The first circuit didn't get its fair day in court. I put its output through my recently-made IF amplifier and I am only now discovering it to be a source of distortion. I'll have to fix it.

I'll play with the first circuit today after my chores and see what it can do. I think it should actually be a high fidelity circuit. Have you noticed that both circuits are purely passive? And, except for the LC tank at the output of the second one, they're not tuned. This is an advantage for me since I intend to put an AM modulator in its own aluminum box and use it for experiments. I'm waiting for a couple ICs in the mail so I can breadboard them and then pick the best modulator for the box.

I set up a simulation using a very similar design to yours with some minor modifications.

The trick is (as suggested) to correctly filter the output. I used a PI section high pass matching filter between the output transistor collector and a 1KΩ load resistance. One can match to a range of load resistances.

The resulting AM was good quality with some very minor distortion. This showed up as low level residual sidebands in the frequency spectrum.

Of course adding the matching network limits the operation to a small frequency band - something you seem to be wanting to avoid.

One doesn't need to drive the signals into distortion - the differential pair gain modulation effect works quite well in the linear region.

Another approach, which seems to have promise, is to replace the output transistor collector resistor with a RF transformer with a suitably loaded secondary. This would give a potentially greater bandwidth as the transformer needs only to reject the low frequency modulation signal component in the composite waveform. So this might fit with your requirement for an "untuned" device.

 

I set up a simulation using a very similar design to yours with some minor modifications.
The trick is (as suggested) to correctly filter the output. I used a PI section high pass matching filter between the output transistor collector and a 1KΩ load resistance. One can match to a range of load resistances.

The resulting AM was good quality with some very minor distortion. This showed up as low level residual sidebands in the frequency spectrum.

Of course adding the matching network limits the operation to a small frequency band - something you seem to be wanting to avoid.

One doesn't need to drive the signals into distortion - the differential pair gain modulation effect works quite well in the linear region.

Another approach, which seems to have promise, is to replace the output transistor collector resistor with a RF transformer with a suitably loaded secondary. This would give a potentially greater bandwidth as the transformer needs only to reject the low frequency modulation signal component in the composite waveform. So this might fit with your requirement for an "untuned" device.

I thought about a transformer at the output but without a capacitor across it so as to make it pass any reasonable frequency. Are you saying that with a tuned circuit you get the correct AM signal? I've been wanting to make a versatile modulator such that it could just combine any audio signal with any rf signal from about 300kHz to 30MHz. This morning I found the solution after my mail came and included an MC1496 I had ordered from Jameco. I was surprised at its performance. The output is just as nice as the diagram drawn in a book and it performs well at least up to 35MHz (the highest frequency my scope will display). It reproduced perfect 1kHz sine waves, triangles, ramps, square wave and pulses. Very sharp. No distortion at all. I downloaded Motorola's data sheet and simply bread boarded the AM modulator as per the drawing. This modulator will suppress the carrier if wanted such that DSB comes out, it will demodulate PM and FM, act as a frequency doubler and a couple other things as well. Its only drawback is that it requires two supply voltages -- positive and negative. The IC includes only eight transistors set up as a cascoded differential amplifier array. Not that sophisticated, but the fact that all those transistors are built on the same substrate means they're matched perfectly and they're fast. It will double a 150MHz signal to 300MHz. I'm impressed.

 

I used to work on board that used the 1496 (also the 1596). It is an oldie but goodie. No one uses old telephone tech any more, but there was a time all telephone channels were SSB 3.4Khz wide stacked in 4Khz slots, up to 8Mhz. We used lots of those chips back then, along with Mr. Art Collins signature invention, the mechanical filter.

 

Hello,

Here is an example of the 1496 schematic Bill was talking about:
http://hobbyelectron.blogspot.nl/2011/08/am-modulator-circuit-diagram.html

It is also mentioned in this thread on the AAC:
http://forum.allaboutcircuits.com/showthread.php?t=38171

Bertus

 

I used to work on board that used the 1496 (also the 1596). It is an oldie but goodie. No one uses old telephone tech any more, but there was a time all telephone channels were SSB 3.4Khz wide stacked in 4Khz slots, up to 8Mhz. We used lots of those chips back then, along with Mr. Art Collins signature invention, the mechanical filter.

No wonder telephones sound like people are wearing close pins on their noses. I guess as long as we can still understand one another a narrow audio bandwidth is alright.

As for the mechanical filter would that be based on a ceramic crystal? I just bought a 455kHz ceramic thinking it would be useful for an IF amplifier or mixer output.

 

Hello,

Here is an example of the 1496 schematic Bill was talking about:
http://hobbyelectron.blogspot.nl/2011/08/am-modulator-circuit-diagram.html

It is also mentioned in this thread on the AAC:
http://forum.allaboutcircuits.com/showthread.php?t=38171

Bertus

I got an MC1496 this morning and bread boarded it as per Motorola's
application sheet and got perfect modulation up to the limit of my scope's 30 MHz ability. It might be that guy who had problems just didn't set it up right or that he had a bad IC. Some defective ones do get by quality control, you know.

 

No wonder telephones sound like people are wearing close pins on their noses. I guess as long as we can still understand one another a narrow audio bandwidth is alright.

As for the mechanical filter would that be based on a ceramic crystal? I just bought a 455kHz ceramic thinking it would be useful for an IF amplifier or mixer output.

I opened mechanical filters up several times, they were odd to say the least. They came in a plastic box around 3" long, 1½" X 1½". Inside was a 3 or 4 metal disks, about 1" wide and ¼" thick, with a bunch of wires spot welded in odd places holding them together. I have no idea what the theory of operation was, but they had an extremely high Q and were really flat in their pass band range, with tight rejection characteristics. They were delicate, drop one and it was ruined. They were the IF filter for their modems, and only two per channel was needed (TX / RX), one modem per voice channel. They used the 1458 (a dual 741) for the audio amps, and had really nice specs, showing that even old parts can do a good job in a good design. The modems were fed with synthesized frequencies in 4Khz steps.

 

I thought about a transformer at the output but without a capacitor across it so as to make it pass any reasonable frequency. Are you saying that with a tuned circuit you get the correct AM signal? I've been wanting to make a versatile modulator such that it could just combine any audio signal with any rf signal from about 300kHz to 30MHz. This morning I found the solution after my mail came and included an MC1496 I had ordered from Jameco. I was surprised at its performance. The output is just as nice as the diagram drawn in a book and it performs well at least up to 35MHz (the highest frequency my scope will display). Snipped ...

Glad to hear you've found a suitable solution. I'm not surprised the purpose designed IC works far better than the original proposed discrete component based setup.

Yes, the PI (L:C:L) filter "worked" fine (in simulation) to produce the "traditional" AM signal - not DSBSC.

 

I opened mechanical filters up several times, they were odd to say the least. They came in a plastic box around 3" long, 1½" X 1½". Inside was a 3 or 4 metal disks, about 1" wide and ¼" thick, with a bunch of wires spot welded in odd places holding them together. I have no idea what the theory of operation was, but they had an extremely high Q and were really flat in their pass band range, with tight rejection characteristics. They were delicate, drop one and it was ruined. They were the IF filter for their modems, and only two per channel was needed (TX / RX), one modem per voice channel. They used the 1458 (a dual 741) for the audio amps, and had really nice specs, showing that even old parts can do a good job in a good design. The modems were fed with synthesized frequencies in 4Khz steps.

I'm going to look for a mechanical filter so as to convert DSB to SSB. As I said, I bought a 1496 and found it to be the cat's meow for AM modulation: just a turn of a pot and the carrier is gone. As I understand it all I have to do to get SSB from an IF at 455KHz is tune the carrier about 5kHz higher than the IF (460KHz) to get the USB or 5kHz lower (450 KHz), then pass this through a sharp filter tuned at 455KHz. This signal is then transmitted.

Incidentally, I also got a 565 PPL in that batch of mail. I set it up yesterday and set it up for a free running frequency of 455 KHz. It works great -- just exactly like expected. Using dual trace mode, I can watch the DC at pin 7 change with frequency through a center point with the expansion or contraction of the wavelength. Now all I need is an FM audio wave to test it. I'll try to mix down a broadcast band wave through my 455kHz IF amp today. This is a whole lot easier than building a working discriminator from toroids! I'm finally catching up with the 20th century.

 

OK, I drew this from memory, hope I got it right.

So do you mean add a DC offset to the audio input?

Bottom two diodes are backwards...they need to go around ina ring (which is why it's sometimes called a ring modulator!

Yes..you just add a touch of DC to get a carrier out of the thing.


eric

 

I opened mechanical filters up several times, they were odd to say the least. They came in a plastic box around 3" long, 1½" X 1½". Inside was a 3 or 4 metal disks, about 1" wide and ¼" thick, with a bunch of wires spot welded in odd places holding them together. I have no idea what the theory of operation was, but they had an extremely high Q and were really flat in their pass band range, with tight rejection characteristics. They were delicate, drop one and it was ruined. They were the IF filter for their modems, and only two per channel was needed (TX / RX), one modem per voice channel. They used the 1458 (a dual 741) for the audio amps, and had really nice specs, showing that even old parts can do a good job in a good design. The modems were fed with synthesized frequencies in 4Khz steps.

The Collins mechanical filter is a classic. The filter works on MAGNETOSTRICTION, and the coupling wires transfer energy to each of the resonant disks. Let me see if I can find a complete description....it is also in the ARRL Handbook
Eric

 

The Collins mechanical filter is a classic. The filter works on MAGNETOSTRICTION, and the coupling wires transfer energy to each of the resonant disks. Let me see if I can find a complete description....it is also in the ARRL Handbook
Eric

http://www.wa3key.com/filters.html
Here's a nice little rundown on the gems.

 

Bottom two diodes are backwards...they need to go around ina ring (which is why it's sometimes called a ring modulator!

Yes..you just add a touch of DC to get a carrier out of the thing.


eric

Thanks, I may wire one up and try it out.

 

Glad to hear you've found a suitable solution. I'm not surprised the purpose designed IC works far better than the original proposed discrete component based setup.

Yes, the PI (L:C:L) filter "worked" fine (in simulation) to produce the "traditional" AM signal - not DSBSC.

I am beginning to suspect that really understanding how the modulation comes about is the key to design. I will admit my ignorance. But I plan to get a handle on it.

 

The Collins mechanical filter is a classic. The filter works on MAGNETOSTRICTION, and the coupling wires transfer energy to each of the resonant disks. Let me see if I can find a complete description....it is also in the ARRL Handbook
Eric

Thanks, KL7AJ, I just realized you are an affiliate of ARRL. That means you are big time. I respect ARRL as the amateur's Bible for radio electronics. So I am really honored that you should speak to me.

I shall get me a Collins mechanical filter at 455 kHz. Do you know why this frequency was settled on as a standard? Whatever dictated its origins, the industrial supply of stuff for that frequency has consolidated that proposition.

 

http://www.wa3key.com/filters.html
Here's a nice little rundown on the gems.

So what happened? Did the magic number, 455KHz become enshrined? The AM commercial band is happy with it; but, the FM commercial band didn't like it. I suspect that was because of the distance between stations is so much greater on FM. And so you use a 455KHz IF strip and you have too close tuning using a variable capacitor?

 

Hello,

The ring modulators are available in small packages from minicircuits.
I have attached an overview of them.

I also have attached a picture with a homebrew version.

Bertus

 

Hello,

The ring modulators are available in small packages from minicircuits.
I have attached an overview of them.

I also have attached a picture with a homebrew version.

Bertus

Thanks Bertus. I have already constructed this circuit from discreet diodes and from an IC can. They are passive and non-tuned, which makes them desirable for me since I am just experimenting with them.

 

Thanks, KL7AJ, I just realized you are an affiliate of ARRL. That means you are big time. I respect ARRL as the amateur's Bible for radio electronics. So I am really honored that you should speak to me.

I shall get me a Collins mechanical filter at 455 kHz. Do you know why this frequency was settled on as a standard? Whatever dictated its origins, the industrial supply of stuff for that frequency has consolidated that proposition.

Well, I don't know how "big time" I am, but I have written articles for QST for about 35 years.

The original Collins mechanical filters were at 455, because that was the standard I.F. frequency for single conversion receivers long before then. 455 was historically chosen as a good compromise between bandwidth and image rejection...the images were well above the standard broadcast band. It was also low enough so there was no feedthrough ("blow-by") at 455 because there were no licensed broadcasters there.

In a way the 455kc standard is kind of like how we ended up with 4'6" spacing between American railroad rails.

Eric

 

I am beginning to suspect that really understanding how the modulation comes about is the key to design. I will admit my ignorance. But I plan to get a handle on it.

Hi Paul:

Here's the cheap and dirty rule: Modulation is a MULTIPLICATION process. What you see in your oscilloscope plot is mere ADDITION (superposition between the carrier and the audio. Multiplication requires a non-linear element....addition does not. This is why you always want to saturate (overdrive) your non-linear elements with your carrier.

Eric