Coupling cap in input stage

[Enzo]

Don't you ever hear the beat note when tuning two strings? That pulsation that slows to a stop as the strings come closer to tune? Those lower frequency pulsations may not qualify as "tone," but they certainly demonstrate that sub-freqs are generated when multiple notes are played.

The beat freq is the difference between the two signal freqs, but there also is a sum freq generated.

[mooreamps]

I have heard that. But I'm not convinced it's a "sub-harmonic" frequency coming out of the amp. I would be more convinced it's the perception of what two tones very close in frequency sound like at the ear. One could take two separete amplifiers, and two separete guitars , and probably produce the same kind of effect, and yet provide enough electrical isolation there could be just no way any kind of electrical signal mixing effect inside the equipment is going on.

-g

[Leftymatson]

Coupling cap in input stage

I'm new to the forum and thought I'd add my $0.02.

Here's an experiment: take two signal generators or other tone sources that are close in frequency, run them through separate amps and speakers (or use the same string on two different acoustic guitars), and tune one slowly to match the other in frequency. Hear the beats as you get closer? The beats are happening in your ear/brain circuit and not in the guitar amp front end.

While nonlinear distortion mechanisms can generate sum and difference tones that add to the original signals, that's not happening in the typical 12AX7 preamp circuit.

Filtering out frequencies below the lowest fundamental your instrument is capable of is a good idea as it tends to tighten up the lower registers and minimize mushy bass, but it won't eliminate the beats you referred to.

Sorry, Mooreamps, just saw your post.

[Enzo]

Yes, notes can beat acoustically, but that does not bear on whether or not they also do in the amp circuitry. proving that they will acoustically does not address teh isue of doing so in the circuits. Any distortion in the amp - and these amps are full of it even in "clean" channels - will result in intermodulation.

It would be easy enough to run an amp into a load and scope the output to see.

[Enzo]

Oh, and welcome to the forum, Lefty.

[Steve Conner]

I always thought that those beat tones were ear/brain artifacts, but you can also get them from intermodulation in circuitry that distorts. Notice how much stronger and more noticeable the beat tones are when you're using distortion on a guitar.

Beat tones are intimately related with consonance and dissonance in musical harmony. When you bring distortion into the mix, the harmonics and beat tones make consonant intervals sound stronger, and dissonant ones sound more dissonant.

I've heard some metal bands that deliberately use dissonant chords with distortion, I can only assume it's to make the listener suffer more. Next time you have your Boss Metal Zone plugged in, try some chords with intervals of a minor second. This is a pitch difference of one semitone, the most dissonant interval you can get. If you play them high enough up the fretboard, your dog will probably leave home.

As a teenager, I used to love that tone that Bruce calls a "wanker woofie drone". I think it's supposed to imitate the punch and power that you hear on metal records. But one day I realised that actual metal guitar tone is quite bright and bass-light, and all that punch comes from the bass player doubling the guitar riff. That gives you a massive bottom end with no intermodulation worries.

[kg]

I'd have to see it on an audio spectum analyzer first, before I could believe it.

-g

well, you are free to believe what you want to believe. i'm just relaying the science... i'm not making it up!

Don't you ever hear the beat note when tuning two strings? That pulsation that slows to a stop as the strings come closer to tune? Those lower frequency pulsations may not qualify as "tone," but they certainly demonstrate that sub-freqs are generated when multiple notes are played.

The beat freq is the difference between the two signal freqs, but there also is a sum freq generated.

the important thing is that the AMP is NOT generating the sub-fundamental frequencies... you have two signal generators (two strings) each with a different output frequency.

assume for a moment that the strings are producing nice clean sine waves, and both have equal amplitude, but slightly different frequencies. put the output of the pickup right into an oscope.

you will see that there exists an output which has greater amplitude than either of the two waves alone.

here is a picture of the effect from the wikipedia article on beats:

A 110 Hz A sine wave (cyan), a 104 Hz G# sine wave (magenta), their sum (blue) and the corresponding envelope (red)

important is the fact that we have a peak amplitude 2x as large, and a very low beat frequency (the large swooping red line in the bottom frame) that are a byproduct of the simple linear addition of two sine waves. when the two waves are in phase, they combine (aka interfere) constructively and reinforce each other, and when they are out of phase they combine destructively and cancel each other out.

you can also see this on the surface of water that has waves moving across it.

I have heard that. But I'm not convinced it's a "sub-harmonic" frequency coming out of the amp. I would be more convinced it's the perception of what two tones very close in frequency sound like at the ear. One could take two separete amplifiers, and two separete guitars , and probably produce the same kind of effect, and yet provide enough electrical isolation there could be just no way any kind of electrical signal mixing effect inside the equipment is going on.

i didn't bring it up because it has to occur "inside the equipment." in fact, since we were talking about the input stage, this signal mixing would have to occur in the pickup or in the instrument cable to be present at the input stage grid.

Here's an experiment: take two signal generators or other tone sources that are close in frequency, run them through separate amps and speakers (or use the same string on two different acoustic guitars), and tune one slowly to match the other in frequency. Hear the beats as you get closer? The beats are happening in your ear/brain circuit and not in the guitar amp front end.

no, the beats aren't happening in your ear/brain.. they're happening in the air which is the transmission medium for sound energy. you can observe it by monitoring the output of a pressure transducer--no ears or brains are necessary to perceive it, and it will look just like the idealized/simplified version referenced above from the wiki article.

While nonlinear distortion mechanisms can generate sum and difference tones that add to the original signals, that's not happening in the typical 12AX7 preamp circuit.

you do not need nonlinear distortion nor any amplifying devices at all to create beat frequencies. two devices emitting sound energy are all that's necessary. the beat frequencies are already in the signal that is coming from the pickup before it ever hits the input stage.

Filtering out frequencies below the lowest fundamental your instrument is capable of is a good idea as it tends to tighten up the lower registers and minimize mushy bass, but it won't eliminate the beats you referred to.

what beats am i referring to?

if i make a beat at 7Hz in my guitar, then sure, you can filter it out pretty easily, and you probably should.

if you make a beat at 70Hz, do you think it's a good idea to roll that off at the head end? speaking for myself and my own designs, i am not so sure.

(in either event, i wouldn't do it pre-input stage, i would do it downstream where i had control of both source and load impedances.)

ime this is the most challenging aspect of amp design--getting the entire guitar, low string to high string, to sound cohesive and pleasant, as you strum a full chord. there are an infinite number of beat frequencies that are present, though obviously some are much stronger and others are inaudible. getting them all to "play nicely together" is immensely complicated and basically a black art because of the sheer # of variables.

Yes, notes can beat acoustically, but that does not bear on whether or not they also do in the amp circuitry. proving that they will acoustically does not address teh isue of doing so in the circuits. Any distortion in the amp - and these amps are full of it even in "clean" channels - will result in intermodulation.

this intermod distortion is a critical component of the amp's tone, and what we as musicians have come to expect as the output of a MI amp. shaping it is the name of the game and has a huge effect on the way the amp behaves, and is also very difficult and/or time consuming.

I always thought that those beat tones were ear/brain artifacts,

as above, they are an acoustic phenomenon, not a perceptual, psychological or physiological one.

but you can also get them from intermodulation in circuitry that distorts. Notice how much stronger and more noticeable the beat tones are when you're using distortion on a guitar.

Beat tones are intimately related with consonance and dissonance in musical harmony. When you bring distortion into the mix, the harmonics and beat tones make consonant intervals sound stronger, and dissonant ones sound more dissonant.

completely agreed. i think the distortion has two major effects:

great compression of signal amplitude which brings out lower level harmonics that then contribute to their own beat combinations.

large amount of clipping and nonlinear distortion which tends to heterodyning inside the amp circuits itself.

As a teenager, I used to love that tone that Bruce calls a "wanker woofie drone". I think it's supposed to imitate the punch and power that you hear on metal records. But one day I realised that actual metal guitar tone is quite bright and bass-light, and all that punch comes from the bass player doubling the guitar riff. That gives you a massive bottom end with no intermodulation worries.

well, sure--as long as the two guitars are in tune!

[Steve Conner]

you can observe it by monitoring the output of a pressure transducer--no ears or brains are necessary to perceive it, and it will look just like the idealized/simplified version referenced above from the wiki article.

Sure it will. But the question is whether a waveform looking like that contains any beat tones. The math says that it doesn't: it's just a sum of two sine waves and contains only those two frequencies. If you fed it into a spectrum analyzer, you'd just see two peaks.

The beat frequencies only come into existence when the wave undergoes some sort of non-linear process. People explain our perception of beat tones by saying that our ears are non-linear, maybe that they perceive sound the way the analyzer in a vocoder does: a filter bank followed by a bank of envelope detectors. If you hear two tones in the same "bin", you get a beat tone.

I like to think that our perception of sound has about 12 log-spaced bins to the octave, and the whole structure of music is a consequence of this.

[gmoon]

An excellent graph, and illustrates well that phase-effect "beat frequencies" are amplitude modulation, rather than frequency. That distinction isn't terribly relevant in terms of perception, however (especially when distorted.)

I'm still curious about the original statement--any measurable subharmonics? (1/2f or 1/3f, rather than the overtone 2f, 3f, etc...)

Do they exist a high enough percentage to be perceptible? From either the instrument or the amp?

[Arthur Dent]

Next time you have your Boss Metal Zone plugged in, try some chords with intervals of a minor second. This is a pitch difference of one semitone, the most dissonant interval you can get.

Nah, the diminished fifth (six semitones) is slightly more dissonant.

[Steve Conner]

I don't buy that. Look at the uses of the tritone in music:
http://en.wikipedia.org/wiki/Tritone#In_popular_music

Now check out the minor second. It sounds so bad that only Schoenberg could stand it
http://en.wikipedia.org/wiki/Minor_second

[Arthur Dent]

The tritone is traditionally held to be the most dissonant of intervals, and was often associated with the devil.

Either way, both intervals are used frequently, it's just that their root is usually not the root of the chord. A major 7 chord contains a m2 interval (M7-1), and a dominant 7 chord contains a dim5 (M3-m7).

Whatever, they both sound goofy!

[mooreamps]

i didn't bring it up because it has to occur "inside the equipment." in fact, since we were talking about the input stage, this signal mixing would have to occur in the pickup or in the instrument cable to be present at the input stage grid.

Sorry, not possible... Overtones perhaps, but you would need a pn junction to generate the sum, the difference, and the original two fundamentals.

-g

[kg]

Sorry, not possible... Overtones perhaps, but you would need a pn junction to generate the sum, the difference, and the original two fundamentals.

-g

gary, as i've said you are free to believe that cutting off all frequencies <83Hz on a standard tuned guitar won't change the tone of the amp.

i however know differently.

[Steve Conner]

Some interesting points there. I once experimented with a 4-pole highpass filter at 20Hz in front of an amp. It sounded like crap! I think the problem was that it removed all the little transient shifts and jumps you get from pick attacks, palm muting, etc. which would otherwise kick the operating points of the stages around. I don't think it has anything to do with difference frequencies.

BTW, the point about dissonant intervals in chords is a good one. All I can say is that no guitar chord I know of contains an interval of a semitone, because the fingering would be hard. The major seventh example you gave will always have the lower note in the m2 moved up by an octave, or the higher one moved down. This inverts the interval so it's a semitone short of an octave. Which oddly enough is a major seventh.