Pulled out of tune by distortion?

Here is a forum for everything between the guitar (here) and the amp (there). If you want to know about cables, effects pedals, multieffects, noisegates, line drivers, phasers, distortion post your questions in here.
Post Reply
rnrsh
newbie
Posts: 2
Joined: August 12th, 2007, 12:24 am

Pulled out of tune by distortion?

Post by rnrsh » August 12th, 2007, 12:31 am

I recently had 2 electrics set up by a respected tech in my area. However, as before they were intonated, playing a 3rd interval on the 5th and 4th strings, with distortion, gave a heavy motor-boat pulse. I took them back to the tech, and he showed me how the effect was far less noticeable with a clean sound, so he believes the distortion is creating a harmonic which is pulling the chord sharp. His recommendation was to use an EQ to notch out frequencies until the warble went away.
I spent some time doing that this evening, starting around 4k with a very tight Q, and had little positive result.
Does anyone have experience with trying to "tune out" harmonics form a distorted guitar sound? Any suggestions as to problem frequencies to look for, or setting to use?

Thanks!

User avatar
NoteBoat
Musically Insane
Posts: 5675
Joined: August 9th, 2003, 8:48 pm
Location: SW of Chicago
Contact:

Post by NoteBoat » August 12th, 2007, 3:50 am

First off, nothing you can do in the signal chain will "pull a chord sharp".

Second, your tech is right about distortion causing the problem, but wrong about the solution. Get some coffee, a full explanation will take a little bit.

What's happening is a conflict between the third that you're playing and the overtone brought out by the distortion. To illustrate, I'm going to use the pitches of the open 5th string and the third, C# - even though that's not available on the 4th string, the math works out easiest using A (the effect will be the same on any third, but with different frequencies... which is why EQ is not a solution)

When you hit the open A, the fundamental vibration is 110 times per second. The string is also vibrating at multiples of 110, but with less intensity - these are the Pythagorean harmonics: 220, 330, 440, 550, etc.

550 is a key number here.

You get those fundamentals on any plucked string, amplified or not, distorted or not, guitar or not. But we don't hear them, because they're weak. Just for illustration, let's say each harmonic is 1/2 the volume of the one before it:

110 = 100%
220 = 50%
330 = 25%
440 = 12.5%
550 = 6.25%

(Actual relative strength varies. To hear the actual strength, play the open string without amplification, and the 4th fret harmonic with the same pick force - that's the difference in intensity on your setup)

When you add distortion, you reach a point of "saturation" in a circuit. This is the point where a vacuum tube can't generate more output in relation to the input signal... it doesn't matter if your amp is tube or solid state, the concept here is going to be the same.

You reach saturation, and the 110 can't get any louder. But the overtones aren't at saturation - as you continue to increase power, they can still get louder. So your distorted sound may look like this:

110 = 100%
220 = 100%
330 = 100%
440 = 100%
550 = 75%

Now that 550 is loud enough to be heard - not as a separate tone, but as a prominent part of your distorted sound. Now look at the third you're playing:

C# = 554.37

The overtone series for C#will be 1108.74, 1663.11, 2217.48, 2771.85, etc. These overtones, plus higher ones from the A note, will also be amplified. But the big problem comes from the conflict between 550 and 554.37.

Signals can be thought of as additive. When signal A and signal B are both at their peak, with the same volume, the result will be A+B. If the value of each input sound is 1, the result is 2. This is equivalent to your speaker cone moving forward.

When both signals are at a wave trough, you get -1 + -1 = -2. Your speaker cone has moved backward.

But when one signal is +1 and the other is -1, you get 0... the speaker cone doesn't move, and you get no sound.

Your A550 is going +1 to -1 and back to +1 550 times each second. Your C# is doing the same thing... but slightly faster. Each second, it's doing 4.37 more full waves.

And that means 4.37 times each second, you'll have zero output. In between those times, the additive signal will be rising until it reaches 2 (or falling until it reaches -2), and you hear the "motorboat pulse". The speed of the pulse in my example is 4.37 pulses per second; with higher thirds it'll be faster. Classical musicians call this "beats" - as I said, they occur on all plucked strings, and violinists use beats between strings to do fine tuning.

You'll never be able to cure this with EQ. Your Q level won't be fine enough - you can't screen 554Hz and still leave 550. Even if you could kill the 554, you wouldn't want to - it's a note you're playing.

On top of that, the conflict will be at a different frequency for every third you play.

The solution: don't play thirds. Seriously. If you're using that much distortion, the solution is "power chords" - they're called power chords because they still sound ok at distorting power levels. You'll still have conflict: the 330 overtone from the A note won't exactly match the E, but the E fundamental will be at 329.63. That's a difference of only .27 vibrations per second, or about one beat every four seconds. The effect is there, but as a practical matter it isn't; you rarely hold one interval for four full seconds without a second pick attack, which starts the whole thing over.

EDIT: too early here when I wrote that... the fretted C# would be 138.59; it's fourth overtone is the 554.37 that creates the conflict with the fifth overtone of A.
Guitar teacher offering lessons in Plainfield IL

rnrsh
newbie
Posts: 2
Joined: August 12th, 2007, 12:24 am

Post by rnrsh » August 12th, 2007, 9:21 am

Thanks for the response, that's very thorough. I suppose I can rearrange the parts to have the cleaner guitar do the lick with the 3rd.

User avatar
jeffster1
Full Member
Posts: 231
Joined: June 27th, 2007, 5:13 pm

Re: Pulled out of tune by distortion?

Post by jeffster1 » November 24th, 2007, 2:38 pm

Noteboat, excuse my ignorance and thread hijack, but this is interesting to me.

Will the same effect be more prominent while playing through a compressor?

gnease
Musically Insane
Posts: 5523
Joined: March 2nd, 2004, 10:38 am

Re: Pulled out of tune by distortion?

Post by gnease » November 24th, 2007, 2:57 pm

Not in normal operation. A compressor works by essentially adjusting the gain of a linear amplifier to compensate for loudness changes in the audio. The harmonic generation Tom (Noteboat) describes is result of nonlinear processes. The only situations in which a properly-configured and properly-driven compressor's operation might be somewhat nonlinear is during the moments of the actual change of the gain -- the brief period during the attack (gain goes down) or the longer period of release (gain returns to "nominal"). The attack is usually a short duration process, so those nonlinearities will be brief and mostly unnoticeable. The release gain change is usually slow enough to result in a very low level of nonlinearity.

finndesolace
newbie
Posts: 1
Joined: May 9th, 2017, 1:15 am

Re: Pulled out of tune by distortion?

Post by finndesolace » May 9th, 2017, 1:18 am

Is there a digital distortion pedal that compensates this behaviour?

Retters
newbie
Posts: 1
Joined: November 25th, 2017, 2:03 am

Re: Pulled out of tune by distortion?

Post by Retters » November 29th, 2017, 11:33 pm

finndesolace wrote:
May 9th, 2017, 1:18 am
Is there a digital distortion pedal that compensates this behaviour?
No, I think you have to do it manually just with skill and practice.

Post Reply