Do you build your tops to a certain pitch or body resonance?

[Alan Carruth]

I really liked the Gore/Gilet books, but take all such stuff with a grain of salt (as anybody should with any of my pronouncements). I have not heard any of their guitars, so I can't say what sort of sound they're going for, or how that compares with what I like.

What I've found in my building is that I tend to like the 'back' and 'top' resonances fairly close to each other. Having the back a semitone above the top when the guitar has had some time to settle in gives the strongest practical coupling between the two, yielding a lower pitched and stronger 'air' resonance without risk of 'wolf' tones. The higher pitched back might well help with the high end sound, but I don't seem to need it: I believe (but can't 'prove') that's related to the 'free plate tuning'method that I use on the tops, which seems to help high-end response.

The whole point is that any such prescriptions must be related to the specific system that the builder uses, and my not be so generally applicable as they might seem.

OTOH, _if_ you accept the notion that all good guitars have a lot in common, then we must all be trying to do the same things with our different systems. If that's the case, then it would hardly be surprising to find that we would tend to converge. In that case, it might well work out that a guitar that was built using one system would turn out to have many of the desirable markers of another. An example of that would be the guitar I sent out to David Hurd to measure, using his deflection mapping system. The top had been 'well tuned' by the 'free plate' method, and also had the sort of stiffness distribution that Dave likes to see. If that's the case, then the difference between my 'semitone' and G/G's 'four semitone' recommendations could be mostly a matter of taste. They can both work.

[Jeff Highland]

Jeff, I/m still reading the gore/gilet book .What meter or tester do you use to measure the live or inactive back frequency with ??thanks.

Vitual Analyser, a drum microphne with a usb conversion cable amd a stick with a rubber eraser to tap
Tapping is not done directly on the back, only on the top and the coupled back response is found on the resultant graph

I use the settings sugested by Trevor on this page http://www.goreguitars.com.au/attachmen ... ata_R1.pdf

[Trevor Gore]

What I've found in my building is that I tend to like the 'back' and 'top' resonances fairly close to each other. Having the back a semitone above the top when the guitar has had some time to settle in gives the strongest practical coupling between the two, yielding a lower pitched and stronger 'air' resonance without risk of 'wolf' tones....

...If that's the case, then the difference between my 'semitone' and G/G's 'four semitone' recommendations could be mostly a matter of taste. They can both work.

I strongly encourage people to define how they measure things when discussing pitch differences between top and back. My 4 semitones separation is the separation between the T(1,1)2 peak and the T(1,1)3 peak when looking at the frequency response of a tap tested top on a completed guitar. It is that separation that matters because that is what you hear when the guitar is played. It is a different measurement from tapping the top and measuring its frequency then tapping the back and measuring the back's frequency. However, strange as this may sound, Al's one semitone difference and my 4 semitones are a lot closer than you might think. 3 or 4 semitones between the T(1,1)2 and the T(1,1)3, depending on the strength of the coupling, can be much the same as 1 semitone between the tapped top compared to the tapped back. The reason I use the measurement method I use is that it incorporates the degree of coupling between top and back that you hear in the finished instrument. And, as you should all know by now(!) coupled resonances "repel" each other in frequency.

[senunkan]

Trevor,

Just a question regarding the T(1,1)2 peak and the T(1,1)3 peak, what would the result if the peaks are closer to each other say about 2 semi-tones instead of 3 o 4?

I measure my guitar (active back) and the peaks are T(1,1)2 peak: 195 Hz G2 and T(1,1)3 peak: 225 Hz A3
If I need to raise the T(1,1)3 do I simply glue a piece of wood to the center scallop in the 3rd back brace (@ lower bout) as what you have mentioned in your book?
Thanks!

Sen

[Trevor Gore]

Sen, If you made that plot by tapping the top (bridge on, saddle in), you may have some other issues to consider. It looks like your peak at 225Hz is higher (greater in magnitude) than the one at 195Hz, which suggests to me that the back is pitched lower than the top, (i.e. the peak at 195Hz is the T(1,1)3) with the peak at 225Hz being the T(1,1)2) and that situation results in a sound that I don't generally like (but you may...). You can check whether that is the case by tapping the back and measuring its main frequency. If the result of the back tap is a main peak at ~200Hz, then the back is pitched lower than the top. If I had that situation, I would be stiffening up the back by capping the main transverse brace in the lower bout.

If you do that, you will see (and hear) a few interesting things. With the back now stiffer than the top, due to the coupling repulsion effect now acting "the other way round", the 225Hz peak (what I think is the T(1,1)2) will "jump" down to ~200Hz and the "back" peak (T(1,1)3) will appear at some higher frequency depending on how much you stiffened it and the second peak rather than the third peak will be the highest and will now be the true T(1,1)2. You should then be back in the "normal" zone (or, at least, how I like to build my guitars) and you should get much better volume and projection than you had before. If this sounds a bit cryptic, read the relevant sections of the book and all should become clear.

Let us know how it goes!

[Alexandru Marian]

Sen's guitar is a small size Torres build. I would look at around 250 Hz for the back resonance. Try to get a graph by tapping directly on the lower bout of the back.

[senunkan]

Yes like what Alex has mentioned mine is a small body Torres size (LB width 360mm).
I believed the back is indeed lower than the top.
But first let me get more data 1st.

Thanks Trevor and Alex.

[senunkan]

I measured the main back response and it is indeed the third peak at 220 Hz
With the strings and saddle on the freq response for T(1,1)2 191 Hz and T(1,1)3 220Hz looks about the same level.
T(1,1)1 103 Hz

[Alexandru Marian]

I am really surprised the back goes so low, considering the small size of the guitar, back thickness 2.2mm, and hefty braces (Sen told me they are about 15mm tall and 10 wide!) Compared to mine I would have expected at least 260 Hz, even more. It is also interesting that the back shows so clearly on the soundboard tap too. On mines the top mode always shadows the back.

[DennisK]

Sen, how are the ends of your back braces shaped? Tall and notched into the linings, scalloped to nothing, or something else? And is it significantly domed, or relatively flat? My small Torres style (425mm body length, 285mm lower bout width) appears to have resonances at 112Hz (A2) for the air, 212Hz (G#3) for the top, and 258Hz (C4) for the back.

Bridge.png

Those are taken while holding the guitar up by the neck, strings off, and tapping the bass wing of the bridge 3 times with my index finger. I also did it tapping the middle of the back, and the peaks came out within 1Hz of the same. There's also that 4th peak with a sort of sweep up to it, which I'm not sure what is. Cross dipole, perhaps? Would make sense with me tapping the wing. it's at 394Hz. When tapping the back, there's a 4th peak at 344Hz, but that's probably some unrelated higher mode of the back.

With strings on, all the peaks drop by 3-4Hz, except the 4th peak which goes to 382Hz. The back's 4th peak is unchanged.

So, I'm not sure what it all means exactly, but the guitar sounds great. It only has two back braces. One at the upper bout, which is structural only and very rigid, so really only one involved in the vibrating portion of the back, which is fairly dense redwood, about 4-5mm wide, 15mm high in the center, sweeping down to 3mm at the linings. The back is Honduran rosewood, approximately 2mm thick, although I didn't actually measure it. Slightly domed for humidity tolerance, but not enough to stiffen it very much.

BackBrace.jpg

[Alexandru Marian]

That's quite a nice HRW set!!!!

[Jeff Highland]

Hi Sen, I'm not going to comment on your spectrographic results other than to say-
-tapping only on the bass wing of the bridge may not be exiting the monopole sufficiently
-using your index finger to tap is not the best

It may be the perspective of the photo, but it appears that there is a large expanse of unsupported panel beyond the lower bout brace, which is probably doing whatever it wants in terms of vibration.

[Trevor Gore]

Sen, the T(1,1)2 and T(1,1)3 are rather close and show the classical "split peak" characteristic which gives a sound that I don't particularly like on the guitars with this characteristic that I have heard. The puzzle, though, seems to be why the T(1,1)3 is so low given what others have said about the back's construction. My experience has been more in line with what Alexandru has seen.

If the diagnosis is correct so far (hard to tell using different apps and different techniques than I use) then I would be inclined to stiffen the back to satisfy my sonic tastes.

[Alexandru Marian]

On tapping, I am giving it a smack with the knuckle of the index. Need to move the thing not caress it.

Sen, first stop thinking about graphs too much and fret it, see how it sounds. That's all that matters in the end

[senunkan]

Yeah I think I will fret it up and see how it sounds before I tune further.

DennisK,
Regarding the height of the LB back brace; it started pretty tall about 15mm but I think plane it down to about 12 mm or so (I didn't measure this part)
Center scallop is about 7-8 mm tall
The end scallop is about 5mm tall (1/2 of the lining height)
The doming is pretty significant about 15' or less and length wise even more I believe.
I'll measure it and let you know again.

Although my body is a small, I make the depth it deeper than usual about 100 mm or so at the tail end.
The heel is about 70mm due to the "dropped top" elevated fretboard where the top curved down.
This is to compensate for the loss of volume due to the "dropped top" elevated fretboard.

Here is a pic of the back before I closed it up.


Here you can see how wide the brace were.


Jeff,
As for exciting the top, what do you use to knock on the top?
Also regarding unsupported back do you mean the rim is not fitted as tightly as it should?
That could explain why it free to vibrate so much.