# Audio gen. RF harmonics

Discussion in 'Electronic Design' started by Genome, Aug 31, 2007.

1. ### GenomeGuest

Stop being a twat and get yourself of down the local toilet and snork
off blokes.

DNA

2. ### Tim WilliamsGuest

So, you want to add the frequency 1.6MHz, or a sideband at 1.6 (4.6 + 1.6 =
6.2MHz)?

The first is a linear mixer (resistor, say), the second is a linear
multiplier.

I have no idea what you mean by audio frequency. The farthest first order
sideband that can be produced is +/-20kHz. (Higher order IMD products, from
a nonlinear multiplier, will give spurs at multiples and so on.)

Tim

3. ### LVMarcGuest

The process that matches your description is Frequency modulation of the
RF caRRIER. Say you have an rf carrier of 4.6 Mc, and you take a 20 Kc
tone and frequency modulate the carrier. And, further, you arrange the
amplitude and sensitivity such that the 20 KC tone produce a deviation
of 800 kc. the resultant spectra would look like:

the carrier,

surrounded on both side by 20 kc "harmonics" on both sides
further there will be "harmonics" at 20 40 60 80 etc , spacing, on each
side of the carrier
further, they will extend + or 800 Kc on each side, 1.6 mC all together.

The amthemactical function that describes the placemt and amplitude
distribution of this "harmonics" often called sidebands, are the
bessel's function.

Happy modulating

marc

http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&rd=1&item=300142128658&ssPageName=STRK:MESE:IT&ih=020

www.fwt.niat.net amazing antennas

BS benevolent Stuff for engineers

Furthermore, if you modulate the frequency and the amplitude at the same
time, you can get very peculiar spectrum. However the 20kHz modulation
is not going to produce a pure 1.6MHz tone in addition to 4.6MHz carrier.

DSP and Mixed Signal Design Consultant
http://www.abvolt.com

5. ### Mark JenningsGuest

I am hoping some mathematical genius can help me with this. I read of
a process whereby _specific_ harmonic frequencies can be added to a
fixed RF carrier by mixing it with an audio frequency. Both are
initially sinewaves, of course.

What is the mathematical basis for achieving this?

As an example, say one wants to add a 1.6MHz spectral component to a
fixed 4.6MHz carrier. What frequency within the audio range would
achieve this, and how is it determined.

Thank you greatly for any insight on this matter.

Mark Jennings

6. ### Don LancasterGuest

When applied to electronic music, this was called the Chowning method.

Summary at http://www.indiana.edu/~emusic/fm/fm.htm

--
Many thanks,

Don Lancaster voice phone: (928)428-4073
Synergetics 3860 West First Street Box 809 Thatcher, AZ 85552

Please visit my GURU's LAIR web site at http://www.tinaja.com

7. ### ChuckGuest

If you want to see a 1.6MHz sine wave
and a 4.6MHz sine wave on one terminal,
you can combine the two inputs with a
simple resistive combiner, as has

No harmonics at all will be present at
the output of the resistive combiner.

If you want harmonics, the two (or more)
inputs must pass through a non-linear
device (mixer or modulator), in which
case the output (this depends on the
particular mixer/modulator chosen) will
consist of the original inputs plus
their sum and difference frequencies,
plus sums and differences of harmonics
of the inputs, such as 3n+m, 2m+17n,
n+4m, etc.

In theory, it doesn't matter whether the
frequencies are audio or RF. In
practice, the mixer must be chosen to
accommodate the frequencies to be mixed.

If this sounds like what you have in
mind, do a Google search on "frequency
mixer."

Chuck

8. ### Tam/WB2TTGuest

1.6 MHz is not in the audio range; so, just exactly what are you trying to
do?

Tam

9. ### LVMarcGuest

chuck,

the non linear device ie the mixer, will produce the desired sum and
difference, and then you get the inter modulation products.. I consider
the IM the "bad" result and distortion. I had thought that the poster,
really wanted to produce a picket fence of evenly spaced spectra, and he
described what he wanted best way he could. So the mixer and IM
distortion will produce sorta what the poster wants but in a "bad" way.
Te relatively wide BAND FM prodUCes A PICKET FENCE SPECTRA, AT THE
SPACINGS AND ....DOESNT COUNT ON THE NON LINEAR TRANSFER CURVE OF A
DIODE FOR EXAMPLE.

yOU CAN GET INto A WHOLE thread on making wieband, even ly spaced and
equal(almost amplitude, picket fence spectra creations.. all depends
what you really want to do, and considering implementation constraints
to guide the final design optimizations.

best regards,
Marco

10. ### ChuckGuest

LVMarc wrote:

snip
You can call them IM products but they
are always present in the output of a
nonlinear mixer. Usually unwanted in a
mixer's output, but always bad in a
device intended to be linear.

I had thought that the poster,

No way of knowing the OP's intent.

Chuck