envelope detection and mixing ARE the same thing...

I have heard a clear explanation why both descriptions of the diode detector are correct. Saying envelope detection is "the same" as mixing misses the point.

There are different ways to design "mixers" with different behaviour. As said any non-linear characteristic will generate intermodulation products of various orders or in other words "mix" input signal components. In so far, an envelope detector is a mixer, but not the same as a mixer designed with different behaviour.

Syncopator · Nov 16, 2012

neazoi said:
I think that is not true. The carrier contains no information.

An unmodulated carrier contains "no information".

If, as you think, the carrier of an a.m. transmission (a transmission which is modulated) contains no information, then both the rectifier and the mixer explanations fall down.

What is modulated are the sidebands.

No. It is the carrier which is modulated. The sidebands are a result of modulation.

I still haven't got a technical explanation that describes WHY envelope detection and mixing ARE the same thing...

I am disappointed then. I thought that I had given an explanation.

Ok. I rest my case and give up.

One last observation though. An unmodulated carrier is often said to carry no information. Taking a wider view, it gives me the information that somewhere there is a transmitter with a viable power source. And it could serve as an, albeit unidentified, n.d.b.

neazoi · Nov 17, 2012

FvM said:
In my understanding, the term rather means the envelope of the whole signal, the modulated carrier, because the components don't appear isolated in time domain.

That is exactly the point I was referring to. The carrier that stays put and the two sidebands that go up and down in amplitude and in frequency. All these three components are seen as one on an oscilloscope, with the envelope varying in amplitude when the signal is modulated. The way it seems, it doesn't mean that the carrier contains information, only the sidebands contain the actual information.

- - - Updated - - -

Syncopator said:
An unmodulated carrier contains "no information".

If, as you think, the carrier of an a.m. transmission (a transmission which is modulated) contains no information, then both the rectifier and the mixer explanations fall down.
No. It is the carrier which is modulated. The sidebands are a result of modulation.

This is just what I described above, the AM signal is composed of the carrier and the two sidebands, which are seen as one signal in an oscilloscope, but the sidebands actually are the ones that cause the envelope to vary in amplitude.

Syncopator said:
I am disappointed then. I thought that I had given an explanation.

Don't take it personally, I always enjoy your help, it is just so simple thought things and it seems there is a lot of misunderstanding by many radio amateurs out there. Knowing exactly how things are made avoids errors and false knowledge. (my bad English again...)

Envelope detector, how does it actually work?

Advanced Member level 6

Super Moderator

Advanced Member level 6

Super Moderator

Advanced Member level 6

Super Moderator

Advanced Member level 6

Advanced Member level 4

Full Member level 6

Advanced Member level 6

Full Member level 6

Advanced Member level 6

Super Moderator

Full Member level 6

Advanced Member level 6

Similar threads

Privacy & Transparency

Privacy & Transparency