Simpler frequency shifter?

[Dr.EM]

I found his chip:

**broken link removed**

It claims to produce the "product of an input voltage (signal) and a switching function (carrier)." Which I assume means that it adds two signals frequencies together. Would this mean that if I fed one input with an audio signal of a suitable level, and the other input with a sine wave of variable frequency, then the output would be that audio frequency plus the frequency of the sine wave? There are probably 1001 reasons why thats not the case, I know very little about the chip, just looked at the datasheet. All I can think is that you'll be able to hear the sine wave even without an input, which would be very annoying :lol:

 

[Nigel Goodwin]

I found his chip:

**broken link removed**

It claims to produce the "product of an input voltage (signal) and a switching function (carrier)." Which I assume means that it adds two signals frequencies together. Would this mean that if I fed one input with an audio signal of a suitable level, and the other input with a sine wave of variable frequency, then the output would be that audio frequency plus the frequency of the sine wave? There are probably 1001 reasons why thats not the case, I know very little about the chip, just looked at the datasheet. All I can think is that you'll be able to hear the sine wave even without an input, which would be very annoying :lol:

It's a double-balanced mixer, the double balanced part means that the two inputs should be at reduced levels (perfect balance would be none at all!) on the output.

So what you get out 'should' be the sum and difference only - you then have the problem of seperating the sum and difference!. This is why you commonly use a 9MHz modulation frequency, and follow the modulator with a sharp 9MHz crystal filter - producing an SSB signal, either the sum or the difference, depending on the exact modulation frequency.

 

[stevez]

I know that Nigel knows this but I am adding it for further info. The mixing of an audio signal and RF will produce the sum and difference. If the audio were a microphone the output would be a common AM signal but absent the carrier (double sideband or DSB). As Nigel explained, quite often we only want one sideband so we filter the other sideband and the carrier if any is left.

Another thing that can be done - input a modulated RF signal to the input as well as a pure sine wave or carrier to the other. The output ought to be audio frequency and some more RF (the RF is usually undesired and filtered). This is called Direct Conversion - so this would be the heart of a direct conversion receiver. Quite often radio amateurs or hobbyists use this for relatively simple but effective receivers.

 

[Dr.EM]

Think I understand you, so its not any use as a frequency shifter, as you will be getting the difference frequency in there too, at least without some clever filtering. If it produces sum AND difference, then it is the exact same effect as a ring modulator right?

 

[stevez]

I do not know what a ring modulator is but there are any number of ways to do accomplish mixing. A hybrid combiner is yet another way. For what it's worth, mixing of two RF signals was used in receivers - heterodyning was what they called it - leading to the 'superheterodyne' description in radios. The different methods each have their own advantages and disadvantages so it might be incorrect to conclude that they are exactly the same. I know my amateur radio references have quite a bit of information on mixers - as they relate to receivers or the generation of SSB. Its interesting stuff.

 

[audioguru]

Hi Dr. EM,
That's the same MC1496 balanced modulator IC that I used for my wireless tele-conferencing system voice encryption.
I used a 3.5kHz carrier and the output was the sum and difference frequencies with the carrier suppressed by the IC's balanced circuits. I used a switched capacitor lowpass filter to sharply cutoff the sum frequencies and the remaining frequencies were audio from about 100Hz to about 3400Hz but the frequencies were the opposite of the original:
100Hz in, 3400Hz out.
500Hz in, 3kHz out.
2kHz in, 1.5kHz out.
3kHz in, 500Hz out.
The quality was so good that when I used an identical circuit to un-encrypt it, it sounded identical to the original.
I think if you change the carrier frequency of one of them then you will have a frequency shifter. But since I think it would shift all frequencies by the same amount, their harmonics wouldn't be musically related anymore. :lol:

 

[Dr.EM]

Ah, I get ya , that makes sense. I guess with relation to harmonics, only musical "5ths" and octaves are going to sound tolerable, in much the same way as a ring modulator does :lol: . I guess the only way to actually, factually shift frequencies by any amount is using digital circuitry, but that takes the fun out of it, boolean maths, yawn... :lol:

 

[David Bridgen]

If it produces sum AND difference, then it is the exact same effect as a ring modulator right?

Yes.

I do not know what a ring modulator is but there are any number of ways to do accomplish mixing. A hybrid combiner is yet another way.

A hybrid combiner isn't a mixer. It's linear. It's an adder.

 

[mechie]

Ring Modulator

then it is the exact same effect as a ring modulator right?

This is correct - a synthesiser's ring modulator is a double-balanced mod, if you search for circuits you will find a simple ring modulator circuit can be built out of four diodes, similar to a bridge rectifier but one diode is back-to-front, creating what was called a "twisted ring modulator", originally used in telecomms I believe to allow more traffic along a single pair of wires (sum frequency only), and stolen for synthesisers because of their ability to create extremely harmonic-rich tones that sound like bells, enforcing the name as ring-mod.

Feed both inputs with a triangle wave and it does its magic with each and every harmonic in both signals - if the inputs track each other musically the output has a distinct bell-tone, not all the resultant harmonics are musically related.

 

[Dr.EM]

I have that simple ring mod you speak of in my synth. I think it might be limited by the cheap transformers, they have an upper frequency limit of 3.5khz, though I think that just means that they will be a lot quieter above that, not that they cease to work. I can get a pure triangle wave from one of the LFOs, if I turn the frequency up high enough, and will have to try the oscillators kind-of sort-of sine wave with that.

 

[audioguru]

I think a triangle waveform sounds rough, and a sawtooth sounds like a violin. I have heard real and simulated bells with some harmonics slightly out-of-tune, giving them slow beating something like wah-wah-wah, you know. :lol:

 

[mechie]

Flute sound

I think a triangle waveform sounds rough

A gentle low-pass filter will make it sound very much like a flute :wink:

I know -- off topic -- Sorry guys :roll:

 

[mechie]

Ring modulators

I can get a pure triangle wave from one of the LFOs, if I turn the frequency up high enough, and will have to try the oscillators kind-of sort-of sine wave with that.

The more harmonics you put in the more you get out - try square and ramp (sawtooth) - sines aren't going to give a rich output.

 

[Dr.EM]

Because my synths oscillators use waveshapers, they have the advantage of continuously variable waveshapes (and the disadvantage of the waveform changing over the frequency range :roll: ). So, I have tried using one input fixed as a sort of "soft wave", and sweep the other from soft, to saw shapes. You get a huge dynamic change, but it sounds pretty interesting too. Will try toget a goodbell sound later

 

[stevez]

David - I happened to read about a combiner the other day and the relatively simple explanation left me with the sense that the output was the sum of the inputs - therefore my perception that it was a lot like a mixer. When I have a spare moment I'll read up on the subject - too much going on at the moment. Great to either adjust my understanding or reaffirm my understanding via the forum.

 

[Nigel Goodwin]

David - I happened to read about a combiner the other day and the relatively simple explanation left me with the sense that the output was the sum of the inputs - therefore my perception that it was a lot like a mixer. When I have a spare moment I'll read up on the subject - too much going on at the moment. Great to either adjust my understanding or reaffirm my understanding via the forum.

The confusion arises because of the use of the same word for different purposes, an audio 'mixer' is totally different from an RF 'mixer'.

The chip in question is an RF mixer, which produces sum and difference at the output. An audio mixer simply adds them together, in fact a virtual earth mixer is just an analogue computer adder circuit.

 

[stevez]

Thankfully I had my trusty ARRL Handbook nearby this morning and took a minute to read about mixers. A few sentences helped ease my confusion -

The combiner simply adds the instantaneous voltages of the two signals together to produce the output at each point in time. The mixer, on the other hand, multiplies the instantaneous voltages of the two signals together of the two signals together to produce its output signal from instant to instant. Comparing the output spectra of the combiner and mixer, we see that the combiner’s output contains only the frequencies of the two inputs, and nothing else, while the mixer’s output contains new frequencies. Because it combines one energy with another, this process is sometimes called heterodyning, from the Greek words for other and power.

The rest of the chapter does a great job with a bit more detail.