An AM modulator is very linear and varies the amplitude of the carrier at the rate of the modulating frequencies. An AM demodulator is just a rectifier and the demodulation is also fairly linear.

An FM modulator is very linear and varies the frequency of the carrier at the rate of the modulating frequencies. An FM demodulator is just a linear f-to-v converter.

A non-linear distorted signal is different. It contains added harmonics. It doesn't have the sum and difference frequencies of a modulated signal.

__________________
Uncle $crooge

 

Linear: having or being a response or output that is directly proportional to the input.
I don't know how else to say it other than you are wrong. To simplify, you supply two frequencies to the input of a modulator, and get two completely different frequencies on the output. That is not linear. Mixing is the function of multiplying the inputs which is not a linear function. The rectifier you refer to does not have an output that is directly proportional to the input. maybe the attached pictures will help you understand.

Attached Images

	mixing (Small).JPG (54.7 KB, 10 views)
	summing (Small).JPG (53.0 KB, 6 views)

 

You showed an AM modulated carrier. The peak voltages of one polarity of it (the modulated carrier that is rectified) are a perfect copy of the modulation. You can even see it.

The multiplication and non-linearity occurs on the carrier and its sidebands, not to the baseband.

__________________
Uncle $crooge

 

The baseband signal will be mixed and, therefore harmonics of the baseband signal will be present in the output. I would not be so adamant if I did not have the information right here in front of me. I feel my resources are reputable and accurate. Mathematicaly, the equation that describes FM AND AM signals clearly show that the relationship between the carrier and the modulating signal is not linear. Look it up.
At any rate, I hope that any beginners that are interested in learning about communications techniques, will study up for themselves so as not to be given false information, perhaps even from myself. I'm sorry that we can not agree, I still value the knowledge and experience you present on this forum and hope that you continue to do so, as I still have much to learn.
Thank you for the stimulating (conversation??). I enjoy hashing out different concepts and I'm sure this will cause me to reread and restudy everything I've learned about amatuer radio thus far. Perhaps some radio guys will chime in here and clarify some things for me, as I do not want to go spreading false information.

 

The book seems very badly written?, to suggest that the two original frequencies disappear is nonesense!. The output of an RF mixer is the sum, the difference, and the two original signals. IF it's a double-balanced mixer then the originals will only have very low amplitude, and (theoretically) could be non-existant. However, the crude diode mixer ISN'T balanced at all.

As an example of AM, if you have a 1MHz carrier and modulate it with a 1KHz sinewave you will get three output signals.

1,000,000Hz - carrier
1,001,000Hz - sum
999,000Hz - difference

There will be no 1KHz output as it won't pass through the RF stages.

__________________
PIC programmer software, and PIC Tutorials at:
http://www.winpicprog.co.uk

 

Hi Sig,
I don't bother with lousy-sounding AM radio.
If "harmonics of the baseband signal will be present in the output" of my favourite FM station or my FM radios, then they wouldn't have 0.1% or less harmonic distortion, would they?

The harmonics occur at radio frequencies and are reduced with tuned circuits to avoid interference.

__________________
Uncle $crooge

 

OK, I think I see what is happening here. We are talking about two different things. I am saying that it is not linear because the function that is applied to the signal is not a linear function(multiplication) ie. For an FM modulator: An increase in the AMPLITUDE of the baseband causes an increase in the FREQUENCY of the carrier. Though this happens at a linear rate(which is what audioguru is refering to right?), it is not caused by a linear function(which is what I was refering to). To clarify what I believe you are talking about... a given percentage of increase in amplitude of the baseband signal gives a linear responce in frequency of the carrier.

This was my fault. In an attempt to Illistrate what I was talking about, I mistakenly posted an example of a mixer. Modulators are all mixers but that type of mixer was the wrong illustration for an AM or FM modulator. IMHO the book is well written and very informative. If you care it is the ARRL handbook for radio communications. As you correctly stated, in an AM signal, the original frequency is obviosly present!

To clarify what I was talking about, the act of mixing or modulating is not a linear function, but it obviosly occurs at a linear rate.