# Modulation Theory.

Discussion in 'General Science' started by BR-549, Mar 3, 2016.

1. ### BR-549 Thread Starter Well-Known Member

Sep 22, 2013
2,178
419
Skinny modulation(no side-bands). I have a new idea for a modulation method, but I might be crazy, so I thought I would try here first.

If this is possible, it might be handy for weak and/or noisy environments.

My idea is to physically or electronically move the origin of the electric field reference or the magnetic dipole reference, during one cycle of the carrier frequency.

This means that the modulating(base-band) frequency must be at least equal to or multiples of, the carrier frequency.

Think of moving an antenna during one carrier cycle. If we move the antenna in any one direction, during the first pie, then move it back to original position on the second pie.......what would the received signal look like?

You would see a bow(an out of round) in the first sine pie and a counter bow in the second pie. The amplitude, the frequency and the phase do not change.

For every one complete displacement of the origin, we get two bows. If we did four complete displacements per cycle, we would get four bows and four counter bows intertwined.

The modulating F would be 4 MHZ for a 1 MHZ carrier.

For detection, we compare a un-bowed sine.

Of course this has problems trying to implement. I haven't been able to figure out a way to electronically do this. I kinda gave up on the electric side, but have been thinking of a sort of magnetic modulation.

Normally the magnetic dipole is perpendicular to a current loop. What would happen if during the fist pie, we diverted that field by 45 degrees(with an external, same polarity, but varying field), and let it revert back to perpendicular during the second pie? We would be changing the electric/magnetic relationship, within the cycle. I would think this might cause a bow or wobble also.

Ok then, what do you think, any suggestions? Or should I try medication?

The goal is to cause a wobble modulation, without sidebands.

What can we vary on a constant carrier, without causing sidebands?

I believe that magnetic harmonics can be placed on a fundamental electric, without the electric harmonics.

This would cause a widening in the spectral line, but no sidebands.

What if we made the feedpoint stationary and on a fulcrum. And during the first pie, we rotate the end of the dipole in a circle and the other dipole end follows symmetrically.

What if we rotate at 10 times in a pie?

Last edited: Mar 3, 2016

Aug 27, 2009
3,011
2,357
3. ### BR-549 Thread Starter Well-Known Member

Sep 22, 2013
2,178
419
Ok, I looked at that article, but I fail to see any similarity. There is no mixing going on here. Mixing or beating would cause products(sidebands). Requiring bandwidth. I am trying to acquire a one period bandwidth.

Let me try again, only this time instead of using a sine graph at the receiver, let's use a polar(circular plot) graph.

An unmodulated signal at the receiver would look like a circle(symmetrical). The radius of that circle, represents the amplitude of the carrier. The period of rotation around the circle, represents the frequency of the carrier. The clockwise or counter clockwise direction of rotation represents the phase of the carrier. The circumference of the circle represents the wavelength.

The amplitude, frequency and phase of transmitter NEVER changes. And NO mixing or beating.

Now, for the hard part. The circumference of the polar display, represents and is proportional to the wavelength of the transmitted carrier(length of period). Let's say that the wavelength is 10 inches. And let's say that we have a calibrated display and the circumference of the plot is 10 ins. So that 5% of wavelength is easy to see. It would be about a half an in. on the arc of the circumference.

For simplicity of explanation, let's sync the receiver display to the Transmitting antenna signal.
The cycle starts at the 12 o'clock position and rotates clockwise.

During the first pie of rotation, we will physically lower the antenna structure by .5 in. at a constant rate. This will cause the pie that is swept at the receiver, to become a little squashed and a little longer.

During the second transmitting pie, we will return the antenna to the original position in a like manner. This causes a mirror of the first pie in the receiver display. We now have slightly squished and slightly oblong display. Sorta football shape. Vertical.

Even though we have a distorted display, the amplitude remained constant, because the origin moved with the distortion, keeping amplitude constant. The period and phase also remained constant. A circle could represent a "0" and a football could represent a "1".

I would think that we could do multiple movements within the period, as long as we return to the original antenna location(before end of period). Which would be the original origin on display.

Of course until I learn how to move antennae very quickly and very precisely, it doesn't do any body any good.

But all the receiver sees, is a signal that has it's origin vary during the period.

So my question is, can anyone think of a way, to electronically vary the origin(of resonance) during the period?

4. ### Wendy Moderator

Mar 24, 2008
20,772
2,540
The rotation you refer to is actually a form of FM modulation, and is also known as phase modulation. Any modification of a pure sine wave will generate side bands. It is pretty core theory.

5. ### KL7AJ AAC Fanatic!

Nov 4, 2008
2,046
293
Yes...even phase shift modulation implemented by physically moving the antenna creates sidebands!

6. ### BR-549 Thread Starter Well-Known Member

Sep 22, 2013
2,178
419
Wendy, Thank you. My premise is that if the modulated signal is completed before end of carrier period, and the carrier is returned to the same state as the start of the period, there will be no sidebands generated. How can I test that?

KL7AJ, Thanks and I agree with you, if the antenna movement duration is longer the the carrier period.
Think of it as inducing a Doppler shift during the first pie, and then, inverting that Doppler shift, during the second pie.

"Any modification of a pure sine wave will generate side bands. It is pretty core theory. "

I want to disprove that theory. All of our modulation periods are much longer than the carrier periods. This produces either amplitude or period sets(sidebands).

I want to use modulating periods that are multiples of the carrier period. This insures that the carrier is back to it's original state at the end of each period.

My problem is that we use alternating current to represent the two pies. This necessitates using zero voltage and zero current as reference for oscillation. I can't see anyway around it.

Perhaps there's another way to consider. All the products of conventional modulation are sines.
This means the electric and magnetic are equal and at 90 degrees.

What if we hold everything constant(amplitude,phase,frequency)...vary that magnetic angle a few degrees during the first pie, and then reset it during the second pie. This angle variance is not carried over multiple carrier cycles. It is completed(or not completed) within each cycle.

This shouldn't create any sidebands. It can be detected with a 90 degree sine at receiver.

Or does everyone still believe that this would create sidebands?

Does anyone know of a modulation scheme where the modulating frequency is harmonic multiples of the carrier?

7. ### nsaspook AAC Fanatic!

Aug 27, 2009
3,011
2,357
Lets take a simple look.

I'm not sure exactly what you expect to happen. Varying the antenna magnetic angle (not exactly sure what that means physically in a practical manner) would be roughly equivalent to an rf envelope from an another signal source (the antenna movement adds it's energy to the antenna surrounding exciting fields in a linear manner, no new frequencies are generated so it's not modulation or classic non-linear mixing) The harmonic content of the rf energy from the antenna movement (yours would be more rectangular and pulse like) added to the 'carrier' would simply be a superposition of fields with a envelope that could be 'demodulated' with a fancy envelope or synchronous detector.

8. ### WBahn Moderator

Mar 31, 2012
18,085
4,917
You can want to disprove whatever you like -- that doesn't change reality.

If the signal at the receiver has no side-bands, then the signal at the receiver is a pure sine wave. Period. If the receiver is able to detect the kind of change you are talking about making to the signal, it is because there is information present in a sideband to be detected. This is true even if you do nothing but change the amplitude of one cycle (including by reducing it to zero).

thumb2 likes this.
9. ### BR-549 Thread Starter Well-Known Member

Sep 22, 2013
2,178
419
When we modulate, the modulation duration is longer than the first pie of the carrier, therefore, the second pie is not able to mirror the first pie. This causes multiple periods of the carrier sines to change with the modulation asymmetry. Generating sidebands.

My premise is that if our modulation duration can be completed within the first pie, and mirrored in the second pie, no sideband generation will occur.

Changing the physical origin of the carrier during the period(or imitating such electronically) is only one way a sine might be altered without sidebands. (I still think there ought to be a way to do this)

As I previously stated, in a sine, and in all the mod techniques we use, the magnetic and electric fields are always 90 degrees to each other. Don't confuse current angle with magnetic angle.

The current angle is a result of the electric force and the magnetic force on current.

The magnetic field angle is always at 90 degrees to the electric field. These perpendicular interwoven fields have a spring character. This 90 degree relationship is responsible for sine symmetry.

[PREMISE: energy may be stored and retrieved from this 90 degree spring without loss.]

What if by some means, during the first quadrant we change the magnetic angle from 90 degrees to 85 degrees and during the second quadrant we let it relax back to 90. We mirror this effect during the second pie.

There would be no net change to a pie, let alone the period. We are only changing the angle, not adding power, and what power we do add to change the angle, is returned before end of pie.

There is no power available for sideband generation.

The minimum modulation frequency is equal to the carrier frequency, 1 deformity per pie.
The next would be 2 deforms per pie(2X the carrier), then 3 deforms per pie and so on. The modulating period is a sine also, not a pulse.

We can detect the deforms at a receiver by comparing to an un-deformed sine or pie.

The net power level of carrier is constant.

And now a word to our viewers. If this thread or my posts seem confusing or contradictory to you, it is only because of your point of view. And I don't mean that you are a democrat or republican, or a modern scientist or a classical one. Or a mad one.

What I mean is where you are located physically in relation to the phenomenon.

When we measure the amplitude of a wave, 90 degrees to the waves direction, it appears to change continuously and periodically. A sine graph function. But if we change our location to the front or back of the wave, 0 or 180 degrees to wave direction, we measure a polar function. At this location, we see that the amplitude is really constant, and just changing angle relative to the origin, as it rotates.

If we move our point of view(location) to 45 or 135 degrees, where we can see both actions work together(sine and polar), we can finally see the whole picture. Which is a helical structure.

To further complicate matters, when we imitate the structure electronically, a current vector, can only represent one pie.(one half of rotation) We have to reverse current to imitate the second pie.

But the actual wave helix does not alternate, it rotates.

Most are taught the sine and polar descriptions, so I used both points of view for my premise, along with the physical antenna movement.

Hopefully this clears up some of my comments.

Now if you are of the mind that any distortion to the sine, not matter how nimble or symmetric, will cause sidebands, then this wouldn't interest you. Unless you have another need to offset a magnetic field.

As for the 90 degree problem, what if we put a non uniform permeability core in the resonating coil. The permeability would be such that at one end of the coil, the high u would be at 12 o'clock and at the other end, the high u section would be at 6 o'clock. Could an angled permeability gradient, possibly change the angle of the dipole(center line of coil)?

We know that we can bend and guide a dipole, because we can bend and guide a coil. But when we do, the electric bends with it, and stays at 90 degrees to the center line.

If we had an air core coil, then inserted a high u, but skinny rod, at an angle, would that change the 90 degree angle on an alternation of current?

Would a flat spiral coil, with an angled high u core, change the angle at some time on an alternation?

Any ideas or suggestions?

Now, it is quite possible that even if we could produce this effect in a resonant tank, that "current" could not propagate this effect. Because current(charge) itself has this 90 degree property. On the other hand, at resonance, the current is tightly unified, this effect might carry thru quite well. Sort of a quantum wave angle effect. Hey, you never know.

I haven't heard of anyone trying this. And I'm not sure what anyone would call this.

10. ### WBahn Moderator

Mar 31, 2012
18,085
4,917
You can have whatever premise you want. You can assert the premise that this "pie" you keep talking about is cherry in one case and apple in the other. None of that matters a hoot unless you can prove it -- prove it mathematically or proof it experimentally. Short of that, you might as well assert that modulation is the result of remote viewing.

thumb2 and nsaspook like this.
11. ### profbuxton Member

Feb 21, 2014
233
68
BR-549, from your description it sounds like you are suggesting some sort of movement of the transmitting aerial to modulate a signal?
I believe it is already in use. When there is a local storm it shakes the local tv transmiter aerial and I end up with a "no signal" message or the display gets "pixilated". I just can not move my receiving aerial in synch or fast enough to regain the signal.
Apart from that how do you propose to move the receiving aerial in synch with the transmitter so demodulation can be achieved?

12. ### ErnieM AAC Fanatic!

Apr 24, 2011
7,436
1,626
Since a wave without side bands carries no information suggesting a scheme to transmit information without side bands belongs in the same crazy box as anyone who believes in over unity power generators.

They all may be discarded out of hand, only general politeness prevents further derision.

13. ### BR-549 Thread Starter Well-Known Member

Sep 22, 2013
2,178
419
Many CW operators would disagree with that. I do too.

A wave without sidebands has a lot of information.

It's presence is information.

It's direction is information.

It's frequency is information.

It's polarization is information.

14. ### WBahn Moderator

Mar 31, 2012
18,085
4,917
You are splitting semantic hairs.

Even if we accept that the presence of an unmodulated signal conveys information, that information is static and unchanging. It is exactly the same information now that it was a year ago and that it will be a year from now. As soon as you change ANYTHING about that signal that is detectable ANYWHERE else, you have created sidebands.

ErnieM likes this.
15. ### BR-549 Thread Starter Well-Known Member

Sep 22, 2013
2,178
419
profbuxton, No, I was trying to use the moving antenna, to show what moving the origin of resonance, during the period of carrier, would do.

The Effect of moving the origin is like moving the antenna.

I see now that it's doubtful anybody was able to follow anything I said.

16. ### WBahn Moderator

Mar 31, 2012
18,085
4,917
That's because what you claiming is nonsense.

You are admitting that what the receiver receives is NOT a pure sinewave -- that it is somehow "bowed". If it is ANYTHING other than a pure sinewave, it HAS sidebands!

All you have to do is take the mathematical description for you "bowed" sinewave and take the Fourier transform of it. Guess what? Sidebands!

17. ### profbuxton Member

Feb 21, 2014
233
68
Yes, you're right, I don't quite understand(sigh). I'm not sure what the difference is between moving(physically) the antenna or the origin of resonance although I would love to see that done at any useful frequency. Even if were possible how would the signal be received? wouldn't the receiver antenna have to move in unison to get a useful signal?
Many years ago I read a proposal to invent a radio transmission system which was meant to be totally secure. This involved switching the polarity of the antenna in some precoded fashion and also the receiver antenna so that transmission and reception were always in phase but no one could listen in. Don't know if it was feasable, sounded clever but probably play merry hell with the RF systems.

18. ### ErnieM AAC Fanatic!

Apr 24, 2011
7,436
1,626
Quite the opposite. I see what you did there, though you entirely miss the point that as soon as you modulate a signal, as soon as you add information to it, you get side bands. Have to, 100%, it is simple physics.

Even keying the carrier on and off for a simple morse code transmitter adds side bands.

19. ### BR-549 Thread Starter Well-Known Member

Sep 22, 2013
2,178
419
In this modulating system, you can not use voice, music or video for the modulating signal.

The only modulating signals allowed are integer multiple tones of the carrier.

20. ### WBahn Moderator

Mar 31, 2012
18,085
4,917
And modulating it with integer multiple tones of the carrier will add sidebands.

Do .... the .... math !!!!