Mike's Oud Forums
Not logged in [Login - Register]
Go To Bottom

Printable Version  
 Pages:  1  ..  3    5
Author: Subject: Something Completely Different - Metal Bowls - and a Colascione
jdowning
Oud Junkie
*****




Posts: 3299
Registered: 8-2-2006
Location: Ontario, Canada
Member Is Offline

Mood: No Mood

[*] posted on 7-24-2015 at 11:14 AM


My pleasure!

For the 'ring sound hole' tests the circular discs will be mounted centrally in the sound hole on a thin, deep support glued on each side of the sound hole. Fish glue will be used as it is fairly strong when dry but easily softened and weakened with water and heat - so the support beam may be easily removed after the tests have been completed.

Each disc will be clamped to the support beam with a screw.
View user's profile View All Posts By User
jdowning
Oud Junkie
*****




Posts: 3299
Registered: 8-2-2006
Location: Ontario, Canada
Member Is Offline

Mood: No Mood

[*] posted on 7-25-2015 at 11:05 AM


The support beam has been glued across the sound hole using water soluble fish glue that hopefully will be strong enough for the task in hand.
Clamping force for gluing is applied with a temporary flexible 'strong back - both strong back and support beam have reference marks to ensure central positioning of the disc mounting screw in the sound hole.
Will leave the glue to cure overnight before starting trials.

It will not be necessary to remove the strings for the trials so measurements will be recorded with the sound board under full string tension (but with strings damped with felt to prevent them from vibrating).

[file]36025[/file]
View user's profile View All Posts By User
jdowning
Oud Junkie
*****




Posts: 3299
Registered: 8-2-2006
Location: Ontario, Canada
Member Is Offline

Mood: No Mood

[*] posted on 7-28-2015 at 11:31 AM


The 'ring' sound hole test has been completed with solid dics mounted centrally inside the colascione open sound hole of 7.6 cm diameter (D). Six fibre discs with diameters ranging from 4.2 cm to 7.4 cm (d) were used the air resonance signal in each case recorded with a Zoom H2 digital recorder positioned over the sound hole area. With strings damped air resonance was initiated by tapping the sound board just in front of the bridge. The recorded signal in each case was then spectrum analysed with 'Audacity' to determine the resonance peak.
The attached normalised graph of frequency f/f0 (f0 =198 Hz open sound hole) and d/D now includes the meaured results.

The curve of the measured results is a bit closer to the curve obtained from resonance chamber tests than I might have expected.

However the calculated curve for the colascione has not been validated by this latest test so can be ignored.
Here the original assumption that the measured air resonance frequency of a circular sound hole with active area A is identical to that of a ring sound hole of the same area with outside diameter D measuring 7.6 cm - is shown not to be valid.



[file]36068[/file] [file]36070[/file]
View user's profile View All Posts By User
jdowning
Oud Junkie
*****




Posts: 3299
Registered: 8-2-2006
Location: Ontario, Canada
Member Is Offline

Mood: No Mood

[*] posted on 8-11-2015 at 06:50 AM


The most recent paper from MIT about the evolution of air resonance power efficiency may go some way to explain why the calculated curve (red curve) f/f0 against d/D is lower than the measured (green) curve in the graph previously posted. Refer to Figure S1 on the first page the Supplementary Information here

http://rspa.royalsocietypublishing.org/content/royprsa/suppl/2015/0...

Although this compares air resonant frequency of narrow and long violin f shaped sound holes to circular sound holes with the same area and circular sound holes with the same peripheral length the same basic consideration applies - area of round soundhole compared to a slender, long ring sound hole of the same area. I must do the calculation for peripheral length for comparison.

I plan to undertake two more sound hole acoustic trials while the colascione is set up for air resonance testing.
The first will be with half ring sound holes - using the same central discs as before but with half the sound hole area blanked off with a thin metal plate.
The second will be to test sound holes of 'half moon' configuration. The circular sound holes used for previous tests being reworked accordingly.

The additional quantitative data from these tests may help in further understanding of how sound hole shape influences the air resonance frequency.
View user's profile View All Posts By User
jdowning
Oud Junkie
*****




Posts: 3299
Registered: 8-2-2006
Location: Ontario, Canada
Member Is Offline

Mood: No Mood

[*] posted on 9-1-2015 at 02:04 PM


The test for the 'C' type sound hole (i.e half ring sound hole) has been completed and a normalised plot of resonance frequencies f/f0 against d/D is attached for information - very similar to the results for a ring type sound hole previously posted but with lower air resonance frequencies (smaller area of sound hole).

For comparison with previous results, the spectrum analysis for a d/D of 0.7 is also attached - the point where with the central area of the sound hole blocked the air resonance frequency has been reduced by less than a semitone.

Next to remove the temporary support in the sound hole in order to test a 'D' or 'half moon' sound hole configuration.

[file]36560[/file]
View user's profile View All Posts By User
jdowning
Oud Junkie
*****




Posts: 3299
Registered: 8-2-2006
Location: Ontario, Canada
Member Is Offline

Mood: No Mood

[*] posted on 9-6-2015 at 02:22 PM


The temporary central sound hole bar has been removed (with water brushed on and a hot thin artists spatula) to measure the air resonance frequencies for 'D' shaped sound holes of different diameters.
The results posted here for information have been computed for comparison with a round sound hole based upon sound hole area and sound hole periphery (i.e. the length of the sound hole boundary or edge).
As the MIT tests have shown (verified by the more basic testing reported earlier in this thread), most of the air flow at resonance through a sound hole occurs at the sound hole outer edge or periphery. So that a round sound hole is rather 'inefficient' with the majority of the inner part of the sound hole (typically an area of 0.7 sound hole diameter that I call the 'dead zone') contributing little to the oscillating resonant air movement across a sound hole.
From these results it can be seen that the sound hole geometry does have some effect but this is rather small when comparing a circular open sound hole with a 'D' shaped semicircular (or non semicircular) sound hole of the same area or periphery.
No doubt this result will also be similar to a comparison between single round and oval sound holes (the latter found on some modern ouds) - i.e.not much difference as far as air resonance frequency is concerned - all else air volume etc. being equal.


[file]36653[/file]

[file]36657[/file]
View user's profile View All Posts By User
jdowning
Oud Junkie
*****




Posts: 3299
Registered: 8-2-2006
Location: Ontario, Canada
Member Is Offline

Mood: No Mood

[*] posted on 10-1-2015 at 11:54 AM


Following the last of the acoustic tests the colascione has been sitting quietly unattended. Time to move on.

From the test results it has been decided to stick with the G1 G2 D3 tuning (octave and a fifth interval or the first three tones on the harmonic series). This gives a powerful bass sound that I believe is primarily due to the air resonance effect - the sound board being quite small in surface area (see attached image - my medium size hand pretty well covers the surface).

The sound hole will be left open for now just in case the neck requires resetting but eventually a parchment rosette will be fitted - as on the Dean Castle and other surviving colascioni - the rosette, however, having no significant effect on the air resonant frequency.

Here is a brief audio clip to give some idea of the sound - the 'squeal' of the wound basses will require attention! - more to follow.

[file]37004[/file]

[file]37005[/file]
View user's profile View All Posts By User
jdowning
Oud Junkie
*****




Posts: 3299
Registered: 8-2-2006
Location: Ontario, Canada
Member Is Offline

Mood: No Mood

[*] posted on 10-1-2015 at 03:07 PM


The audio clip is a much simplified (!) representation of a theme found in the only known parody of Colascione music written for the large Italian early 17th C extended neck lute (called a Chitarone) by Girolamo Kapsberger (Libro Quarto d'Intavolatura di Chitarone, Roma 1640). Kapsberger provides the then alternative name for a Chitarone as a Tiorba (Italian "Chitarone over Tiorba" - i.e. Chitarone otherwise known as a Tiorba).

Athanasius Kircher in 1650 (mentioned on page 2 of this thread) tells us that the Tiorba was the invention of a Neapoiltan street busker who took a lute and doubled the length of its neck to make a baritone sounding instrument (with fewer courses than a lute) that was called - as a joke - a Tiorba. So this implies that this joke of a street instrument with its only 3 or 4 single courses (?) eventually was developed to become the more sophisticated extended neck, double pegbox, Chitarone/Tiorba with its up to 14 courses.

As far as I can determine so far, in Italy any lute like instrument with a long neck, with single peg box and 3 or 4 single strings was called a Colascione (or variants of that name). Therefore, some Colascione variants in Italy were also known as a Tiorba - an unsophisticated street instrument - a long necked lute used as a bass accompaniment for voice or other instruments (or dance).

Attached for information is Kapsberger's manuscript composition 'Colascione' written in Italian tablature. In Italian tablature the six lines represent the six strings of the lute that are stopped on the fingerboard and the numbers represent the stopped fret positions - 0 open string, 1 first fret, 2 second fret and so on. The bottom line is the top treble string (closest to the floor as seen by the player). The seventh course (open) is represented by an O with a line through it written over the top line. Clearly, this composition is an elaboration of a simpler melody heard by Kapsberger when he heard a 3 string colascione being played.

This interpretation of Kapsberger's composition for Tiorbo is by lute virtuoso Paul O'Dette (who has added a few notes not otherwise written in the tablature)

https://www.youtube.com/watch?v=mhuwIBKkdPc

Eventually I hope to transcribe this composition for this project Colascione to see how it might originally have been heard on a simpler three stringed instrument.


[file]37007[/file]
View user's profile View All Posts By User
jdowning
Oud Junkie
*****




Posts: 3299
Registered: 8-2-2006
Location: Ontario, Canada
Member Is Offline

Mood: No Mood

[*] posted on 10-3-2015 at 03:33 PM


As noted in an earlier posting, historically, with all plain gut (or silk) strings the optimum tuning of a colascione of this string length (79cm) might have been G2 G3 D4 (or G g d' or 98 196 294 Hertz). This represents the extreme practical range of plain gut strings with diameters of 1.04 mm (2.8 Kg tension), 0.56 mm (3.2 Kg tension) and 0.4 mm (3.5 Kg tension) - A440 standard pitch.

For the current tuning an octave lower (G1 G2 D3), historically the G2 and D3 strings might be gut (or silk) but the octave lower bass string would have to be a composite string of higher density (loaded gut or silk, or wound). Wound strings produce an amplified 'squeal' as the finger slides along the string so an alternative smooth bass string may be the necessary alternative. As a first trial I will test one of my lead cored braided strings in place of the Pyramid wound.

An interesting acoustic phenomenon of this style of colascione is not only the pronounced air resonance effect (in this case peak resonance at G3 or 198 Hz) but also the 'missing fundamental' phenomenon where the human ear in processing complex sound waves (ie a non pure (sinusoidal) sound comprising a range of harmonics) detects a fundamental tone where none may exist (or is very weak in intensity) - although the overall tone 'colour' may be different. In detecting the upper harmonics the ear is deceived into assuming the lower pitched fundamental exists.
Attached are previously posted 'Audacity' Fourier series spectrum analyses for comparison. To my ear the G1 bass is relatively strong (loud) yet barely registers in the spectrum graph loudness scale (dB).

Attempting to play the colascione feels very strange at the moment for me as a lute player. Plenty of space around the player is necessary to avoid problems with the long neck accidentally contacting neighbouring objects such as computer monitors or low ceilings etc.! A shoulder strap seems to be essential for adequate control.
As a lute player I am using soft finger tips to pluck the strings that are quite widely spaced at the bridge but a pick, plectrum or risha would be an historical alternative allowing an increased string tension.

One advantage of only having three symmetrically positioned strings is that the instrument may easily accommodate both right and left handed players without any structural modification - just switch the strings over from bass to treble and vice versa.
For the 'long' string length (79 cm) and relatively 'low' string tension (in the 2.5 to 4 Kg lute range) tonal pitch can be significantly altered by 'bending' (stretching a string sideways on the fingerboard) also enabling pitch vibrato -good for vocal accompaniment (?). Under these circumstances there seems to be little point in 'micro tonal' fretting the instrument other than basically 12 tone Equal Temperament (frets adjusted by ear for acceptable intonation).



[file]37043[/file]

[file]37045[/file]
View user's profile View All Posts By User
jdowning
Oud Junkie
*****




Posts: 3299
Registered: 8-2-2006
Location: Ontario, Canada
Member Is Offline

Mood: No Mood

[*] posted on 10-13-2015 at 09:39 AM


The G1 pitch Pyramid wound string has been replaced with the 63# 2 strand lead core prototype string left over from previous trials. Calculated tension is about 2.5 Kg which should be about right.

http://www.mikeouds.com/messageboard/viewthread.php?tid=10010&p...

The string is only just about long enough to fit and will take a while to stabilse in tension no doubt. If this preliminary test is favourable I will likely make another 2 strand core string using a finer weave braided jacket for maximum smoothness.

Comparative sound clips will be posted later.
View user's profile View All Posts By User
jdowning
Oud Junkie
*****




Posts: 3299
Registered: 8-2-2006
Location: Ontario, Canada
Member Is Offline

Mood: No Mood

[*] posted on 10-27-2015 at 11:26 AM


The prototype braided G1 pitch drone string has been on the colascione for about 10 days and is still settling down - requiring tuning up slightly each day - but otherwise not a problem. The thicker string also requires some adjustment to fret spacing in the higher positions to eliminate slight pitch increases. Use of higher string tension might also help.
The first and second melody strings (tuned D3 and G) are Pyramid 0.63 PVF and 1010 wound. Note that the chosen historical pitch intervals of an octave and a fifth are the first three tones of the harmonic series.

The two attached brief audio clips - recorded on a Zoom H2 digital recorder are raw and otherwise unenhanced.
The first is that of the open strings played in succession. I am using a stiff leather plectrum or mizrab as some of the Italian sources have indicated that the Colascione plectra (or tacconi) were made of leather. I have not been able to confirm this independently but leather works OK.
The tonal difference between the braided and wound strings is obvious and as expected with the wound string having that typical metallic 'brassy' sound rich in upper partials.

The second audio clip represents the Kapsberger 'Colascione' theme in order to evaluate string noise generated by left hand fingers sliding along the string. Not as pronounced as on the wound string (previously posted) but still there nevertheless. A bit like the sound of sandpaper sliding on wood I suppose.

On balance I think that the braided string has potential for this instrument so will make another string with smoother braided sleeve and hopefully a tighter wound lead core. I also plan to replace the #1010 wound string with a braided lead core string just to see how that works out.
View user's profile View All Posts By User
 Pages:  1  ..  3    5

  Go To Top

Powered by XMB
XMB Forum Software © 2001-2011 The XMB Group