Official Luthiers Forum!

Disclaimer: I'm attempting to build an active back guitar as described in Trevor's books but didn't take the time to analyze the top & back before thicknessing them. I had a reject Martin top & a Sapele back & side set that were already thicknessed.

So, the box is closed & bound & I've recorded some taps.

I recorded the top then the back with the sound hole plugged by a 3/4" thick piece of styrofoam. I laid the guitar on the bench with a sleeping bag under it to damp out the side I wasn't recording.

Here's the top

Here's the back

And here's the whole guitar - tapped the top, hole un-plugged, top & back free to resonate

First question - the top & back spectrum plots look very similar, does it look like I got accurate recordings?

Assuming the recordings are accurate, the top & back are only 4 Hz apart. Per Trevor's book, we want at least 3.5 semi-tones separation or approx 40 Hz.

I can stiffen the back & pick up approx 16 Hz and adding mass to the sides can drop the top frequency but will I be able to get anywhere near 40 Hz of separation?

Perhaps if I had spent some time thicknessing the top & back I wouldn't be in this situation, lesson learned perhaps?

Any help or insights would be appreciated.

Thanks.
Kevin Looker

_________________
I'm not a luthier.
I'm just a guy who builds guitars in his basement.
It's better than playing golf.

I don't have the answers but I have almost exactly the same questions. I've got a falcate braced dred with live back that I just finished but did not do the frequency analysis to determine the top and back thicknesses. I tried but didn't get good data and needed to move on.

I had to get a better mic and am now getting to where I think I'm getting accurate and repeatable spectral data so I'll be looking to see what answers you get here.

_________________
Steve Smith
"Music is what feelings sound like"

Kevin, do you have the bridge glued in place? If not, that will probably lower the top frequency (at least change it). You might want the bridge in place before making adjustments.

_________________
Formerly known as Adaboy.......

When Trevor sates his four semitone rule I believe he's measuring how the top is coupling with the back, he's not tapping the back. I guessing the 223 Hz measure is what trevor would read as the T1,1-3 mode. Maybe he'll be along and straighten us all out.
My tops typically drop about 10 Hz after I glue the bridge on, for whatever that's worth to you.

_________________
Jim Watts
http://jameswattsguitars.com

I forgot to mention the bridge.

It's on but just with double sided tape. I can easily take it back off & record the top again.

Kevin Looker

_________________
I'm not a luthier.
I'm just a guy who builds guitars in his basement.
It's better than playing golf.

The coupled frequencies are what I mainly look at. The main air is 104.. coupled top is 204..coupled back is 223. Depending on the size of the guitar, these could be good, if the bridge has not been glued on. If it were a OM size, I would like the frequencies a little lower.

Chuck

_________________
_________________

It's an OM.

Kevin Looker

_________________
I'm not a luthier.
I'm just a guy who builds guitars in his basement.
It's better than playing golf.

Trevor and I measure these things differently, and have different targets as a result. I strongly suspect that we are doing pretty much the same things, but just talking about them differently. So:

I test mine the way Fred Dickens taught me to: by tapping on the top and back with the hole (more or less) plugged. This (more or less) cuts down the coupling between the top and back, and gives you some semblance of their actual resonant pitches (more or less).

The closer the 'real' top and back resonant pitches (which you cannot actually see in the assembled box) are to each other, the more strongly they will couple. This will have the greatest effect on the pitch and activity of the 'main air' and 'main top' modes. As usual, you're looking for a balance here; too much invites 'wolf' problems, but you do want strong coupling: I find a semitone difference is about as close as I want. You also have to consider what happens when you string it up and play the guitar in. On very lightly built guitars the stress of stringing can drop the 'main top' pitch and raise the 'main back'. Basically, the back is being 'stretched' by the neck rotation, which acts on it like tightening a drum head, and the top is undergoing the opposite stress. Tis is not a big thing on most guitars, but sometimes... On my guitars I typically find that a month's worth of playing will 'loosen up' the top enough to drop the 'main top 'pitch by about half a semitone. Also, of course, if you're doing your tap tests on a box with no bridge, you have to think about what will happen when you glue on that massive brace. On my guitars, the 'top' pitch usually drops by about 1/2 semitone; YMMV.

So, all things considered, if the assembled box with no bridge has the 'top' and 'back' tap tones at the same pitch when the hole is blocked, I figure I'm in the right range. I don't want to see the top higher in pitch by very much, since almost everything I do is going to drop that pitch, and I'm trying to avoid an exact match. Still if the top pitch is only 1/2 semitone below the back with the bridge on, I expect to get a little 'bark' in the low range (typically at G on the low E string) at first, which plays out relatively quickly. This is a good time to think about bridge mass.

Again, Trevor holds and taps his guitars differently, and that could account for some difference in the result. Also, he's making his sound, and I'm making mine. We might like different things, and who's to say which is 'better'? If you're following his system, though, you should try for his numbers, especially the first time.

If you are using Trevor's method, the last of your tests is the only relevant one (Soundhole open, tapping on top, back undamped)
If you are doing a falcate braced top, the stiffness of the bridge makes a major contribution so it needs to be glued on rather than just taped on.
But don't do any adjustments yet, these measurements will move around quite a lot, mostly down, with final sanding and finishing.
You need to finish the guitar complete with neck, any pickguard, saddle ,strings and pins installed and tuned up before you make plans for any adjustments.
In progress measurements are interesting, but certainly not the basis for any adjustment.

who is "Trevor"?

http://sempleguitars.com/themaker_18.php

this guy?

Trevor Gore
http://www.goreguitars.com.au/main/page_home.html

Nope. This guy: http://www.goreguitars.com.au/

Kevin, FWIW, the most recent OM size that I finished ( actually a Norman Blake style 000), had the following coupled frequencies (with the box closed and bound and no bridge). Main air; 99 Hz......Main top; 193 Hz...... Main Back......217.Hz. The finished guitar with strings on changed to; 98 HZ.... 189 HZ....211 HZ respectively. This guitar sounds really good, according to the guys at the shop selling my guitars. You are not that far away, get the frequencies without the bridge taped on.

Chuck

_________________
_________________

The 4 semitone rule applies to the separation between the T(1,1)2, "top" peak and the T(1,1)3 "back" peak, which are the second and third peaks seen on a live back guitar (the first, the T(1,1) is the main air). The T(1,1)3 peak is absent on non-live back guitars, hence the shorthand for its name. The measurements are taken by tapping the guitar, holding it as you would when playing it, much as Jeff described above.

So taking Kevin's last plot, which was taken more or less that way, the T(1,1)1 (main air) is at 104Hz, the T(1,1)2 (top) is at 204Hz and the T(1,1)3 (back) is at 223Hz. I'm assuming you have a X-braced guitar, which is already stiff across the grain, so the bridge acts mainly as mass (but that it not so with a falcate braced guitar). In this situation, taping the bridge on may give you an indicative result. With the figures you have I would be aiming for a 180Hz T(1,1)2, leaving the back where it is. As you bring the top down in frequency, the back will follow a bit, but not by much. Remember you still need to finish sand, apply finish etc. and when I do that not a whole lot changes until I put the bridge on, but I don't know how a guitar will respond under your finish/application method. So I would suggest using the techniques given in the book for bringing the top down a bit in frequency (which will also increase the monopole mobility) with a view to having the finished, strung guitar with a T(1,1)2 a little higher than the 180Hz target, say 185Hz, then fine tune using side masses to finish where you want. I'd strongly recommend keeping the bridge mass below 20 grms.

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

http://www.goreguitars.com.au

Thanks everyone for your replies.

It's greatly appreciated.

Kevin Looker

_________________
I'm not a luthier.
I'm just a guy who builds guitars in his basement.
It's better than playing golf.