DISSOCIATING WATER, A KEELY Project

[linda933]

G'day all,

Please you electronics guys, have a look at the transmitter circuit on page 2 of this thread and tell me why the modulator part is not functioning as it should. Any bright ideas on how to improve it?

Thanks fellas,

Hans von Lieven

Guys?  Fellas?  Do I hear a sexist attitude?  You handsome devil, you!

I would say that the circuit you show is not an AM modulator in any sense.  It is apparently a PWM modulator which puts out an unfiltered carrier (originally approximately square from first 555) that gets PWM'd to form a PWM replica of the "FM" input signal.  It won't do that very well either, since there is no output anti-alias filter to integrate the pulses into a model of the "FM" input.  The circuit appears rather pointless to me, since at best, if the proper low-pass filter was added, all it would do would be to reproduce whatever you fed into the "FM" input.  Maybe I am not understanding what you are trying to do, but I can tell you that what is shown is not an AM modulator.

An amplitude modulator performs the multiply function.  If the object of the game is to amplitude modulate a single-frequency sine wave carrier then wat you must do is control the amplitude in a linear fashion using a modulating signal...i.e. multiply by a number ranging from zero to one.  There are multiplier chips available.  MC1494, 1495 and 1496 are some real oldies but goodies Motorola used to make.

Analog Devices, Burr Brown (now TI) and others also made/make linear volume control chips of very high quality...these are simply multipliers optimized for low noise audio-band work and can also be had in log input (for "linear voltage to decibel" control schemes).  They are also known as electronic attenuators.  Finding one of wide bandwidth beyond normal audio range might narrow the selections down.  Analog multiplier is a term to search.  Gilbert cell is another, since the differential amplifier with current source known as the Gilbert cell forms the basis for most high quality multipliers.

Of course, the modern "code-head" design engineer might jump immediately to DSP chips given the task to provide one or more AM modulators.  What could be easier than that?  Another modern but non-software-intensive approach might be to use a "multiplying DAC" which is a digital-to-analog converter chip which multiplies a reference input by a digital number.  The reference input would be your sine wave carrier and the digital input would be the modulating waveform "multplier".

Be aware that when you do AM you will generate sidebands consisting of the sum and difference frequencies between the carrier and modulating signal.  

There...how's that for information overload!  And from a girl, too!  LOL

Linda

[hansvonlieven]

Linda darling you have outdone yourself you gorgeous creature you.

No more sexist remarks or omissions from me :-)

As to:  Be aware that when you do AM you will generate sidebands consisting of the sum and difference frequencies between the carrier and modulating signal.

Yes, I know. This is why I cannot use a transmitter that uses a crystal to generate the carrier frequency, pity really as it would have been easy to buy one off the shelf . The carrier frequency must be tuned (at least in my opinion) to a harmonic of the audio waves being fed in which have to be also in a harmonic relationship to each other. According to Keely stray frequencies are to be avoided at all costs.

Thanks for your contribution, this gives us something to work with.

Hans von Lieven

[fritz]

As I pointed out a while ago, AD9835 chip and his "brothers"
are quite nice chips (numerical oscillators).
Can generate sinusodial output from 0.01 to 20Meg with 0.01Hz
steps, locked to a master clock.
Still thinking if I should prototype a board with 8 or 16 such
oscillators.
Don?t forget the CA3080 for use as modulator ...
Or use a ring modulator..... lots of options.

I thought of doing a board with 8 to 16 AD9835 chips -
this gives 8 to 16 completely independant 0.01 to 20 MHz
oscillators, put a TCXCO master clock on the board - maybe
with GPS tuning feature and some options for analog modulation -
and envelopes.
This stuff should be connected via a controller to usb.

I got stuck with the AD9835 retail price - its 21$ a piece -
and 6$ if I take 100pcs.

With such a setup - you don?t need a modulator.

If somebody is interested - pls. contact me.

[hansvonlieven]

G'day all,

What do you electronics guys [ sorry Linda, or chicks :-) ] think of this as a simple way to get to my composite signal?

Hans von Lieven


EDIT   Sorry I missed to post the link

http://www.zen22142.zen.co.uk/spice/ammod.htm

[argona369]

Basically you want to amplitude modulate a carrier signal.

As simple as it  gets, might not fit the bill without a linear
amp output though.
http://sci-toys.com/scitoys/scitoys/radio/am_transmitter.html

And this one too.Though, it seems to need some mods,
http://users.usinternet.com/kyledrake/schematic/

There?s probably a lot of ways to amplitude modulate a carrier signal, this was just a couple
of simple ones.

Cliff,