Mike's Oud Forums - Powered by XMB

Designing a Golden Mean based oud

Pages: 1 2

ALAMI - 8-15-2007 at 10:24 AM

Let me start by saying that I am not an amateur of all that Golden Mean / Fibonacci serie esoteric mania, but I agree with the assertion that there is an aesthetic side related to the golden proportion (some call it the divine proportion)and the beauty of the underlying math.
It all started with a very interesting article I read recently :
The Foundations of Scientific Musical Tuning by Jonathan Tennenbaum:

http://www.schillerinstitute.org/fid_91-96/fid_911_jbt_tune.html

This made me go back to an old book I have discussing the Golden Mean and I was struck by the Fibonacci spiral, I saw it as a oud bowl and I was wondering if oudmakers have used the golden calculation in their oud designs and then I remembered an old thread on these forums about a mysterious sign on some of Manol's ouds.

http://www.arabicouds.com/messageboard/viewthread.php?tid=5788&...

I think that it is a Greek "PHI" and it means (just a theory) that the oud with a phi is designed according to the Golden Ratio. PHI is the symbol of the Golden Ratio (le nombre d'or) and it is attributed to PHIDIAS who designed the parthenon.

That's how I decided to start a new oud design based on the Golden Mean.

All I know about oud making is limited to what I read here (the projects of Jameel, Riad, Jonathan and the others) I also have Dr Oud book and I have talked with some lebanese oudmakers (Nazih Ghadban and Fadi Matta)
With this kind of "virtual" knowledge it is safer to start with "virtual" oudmaking, kind of a "digital" oud project that I hape it will end with a REAL oud:

First step the concept:

ALAMI - 8-15-2007 at 10:31 AM

Step two was the actual 3D virtual making.
I started with the overall proportion obeying the rule of x=y+(x-y) and using the golden spiral as base for the bowl and face shapes.
Here an amazing thing happened, everything was directly falling in place the oud seems to be "naturally" in compliance with the golden proportion:
The images will illustrate better what I am saying

ALAMI - 8-15-2007 at 10:34 AM

Those spirals did the job, I just love the harmony in this image:

ALAMI - 8-15-2007 at 10:38 AM

Now time to see the volume, I started with a single hole version, but as you'll see later I have the position and the size of the 2 small holes, I don't know which is better and in which way

ALAMI - 8-15-2007 at 10:46 AM

This is the initial bracing.
Only for this I had to go to the equations of the of the Fibonacci series I am not sure if it is final or if it is enough to have only 7 braces, please comment on this, ideas and suggestions are most welcomed.
I really want to bring this oud to reality and once ready I think that I will work it out with a professional luthier,.
Work is still in process, will keep updates

One Last Note: Sorry for the big images that are making this thread slow but those thin lines were suffering a lot when reduced and compressed

Ararat66 - 8-15-2007 at 11:02 AM

Hello Alami

I really enjoy your enquiring mind - this is a lovely post. I've looked into Golden sections, Phi, Fibannacci etc in relation to architectural spaces and have wondered the same about the oud. One thing though, the oud is asymetric in terms of the pitch and tension of strings so there may well be another variable here to complement this.

I've noticed that the harmonics on my Turkish oud correspond to lines passing the centres and edges of sound holes and also the ogee of the neck to the sound board. There is a slight flatness in the curve of the back when seen in profile which I have seen in pictures of many other ouds. The geometry is fascinating and I would love more time to look at this (I often try to figure it out visually).

Best wishes

Leon

Microber - 8-15-2007 at 12:24 PM

Very interresting project.
Good luck.

Ronny Andersson - 8-15-2007 at 12:38 PM

Looks very much as a theoretical lute by Arnault de Zwolle

and some here David van Edwards

mourad_X - 8-15-2007 at 02:51 PM

good work alami
and very interesting thread

/Users/Mourad/Desktop/logoleoperga.jpg

what 3d programm are you using for this exambles

regards
mourad

mourad_X - 8-15-2007 at 02:53 PM

good work alami
and very interesting thread

what 3d programm are you using for this exambles

regards
mourad

Jameel - 8-15-2007 at 03:27 PM

Very cool drawings Alami. I'll have to take some time and look at these closer. This looks like it could be a very interesting thread...

jdowning - 8-16-2007 at 09:45 AM

Interesting project and nice graphics ALAMI - good luck.

The Tennenbaum article presents a viewpoint on musical tones and pitch, bel canto singing, 'music of the spheres' etc.
The author discredits the great monumental work by Hermann Helmholz on the sensations of tone claiming that every essential assertion made by Helmholz has been proven to be false but without much further elaboration.
For those interested in reading further about the Helmholz theories and forming their own opinions a modern, low cost reprint of his work "On the Sensations of Tone", by Dover publications translated into English by Alexander Ellis, including detailed additions by the translator, is a comprehensive reference book of great interest and historical value. It also includes a section on Arabic and Persian musical systems.
For those interested in exploring the spiritual and philosophical dimensions of music "The Harmony of the Spheres - a sourcebook of the Pythagorean Tradition in Music' by Joscelyn Godwin traces the history of the idea of a harmonious universe from Plato through Kepler and Newton to modern times.
All fascinating stuff (if one has the time to to read into it!)

ALAMI - 8-17-2007 at 11:57 AM

thank you guys for your encouragements,

thanks Ronny for the manuscript, I've also found the plans of a modern luthier, Joel Laplane, who actually did a luth with a very similar design to the one you posted, a pdf of his the design and final luth he did:
http://assoc.orange.fr/la-passion-du-bois/rencontres2003/conference...

They both seem to think "circle", with the oud I am trying to think "spiral" which is an open shape, if does mean anything when it comes to sound I don't know, and thanks jdowning for the infos, "The harmony of the Spheres" seems very interesting, I ordered it.

The software I am using is Softimage XSI

All the work I've done till now was based on the relative proportions and not on actual dimensions.
I decided that the leading unit will be the scale length, which means that the entire design should end up encapsulated into one formula with one input: the length of the strings.
This way I would have the plans for 58.5 60 or 62 cm without having to touch the design.
The problem I am facing is from where to where I have to measure the bowl and the fingerboard:
Should I consider the whole "theoretical" length of the bowl knowing that it is going to be cut to fit the fingerboard, or consider the realistic "cut" size which may break the spiral proportions ?
Any Ideas?
For now I have a fingerboard of 22.4 cm for a string length of 60 cm.
Here is sketch of my problem

ALAMI - 8-17-2007 at 12:14 PM

And if everything works that's how I hope it will look

zalzal - 8-17-2007 at 12:44 PM

i did not understand much because math and me are like 0 and 1 (guess who is the 0) but the enthousiasm showed by Alami is really contagious.
the pegbox will be really innovating.....i think Jameel will have to adapt a special set of his soft cases once industrial production will start....
ALAMI dont you think strings will have an enormous tension, specially that this will be a seven course oud.

jdowning - 8-17-2007 at 01:00 PM

I built a lute following the Arnault de Zwolle geometry and barring instructions during the 1970's as an experiment. I made the neck shorter to accommodate 8 frets - basing the neck, pegbox and peg design on the lute depicted in the famous "Ambassadors" painting by Hans Holbein (1497 - 1543) in the London National Gallery. I felt at the time that the "Ambassadors" lute - shown in perspective view - must have had a geometry similar to the Arnault lute but was unable to easily transform this to a plan view using simple geometry in order to confirm my speculation. Note the apparent "heavy rounded" shoulders of the lute in the painting.
This instrument was quite successful but I did get a few comments about the very 'heavy' neck section that did not fit in with most people's opinion at the time about how a lute neck should look. Making the pegs was an interesting project in itself!
I no longer have this instrument as I gave it away to an American friend years ago but - as I recall - string length was about 60 cm. It was fitted with six double courses with a single top string.
Of course, the geometry of the Arnault lute is not based upon the Fibonacci Spiral.
ALAMI is your software Softimage XSI also based on finite element engineering software that might also be used to calculate stresses in the shell geometry?
Could the program readily analyse the "Ambassadors" lute to give a plan and section profile?

ALAMI - 8-17-2007 at 02:17 PM

You're probably right about the tension zalzal, I like playing 7 course but the utility behind it is to be able to test a broad range of pitches on this experimental model, floating bridge may be ? (am already contracting a factory in China for a 1000-piece-a-day-mass-production)
however....Nazih has a fixed bridge seven course ( the sada al rouh model).

The 3D software I am using is mainly designed for character animation and effects and is useless for any kind of engineering application like finite elements or stress maps calculations but I think that I could use it extract the plan of the lute from "The Ambassadors", the full painting shows tiles on the floor which would help defining the scene perspective and pegbox seems to be perfectly vertical, so if you know (or can guess) the height of the pegbox in mm it would help to get the dimensions of the luth, a high resolution image of the full painting will also be needed (a 300 dpi scan from a high quality print). A nice challenge!
Any tips on a special gluing for the bridge to sustain a 7 course?

excentrik - 8-17-2007 at 02:22 PM

Wow.

Alami- you really are moving into an interesting realm here- I can see the pick guard reflecting the spiral as well- or would that throw off the math?

man keep updating us- this is really amazing stuff (I wish I fully understood the math).

tarik

jdowning - 8-17-2007 at 06:00 PM

I have a high quality print only of the area of the painting as posted. I will see if I can find my working drawings to determine the pegbox length that I used - which should be around the same depth as the lute body.
A quick check might be to rotate your generated image of the oud to about the same orientation of the lute to see how closely (or not) they match visually? However, I suspect that the artist Holbein may not necessarily have been using a 'true' perspective representation.
As lutes of the 17th and 18th C carried up to 14 double courses, a 7 course oud bridge should not be a problem even when using traditional hide glues?

ALAMI - 8-21-2007 at 11:21 AM

Good Eye Jdowning, the bowl is a match and the overall shape is very close but not the hole and the bridge positions
I can't say it is absolutely accurate because the perspective is not real (as you said) so I did an almost orthogonal view, I used the oud model I am working on, just changed the pegbox (at 96 degrees relative to the neck).
So if you think it is close enough, it would be easy to extract dimensions (they can be relative to the pegbox or to the overall length)
And Tarik you're right about the spiral in the pickguard, I am trying different designs.

jdowning - 8-21-2007 at 03:14 PM

Thank you for taking the time and trouble to do this work
ALAMI . This is a very interesting (if not astonishing!) result. I would judge the results to be very close.
Hans Holbein as a Renaissance artist likely would have attempted to use a 'vanishing point' perspective to represent the lute, rather than orthogonal. In which case both soundhole and bridge on your model (if transformed to a vanishing point perspective) would appear rotated more to the left and would even more closely represent the lute in the painting? I have always thought that the centre line of the lute in the 'Ambassadors' painting seemed to be distorted or curved to the left so your graphic comparison may be a confirmation that this is the case. No wonder I failed to transform the image - years ago - using simple geometry. Thanks to modern computer technology!
In this painting - full of symbolism and allegories (but that is another story open to many interpretations) - Holbein is playing with other perspective ideas like the anamorphic perspective view of a skull in the lower part of the painting. So nothing can be taken for granted.
The painting also represents articles of Arabic origin like the tapestries and - perhaps - the scientific instruments. The painting of the ambassadors from the French court of Francis 1 (dated 1533) may depict an old, neglected lute from an earlier time (one of the strings is broken) so the instrument might be an early example of a transitional instrument - from oud to lute? Just my speculation, however.
For those unfamiliar with the painting I attach, for information, an image of the whole painting as well as an image of the skull when viewed from the required angle.

Ambassadors composite reduced.jpg - 111kB

ALAMI - 8-31-2007 at 09:53 AM

I've been busy fine-tuning the design of the golden mean oud according to 2 simple rules:1- Oud rules and 2- golden mean (phi) and it is working.
Every important construction and bracing point is "in ratio" with phi. In terms of practical oudmaking all what should be done is placing one point and trace the relative golden spiral, and decide for the string length (62, 60 or 58.5) this will determine the constant "s" in the formula which is s = (scale length/3) and by this I am respecting rule 1, and then all the other points are obtained by an amazingly simple formula for each one.
The formulae have only the integers: 1,2,3 and Phi, all multiplied by s to give the dimensions in mm or inches.
on the attached image all the points show their relative Golden ration connection by thin lines of the same color.
Another very interesting issue was raised by jdowning, it is a speculation, a very tempting one, suggesting that old arabic ouds MAY have followed this type of geometry ratios. The match with the 1533 Holbein's painting is really "astonishing".

I plotted the final design on a 1:1 scale and took it to Fadi Matta who is a real fine luthier and he was enthusiastic about the project.
The oud construction will start next week, we are still reviewing the bracing formulae.
A mould will be made to ensure the accuracy to the mm.
This is going to be a big oud with 43cm height for a 60 cm scale length.
Still not sure for the wood, I would go for walnut that was used on old arabic ouds but it seems that no high quality walnut is available, we found a nice palissander.
The rosette is another story to come, it is also going to be another experimental feature.

oudipoet - 8-31-2007 at 11:19 AM

dear alami
i ve been reading your theory and i really like your calculations and designs esp the shape of the oud and the peg box there is only one thing that bothers me that the distance betwen the bridge and the sound hole somehow they look so close to me it looks there is not enough space for a pick guard and enough space to play confortably ofcourse that s my oppinion i am not a mathematician or anything close to that so my oppinion is just an oud players oppinion anyways so i was wondering what would happen if you were to move to sound hole up and bridge little bit down would that be too much out of your calculations?

Here is what happens if folk player tries to count the frequenses up to 440

Marina - 8-31-2007 at 01:08 PM

Here is what happens if folk player tries to count the frequenses up to 440.
SO, DON'T DO IT AT HOME!

jdowning - 8-31-2007 at 01:58 PM

Another early lute image that makes an interesting comparison is depicted in the painting by Italian painter Lorenzo Costa (1495 - 1535) in the London National Gallery. The artist did not seem to quite manage to correctly represent the longitudinal symmetry of the instrument so one might choose either the upper or lower outline of the belly as the intended geometry.
The most obvious difference from ALAMI's computer generated oud image is that the bridge is somewhat lower - not sure about the soundhole position though. However, if the artist incorrectly represented the instrument symmetry he may also have erred in the precise bridge/soundhole postioning?
Interestingly the "rule of thumb" for early lutes is that the soundhole diameter should be 1/3 of the belly width measured across the centre of the soundhole. The computer generated model comes very close to that criterion.
This also suggests that artist Costa may have depicted the lute profile fairly accurately but that the centreline is distorted - offset too much towards the top edge of the belly (about 7% at the soundhole location I would estimate) - affecting the true bridge and soundhole positioning as well as the orientation of the neck?

ALAMI - 9-1-2007 at 05:29 AM

I'd go for the lower part as it looks more in-line with the overall apparent (and inaccurate) perspective, I'll test in 3D later on (I don't have the 3D soft on my laptop).
Marina I like your spirals, a great base for a curly oud designed after the Marina-cci spiral.
And oudipoet you are right about the distance between the bridge and the hole, I personally like to play in front of the hole, but in fact I have another more classical face design and I was talking with Fadi Matta on the possibility of trying the 2 designs (as part of the experiment)
Here is the second design

ALAMI - 9-1-2007 at 05:39 AM

Another experimental idea is the rosette, it has been used on lutes.

The idea is to have a 2 thin layers of the same rosette, a simple flower of life design, also based on the spiral (obviously I am developing a spiralomania syndrom).
The upper layer would be fixed on the inner part of the face and the lower can be rotated to "close" or "open" the hole, I think it would have a considerable influence on the sound.
Here is the idea "in action"
(mp4, open with Quicktime player)

Microber - 9-1-2007 at 06:04 AM

Oh ! I want such a rosette.
It makes me thinking of the electric moucharabie of the Institut du Monde Arabe in Paris that open and close automatically depending on the sunny or cloudy weather.
Once again, very interresting post, Alami. :applause:

Robert

jdowning - 9-1-2007 at 04:55 PM

I have never seen that particular rosette design on surviving lutes.

oudipoet - 9-1-2007 at 06:15 PM

the idea of rosette is great but we should not forget that the face of an oud vibrates a lot as other instruments so incase of any space betwen these 2 layers of rosette might cause some vibrant sound as you play unless they are very well tightened.

jdowning - 9-7-2007 at 03:56 PM

The latest catalogue from Lee Valley Tools of Ottawa
http://www.leevalley.com arrived by mail today.
A new item is a range of scale rulers for laying out the "Golden Ratio" PHI (Catalogue# 06K30.06 to 06K30.50) ranging in price from $2.25 Can. to a set of four rulers at $24.50Can. (currently on special at $19.50).
Interesting, but not sure if these would be of much use for this project!

ALAMI - 9-8-2007 at 12:44 AM

Nice and practical idea, but to draw golden ratios all you need is a ruler and a compass. This project was going until now in the virtual world of 3D cad and it is the advantage (and the problem): there is always a way back, "undo" is so much easier than "unglue".
I found out that there is no "one golden-mean design" and that there is a lot of possibilities so I kept exploring for a maximum geometrical harmony in a design, I am now in my fourth model and this one is a more conventional oud but with an amazing number of relations between the main 9 construction points, they are connected with 10 Golden ratios and ten "equals".

here's where I got: the sections plan and the 3d preview and an illustration of the many "harmonies" in this design

ALAMI - 9-8-2007 at 01:09 AM

Funny, Looking back to ronnie's post, it seems that I am ending up with an oud version very close to the Arnault de Zwolle (theoretical) design, it may be logical as I am using the same rules.

ALAMI - 9-14-2007 at 11:17 AM

After being through 6 different models, I was feeling that this oud project was not going to a new territory as I was hoping.
When I plotted my last model #6 in 1:1 scale I realized that it has the same domensions as my Nahat only with a bigger bowl.
I compared my models with tens of oud pictures and I found out that the ratio 1.6 is widely used, probably remaining from old ways or simply through experience.
I can now say that golden mean was always in a way or another in oud making.
The idea was to try to design something new based on a golden mean approach, so I went back to my starting Fibonacci spiral.
Leon mentioned asymmetry in his post, but I was unable to find a "harmonic" asymetry (from a geometry pov), I discarded a few asymmetrical designs.
Then I found a way to "harmonize" things by a Phi ratio of asymmetry: roughly by saying that 1 on the lower side is "equal" to 1,618 on the upper side.
And then like in the first model, things start falling down easily in their place.
The result is a totally asymmetric oud, it is not only about a cutaway bowl, it is totally assymetric but still harmonious oud.
The axis of the oud is running along the lower side of the fingerboard and rhe lower side of the bridge.
It is a 60cm scale length with 20cm fingerboard (on the upper side of the bowl-Fingerboard intersection).
Can it be done?
How this face can be braced ?
What do you think guys ?

ALAMI - 9-14-2007 at 11:18 AM

Another set of previews

zalzal - 9-14-2007 at 11:38 AM

extended fingerboard?? you??

Ararat66 - 9-14-2007 at 12:22 PM

Het Alami, this is one great tread, I love it - but who could attempt to build it? I'd love to hear it played. That treble curve would be strong on the back - I bet bracing would be minimal, especialy if braced to curve the top like a Karibyan (or my Haluk for that matter!! he has good tricks up his luthiers sleeve, I've been measuring the top).

You're a very talented and imaginative bloke.

Thanks

Leon

Ararat66 - 9-14-2007 at 12:23 PM

Also you would be able to play right up the neck (without the dreaded cutaway Ughh).

Leon

Jameel - 9-14-2007 at 12:58 PM

Whoa. Incredible renderings! I don't think I have the guts to make a bowl in that shape. Heck, the mould itself would be a challenge. Nazih made a bowl similar to that shape once. Not as radical, but similar. I wish I had the knowledge and software to do renderings like that. It would really help in designing ouds.

mavrothis - 9-14-2007 at 02:05 PM

Wow! I want one! Seriously, what a beautiful and utilitarian design. Great work, you are very talented Alami!

:applause:

mavrothis

Peyman - 9-14-2007 at 02:25 PM

:bowdown:

. That's a beautiful design.
If I may think outside the box for a second, one could make this into a solid mold by carving the shape out, and then casting a fiber carbon or synthetic material similar to it around it. Or even plying blcoks of wood then carving the shape, and then using veneers to get the striping effect. Anyway, it's fun to think about the possiblities...

ALAMI - 9-15-2007 at 12:55 AM

Wow, thanks guys, you can't imagine how good it feels to read your comments in the morning.
When attempting such a radical design, one cannot expect reactions, I was expecting something like "what an ugly mutant alien !" (and it is normal if many think so), but at least I know that I am not the only one who finds these curves intriguing and beautiful.
Hey zalzal, it is not an extended fingerboard, it is a retracted bowl

. No seriously, the fingerboard is part of the structure in this design, it is not a bowl with a neck, it is more integrated as the the ribs ends should hide under the fingerboard.
Leon, can you please explain "braced to curve the top", I am unfamiliar with this technique, I agree that bracing should be minimal, I've already tried (on paper) a way of starting with regular braces as on a normal face then by introducing the same general asymmetry ratio, it would result in 5 diagonal braces.
Mav thanks for your kind words, coming from a great musician with a great oud collection such as yours it is a huge encouragement (you'll get yours)

Now I am very excited to go to the construction stage, I am anxious to hear Jameel talking about not having the gut (but having seen his work I don't really believe him).
As Jameel says, the mould itself would be a challenge and Peyman ideas are great (please keep having fun).. I am thinking of 2 possibilities: the first is to extract the data from the 3D model and start by making a perfect virtual mould, when done it can be breaked down as a regular 2d cutting blueprint for each "arch" of the mould (how do we call them?) as well as the base to support them and the markings of each rib position with a side top and front view for each arch.
This will allow to produce an accurate mould. add to this that I can also "unbend" the ribs from the end design and extract an exact planar outline of each rib before bending which I think would help the maker a lot as there are 21 different shapes of ribs and only the first and last one end with a point on both sides like a normal oud ribs.
The second possibilty is the one suggested by Peyman, I'll add to it an easier start:instead of a manual carving, the bowl can be prototyped directly from the 3D file into a plycarbonte real size smooth surface object, there are many service bureaus providing this type of prototyping. So my question is, if we have this accurate prototype, where can we go from there and what would be the next steps ?
Are we talking of an ovation-style composite material bowl ? how would it sound ? I think that I read on these forums that Dr oud tried something like this and the result was not really convincing.

jdowning - 9-15-2007 at 10:45 AM

An interesting design!
Construction of a moulded bowl from plastic composites might not be as quick and straight forward as might be imagined. I have made lute bowls in the past from thin fibreglass reinforced with carbon fibre strands moulded along the 'rib joints' (and using expanding foam neck filler). Nevertheless these bowls were still relatively heavy compared to bowls made from wood in the traditional manner and I was not too happy about the response of the completed instruments. Whether or not this was due to the fibreglass bowl or my inexperience as a luthier I do not know. I did not continue with these experiments anyway as I hated working with fibreglass.
To make these bowls, I first made a replica bowl in plaster of paris. This was polished to a gloss finish and coated with a release wax. This was then coated with a gel coat layer followed by a number of layers of glass cloth and resin to make a rigid master mold.
The bowl was then made from the mold by first coating the inside with release wax, followed by a thin gel coat, followed by a single layer of woven glass cloth and resin with strips of carbon fibre worked into the resin along the 'rib joints'. All pretty time consuming and best for production runs of identical instruments.
Papier mache might be an alternative material to fibreglass or even a thermoplastic like sheet acrylic - heated in an oven until soft and then stretched over a wooden mold using vacuum pressure to pull the plastic in close contact with the mold surface until the plastic rehardens.
I think that the bowl could be built up from wooden ribs in the traditional way although it would not be easy and likely almost impossible to to make with sufficient precision to conform exactly to the theoretical geometry. I am sure that Jameel would be up to the challenge - that is if he has no other projects that take priority!

jdowning - 9-15-2007 at 11:28 AM

I thought I had seen that shape in an instrument before! Here is an Arch Cittern, late 18th C made by Renault, Paris.
Not exactly the correct profile and it is 'flat' backed so a lot easier to build than the oud will be.

Peyman - 9-15-2007 at 11:54 AM

I also was thinking in the lines of using automated machinery to build the oud. But what about the cost!

I think you should take a look at this website:
http://www.beyondthetrees.com/gallery.html
This luthier has used the papier mache jdowning is refering to, to make instrument bowls.

There is also the possiblity of making half of it the traditional way and half of it using a different method, either carving or casting.

I forgot, The rosette looks awsome too!

jdowning - 9-15-2007 at 02:04 PM

I have not been following all of the details of the mathematical analysis of this project. I assume, however, that a mirror image of the layout posted is also an alternative possibility? This might have some practical advantage for the majority of players who are "right handed" given that the 'cantilevered' fingerboard or 'hollowed out neck' design might allow a somewhat better access to the higher fingerboard positions and make the instrument a bit more convenient to hold?
As for barring, my hunch would be that a single bar just below the rose might be all that is required?
Another 'asymmetrical' thought is that the bridge might be set at an angle so that the bass strings are longer than the treble - to improve bass response (but make things more difficult for the player of an unfretted instrument). This geometry can be found on Western historical instruments like the Bandora. Not sure if this is allowable within the mathematical constraints of this project, however. Just throwing out ideas!

abc123xyz - 9-15-2007 at 08:32 PM

Quote:

Originally posted by jdowning
I have not been following all of the details of the mathematical analysis of this project. I assume, however, that a mirror image of the layout posted is also an alternative possibility? This might have some practical advantage for the majority of players who are "right handed" given that the 'cantilevered' fingerboard or 'hollowed out neck' design might allow a somewhat better access to the higher fingerboard positions and make the instrument a bit more convenient to hold?

I don't understand why you would want to reverse the profile for a right-handed player. The cut away on a right-handed guitar is in the same position as in Alami's illustration, as too on the cittern that you showed, that being on the bottom of the instrument as held by a right-handed player.

David

jdowning - 9-16-2007 at 05:25 AM

I don't know about modern instruments but all of the early citterns that I have seen have the cut away in the neck, (underneath the fingerboard), on the bass string side of the instrument - that is on the top side of the instrument as held by a right handed player. The arch cittern is no exception to this - the cut away being on the bass string side of the instrument. On ALAMI's oud the cut away below the fingerboard is currently opposite to this i.e. on the treble string side.
It seemed to me that the 'reversed profile' alternative might be easier and more convenient to hold for a right handed player - the bottom part of the bowl that rests on the legs being pretty much the shape of a conventional oud. This was just a bit of lateral thinking on my part as I can see no particular reason for the profile not to be the mirror image of the geometry given by ALAMI.

ALAMI - 9-19-2007 at 11:41 AM

Having the shorter side of the bowl on lower side of the oud (closer to the ground for a right-handed player) would make it easier to play up the neck as noted by Leon and David, I think it is a practical advantage (for the sound I don't know)
Jameel you were absolutely right for the mould: it is very challenging. I am working on it (virtually), I put 8 blocks plus the tail and neck blocks, I know it is way more than usual but I think it would make the luthier work easier and more precise as he has to bend and glue 21 different ribs.
I think that the role of the designer is not just a concept, he has to go into the smallest manufacturing details and verify that they can be realisticaly done and deliver detailed plans and prints for each part and aspect of the design.
Now I am working an the individual plans of each part of the mould (it looks like an animal), I will be able to have 1:1 papers to glue on the raw mould wood blocks indicating the curve to cut and the curve to taper it, this includes also the markers of the ribs on each block (the thin black lines).
Now I am realzing that I have to design a new neck block to make gluing the ribs easier, I am thinking of 2 options as the neckblock will grow bigger:
Using a hollowed blockneck to reduce the weight, or carving the neck and the block from one single continuous piece (still with hallowing the block), a neck and neck block carved from the same piece will never need setting, right? has it been done or tried ?
Working on this project made me realize that the oud is not a simple instrument as it looks, somehow I have a greater admiration now for the fine work of luthiers.
I don't know if it is common that a luthier works on somebody else's design (other than replicas ) and if they do, how would they like to receive the design ? Who's oud it is it going to be at the end ? What should be written on the label? I really would like to hear your thoughts on this subject.

braces

Ararat66 - 9-19-2007 at 12:29 PM

Hi Alami

Sorry not to reply - I've been away. Many 'flat top' acoustic guitars and I am sure a number of ouds have a slight upward curve or doming of the sound board so the centre of the soundboard is 5 to 10mm higher than at the rim of the sides (if viewed as a side elevation). This is made possible by shaving the braces so they curve (instead of being flat) making them thinner at either end. The sound board is then glued with the braces to form a slight upward bowl.

This does a number of things, mainly it strengthens the soundboard and gives it a kind of springyness that can make for a very bright and resonant sound (if done with skill, otherwise the top becomes stressed and dampened!!). It also allows the luthier to reduce the thickness of the braces and the overall weight of the instrument.

I don't have practical hands-on experience of this but am considering making a guitar so I have read (an looked and measured) quite widely on this and it makes sense. I think others have mentioned this in other threads here - I'd be interested in their knowledge.

I love this project Alami - put me down for one aswell

hehehehe.

Best wishes

Leon :buttrock:

jdowning - 9-21-2007 at 09:09 AM

A solid mold might be better and likely easier to make with greater precision than the 'toast rack' or 'bulk head' alternative. This would be made by gluing together slabs of pine or limewood (say an inch thick) each cut to the exact section profile and then carving away and blending in the 'steps' to finish the mould. A solid mold would also facilitate more accurate fitting of the ribs given such a complex bowl shape.
The only case of an integral neck and neck block that I have encountered is in guitars built in the Spanish style which has many constructional advantages for the luthier but is unlikely to prevent a neck being pulled up over time due to string tension.
In the case of the oud some kind of extended neck would be necessary to provide a solid foundation for the rib ends along the edge of the fingerboard. This arrangement would likely be much more resistant to pulling up of the neck over time compared to a conventional oud neck block. However, carving such a neck/neck block(s) from a solid piece would have no advantage over one built up from blocks glued together.
From a practical instrument making point of view it might be better to eliminate the undercutting created by the curvature of the ribs as they meet the underside of the fingerboard (this is the cut away or hollowing out that I was referring to earlier - not the cut away design of the bowl). This would require only a slight change to the theoretical profile along the edge of the fingerboard and might simplify construction somewhat.

oudipoet - 9-21-2007 at 09:17 AM

hi everybody

i have to say i really like the design but i know that as a fact even the cutaway guitars are not desirable for pro guitar players cos of losing so much sound on cutaways so bye designing an oud like that i think we would lose a lot of volume and sound ofcourse that s my oppinion i might be wrong too.

jdowning - 9-21-2007 at 01:50 PM

I have no experience or information about guitars or other instruments that have bodies with a 'cut away' design. It would be interesting to know how existing instruments of this profile are barred. Is the barring assymetric? Are the soundboards arched or carved (like a violin)?
ALAMI, can your analysis give a possible optimum theoretical barring arrangement for this particular geometry - maybe with complex curved bars with a radial placement as a starting point? The design seems to beg for an assymetric barring geometry - but where to begin?

ALAMI - 9-21-2007 at 03:01 PM

When designing something without having a practical experience in the real life making, it is easy to misjudge the priorities. I've always thought that it is important for the luthier to be able to reach the inner side of the rib he's working on, all the mold I've seen were hollowed, now I am learning that it is possible to work on a solid mold and it is a great news, solid mold is easier to do. I know that usually people start bt fitting the middle axis rib, for this model I think that it should start from the lower one (the shortest rib)
Thinking from a structural point of view I was working on an unusual neck block but I was afraid that a luthier would laugh at me when seeing it, and that's exactly what jdowning is suggesting (see picture), another great news. It is great to have comments and suggestions from the experienced guys, thanks a lot.
Now for the bracing: I agree that to keep a consistent logic, the barring should be asymmetric. The base idea is to take the upper half of the face (the bigger one) and start by considering it as a normal symmetric big soundboard, then put the plan for the ideal straight bracing for this kind of normal face (and I am not sure how it should be).
The second step is to apply to those braces the same progressive and continuous asymmetry factor used for the bowl and the face, they will become curved (probably an arc of spiral), the distance on the lower part will shrink.
Step three is to keep only those needed for the integrity of the structure, taking into consideration the solid neckblock.
My problem with this method is that I am not sure of the initial configuration (braces for the normal face) it is the most well kept secret and nobody wants really to talk when it comes to bracing. I found only very few detailed infos on this subject.

Peyman - 9-21-2007 at 08:38 PM

You come up with cool graphics Alami.
As far as the construction goes, I was also thinking starting from the sides is the best option. The persian setar is always built from the side ribs, then at the end, the last ribe is fitted by taking a piece of paper, putting it on the gap and making indentations, to get the shape of the rib. Then the rib is cut and sanded to fit.

I don't know much about bracing but a good approach is to divide the face into 7 or 8 parts... Lutes have different bracing patterns. They usually have bass/treble braces and have thinner soundboards. I remember reading about Arafati (Iranian oudmaker, Dr. Oud vistied him) and he mentioned that his ouds have 5 "complete" braces and 6-8 "incomplete" ones. Unfortunately, there were no pictures in the article but I am guessing the complete ones touch run accross the face and the incomplete ones don't. He also said he could control bass/trebel response.

Some cool cutaway guitars I have seen had synthetic material on the cut away, obviously for strengthening reasons.

jdowning - 9-22-2007 at 12:26 PM

The design of extended block could be made from a light wood any surplus material trimmed away to keep the weight down. The neck could be fitted with a spline for gluing to the neck block.
The most basic barring arrangement for symmetrical instruments such as guitars, ouds, lutes etc would normally be a single full bar located just above and just below the rose running across the belly and with the bar ends glued to the side ribs.Bars across the area above the rose may not be required with the extended design of neck block? The bridge seems to be quite close to the rose and offset to the bass side so with this relatively 'tight' arrangement maybe only two bars are required - at least as a starting point - with the belly made a bit thicker than usual? Fan bracing and a thinner belly might also be another possibility - as already suggested?
Here is how I made a solid mold for a lute. The geometry of the bowl is symmetrical but is of flattened cross section rather than semicircular. The end block is shown in position and shaped to the exact profile required after screwing in place on the mold The ribs, therefore, are not all of equal size and geometry (as they would be if the cross section was semicircular) so require some judgement to fit as work progresses. Also - because of the flattened section - the ribs cannot be made to meet at the end of the bowl but 'run off' the bottom edge.
Not sure how rib fitting would work out for this oud but it will certainly be more complex requiring twisting as well as bending of the ribs. Computer generated profiles would be of some help in visualising the approximate shape required for each rib. A greater number of ribs than shown in the graphical representation might be required in order to match the mold profile as closely as possible. Also a dark wood fillet between each rib might help to disguise any inevitable slight discrepancies in the jointing of the ribs.
The advantage of starting with the central rib on a symmetrical instrument is that small adjustments can be made to the rib widths as construction proceeds in order to maintain visual symmetry of the bowl. This would be much more difficult to judge if working from one edge due to the accumulation of errors but in the case of the asymmetrical oud - where there does not seem to be a central rib - perhaps this method would be feasible also.
Impressive graphics ALAMI and a potentially challenging but interesting project.

excentrik - 9-26-2007 at 11:17 PM

wow. again- you've truly impressed the eyes of this entire forum- habibi!

I want one by the way- send it to multiple makers for a competition- best one wins the rights to the design...

haha-

tarik

ALAMI - 9-27-2007 at 11:49 AM

Many thanks for the suggestions guys, really helpful.
Tarik, shukran habibi for your nice words, I hope to be able to announce a great news very soon, From a sentimental point of view I am trying to make this oud in Lebanon from concept design to the real finished oud, it has nothing to do with any patriotic #!^, it is only that we've been "exporting" so many bad news in the last years that a tiny good one would be a nice change, the kind of "resistance" zalzal wrote very nicely about once.

Now just a funny story about the project: Today I've met the face of horror and you can imagine my face when I suddenly discovered that in my solid mold-neckblock design:

THE FINISHED BOWL CANNOT BE TAKEN OUT FROM THE MOLD !!

)

)
They would be stuck together and it is so obvious that we all missed it.
Probably it would have been the biggest and most ridiculous error in the history of oud making (not exactly what I am hoping)
OK it is fixed now and I hope that I will not get this kind of surprises in the end

jdowning - 9-27-2007 at 12:11 PM

Hi ALAMI - I don't see why the neck block - as represented by your previous graphics - would be a problem as there are no re-entrant curves to cause any difficulties. The interior sides of the neck block are all vertical so that when the neck block is unscrewed from the mold the bowl may then be lifted out (complete with neck block in place, of course). It is essentially no different from a 'standard' solid mold in this respect despite its complexity.

Jameel - 9-27-2007 at 02:53 PM

Quote:

Originally posted by ALAMI
Many thanks for the suggestions guys, really helpful.
Tarik, shukran habibi for your nice words, I hope to be able to announce a great news very soon, From a sentimental point of view I am trying to make this oud in Lebanon from concept design to the real finished oud, it has nothing to do with any patriotic #!^, it is only that we've been "exporting" so many bad news in the last years that a tiny good one would be a nice change, the kind of "resistance" zalzal wrote very nicely about once.

Now just a funny story about the project: Today I've met the face of horror and you can imagine my face when I suddenly discovered that in my solid mold-neckblock design:

THE FINISHED BOWL CANNOT BE TAKEN OUT FROM THE MOLD !!

)

Just make the mould in 2 separate piece that bolt together with a removable shim between. When the bowl is finished, remove the bolts, slip out the shim (maybe 2cm thick or so) and pull out the mould pieces. I did this on my latest mould, since it returns past 90 degrees. The shim is behind the neck end block and works very well. In this pic you can see the gap where the shim/neck block support gets bolted.

ALAMI - 9-27-2007 at 03:55 PM

Jdowning you're absolutely right but one thing is missing :that the mould blocks overlap OVER the neckblock, they are not cut like the neckblck on the vertical, and once the ribs glued on the neck...this will make it the strongest oud ever.
As Jameel is suggesting, I think that the solution is to cut the mold in 2 pieces (btw Jameel, your mould is as beautiful and well finished as an oud)
I am thinking of making the bowl in 2 separate pieces (Lower pic)
Once the base holding them togreher is unscrewed, the blue part can be removed like from a normal oud bowl making room to the yellow part to be pushed to the wide area and then removed and leaving the neck and tail blocks in their place.
I am not sure if the wood will agree to play the puzzle role, or may be a break is due.

jdowning - 9-27-2007 at 06:50 PM

Yes, you are right - I can better visualise the problem now with your latest graphics ALAMI.
Another solution - if I am thinking straight - might be just to extend the height of the neck block with a loose dummy piece carved to the correct profile - temporarily screwed to the neck block to provide support for the ribs in that area but which could be released after the bowl had been lifted from the mold?

two sections

Ararat66 - 9-27-2007 at 11:51 PM

Hi Alami

I used to make molds in a bronze casting foundry years ago making sulpures and artworks. We had to deal with undercuts all the time. Simply make your mold in two parts, one attached to the other. The first part will be the undercut area, the second part the main 'open' area. One attaches to the ther with screws, to take out simply unscrew, lift out the main open section, then the smaller section in the undercut will easily slide out. If there is a problem because the depth of the undercut section cannot be removed, then make this in two parts with a lateral cut so one can be removed then the other.

Hope this makes sense - if you like I'll post a diagram later.

Best wishes

Leon

jdowning - 9-28-2007 at 06:22 AM

This is a rough sketch of the solution I had in mind. The mold would first be made completely solid, built up from slabs of the correct cross section, and then carved and finished to the required 3D profile.
The section to be occupied by the extended neck block would then be cut out on a band saw and the mold mounted on its baseplate. The upper part of the piece removed would then be cut horizontally on the band saw to create filler piece B. Some shimming would be required to compensate for loss of material in the saw cuts. This piece would be temporarily mounted on top of the blank for neck block A (by means of screws from underneath so that they can be removed after the bowl is completed - in order to release and remove block B). The neck block blank would then be carved and finished to blend into the mold contours. The ribs, of course, would only be glued to the neck block. If there is a slight overhang remaining along the neck block edge of the mould the completed bowl could still be removed by lifting and rotating it slightly.
This solution (if viable!) might be a bit simpler and easier to make than the split in half mold alternative? Either way, a cut out in the mold for the extended neck block must somehow be provided.

ALAMI - 9-28-2007 at 11:40 AM

What a great brainstorming, thanks guys.
The solution is now clearly a solid mold cut in blocks.
John you're right that cutting along the length is not practical nor elegant, I think we have it, like in your sketch but with 3 parts for the molds instead of 2 and for two resons: first the area just behind the extended neckblock might be risky the second one is the weight of the big block which would make very difficult to handle, btw any recommendation for the mold wood ?
Jameel, what's the wood of your mold?

After yesterday's panic I decided to foolproof the procedure (the fool being me and the virtual world in which this oud is still stuck)
So i ran a previsualisation animation for the mold assembly then for how to take the mold out of the bowl, the animation revealed the risky area behind the neck block.
Please give a look to the movie and let me know what you think

(mpeg4 video - opens with Quicktime player)

Microber - 9-28-2007 at 11:58 AM

Dear Alami, jdowning, Ararat66, Jameel and the others,

Thank you very much for this thread.
It's a real thriller for me.
And your final movie, Alami, it's "2007 A Special Oudissey" !!!

:applause:

jdowning - 9-28-2007 at 12:56 PM

Nice animation ALAMI that is very helpful in visualising the concept - not always obvious! Not foolish either!!
I still wonder if it is really necessary to divide the mold into an extra third part (with all of the potential attendant problems of precise registration and alignment of the two parts of the main mold). The small overhang area just behind the extended neckblock might not be an obstacle if the completed bowl is rotated as it is lifted and removed from the mold? If the slight amount of overhang is a problem then modifying the geometry in that locality to eliminate the overhang might be another way to go? Not theoretically perfect but close enough for all practical purposes?
You show an end block in your virtual image. This additional complication can be practically eliminated - replaced by a lute like end plate ( or oud like end plate in the oud that I am currently restoring!) fitted in place after the bowl is removed from the mold.
Very interesting and challenging project.

From Virtual to Real: first step

ALAMI - 11-4-2007 at 11:25 AM

The oud is most probably going to be constructed by Nazih Ghadban. Nazih loves the design and believes that it is going to be a great oud.
We still have to solve his busy schedule and the fact that he is a 100km far from Beirut, and I really would like to be around as often as possible.
The mold is my part of the work.
This weekend I started working on the mold.
The pieces were cut on a router, it took 2 weeks instead of the 2 days promised and 15 pieces are still missing.

I've never worked wood before, In fact all the work I've done in my life ends in digital form, nothing to touch or smell or makes every muscle in your body complain (which is the case this evening). So I reviewed some old threads and talked to a carpenter I know and got myself a set of Made-in-China-tools (at the actual rate of the euro, it was an obvious decision for what could be a one shot): a chisels set, some clamps, some sews, sanding paper, rulers, some other stuff and a noisy monster called electric die grinder with all kind and shapes of stones.
The main mold pieces were cut staight on a router from 10mm thick mdf (50 pieces), I use the grinder to taper them to the correct profile for each piece, my reference is real size ploted drawings.(is tapering the right word ?)
I cut also 21 pieces of wood modeled to the profile of the mid axis of every rib, I call them "Initial pre-benders", they will help to bend seperately each rib to the correct shape along his longitudinal axis, the main mold will guide the transversal rotation bending of the ribs, I thought it would make the work easier to have the rib already pre-bent before they go to the main mold. But what I don't know is that if it is OK from "wood point of view" to have the wood going throudh 2 cycles of water-heat bending.
I am really excited that the oud is finally going out of the computer, a nice sensation.

Jameel - 11-4-2007 at 12:40 PM

I'm really looking forward to seeing this take shape. What a challenging project!

Ararat66 - 11-4-2007 at 01:36 PM

This is great stuff Alami - now you have a master luthier to realise this it is so exciting.

Brilliant

Leon :buttrock:

jdowning - 11-4-2007 at 02:54 PM

Bravo ALAMI - and welcome to the wonderful world of wood! The accurate shaping of the mold will not be a quick and easy task so take it slowly.
The question of pre- bending the ribs, and the method used is one that should be determined with the luthier who will be making the oud. The computer generated rib profiles will, no doubt, be of assistance to the maker but - at the end of the day - each rib will have to be hand fitted to its neighbour which will be no simple task as some of the ribs (on the treble side) will be twisted, as well as bent to a curve, and the rib joints consequently will vary in angle along their length. This means that the conventional method used to accurately join the ribs of a symmetrical, half round oud bowl (i.e. inverted jointer plane) cannot apply ( except on part of the 'bass' side of the bowl).
Likewise the method used to bend the ribs will be up to the luthier - i.e. dry bending or soaking and heating. No doubt the more difficult ribs to fit will require a number of heating cycles to obtain perfection but I don't imagine that this will result in any adverse degradation of the wood.
Just my opinion, though, as I have never made an instrument bowl of this geometry(!).
Good luck and every success with your project.

SamirCanada - 11-4-2007 at 03:57 PM

my best wishes too for this outstanding project.!!

Raby - 11-10-2007 at 04:21 PM

Very Nice project ! Your oud look like a shell ! Just take care with your pegbox lines . Do you look if the strings touch the pegs ? I am curious to see and hear to result! Good luck !

oudipoet - 1-25-2008 at 07:50 PM

hey alami
how you been?
it s been a long time since your last post so i was wondering how is your project going on? i am looking forward to hear about the new devolopments.

Melbourne - 1-25-2008 at 08:32 PM

The Frank Ghery model

. Very exciting project Alami.

ALAMI - 1-26-2008 at 10:21 AM

Thanks guys for asking.
The project is still on, I almost finished the main part of the mold and I am tracing the ribs placement which appears to be very complicated.
The mold is now made of slices of 3cm thick that are not glued together, it looks like a solid mold but it can be easily transformed into a toast rack mold as the slices are not glued together, Nazih was reluctant on using a solid body, he prefers a toast-rack, I worked on it keeping both options open.

The real difficulty where I got stuck is the bottle-neck-block. Carving it from one piece didn't work it is too much for someone with my in-experience in woodworking.

Next week I'll be seeing Nazih to take the project to the next level.

He Melbourne , Frank Ghery !!! Funny and flattering, yeah why not adapting the design into a oud concert hall ?

katakofka - 1-26-2008 at 03:48 PM

Congra Alami, wish you a success with this idea
You remind me of my brother. He was doing a PhD in architecture and came out to a relation between music and dimensions. It didn't come to my attention that PHI would be involved into sound quality. I don't recall any music theoriciens (pythagore e.g) that used PHI for sounds and waves. Do you know any?

Jameel - 1-26-2008 at 04:21 PM

Cool. That looks so organic. Almost like a piece of some exotic fruit. I want to eat it....

gilgamesh - 1-27-2008 at 02:58 AM

Yess, this cool oud is going to seem including a large part of life, like an animal sleeping into a wood structure. I like to imagine that some miracle sound will blossom from it
:airguitar:

nayoud - 1-27-2008 at 08:27 AM

Hi All

This is a very interesting topic Alami ! Chapeau !

I was just in touch with Dr. Ibrahim Karim via email http://www.biogeometry.com/ on a similar topic.

Dr.Karim is a pioneer in Egypt on bio and sacred geometry and this includes how the design of musical instruments relate to the "true" cosmic scale.

I'm not into new age thinking, but I know that according to Alfarabi these issues were important in music instrument design to a great extent in the distant past, and that includes the oud.

I'm currently reading a book published in 1943 on that topic by Alain Danielou with the title of "Introduction to the Study of Musical Scales".

Will get back to the forum if I get any info that might be useful from Dr. Karim

jdowning - 1-31-2008 at 05:39 AM

Your novel approach to construction of the mold is a great example of 'lateral thinking' ALAMI - it seems so obvious! This is a method that can be applied to all molds of complex geometry - including lute molds of 'flattened' section.
Carving the mold as a 'solid' means that the geometry of the bowl can be readily (albeit with a lot of carving work) and accurately created - something that would have been much more difficult - if not impossible - with a 'toast rack' style of mold.
Converting the 'solid' mold to a 'toast rack' mold - at the same time preserving the geometrical accuracy - means that advantage can then be taken of the open mold construction. The advantage over a solid mold is that it helps reduce the risk of the ribs being inadvertently glued to the mold due to fewer points of contact between rib and mold surface. Location of the bulkheads should be such as to provide more support in areas of the mold where there is a greater amount of longitudinal curvature of the ribs - usually in the area around the bottom end of the bowl for more conventional bowl geometry.

SamirCanada - 8-28-2008 at 09:06 AM

what is going on with this oud??

ALAMI - 8-29-2008 at 12:53 AM

It is one tough question Samir,

I thought I finished the mold made of 18 blocks of 3 cm thickness each for a total length og 54 cm.
I handed the mold to Nazih and left it up to him to choose if he wants to use it as solid mold or toast rack mold along with 1/1 blueprints of each part of the design.

When Nazih started to mount the mold he discovered that it is only 52 cm long instead of 54 : a 2 cm difference in length but the width and the depth are exactly as in the plans,
It means that the proportions became corrupted and the mold is not in the real phi proportions.

What I did wrong was in fact is that I used 10mm thick MDF and assumed that when they say 10 mm it is really 10 mm and that 54 slices of MDF would be 54 cm.
This error could have been avoided if I were not thinking like a computer.

Now I have either to rework the mold from scratch or readapt it taking 52 cm as starting point and do reverse (phi)proportional - calculations or use a prototyping machine (3D printing) a service that I was unable to find locally in Lebanon and the European service bureaus I emailed didn't even bother to answer.

And then it was the summer and I got lazy (you know.....those mediterranean genes....)

Thanks Sam for giving it a kick, I really should put this back on track.

zalzal - 9-1-2008 at 01:32 AM

Probably arriving after the battle but there is a link with infos which can be of interest here
http://www.catnaps.org/islamic/geometry.html

It is a long page but with lot of drawings, diagrams, photos, comments.

Inside this page there are comments on fibonacci, golden mean etc with another link on phi
http://www.earthmatrix.com/ancient/phi.html

DaveH - 9-1-2008 at 01:49 AM

Hey Alami - could you not just glue 1mm veneer onto the blocks? You could either cover the whole of one side of each block (expensive, especially as can be hard to get cheap types of wood in veneer form) or just use a couple of strips as spacers, which would leave 1mm gaps between the blocks - shouldn't be a problem. That way, with 18 blocks if you glue veneer onto one side of each block you're only coming up 2mm short. If you need it to be exactly 54cm, you could coat both sides of two of the blocks - this will still mess up your proportions a little, but it seems a shame to chuck away all that work for a couple of mm, and working by hand you're bound to have some small errors anyway.

ALAMI - 9-1-2008 at 08:38 AM

You have a point Dave.

At this point I have nothing to loose with this mold. Instead of veneer I am thinking of getting 1 mm aluminium plates to place between the blocks, they are light weight, cheap, easy to get and 1mm is 1mm no humidity or imprecision risks.
Then I will have to recheck the outer curvature and do some sanding.

Looks like a good and simple idea, thanks man, really.

DaveH - 9-1-2008 at 09:37 AM

no probs. Aluminium sounds sensible. Good luck and please give my greetings to Al Ustadh ras baalbeki!

DaveH - 9-1-2008 at 10:09 AM

Come to think of it, maybe 3 steel washers per section would also do the trick?

jdowning - 5-1-2009 at 03:32 PM

Here is some new information which relates to the discussion in the first part of the thread concerning the apparent faulty perspective of the lute depicted in the early 16th C painting "The Ambassadors" by Hans Holbein.
Rather than repeat details of my observations, I attach a FoMRHI pre-publication article for information of those who may be interested.

ALAMI - 5-2-2009 at 01:01 PM

Very Interesting, John.

Modeling de Zwolle lute geometry in 3D should be simple, but I am not sure if it is possible to simulate the distortion of the camera obscura, I'll have to look it (the pdf you linked to could be very helpful), In fact a 3D software can represent accurately both orthogonal and perspective views of a volume. Let me know if it can help in your article.

BTW, the Golden Mean Oud project is finally back on track, I'll be posting soon the developments.

jdowning - 5-3-2009 at 11:23 AM

Thanks for the information about the software ALAMI. I shall modify the text of the article to include this information (and remove my misinformation!) prior to publication.

If a high tech solution is possible - to work around the perspective distortion problem - that would be very interesting. Perhaps it could be done by using the two vanishing points on the lute and the 'curved' strings to create three images with their different vanishing points and then, using the 'collage' technique that Hockney suggests was used by Holbein, blend together the separate pegbox, neck and soundboard/bowl images to see if that will replicate Holbein's distorted representation of the lute?

Glad to see the project is back on track.

ALAMI - 5-25-2009 at 09:24 AM

This project was on hold for a while due to some technical bottlenecks coupled with a lack of free time.

As explained in earlier posts, the mold was done using slices cut on a CNC router driven by computer files. Due to inconsistent wood thickness (and my total inexperience in woodworking) the mold ended up 4% shorter than it is supposed to be.

My original plan was to extract the flat ribs shapes from the finished mold by drawing lines directly through the markers that I kept on each slice. Each 1 cm slice has a center which is the assymetry axis (and not the midpoint) and all angles from this axis to the sides of the ribs are equal. Once the lines are drawn on the mold surface I'd have cut a paper or carton to the exact shape and I'd have extracted the individual flat shape of each of the 21 ribs (they are all different). In short, a similar approach to what John did in his project (less the skill....of course).

The original plan failed due to the inaccurate mold and the first attempts to virtually "unbend" the ribs by maths revealed to be too theoretical as the nature of wood was not taken into consideration, roughly: pitch and bank can be allowed but not yaw (meaning that wood is not rubber and cannot be distorted in the horizontal plan of the grain).

And to make a long story.....long, I found a software that was able to accept those constraints and the ribs were unbent one by one using extremely lengthy calculations (in fact it uses finite
elements simulations). I had to dig into some long forgotten theories, The annoying side effect is that this proved some professors that I used to hate...right, I ended up needing it one day.

The resulting shapes were kind of surprise. I was expecting a more regular shape.

To test the result, I printed the flat ribs in 1/1 scale and tested them without the mold, only by using an adhesive tape while matching the external shapes, and they naturally bent in place and shape (OK. it is easy with paper).

The funny thing is that now, the carton ribs will guide the correction of the mold ! In this project everything seems to be inverted.

Jameel - 5-25-2009 at 09:45 AM

Cool stuff! That's a brave soul who attempts to build that in wood. I start sweating just thinking about it. But I'm loving it nonetheless.

jdowning - 5-25-2009 at 12:06 PM

Nice work ALAMI!
The rib profiles will greatly assist in the construction of the bowl but each rib, of course, also has a finite thickness so that the angle of the joint surfaces will be constantly changing along the length of each mating rib. This will be a challenge for the luthier but I wonder if this might be fairly easily resolved by first getting the mating rib surfaces to be a reasonably close fit (using the rib templates) and then by running a short, thin file (or sanding strip) between the ribs - to remove any remaining discrepancies - a perfect fit might be achieved? It might also help to have ebony, inter rib, purfling to visually disguise any slight joint inaccuracies?

I know what you mean about some engineering mathematics learned (or not) at University actually having a practical application in the real world!!

msimon - 5-26-2009 at 06:50 AM

I love the idea, ALAMI! I cannot wait to see how it comes along!

Peyman - 5-26-2009 at 07:52 AM

Nice work again. Hope this one comes around.
Maybe I mentioned this before but one could also carve the bowl out of wood!

Oud Freak - 6-4-2009 at 09:49 AM

Quote: Originally posted by ALAMI

This project was on hold for a while due to some technical bottlenecks coupled with a lack of free time.
In this project everything seems to be inverted.

Hi ALAMI, hope all's well. So what about this project? Is it going to be realised?

ALAMI - 6-4-2009 at 12:18 PM

OF, Yes, it is started,
I've been working closely with Fadi Matta those last 3 weeks, he is fixing the mold and doing his tests, Fadi is a great luthier and his workshop,as you know, is just 15 minutes away from Beirut so we were able to meet frequently and discuss and fix things, he helped me a lot by explaining the mechanics of wood and what can and can't be done.

Peyman, I talked to Fadi about carving and he was very reluctant to the idea, he believes that a carved bowl of this size would be very difficult and that the wood near the right-arm-side of the bowl would be very weak due to grain orientation.

Peyman - 6-4-2009 at 12:25 PM

I see what you mean. I am sure the end result will be really neat. From posts here, Fadi Matta seems like a great luthier to tackle this project.

Oud Freak - 6-4-2009 at 01:19 PM

Quote: Originally posted by ALAMI

OF, Yes, it is started,
I've been working closely with Fadi Matta those last 3 weeks, he is fixing the mold and doing his tests, Fadi is a great luthier and his workshop,as you know, is just 15 minutes away from Beirut so we were able to meet frequently and discuss and fix things, he helped me a lot by explaining the mechanics of wood and what can and can't be done.

Ah very good! Am glad that Fadi will do it , your project is in the safest and most artistic hands

Pages: 1 2