This question is directed to luthiers and experimenters such as our dear jdowning.
The understanding seems to be clear and agreed upon that the area we pick on the oud and place pick guards is the most resonant area for sound
production on the instrument.
Recently I removed 2 pick guards from ouds and the dramatic increase in volume was immediately evident.
Some ouds, due to the bridge height and/or dip of the soundboard at that spot, have enough clearance so that a pick guard is not really necessary
(usually ~8-10mm of clearance I'd say), like with the Sandi oud pictured here.
However, for those ouds where the clearance is relatively low (~5-7mm or so), a pick guard is a good idea to protect the top from being damaged over
time by unintentional contact with the pick.
My question is, are there any materials, whether they are used for pick guards already or not, that could be utilized to protect the face, while being
so thin as to minimally affect sound projection?
It seems to me that even .5mm thickness is too much, and I would love to hear of recommendations, even tips on sanding wood strips/veneers to .2mm or
less if possible.
Perhaps 2-ply wood veneers, or any manufactured materials that are very light and strong?
Any ideas?
Thank you,
mavJody Stecher - 1-25-2015 at 08:37 AM
I have found that clear contact paper, such as used for pantry shelves has no effect on sound. It also invisible in most light.
I have also found that for solo playing the most resonant spot to pluck oud strings is between the pick guard and the big sound hole. However for
playing with others a stroke closer to the bridge, right over the pick guard, adds a percussive element that makes the oud sound more distinct. Also
the strings are tauter/tighter over the pick guard. They "fight back" and it feels good to play there. A good analogy is cooking dried pasta,
spaghetti in this case. Each centimeter closer to the sound hole is like another 90 seconds of cooking. Right at the bridge is uncooked. Over the
sound hole is over-cooked.Brian Prunka - 1-25-2015 at 08:42 AM
You were talking about this Thursday night . . . I'm not sure that it is true that this is the most resonant area but I believe it does affect the
sound.
I've thought about this a bit, and while I am not a luthier (curious to hear what jdowning will say) here are my thoughts:
• most wooden pickguards are oriented perpendicular to the grain of the face. This seems like the worst possible solution if your goal is to avoid
damping the sound.
• the actual area that needs protecting is much much smaller than most pickguards. It could probably be restricted to about 2x3 inches underneath
the first 3 courses.
• My gut guess is that tortoiseshell or horn would be the most transparent acoustically . . . you can get them very thin and they are organic
materials.
mavrothis - 1-25-2015 at 10:57 AM
Thanks for the quick replies so far guys.
I think I just might experiment with contact paper Jody.
I have found that clear contact paper, such as used for pantry shelves has no effect on sound. It also invisible in most light.
Hey Jody, I don't see what it is.jdowning - 1-25-2015 at 11:53 AM
An interesting observation that the sound volume is noticeably increased when the pick guards are removed. However I currently have no experimental
data to quantify this observation one way or the other.
Jody's suggestion to use self adhesive plastic shelf covering film would seem to be one practical solution (provided the film may be removed easily
without damaging the sound board).
Modern wood veneer can be pretty thin stuff (about 0.25 mm if memory serves correctly) and is available with a 'tenderising' treatment to make it
flexible so that it can bend without cracking when applied to curved surfaces (such as furniture edging - a convenience for the do-it-yourself
hanyman). It may also be available with a self adhesive backing - so this might be another alternative if wood is preferred to plastic film.
According to Farmer's translation of the Al-Kindi manuscript fragments (9th C) the strings were struck at the widest part of the oud sound board
location where a strip of tortoise-shell was fixed to prevent damage from the midrab. This place on Al-Kindi's oud was positioned at one tenth of the
string length from the bridge. A strip of tortoise-shell might imply a pick guard of quite small dimensions compared to some modern versions? jdowning - 1-25-2015 at 12:12 PM
Tenderised wood veneer is available, paper backed, with pressure adhesive in thicknesses down to 0.127 mm from this company:
Not sure if the thickness includes the paper backing thickness or just the veneer?mavrothis - 1-25-2015 at 02:27 PM
Thank you for the added recommendations!
I do believe that the best option going forward is no pick guard at all, and just allowing enough string height in the area between the main sound
hole and the bridge, so hitting the sound board is minimized.
Now that I think about it, the pick guard on the Sandi oud pictured in my first post was very thin veneer, most likely around .2 or .3mm, and yet the
volume was still considerably increased when it was removed.
mavadamgood - 1-25-2015 at 07:18 PM
Mav if you get to experimenting with some contact paper let me know how it goes!
Cuz, ya know
Adammavrothis - 1-25-2015 at 09:06 PM
Will do Adam!jdowning - 1-26-2015 at 07:12 AM
Eliminating the pick guard altogether may be the way to go - one can always be added later. This would be a chance to obtain some before and after
data by making audio recordings (one with a pick guard and the other without) that may then be analysed with Audacity software to determine the
acoustic differences.
I have to wonder about Farmer's translation of Al-Kindi which seems to be more of a summary rather than a direct translation? So did Al-Kindi write
that "usually a strip of tortoise-shell was affixed to the belly (at the beating place of the strings)" or is Farmer just adding information based
upon his observation of ouds of his time (early 20th C)?
I do not recall seeing pick guards depicted on European lutes in paintings prior to the 16th C (when lutes were played with a pick or plectrum) and
the same would seem to apply to ouds shown in pictures of the same time period. See attached images.
No ouds survive that were made prior to the 19th C and the survivors all have pick guards. So this begs the question about when pick guards were added
to the oud and what was the original reason if the luthiers knew how this addition might affect the instrument acoustically (as no doubt they did).
Some pick guards cover a substantial area of the sound board and others (as on the oud arbi for example) appear to be large and massive - adding
significant mass to the main vibrating surface of the sound board. antekboodzik - 1-26-2015 at 10:11 AM
Thanks everyone! Please keep the recommendations coming!
I've ordered some rolls of contact paper to experiment with. The Sandi would probably be a good oud to use for the experiment.
I will try and record first without and then with the contact paper to see if there is any noticeable difference.
jdowning - 1-26-2015 at 01:01 PM
The New Grove Dictionary - on line version - mentions that pick guard material (for 19th C ouds) might also be fish skin or leather. Both are flexible
materials but,of course, a flexible glue would also be required if the (desirable?) flexibility property is not to be lost. Thin hot hide glue would
very likely not work in this respect.
These days a synthetic 'contact' adhesive might provide flexibility (?) and calf skin leather can be very thin (but costly). Can't say I would like to
try these though.Brian Prunka - 1-26-2015 at 03:00 PM
Wouldn't a flexible material inhibit vibration more than a stiff material?adamgood - 1-26-2015 at 08:00 PM
https://www.lasonanta.eu/en/guitarras-flamencas-clavijeros-golpeadores-unas-zapatos-para-flamenco/guitar-tap-plate
?
Especially the black one...
So there are some good ideas! Actually on my nice Bernabe nylon string guitar I have some of the Kling-on "tap plate" protectors.
"protectors made from clear static cling vinyl"
They are awesome actually. There is no adhesive and it sticks on like Colorforms (remember those from when you were a kid?) Just a hunch but maybe the
stick on is enhanced by a glossy finish like we typically find on a guitar? I'm going order some more and see how it works!adamgood - 1-26-2015 at 08:01 PM
See that there are two types of static kling-on tap plates - one, really tought, used as a golpeador for flamenco guitars, and one thinner, just for
basic protection against scratching with fingernails or pick.jdowning - 1-27-2015 at 05:55 AM
Brian has posed a fundamental question about the stiffness of material added to the sound board surface inhibiting its vibration.
Any plate of material stuck to the most active area of a sound board - as most 'pick guards' on an oud are - will, more or less, add mass and
stiffness that will in turn have an effect on the acoustic response Whether or not this effect is considered advantageous acoustically will depend
upon the preference of an individual. An oud with a 'pick guard' may sound just fine to modern ears but it may not have done so to those living in in
earlier times when oud 'pick guards' were not a feature.
I would speculate that oud 'pick guards' came into vogue around the end of the 18th C due to a change in string technology when wound bass strings
became generally available to replace the duller sounding gut basses. In order to moderate and better balance the overwhelming volume of the wound
basses luthiers may have added weight and stiffness to the sound boards of all gut strung ouds in the form of plates glued to the most active area -
about mid way between the bridge and sound hole(s). These added plates would have been made more appealing to the eye by inlay decoration or being
made into fancy shapes and - because of their location - gained the nickname 'pick guard' (or whatever the equivalent word is in the languages of the
Middle East) and so their original function may have since become obscured. Indeed it is hard to believe that their function is primarily to guard
against sound board damage when many players strike the strings above any protective area that might be provided by these plates and the strings
themselves cover much of the plate area. Compare this to modern guitars where the pick guards are positioned outside the area covered by the strings -
on the treble side - where they will presumably be most effective in protecting against damage from a pick.
That the addition of these sound board plates may have originally had an acoustic rather than protective function is suggested by the huge plate (now
missing) on what may be one of the earliest surviving ouds dating from the end of the 18th C (?) - oud #0164 in the Brussels Musical Instrument Museum
and the engraving of a similar oud from Napoleon's 'Description de l'Egypte' made around the beginning of the 19th C.
The observation made by mav that removal of even a relatively thin plate results in a noticeable increase in sound volume is further evidence in
support of the acoustic function and would account for there being no plates when ouds were fully gut strung when maximum sound volume would have been
desirable.
Curious about the acoustic effect of adding localised mass and stiffness to a sound board I plan to undertake a quick 'before and after' trial on one
of my lutes over the next day or two. I will report back the results here - for general information.
What is the name given to these plates in the Middle Eastern regions and how does it translate literally into English?
Farmer translates 'madrib al-awtar' as the 'beating place' of the strings - in other words the position (on a string) where it is to be struck by the
midrab. This does not suggest that there was a protective pick guard placed on the sound board underneath although Farmer states that there was. BaniYazid - 1-28-2015 at 07:00 AM
If I remember correctly, the name in Arabic is "raqma" or "el raqma"
A translation should be "plate"
To be confirmed by others.bulerias1981 - 1-28-2015 at 08:45 AM
Here's an interesting bit of information from my end. I've had it where I remove the pickguard and I noticed a better sound, and I've also noticed
adding a pickguard to certain instruments snuffed out some annoying "hollow" or "echoey" overtones. So at the moment I don't have strong opinions
about this feature. However, I prefer thinner pickguard in general. I have seen ouds with incredibly powerful tone and thick pickguards as well. I'm
really okay with .5 mm or less. Some ouds if the pickguard is in a bad location and if its significantly chunky, this can prevent the rishi from an
upstroke on the first course, which is incredibly annoying.
As far as making thin veneers, if you have a good 14" bandsaw, get a nice blade (resaw blade) at least 3/4" wide or more. You need a stable rip fence,
a featherboard helps. And with this setup and a steady feed into the bandsaw, you can make a nice veneer of practically any thickness. Then you can
take it to a drum thickness sander, and level any high spots. You'll have to use thin double sided tape and tack it to a thicker piece of wood
because I think the tolerance on most thickness sanders are 3mm.
Thicknessing veneer can be done by hand, but not so easily and precise.bulerias1981 - 1-28-2015 at 10:14 AM
To thickness veneers for pickguards by hand, you could do the following:
Apply thin double sided tape to the back of the veneer.
Tack it to the back of a flat surface, countertop, or MDF board.
Use a sanding block, attach sandpaper of your choice and sand it in 2 or 3 different grits.
When the desired thickness is achieved, apply masking tape to the surface which was sanded.
Carefully remove the pickguard from the MDF board and the tape from the pickguard, this will be tricky since the veneer is so thin and prone to break.
That's why I recommended taping the top surface.
You can keep the tape on for the gluing process, and once the pickguard is joined to the soundboard, the tape can then be removed.jdowning - 1-28-2015 at 12:27 PM
I do not build instruments with pick guards so the tests carried out on one of my lutes this afternoon were of personal interest.
The test lute is one made some years ago a reconstruction deduced from surviving lute fragments by early 16th C luthier Laux Maler - six courses, the
basses octave tuned pairs, Pyramid wound and plain nylon, string length 67.5 cm tuned in 'D' at A440 i.e. D, G, c, e, a, d'.
Sound recording made with a Zoom H2 digital recorder positioned 18 inches (46 cm) directly in front of the sound board. The strings were plucked in
succession from bass to treble with soft thumb tip to sound D G d g b d'.
The audio signals recorded were analysed with 'Audacity' spectrum analysis software to determine the sound intensity or pressure (in dB) of the
individual G-Major resonant tones.
Four comparative tests were made:
1) with no plate.
2) with contact film measuring 12cm x 6cm, weight 1.2 grams placed midway between bridge and rosette.
3) with a larger piece of contact film measuring 10cm x 16cm, weight 2.4 grams placed over the smaller film. Combined weight 3.6 grams.
4) with a copper plate measuring 12cm x 6cm x 0.53 mm thick, weight 36.2 grams positioned midway between bridge and rosette.
A copper plate is about 10 times 'heavier' than one made from East Indian Rosewood made to the same dimension and thickness.
The contact film (Jody's shelf covering) is stuff that I have had lying around for about 40 years but still has some slight adhesive quality but not
enough to damage the sound board surface on removal.
The results of the spectrum analysis showed a small increase in sound pressure level for individual resonant frequencies with the addition of the
plates in each case. However in most cases the measured increase for each resonance peak was within 4dB so would represent a difference in sound
volume generally undetectable by the average human being.
Further tests were made by placing the copper plate close to the bridge and then close to the rosette - all without making any significant difference
to the individual sound pressure levels.
So, as far as this test on this particular instrument is concerned the addition of the plates to the most active area of the sound board - regardless
of mass - made no significant difference to sound pressure levels throughout the spectrum although I thought that I did perceive some slight
'improvement' in the tone colour with the copper plate addition (a greater richness of sound with less of the brassy overtones of the wound strings).
This could be an indication that I may have originally made the sound board a bit too thin for this pitch and string configuration.
More aggressive adhesive than the contact film used here for sticking plates to sound board may result in increased additional local stiffness that
might affect results?
So although this test does not support the observation that removal of a pick guard will result in a significant increase in sound volume this result
may not necessarily be valid for another instrument heard through different ears. Over to you mav!
[file]34215[/file]jdowning - 1-30-2015 at 10:49 AM
For information and interest a second comparative test was run yesterday this time with a larger piece of contact film measuring 17.5 cm x 14 cm
i.e.covering an area as wide as the bridge and from bridge to sound hole. This compared to no film on the sound board.
I could not detect any noticeable change in sound volume (a difficult subjective judgement however when the two cases cannot be compared side by
side).
The Audacity software allows the sound pressure for each resonant frequency peak to be measured by moving a cursor across the spectrum analysis
image.
This time the sound pressure of the resonant peaks in the 2000 Hz to 5000Hz range were compared - this being the frequency range where human hearing
is most sensitive. The frequencies compared were G7, A7, B7, C7/8, D7, E7 and F#7 these being the notes of the G Major chord sounded for the test. The
sound pressure differences measured were 6 between 1 dB and 6 dB and two at 8 dB.
Perceived loudness of sound is subjective but a difference of 10dB is said by psych-acoustic analysts to equate to a doubling of perceived loudness
(whatever that may mean as different people have different perceptions of loudness). At the same time a 3 dB difference is said to be equivalent to a
perceived loudness increase of 23% and 6 dB an increase of 52%.
So if anything addition of the film may have slightly increased perceived loudness to a person with particularly sensitive hearing.
The same analysis was repeated with the previous results for the copper plate and indicate a slightly greater increase in sound level compared to no
plate based upon the differential sound pressure measurements that ranged from 0 to 14 dB (6 between 0 dB to 7 dB and 3 between 8 dB to 14 dB). A
perceived loudness due to a sound pressure difference of 14 dB equates to an increase of about 150%.
A commercial loudness meter might give a better quantitative measure of loudness? mavrothis - 1-30-2015 at 12:16 PM
Do you think the fact that you are playing quietly is diminishing the outcome?
Have you tried picking with an oud pick as hard as possible?Brian Prunka - 1-30-2015 at 12:32 PM
Yes, I would think that discrepancies would be magnified at higher volumes. Due to the subjectivity of experience, however, I think it might be hard
to play with exactly the same force on successive trials.
Perhaps playing one string about a dozen or so times in succession and then averaging the results would be a way to compensate for variation.jdowning - 1-30-2015 at 04:20 PM
My lute is designed for fingertip plucking so the course pairs are relatively widely spaced and would not work well with a risha. However, consistency
of plucking force should be more important than striking intensity. Ideally some kind of mechanical device for plucking the strings would be best but
that would mean more complexity than I currently have time for. In this case I tried - with experience - to be consistent in striking each string
using about the same force and for each test the chord was sounded three times in succession i.e. there were 18 strikes to be averaged for each
analysis. Not laboratory standard testing by any means though - just using what I had readily to hand to obtain some quick results.jdowning - 1-31-2015 at 12:47 PM
Another promising audio analysis software is the free 'Sonic Visualiser' available for Linux, Windows and Mac. - designed for musicians to visualise
and analyse audio recordings. It is more versatile (and complex) than Audacity and supplemented with many third party 'plug-ins'.
I have not yet spent time in reading the operating manual or tutorials to be able to use the software effectively but it looks as though it may be
very useful for the current objective of this thread in analysing, from audio recordings, the effect acoustically of adding or removing sound board
plates. jdowning - 2-10-2015 at 03:08 PM
Before I hang this test lute back on the wall, I will do another series of tests - as suggested by Brian - just plucking the 5th course wound string
a number of times using a plectrum (to obtain a louder response than with fingertip) but also to attempt to measure sound pressure levels with an SPL
meter.
Basic SPL meters are available on EBay for between $15 and $30 - not laboratory grade (meters that can cost hundreds if not thousands of dollars) but
good enough for comparative rather than absolute measure. They are not loudness testers but have filters (A filter) that approximate (roughly!) to the
human perception of 'loudness'
The meter that I will be using is this one that records maximum values in sound pressure recorded in dB. Should be interesting to have another
reference for comparison.jdowning - 2-17-2015 at 11:29 AM
The SPL meter is a low cost one of Chinese manufacture - a 'Wensn' ws 1361. It has a digital read out and appears to be quite well made. It measures
sound pressure levels in dB for frequency weighting A and C - the dBA weighting is supposed to represent human hearing that is more sensitive to
frequencies in the 2000Hz to 5000Hz range. The meter comes with software that records readings to a PC via a USB connection. Unfortunately the
software will only work with either Windows 7 32 bit or 64 bit operating systems not Linux so I can not take advantage of this feature at present. The
meter has a stand alone maximum level feature so will record the highest measured dB value over a period of time.
This feature was used to determine dB levels for a series of tests on my lute this time using a stiff leather plectrum to repeatedly pluck the wound
string of the fifth course in an effort to produce a 'louder' sound than with using soft fingertips.
Two trials were made one without contact film the other with film measuring 10 cm X 16 cm (as used in the previous trials).
The SPL meter and Zoom H2 recorder were both placed at a distance of 18 inches facing the sound board of the lute (constant distance is important for
comparisons as sound pressure diminishes with increasing distance from source). The recorded signal was subject to Audacity spectrum analysis for
comparison.
The SPL meter measured a range of 57dBA to 59dBA (slightly louder) for the lute without the film and 53 dBA to 55dBA for the lute with contact film.
Comparison of the spectrum graphs indicated that for the lute with film, in the lower frequencies (94 to 472 Hz range where the resonant peaks were
most prominent) sound pressure levels were slightly greater but only by about 4 dB or less. In the 1000Hz to 3000Hz range it was difficult to judge as
the dB levels at peak frequencies varied greatly in some cases in others not so much. However, on balance the lute without film appeared to have
somewhat higher sound pressure levels so would tend to confirm the maximum SPL meter readings recorded.
So I have to conclude that given the relative inaccuracies inherent in the test equipment and procedures, the addition of contact film to the sound
board appeared to make little difference, if any, to perceived loudness levels for this instrument.
Time to move on!
