Official Luthiers Forum!

I have a D-28 with a resonance problem. When I pluck the open G-string, there is a bright and strong overtone at pretty much exactly 2165 Hz. I can hear it when I mute the G-string and play the 5th fret on the D-string as well, though not quite as clearly. It is really intense. Any idea what this might be?

Thanks,
Sondre

No idea what it is, but I get that from time to time as well. Different freq mind you...

That's the 11th partial of the open G string. (2165/11=196.8 which is G3, 7 cents sharp. You would expect an upper partial to be a bit sharp.) So some part of your guitar must have a particularly strong resonance right at this frequency. You can go hunting for it with a bit of modeling clay or blue tack. Up that high, I think it could be almost anywhere: top, back, side, neck. Maybe Trevor or Alan could suggest some likely spots.

Thanks Eric. I have tried muting some arbitrary areas of the guitar with my finger. No luck so far

Conveniently, the cooling fan in my laptop is producing exactly same pitch right now. Maybe I'm just going bonkers

A piece of tubing can be useful when trying to find a localized, high frequency noise source. Also, remember that it could be an air cavity resonance as well.

Have you tried muting the strings above the nut?

_________________
John

If it's loud enough to annoy you, and it's coming from the box, you should be able to see it on a spectrogram and using Chladni patterns. The size of the nodal areas will be getting down toward the size of the gaps between the braces. But you'll need to Chladni it to figure out which areas to work on.

On the other hand, JA may well be on the money and that's a pretty simple fix.

_________________
Trevor Gore, Luthier. Australian hand made acoustic guitars, classical guitars; custom guitar design and build; guitar design instruction.

http://www.goreguitars.com.au

Being an amateur and a newbie, I'm really curious, how did you determine that you had an overtone at "pretty much exactly 2165 Hz."? I can imagine being able to detect an overtone around a particular string or position on the fretboard (and therefore the note), but the fact that you specified the exact frequency implies to me that you used something to measure the problem "area". Are you using a tuner that displays the frequency? Did you detect the overtone by ear or was that also measured?

Thanks...

I just played it through a spectrum analyzer, and I can see a slight peak around that frequency. If I understand you correctly, what I need to do is e.g. stiffen up the antinode areas that show up in the Chladni pattern when the top is driven at 2165 Hz. Is that correct-ish?

I have tried muting the strings above the nut, that's not where the problem lies, unfortunately.


I can hear the overtone clearly, so I just used a sine tone generator to help determine the frequency by ear.

Thanks,
Sondre

The idea is to use Chladni patterns to find the problematic mode, then shift its frequency. The frequency shifting is done by altering the mode's effective mass or its effective stiffness, so you can shift up or down in frequency, doesn't matter which, so long as you get the frequency off the string harmonic frequency. The usual trick is to mass load one or more of the anti-nodes of the troublesome mode, which will then drop the frequency of that mode. Generally, not much mass is required and the skill is in finding a place to stick the mass which has the effect you want, but doesn't "kill" any good stuff that happens to be going on. Changing stiffness can be done by gluing on more wood, but you need to be sure you're adding stiffness rather than just mass. At the frequencies you're looking at, I suspect that local plate thinning (to reduce stiffness) might be more effective than shaving braces, if you choose not to add mass.

_________________
Trevor Gore, Luthier. Australian hand made acoustic guitars, classical guitars; custom guitar design and build; guitar design instruction.

http://www.goreguitars.com.au

Thanks a bunch! PS. Your books are awesome