Official Luthiers Forum!

I'm taking a big step here with the koa 000 I'm working on. I'm trying to use Alan Carruth's method of comparing resonant pitches of the top and back to get them about the same before they're glued to the rim. If I understand correctly, the hope is that the addition of the bridge will leave the top about a semitone below the back. My problem is with figuring out which of the many pitches I'm getting from the top and back is to be considered the resonant one. They change with where I hold the plate, where I tap and even how hard. My left arm is gettin' real tired holding up the top or back for all this tapping that has me going around in circles. Sorely needed guidance appreciated!

_________________
now known around here as Pat Foster
_________________
http://www.patfosterguitars.com

I know Al likes to use Chladni patterns as a tool for his tuning. He sprinkles some glitter on the outside of the plate and compares the patterns created when a sound wave is applied using a stinger speaker.

Pat, if you check the archives, I know he's referenced this process before.

_________________
Joe Breault
Merrimack, NH
Perpetual novice

Thanks, Joe and Hesh.

Yeah, my family watches in silence out of the corner of their eyes while I'm doing this. Then they look at each other solemnly and leave the room.

I'm somewhat familiar with the use of Chladni method to determine where the plate flexes (or perhaps more accurately where it doesn't flex. If I'm correct, he uses these two steps (glitter patterns and plate resonance) in his building process separately, in a sense. The glitter patterns is something that I think I'm not quite ready to try to get my brain around yet. Maybe after a few more guitars.

But the top and back resonances, I think I can tackle at this point. I've only listened to tap tones to get a sense of whether the top is loose enough, and maybe not a very good sense, though my first sounds just fine (except for a bit of a wolf note). This would be the first time I've tried to hear a specific pitch, though I'm not working toward anything specific, so that I can get the top and back about the same before they're glued on. Since I was only listening for a deepening pitch before, as I carved braces, along with trying to get more sustain, I was only tapping in one place more or less and listening to one pitch. I never noticed just how many different pitches there are on a top. So far, my untrained ear has detected about eight, and I'll bet there are many more. What baffles me is, which of these is the resonant pitch?

_________________
now known around here as Pat Foster
_________________
http://www.patfosterguitars.com

I tap my plates for my tuner. It gives me a note and plus or minus 50 cents for that note. Then I mark that on the top and tap the top again once it is fastened to the rim. I'm just keeping records to see which guitars sound the best, and which top and back notes these guitars have. ANYTHING to get some objective feedback during the building process. I do other things as well, flexing, etc. but I like taking notes on the notes.

**DISCLAIMER** I do not necessarily know what I am talking about
I use chladni patterns to tune my classical guitars, but I do it in a different manner than described in various threads on OLF. I use my computer, a free program for creating various sine wave frequencies (http://www.nch.com.au/tonegen/index.html), a 20 watt amplifier, and a 6" speaker.

The only information that I am interested in getting (at this point)is the resonant frequency of the top. I tune the soundboard at various stages, but the ultimate goal is to find the resonant frequency of the top when the guitar is completely assembled (with the exception of the bridge). I believe that everything on the guitar will ultimately affect the frequency of the soundboard, so I try to incorporate these variables when sanding, scraping, etc.   

I sprinkle saw dust on the top and start at a low frequency (around 120hz or so) and start moving up in frequency until the sawdust begins to bounce. The sawdust will vibrate the most at a certain frequency (I aim for about 180-190hz), and will not vibrate as much at lower or higher frequencies. This is the resonant frequency of the top.

I try to tune the top so that the pitch is in between any two notes on the guitar (this will decrease the likelihood of resonant notes and will create a much more even sound).

There is a lot of information on the web about standing waves, resonant tones, forced resonances etc. that are very relevant (IMHO) to guitar building and these principles have guided my building to a large degree.

I am eager to learn about how these patterns can be manipulated to control various frequencies and ultimately get a "better" sound from the instrument. My purpose is much more simplistic, but it seems to be working somewhat.

In short, this seems to be a very easy way to find the resonant frequency of the soundboard.        

There are several issues here. For one thing, when you hold and tap, you will be getting all of the modes that can be driven from the tapping point and that are not moving too much at the holding point. There's a lot going on, it changes with every little shift of position of one thing or theother, and if you don't know what the modes look like you have little idea what you're hearing.

Another problem is that we aren't altogether sure how the resonant modes of the 'free' top relate to what the assembled instrument does. Once you have some experinece with a given design and certain types of wood you can make some ballpark predictions, but it changes with every shift in variables. And _everything_ is a variable.

FWIW, the pitches of the 'free' plate modes give you some idea of the overall relationship between the mass and stiffness. If the mass of your top is 'normal', and the modes are really low, then it may not be stiff enough.

The shapes of the modes tell you something about how the mass and stiffness are distributed. You can't seperate mass and stiffness out very well from just knowing the mode shapes, unless you're really good with computerized modal analysis.

I use the Chladni method in part because I never could make much sense out of the various systems of 'tap tuning'. However, any system you use will require some learning time. I think it's easier to communicate the Chladni data, but that's just my opinion.

Alan,

Thank you for your response. Do you wait until the plates are glued to the rim to determine whether the back is about the same pitch as the top before the bridge goes on?

_________________
now known around here as Pat Foster
_________________
http://www.patfosterguitars.com

I look at plate pitches at several stages.

When I'm tuning the 'free' plates, not glued to anything, I'm more interested in the shapes of the patterns than I am in the pitches. However, I know from experience, using _my_ designs and bracing systems and woods, about the ranges of pitches that will end up putting the assembled modes where I want them. YMMV, and that's where you will need to do some work to develop a data base.

I test the assembled modes as soon as I've got the box closed, before the binding routs get done. That way, if there's anything really flakey going on, I can work a little on edge thicknesses without making the bindings look funny. Again, the pitches do change when you rout for the bindings, when you glue them on, and when the bridge gets glued to the top. Thus, there's more data base development to do if you want to get things really closely under control, although it's possible that copying somebody else's numbers would get you close.

What realy counts is what the thing's doing when it's all together and strung. Knowing what you want that to be (and you might want something different than what I do) you have to work backward through the manufacturing steps, finding out what the changes are for each one, and factoring them in.

The biggest jump, and the one we really don't understand, is when you glue the free plates to the rim: as yet there's no way to predict exactly what will happen to the modes at that step. Again, after a while you get a feel for what _might_ happen, but sometimes the wood displays what my father-in-law calls the 'innate perversity of inanimate objects'. These surprises get fewer with experience, but they never cease entirely.

Aha! A light just went on after reading your last post. Now I see how you use the glitter patterns together with plate resonance frequencies. I hope I can keep that light on.

Thank you once again, Alan.

_________________
now known around here as Pat Foster
_________________
http://www.patfosterguitars.com