Official Luthiers Forum!

I have a guitar that just got it's strings.
OM size

According to Audacity, the top rings at 106, 181 and 215
The back rings at 106, 213 and 293
Plots are attached to this post.

I built this with the back bracing suggested in the Gore Gillet Books with the four back braces in an x configuration pointing at the center of a scalloped lower brace.

I've built two guitars with this bracing and notice that the Low E string doesn't ring as freely as the rest of the strings. It just doesn't have the overtones that the other strings do.

What should I carve and what target should I set to open the tone up for the low E string?
This is new territory for me. Please help if you can.

Thanks,

dave

Since you got no other takers…

First thing I’d recommend is let it sit for a bit before trying to change anything. Your air at 106 seems pretty high for an OM sized guitar. The two OM’s I have on hand are around 90hz, but I build my OM’s at 4 1/12” at the tail.

Your top at 181 is pretty close to Gore’s suggested number, but you could probably scrape another hz off. Your back at 215 is about a half step lower than Gore’s ideal for a 180hz top, but, that’s all quite subjective to taste.

Before modifying anything, I’d also want to measure top mobility and bridge rotation. If the top is still pretty stiff (it’s possible to get the same freqs in a wide array of stiffness, so the actual freqs are guidelines only), you may get away with taking the top down to 170, which would have a distinct effect on how ‘bassy’ the tone would be and would also give the 4 recommended semitones between top and back, which again is just a suggestion.

YMMV though. Measuring your mode freqs is a great reference in keeping your own work consistent, but, since there’s so many different ways to achieve those numbers, how they reflect in one persons work may not reflect how they correspond to another’s, especially when you add in subjective taste.

Are you taking those 'top' and 'back' pitches with the hole closed? When it's open they push each other apart in pitch, so you don't get the 'real' pitches.

Those were with the sound hole open.
As near as I can tell, those two modes sit right apart what I'd always used as the sweet spot before trying the Gore/Gillet back bracing.
Pretty much nailed that semitone between modes.

I realize the G/G books say to shoot for 4 semitones. But if it ain't broke.....

I've got another one, with the same issue, dead-ish E string. Just for fun, I measured that one today as well. The top and back sit right at the 4 semi-tones apart. 191-242hz.

So I'm a little stumped.

Maybe I should stiffen the back of the first one and lighten up the back of the 2nd. Don't yet know.

Here it is with sound hold closed

It's been on the tonerite for 48 hours and I have to admit, it sounds better already.

(by tonerite, I mean an aquarium pump sitting right on the bridge so is make a heckuva racket. It's like being played constantly.)


thanks for the tips

Also FWIW, I do my measurements when the guitar is strung up and working as a system. Not sure of the value of measuring them independently when they will be in use as a system.

I test all the way through the process.
Have been using Al's Free plate tuning for many years, with much success.

I've always tried to get the top and back to within a semitone. So this 4 semitones thing is weird for me. As is the back bracing scheme. It seems the spruce I use for braces doesn't work the way the Aussie's bracing stock does and I need to make the scallop a bit shallower if I use this method again.

What I've learned so far is that testing and carving after the string up does make an incredible difference and Small changes yield big results.

The Cedar/EIR OM went from Meh, to OMG with just a little carving. Turns out I DIDN'T use the G/G back brace scheme. It was more like a Martin scheme with 4 transvers back braces. The two upper being 1/4" the two lower being wider and shorter.
I shaved a little off the lower braces and brought it to within 4 semitones. The E string opened up and the 3rd fret G, which had 5 seconds of sustain with a sharp drop off, Now has over 10 seconds of sustain, all gradually fading out. I'm happy with that.
As the back dropped in pitch, so did the top. That was interesting to me. I hadn't observed that before.

The other guitar, the one from the OP, I determined was too close in pitch to the top. I flattened out the scallop and put a 3mm cap on it.
That raised the pitch enough that the sustain on the 6th string improved greatly.

So, learning experience.
Mission accomplished

One has to be very specific on measurement technique when quoting resonance numbers for the "top" and the "back".

Believe it or not, the 1 semitone separation and the 4 semitone separation can be almost the same thing. How the...???

It you tap the top, with the guitar in playing state and position, and mic the top you will be able to measure the T(1,1)1, the T(1,1)2 and the T(1,1)3, if it exists. The T(1,1)3 is an expression of back activity measured in the top, as part of a fully coupled system. Say the T(1,1)3 turns out to be 4 semitones higher than the T(1,1)2.

Now flip the guitar, tap the back and mic the back, you will be measuring the B (for back) B(1,1).

The difference is that the system has changed, because you are now "driving" the back rather than the top, so the level of coupling changes (amongst other things). The result is that you will likely measure a "back" frequency that is somewhat lower than the T(1,1)3 frequency and hence closer to the T(1,1)2 frequency.

Al C's measurements and recommendation of 1 semitone is between the (coupled) T(1,1)2 frequency (I don't think he plugs the sound hole to uncouple it; Al??) and the B(1,1).

My recommendation is 4 semitones between the T(1,1)2 and the T(1,1)3. These different results can be obtained on the same guitar, and that's why 1 semitone can be (almost) the same as 4 semitones, depending on which measurement method you used.

If the T(1,1)3 gets closer to the T(1,1)2 than ~2.5 semitones, you really start to get a muddled response with loss of volume and definition. If you go out to ~5 semitones separation, you begin to lose the live back effect. That's why 4 turns out to be about the right number for an effective live back.

If you want to kill the bass response of a guitar, raise the T(1,1)1 frequency and drop the T(1,1)3 frequency to ~1 semitone above the T(1,1)2. I'm sure you can figure out how to improve the bass response.

It's not the only thing that could be done to improve bass response, but the above is a good base to start from.

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

http://www.goreguitars.com.au

Thanks Trevor.

As with all responses that I used to get from Al, this one causes my eyes to roll back in my head and my body starts to convulse a little.
It'll take me a while to decipher what you said as it creates more questions than answers.

It goes back to Plato's statement "you don't know what you don't know

Like, Can you see the B(1,1) in a spectral analysis of T1,1?
It is all a connected mechanism. So it stands to reason that the numbers are there, and one just doesn't know how to interpret them.

How does one manipulate the T(1,1)2 and T(1,1)3 modes?

Gotta dig those books out and see if there are some sample graphs to help decipher all this

thanks
dl

The main control for the T(1,1)1 is the sound hole area and the volume of the box, for T(1,1)2 the top braces, for the T(1,1)3 the back braces. On the back brace design you've used, you would shave the horizontal center bar to adjust the T(1,1)3/B(1,1).

There's a fun demonstration of how these are all connected (coupled) here. You can play with the sliders and see how the frequencies move. It should give you an intuitive understanding of how a guitar works.

Greg