Resonance feedback loop on guitar

[Normalization]

Hey guys, I was messing around with my guitar (classical guitar) and my amp lately. I put the treble on lowest, bass highest, reverb highest, mid highest and mid frequency lowest. When I pluck a string when my guitar is close to the amp, the amplified sound from my amp causes that string to resonate because the frequency of the amp is the same as the natural frequency of that particular string (Correct me if I'm wrong) until the maximum volume of the amp (restricted by the settings) has been reached.

There are just a few curious things I noticed:
1. Only certain frequencies tend to be susceptible to this resonance feedback loop. Has that got to do with the natural frequency of the amplifier itself? When I've tried this in the past different frequencies tended to be more susceptible. Why are certain frequencies more susceptible to this feedback loop?
2. (E string tuned to D) When I plucked the D string it started to resonate but then when I plucked the E string, which was also resonating but less susceptible to the resonance feedback loop, the amplitude of the D string started to oscillate and the resultant frequency emitted started to oscillate as well. I'm guessing the amplitude of the D string is oscillating because the maximum volume has already been reached, so the amp needs to compensate by reducing the amplitude of the D frequency and then it needs to compensate for the amplitude for the E string etc. But why is the resultant frequency oscillating (from low (~D frequency) to higher frequency)? I'm guessing some kind of interference is going on, but if someone could enlighten me on this that would be great.

I'm only a senior high-school student, so maybe my questions just have to do with simple physics I haven't met yet.

Thanks to everyone courteous enough to reply.

I'm sure there are many things I haven't explained/worded properly so please don't hesitate to ask me to clarify on this and that.

 

[CWatters]

For positive feedback to occur the time it takes for sound to go around the loop...

Guitar string
Pickup
Amp
Speaker
Guitar string

..is critical. Imagine a string is moving "up" and that causes a voltage in the pickup that eventually causes the speaker to move out and a sound wave to travel across the room to the guitar. If that arrives at the string in such a way that it is "in phase" (eg it also causes the string to move "up") then it will amplify the motion of the string and the system may resonate.

A lot of factors effect if the sound wave arrives at the string "in phase". For example the distance between the speaker and the string will affect which frequencies arrive in phase and which don't.

There is another condition that also has to be met for feedback to occur.. The gain around the loop must also be more than one. If less than one the sound will die away. If greater than one it will grow stronger.

Guitar strings act like filters and will tend to attenuate some frequencies and amplify others. This will effect which frequencies meet the two conditions needed for feedback to occur.

 

[rcgldr]

Redo the experiement with the reverb shut off. It's possible that the reverb circuity is frequency sensitive.

 

[bahamagreen]

First off, it is the guitar that is where the resonant things are going on with these sounds, not the amp; the amp is just playing what is happening on the guitar.

When the volume is high enough, and especially with bass tone set high on the amp, the entire body of the guitar vibrates, and any undamped strings, including fretted strings will vibrate. The body of the guitar has a collection of natural resonant frequencies, and any string open or fretted whose note is at or close to matching those resonant frequencies of the guitar body are going to be the ones that feedback the most easily... the position and orientation of the guitar with respect to the amp will also influence which resonances of the guitar are being most driven, since the sound is both hitting the exterior of the body, but also entering the sound hole, so pointing the sound hole at or presenting a flat body surface toward the amp will change things.

At high enough levels, any string at any fretted position will feedback because the sound is acting directly on the string and will drive it, even if the body is damped.

The warbling oscillation when you let two notes play is not the amp, it is the interaction between the notes. The D (low E tuned down to D) is generating overtones that are multiples of the frequency (octaves). The open D string is an octave above, so that is a match. The pace of the oscillation is the difference between the two frequencies, so when they are very close, the warbling sound is slower.

For example, your low D (tuned down low E string) is 83Hz and your open D string is 146Hz. When those play together with feedback, the low D is using it's second harmonic at 146Hz like the open D an octave above to match the open D string. Any slight difference in tuning between these two will cause one to modulate the other and create a "beat frequency", which is the difference between them - cyclic constructive and destructive interference.

When feeding back, the string may endure an excursion (how far is moves side to side as it vibrates) much greater than normal playing, and this actually increases the time average length of the string between the nut and bridge, so the pitch goes slightly flat. This can also be part of the difference between the two frequencies.

So, same example, you have the low D sound at 83Hz, and its second harmonic at 146Hz driving the open D string at 146Hz. If the low D is going a little flat by a couple of Hz, then so is the second harmonic, and there will be a few Hz (cycles per second) of beats... the oscillation.

Adding reverb is adding some additional parallel signal gain, which increases feedback, and if your reverb is the old style that uses a metal tank with the limp springs stretched between the little driver transformer and the recovery transformer inside the tank, then all kinds of additional things can be happening with the feedback frequencies. Those springs lengths have their own resonant frequencies (pretty low frequency, so slow beats) and being mechanically located in the amp cabinet, they get driven a bit at loud levels.

If you want to do some more experimenting at not so loud levels, you can induce feedback at much lower level by skipping the air path and touching the head of your guitar directly to the side or top of the amp.

If you are trying to prevent feedback, you need to use a long cable or extension to get some distance away.

 

[sophiecentaur]

If you are trying to prevent feedback, you need to use a long cable or extension to get some distance away.

Alternatively, you can often find a position, right behind the speaker where less of the amplified sound will impinge on the guitar. There is often a 'null' in the radial pattern of loudspeakers and it should be possible to find it. Bands that perform with very high levels of sound, in arenas etc., usually are positioned well behind the massive banks of main PA speakers. They often operate with small personal, foldback or monitor loudspeakers, placed in front and facing towards them, so that they can actually hear what they sound like. (Yes - that is a possible source of howl round)

Solid electric guitars are much less susceptible to howl round than electro acoustic guitars, which can pick up a lot more of the ambient sound on the body of the instrument.

There are systems for reducing howl round which involve frequency shifting networks which kill the build up of oscillations. This link refers to it and other methods of dealing with howl round. The shift is usually not more than a very few Hz but I don't know how it affects the subjective quality of the sound. It's very suitable for speech, though.