suz_i_dil - 8-4-2015 at 12:54 PM
Hello
Make a while i'm getting interest on luthery, and wondering of how a soundboard works physically speaking.
As a player I noticed on several of my ouds that the main harmonic is a perfect fifth on the middle of the main soundholes. So wondered, what if we
study brace position on string notes ?
I don't know if it make sense physically but here is an exemple on the picture of Michael Cone Nahat that has been shared on this forum while ago and
analysed by FastForward member.
Here we go
If the download doesn't work the picture is here:
http://www.mikeouds.com/messageboard/viewthread.php?tid=10882
[file]36177[/file]
suz_i_dil - 8-4-2015 at 01:01 PM
taking simply wikipedia here the intervals of notes i get in ratio of stringlength:
fundamental : full stringlength
augmented unison: 24/25 of stringlength
minor 2nd: 15/16
major 2nd: 8/9 (major tone) or 9/10 (minor tone)
minor third: 5/6 or 16/19 or 27/32
major third: 4/5
diminished 4th: 25/32
perfect 4th: 3/4
augmented 4th or diminished 5th (tritone) : 18/25 or 25/36 or 32/45
perfect 5th: 2/3
augmented 5th: 16/25
major 6th: 3/5 or 7/12 or 16/27
minor 7th: 9/16
major 7th: 8/15 or 27/50
augmented 7th: 64/125
octave : 1/2 of stringlength
suz_i_dil - 8-4-2015 at 01:02 PM
of the Nahat exemple, taking a stringlength of 60 cm, that means those octave along the stringlength
60 cm: fundamental
30 cm: 1st octave
15 cm : 2nd octave
7.5 cm : 3rd octave
3.75 : 4th octave
suz_i_dil - 8-4-2015 at 01:05 PM
and here is what I get if I position tones on the bracing maping made by fast forward
for that I reversed the mapping to work on stringlength, meaning the bridge is the 0 cm position
accepting some imprecision from the way of mapping the bracing...isn't that surprising ?
suz_i_dil - 8-4-2015 at 01:12 PM
if the picture not good here is written:
0 cm: bridge
brace 1: 4.2 cm -> minor 7th of the 3rd octave at 4.21 cm
brace 2 : 7.3 cm -> 3rd octave at 7.5 cm
brace 3: 13.8 cm -> major 2nd of the 2nd octave at 13.3 cm
brace 4 :20.4 cm -> perfect fifth of the first octave at 20 cm
brace 5 : 26.9 cm-> major 2nd of the first octave at 26.7 cm
brace 6 29.8 cm -> 1st octave at 30 cm
brace 7: 32.8 cm -> major 7th of the fundamental at 32 cm / minor 7th at 33.75 cm
suz_i_dil - 8-4-2015 at 01:18 PM
I think we can take in consideration for imprecisions picture distorsion and the measurement of the middle of the brace.
I mean with this method of notes intervals we should, I think, determine precisely the position of the brace taking the border of the brace toward
its bridge size and not the middle of the brace. As would do the nut for the stringlenght. I mean we measure from the border of the nut and not the
middle of the nut.
Another point
I don't mean there is ONE receipe of course.
But what about bracing a soundboard on a choice of precise tones ? Let say I will brace my oud on a major 7th chord, or a minor 7th, or something more
complex...I don't know . What ever. Let the ear decide
Really looking to read your thoughts and feedbacks about this issue.
Brian Prunka - 8-4-2015 at 01:30 PM
Interesting idea. I'm not sure that the position of the brace relative to the string is meaningful, since it really has more to do with the way the
soundboard vibrates.
I'm also thinking that almost any brace location would correspond roughly to some harmonic or another.
It would be great if an experimental soundboard with movable braces allowed for testing of different hypotheses.
suz_i_dil - 8-4-2015 at 01:43 PM
Indeed Brian, that's exactly the problem. In any position a note. And always near from a precise interval.
But here octave appear 2 times...on 7 braces.
With a precision of 2 mm on a Nahat I cannot believe this is by chance.
Something like this that may be interesting because of precision. I mean if one decide to position a brace on an octave it would be a very precise
information. The octave not a millimeter more or less.
Your idea of an experimental soundboard is nice.
or a series of Abdo Nahat soundboard studies 
to see if this receipe is
consistent from one oud to another for this maker -
suz_i_dil - 8-5-2015 at 02:04 AM
I agree it's not enough. But being accurate on how you decide to position the brace is a beginning.
then come a work to make more or less flexible areas on the soundboard
let's say for exemple more flexible in bass areas and more rigidity in treble areas are the usual standards as far as I understood on guitar luthery
forums
Maybe some braces are a must have, like octave braces.
then others one maybe be decided on a matter of simetry depending on where we decide to make the main sound hole ?
Another track maybe the kind of tuning and take the tones on the tuning you wish.
for exemple my oud will be dedicated to
DO SOL LA re sol do tuning
so I will make octave (do) fifth (sol) sixth (la) 2nd (re) braces
or if I want to make the more equilibrate soundboard ever:
Octave 2nd 3rd 4th 5th 6th and 7th brace ? that's worth a try !
Does someone know what was the usual tuning for an Arabic oud in the time of Abdo Nahat ?
Brian Prunka - 8-5-2015 at 06:31 AM
I'm just saying it's not a foregone conclusion that the position of a brace relative to the node of the string is meaningful. It isn't clear to me
that the vibrational relationship between the top and the string is 1:1.
jdowning - 8-5-2015 at 02:17 PM
Its an interesting idea and there is much to be understood about all of the 'ins and outs' of oud or lute sound board acoustics - regardless of maker
or time period. So by all means test the idea through practical application on real instruments to see how it works out.
The old lute (and therefore presumably oud) makers worked on geometrical proportions for generating the instrument geometry including brace placement.
Early writers about music and instrument theory such as Marin Mersenne (1636) provide some detail about lute bracing proportions in their time
(16th/17th C) and surviving lutes from that period show evidence of proportional placement of braces dependant upon sound board geometry in the form
of scratch marks left on the underside of the sound board surface by dividers used to lay out brace positions. The famous engraving of Gaspar
Tieffenbrucker by Pierre Woeiriot, 1562, shows the renowned luthier holding the most important tool of his trade - not a chisel, plane or saw - but a
large pair of dividers essential for his instrument design and layout.
More recently there has been an experimental study on braced lute sound board top vibrations - Ian Firth, 'Acoustical Experiments on the Lute Belly',
Galpin Society Journal. I don't have the detailed reference immediately to hand but will look it up and post the information later for those who may
be interested.
I understand that there are also computer simulation programs that are used by some makers to predict sound board response for different bracing
configurations etc. How well these programs can model the real acoustic situation I am not in a position to say.
[file]36186[/file]
jdowning - 8-5-2015 at 03:19 PM
The article by Ian Firth was published in the Galpin Society Journal Volume 30 (May 1977) pages 56-63 'Acoustical Experiments on the Lute Belly'.
Articles from the Galpin Society Journal (and many other publications) are available on-line from the JSTOR organisation. Those who are not affiliated
with University Libraries etc. may have limited free access to articles by registering as a MyJSTOR member on the JSTOR Website. Registration is free
and allows members to read 6 articles a month on line. A great resource facility.
suz_i_dil - 8-6-2015 at 02:09 AM
Interesting. I'll try to put the hand on this article in the weeks coming.
I'll try to make a mapping of and oud soundboard, first using the commun lute position of the soundholes (a story of 3-4-5 triangle). Don't remind
exactly but I shall have this somewhere at home.
then another mapping positioning the rosette another place.
Actually, maybe that the first path is positionning the soundholes. Then making the bracing.
I say that because it is very commun standard that the brace around the main sound hole appear symmetric, around 5 mm ahead the main soundhole and the
other brace around 5 mm behind. As far as i remember i always see that on ouds.
An experimental soundboard is really something to think. I totally lack of competence to imagine how we may achieve something like that
jdowning - 8-6-2015 at 03:17 AM
You will probably find what you are looking for about the 3:4:5 triangle construction for sound board geometry - something that I proposed here nearly
5 years ago. I haven't read through the old thread recently but you will likely find some reference to bracing geometry as well.
http://www.mikeouds.com/messageboard/viewthread.php?tid=11186
The Ian Firth article describes his experimental sound boardl set up.
suz_i_dil - 8-6-2015 at 10:08 PM
Indeed , thanks !
suz_i_dil - 8-8-2015 at 02:03 AM
Hello
I give a try with the same method at the old egyptian oud from the thread you linked in your last message J. Downing. I mean the one for which you
mapped the bracing.
It would be interesting to know more about how accurate you think the maker was and to get an idea of its sound
results are:
for a stringlength of 622 mm
0 octave: 622 mm
1st octave : 311 mm
2nd octave 155.55 mm
3rd octave 77.75 mm
4th octave : 38.875 mm
suz_i_dil - 8-8-2015 at 02:12 AM
for bracing and nearest overtones:
bridge 0 mm position
brace 1 at 42.07 mm -> major 7th at 41.46 and minor 7th at 43.73 mm (3rd octave)
brace 2 at 85.6 mm -> minor 7th at 87.46 mm (2nd octave)
small soundholes 116.9 mm -> perfect 4th at 116.62 mm (2nd octave)
brace 3 at 142.2 mm -> major 2nd at 138.2 mm (2nd octave)
main soundhole 207 mm -> perfect fifth at 207 mm (1st octave)
brace 5 at 267.1 mm -> minor third at 259.2 mm or 262.4 mm (pythagor interval) (1st octave)
brace 6 at 327.9 mm -> major 7th at 331.7 mm ( 0 octave)
I took the measure from the border of the brace, toward the bridge.
jdowning - 8-8-2015 at 11:24 AM
I have yet to complete work on the sound board restoration.
With the braces removed it can be seen that the student luthier marked out and numbered the brace positions (both faces) in pencil. Brace #1 is
between the bridge and bottom of the bowl so presumably his reference point for locating all braces was the bottom of the bowl/sound board? There are
no visible scribe or prick marks so dividers may not have been used for the bracing layout?
Note that the neck appears to have been shortened by about 4 mm so this should be taken into account in determining original string length.