Official Luthiers Forum!

As a novice to the fine art of guitar building, I find myself at that juncture where I must start voicing the tops of my two OM's, which I'm building simultaneously. One is good grade sitka, and the other is top grade engelmann.

Problem is, I don't really know what I'm listening for. I know from this forum and the books I've read that this is a skill that's developed over years of building, but I'm curious if anyone has attempted to record tap tones of various body styles, at different stages in the voicing process, either as a database sort of reference, or (ideally) in mp3 form that could be shared over the internet?

Is there a way to post audio on this site? Seems like a great way to share the experience of all the great instruments built by the members here. After all, the photos are always amazing, but isn't the tone what it's really all about?

Ken

_________________
Mountain Song Guitars www.mountainsongguitars.com

AlexM posted some tap tones he recorded. you can find them Here.  I think he also discussed his methods.  Uses a Samson USB mic.

_________________
Waddy

Photobucket Build Album Library

Sound Clips of most of my guitars

There's a lot going on when you tap a guitar top, and how you hold it, where and how you tap, and where you listen will all change the sound you get. That's why so many of the people who use tap tones tend to disagree about what they are hearing.

It gets more difficult when you try to make recordings. Your ears and brain are really good at extracting information from a signal, but there is no microphone and computer setup that can do that nearly as well yet. It's quite possible to miss something very important by simply having the mic in the 'wrong' place.

All of this is one reason some of us have gone to looking at 'glitter patterns' and the like. It breraks the information down into something you can easily record and communicate.

I'm not saying that tap tone systems can't work; obviously they do, for a lot of people. It's just that everybody either seems to work out their own system, or else to learn how from some 'master' through hands' on experience and training. I always got hung up reading about those systems: what does it mean, a 'full, rich tone', or that 'the fundamental goes away'? At least I know what 275Hz is.

Cumpliano talks about this (in a couple paragraphs.) on his site

 - Ill be there in a couple months, so I dont know. I know I like the tap tone of my carpathian top, but I have no bracing on it, and Im not sure I want engleman braces now.

I build mandolins and usually use tapping to see if i am getting the whole
board to respond from various points in various ways. Depending on where
you hold it, you will get different modes. In a very non-technical way I am
searching for the board to vibrate in many different modes (evidenced by
different frequencies) when I hold it in different places.

It never made sense to me to aim for the fundamental to be in a certain
range and then call it good. That thing is going to produce sound over 3
octaves, its gotta be able to move in a lot of different ways.

I'm sure there are way better ways to tap tune, but that's how I approach it.
Deflection makes more sense to my head, but I can't quite determine why
yet.

also, freeware sometimes has FFT functions where you can record the tap,
and get a visual of the frequency peaks up to 20,000 Hz or so. I use
Audacity on my mac and its been pretty cool for visualizing.

For instance, when I tap the bridge of a mandolin I get this:
2007-08-18_234511_Picture_3.jpg
">
The peak at 276 Hz is the main body mode.

Then when i tap the back I get this:
2007-08-18_234557_Picture_5.jpg">
The peak is at 288 now. Why? I'm not sure, but maybe 400 instruments
from now i can tell you.

Thanks Waddy for the link -- I took a listen, and I do think there is something that can be learned from this type of recording. Granted, recording setup greatly affects the result, but in this day and age, even a usb mic can provide a good sonic image of a singular sound source.

Allan, could you point me to a source of info on glitter patterns? I've seen your mention of this in other threads, and I'd like to learn more about it. A few years ago I worked with an architecture student investigating distribution of powdered media over tensile surfaces in response to varying tonal frequencies. Pretty interesting stuff, both from an academic, as well as artistic standpoint. I wonder if those ideas are at all related.

Blackheart -- I haven't been to Cumpiano's site, though I have his book. I'll check out his site.

Once I get the second top braced, I'm going to try recording the taptone of both as I begin voicing them. Hopefully by referring back to the series of recordings I can learn something more about the process.

Thanks for the input!

Ken

_________________
Mountain Song Guitars www.mountainsongguitars.com

Ken wrote:
" A few years ago I worked with an architecture student investigating distribution of powdered media over tensile surfaces in response to varying tonal frequencies. Pretty interesting stuff, both from an academic, as well as artistic standpoint. I wonder if those ideas are at all related."

It's probably the same thing.

I wrote a series of articles for the GAL (American Lutherie magazine) some years ago that I tried to make fairly comprehensive. They're out in the 'Big Red Book' series. There's other stuff floating around, by various authors. It's a little hard to give a full list offhand. I'd bet there's some stuff on line, too.

gburghardt --

That's a terrific idea using audacity to analyze the spectrum. While my lack of experience would keep me from making any informed decisions, I'd be very interested to see how the frequency profile might change throughout the brace carving process. By creating a database of those changes, I could eventually go back, once the instrument has settled, knowing what is or is not working, and use that data to inform future voicing decisions. Good stuff!

Ken

_________________
Mountain Song Guitars www.mountainsongguitars.com

I may be wrong but I think the guitar will go 'boinggggggg' and what you want to do it make it do that with the most resonance you can get, so you keep taking off wood until it starts to go higher. This is determeined by the mass of the wood and glue. Backs are different.

 

If you have a good ear hopefully you will start to hear it. Try to make both tops sound the same, and then voice the first one and compare it to the second. Then you have a base to go from.

 

 What ever you do Im sure it will sound at least "real good".

The boing is going up in pitch when wood is removed. It becomes looser and responds well to lower frequencies.  For example a 3mm thick top roughly has a peak (stongest resonance) of 300 hz and a 2mm thick will can be as low as 100 Hz. 

_________________
Build log

E D I T: going down

_________________
Build log

Blackheart --

Great advice. I'm going to record both tonight, with the braces rough-shaped with band-sawn scallops in the x-braces. I'm curious to see which frequencies are most energetic using FFT analysis, per gburghardt's suggestion.

AlexM --

Do the frequencies you mention refer to braced tops, and if so, at what stage of voicing? I'm sure it varies greatly from one top to the next, but is there a general target to aim for? If tops of different thicknesses emphasize different frequencies, do we try to manipulate that by carving braces differently? If my sitka top is more ringy than the engelmann, will thinning the sitka's braces more result in similar frequencies? Sorry for the newbie questions, guys. I'm just hoping to improve my chances for success. Thanks a million.

Ken

_________________
Mountain Song Guitars www.mountainsongguitars.com

Ken, first, I am by no means experienced. I did read a lot and bugged a lot of ppl with my questions. But so far i only built 1 guitar and now working on my second. And classicals, a different animal.

Anyway, someone far more seasoned is helping me along, and the method is to aim for certain resonance peaks at every stage of the build.

Basically, a raw top (as most objects) will have a certain main resonance - subjected to a certain frequency it will vibrate a lot more.  A stiff top should have a higher peak than a floppy top, if they weight about the same.  A very stiff but also very heavy piece (like Sitka) will probably react about the same as a less stiff but light Engelmann.

As you thin it and it is losing stiffness, the resonance will go down. At this point the loss of weight does not play a big role.
If you go by ear, you try to change the sound of the top from that of a high pitched bongo to a nice resonating lower pitch thud.

Next phase is to brace it.  With the raw braces on, the main peak goes up again, but here mass plays a bigger role.  The top sounds again like a bongo for that matter.

You start carving the braces, again trying to lower the pitch and make it sound nice. This is far trickier than the simple raw top sanding, because as i said, with bracing, mass is very important. In classicals you can find a bridge patch, similar to that in SS but made of spruce. In most cases what it does to the resonance is nothing. Alan Carruth posted about this, and my tests resulted the same. In this case, the brace is adding some longitudinal stiffness pushing up the peak, while at the same time its mass is pushing the point down, ruling out each other.

I guess that for this exact reasoning a very heavy guitar top, like a Fleta can work.  The plate is thick, the braces are many and strong. Against all immediate reason, it can work. Why? I think, because it simply has just enough mass to it to keep the response down and not sound like the above mentioned bongo.

I have tried today to subject my tap tone samples to FFT in Audacity and there does not seem to be a direct relation to the resonance testing.  It is all to variable and different, I mean the effect of a knock compared to the effect of 20W of clean sinus. :)

_________________
Build log