AM modulator

Sep 15, 2025 Last reply: 10 mesi fa 14 Replies

Hi



I am looking to do a circuit that realize an amplitude modulation of an audio signal. The carriers would be:


1= 178 kHz, 2= 211 kHz, 3= 244 kHz,
4= 277 kHz, 5= 310 kHz, 6= 343 kHz

but at start will need only one of those.



the output wouldn't on antenna but on a telephone cable, the output level is from 10 to 20 mV.



In your opinion what would be the better circuit for this requirement?



My main concern is to no emit radio frequency.



Umberto



200kHz *is* a radio frequency! Long wave band on classic radios. BBC Radio 4 still transmits on 198kHz Long Wave. (due to stop in about a years time)

Various time signals transmit on 60kHz too.

But at low signal levels and in twisted pair cable you probably are not going to radiate much anyway. Look for tone burst generator circuits which might actually be what you want. How much modulation?

If your modulating signal contains high frequencies, these can produce sidebands at frequencies a long way from the carrier frequency. You may find you have to filter the modulating signals before applying them to the modulator. You may also have to restrict the amplitude of the modulation so as to avoid overloading the modulator (in particular, it must not cut off the carrier completely).

If you want to be absolutely certain of keeping the output within the wanted bandwidth, you really should put a filter on the output. This is always a good idea because a modulator that is non-linear (due to bad design, overload or a component fault) will produce a very wide band of unwanted output frequencies.

Yes, it is a radiofrequency, but in the past in Italy and in Swizzerland, where this system was invented, was a media by wire on the telephon line, it is for this reason that I would like a system that causes as little mess as possible.

if you know italian language:

formatting link
My goal is to give back voice on an old device for "filodiffusione"

I don't know how much modulation.

Look up "Bell Systems Journal", there is certain to be a description of the system design and parameters in that publication.

Here's a clever circuit constructed from a couple of components. And you don't need to know much about modulation to build it:

formatting link

The circuit's voice input signal varies the oscillator's power rail to automagically achieve amplitude modulation.

Thank you for your link to "filodiffusione." My browser automagically translated it to English. The commercial application of audio over twisted pair is very interesting.

Will the carrier frequency transmissions need to be simultaneous, or just selected?

You need two parts - a local oscillator to generate the carrier signal and a modulator to intrduce AM.

Something like the MC2060 has an amplitude controlled sine output and can handle switched crystals. You can also switch c values or crystals in an LC type oscillator, if the amplitude can be stabilized.

For simultaneous transmission, you'd need six odcillators and modulators.

RL

On 15/09/2025 6:22 pm, Umbe wrote: > Hi >

The neatest way of modulating a carrier at those kind of frequencies is a four quadrant multiplier.

You set up you carrier a as a sine wave - ideally with no DC content - and multiply that by the modulating signal.

There are loads of four quadrant multipliers. Analog Devices had some very nice parts, but when I last looked they'd gotten hideously expensive.

Jim Thompson's MC1494 and MC1495 were cheaper - but not nearly as good - and they are now obsolete.

Here's a clever circuit constructed from a couple of components. And you don't need to know much about modulation to build it:

formatting link

The circuit's voice input signal varies the oscillator's power rail to automagically achieve amplitude modulation.

Thank you for your link to "filodiffusione." My browser automagically translated it to English. The commercial application of audio over twisted pair is very interesting.

*** Adendum ***

Here's the cheap check circuit:

The One Apocalypse Radio That Will Never Fail You No Matter What Build yourself a basic AM receiver with scavenged parts for zero cost & zero power!

Antenna ---'|---- [Coil: 100+ turns] -----|' Detector (Razor blade/pencil lead) | | | | | (Tuner: paper clip on coil) | | | Ground ----'| |'----- Earphone |-----------(shared)----------| This is the bottom line when you can’t source a diode. A razor blade, a pencil lead and a bobby pin. Screw the bobby pin eye down just tight enough it can be adjusted back and forth. Believe it or not, you just made a replacement for a diode! With a tuner of sorts to boot!

formatting link

Il 15/09/2025 18:58, Don ha scritto:

Thank to all for the interesting consideration.

Meanwhile I found this solution:

formatting link
I think I will use it with the necessary modifications.

At those frequencies, I’d probably use an OTA such as the LM13700, with the input diodes biased with 500 uA or so.

Two modulators per package!

Cheers

Phil Hobbs

Nice and linear

<snip>

An excellent choice! Its most modern modulation uses a Class D switch to mince the input signal. Such digital adaptation makes analog methods (modulation by mixer) obsolete.

Of course, Class D operation is multiplying by a square wave which has all the odd harmonics up to a limit set by the edge speed of the switch.

You do have to filter them out, and no filter does a perfect job.

An ideal four quadrant multiplier doesn't generate higher harmonics. Real ones can obviously do, but good ones can get 60 or 70dB below the fundamental.

Join the Discussion

Have something to add? Share your thoughts — no account required.

Didn't find your answer?

Ask the community — no account required