jdowning
Oud Junkie
Posts: 3485
Registered: 8-2-2006
Location: Ontario, Canada
Member Is Offline
Mood: No Mood
|
|
Burnished Soundboards
Lute sound boards four centuries old or more have a uniform, polished, dark patina but do not appear to have any evidence of a varnish or other
applied finish.
There are no ouds that survive of this vintage but tradition has it that their sound boards never had any kind of applied finish - applied finishes
being considered to be detrimental to the sound board acoustic properties.
It is said that Spanish guitar makers prior to the mid 19th C did not apply a finish to their soundboards either but simply burnished the wood of the
sound board to a high polish using a silk cloth.
Burnishing is an ancient method of bringing wood (or metal) to a high polish by simply rubbing the surface with a hard (or not so hard) 'burnisher' -
the process appears to generate local heat and compression forces causing the wood cells (or metal molecules) at the surface to compress or
consolidate creating a smooth polished surface. Wood turners will polish their work on a lathe by simply holding wood shavings against the work as it
rotates. Gold leaf work is polished by hand to a mirror finish using smooth bone or agate 'burnishing' tools.
On the down side the slick, shiny surface of burnished wood is to avoided for glued joints or varnished or painted surfaces.
Is it possible that the old lute and oud soundboards were simply burnished to a polished shine - a finish that was maintained by the owners, over the
centuries, regularly cleaning their lutes or ouds by rubbing with a dry cloth? Time and natural oxidation of the wood adding the darkened patina?
Curious about the practicality or otherwise of burnishing a sound board a few trials have today been undertaken.
Using a sample of quarter sawn Sitka Spruce the surface was first sanded to a smooth finish with 320 grit paper. One half of the sample was then
burnished with a raw silk cloth and also a small smooth piece of wood (spruce) - rubbing vigorously in the wood grain direction.
This resulted in a polished surface that was - in turn - somewhat resistant to absorption of a water droplet. A water droplet left on the surface for
several minutes and then blotted dry showed some moisture absorption and consequent localised 'raised' wood grain but this - on drying - was easily
corrected by re- burnishing.
|
|
|
fernandraynaud
Oud Junkie
Posts: 1865
Registered: 7-25-2009
Location: San Francisco, California
Member Is Offline
Mood: m'Oudy
|
|
Well, well, well. THAT's interesting. Maybe in a way the chemical reactions that take place are not all that different between a touch of varnish and
burnishing. As you point out, the heat causes some changes. There are oils and resins in the wood. So maybe this burnishing brings out these resins
into a surface film?
|
|
|
jdowning
Oud Junkie
Posts: 3485
Registered: 8-2-2006
Location: Ontario, Canada
Member Is Offline
Mood: No Mood
|
|
The burnishing operation seems to compact the wood cells at the surface but does not seem to spread any resins (softened by the slight heat generated
by friction?) contained in some of the cells (at the surface) in a uniform manner as would be the case with a varnish coating.
Here are some macro images of the unburnished and burnished samples of Sitka Spruce - end grain and normal to the surface for comparison.
|
|
|
jdowning
Oud Junkie
Posts: 3485
Registered: 8-2-2006
Location: Ontario, Canada
Member Is Offline
Mood: No Mood
|
|
Both hard and soft burnishers were tested - boxwood, burlap (sack cloth) and raw silk cloth. The boxwood burnisher gave the shiniest results quickly
so is probably the way to go. A hard burnisher must be very smooth and well rounded and used with caution to avoid scoring or marking the sound board
surface.
(The makeshift burnisher shown here is the head of a boxwood peg blank)
|
|
|
jdowning
Oud Junkie
Posts: 3485
Registered: 8-2-2006
Location: Ontario, Canada
Member Is Offline
Mood: No Mood
|
|
I don't think that the resin in softwood can be transformed into a kind of varnish by the heat of friction generated by burnishing.
The most basic of varnishes (used by the ancient Egyptians) can be made by dissolving Rosin - the distilled and purified product of a raw wood resin -
in a solvent (traditionally either alcohol or turpentine) but not from the raw resin alone.
The attached image shows a sample of raw resin of Northern White Pine on my property naturally exuding from the tree to seal a wound - in this case a
cut branch.
The second image is a piece of of refined Rosin - transparent amber coloured and brittle. This is the basic resin ingredient for varnish making. I use
Rosin - diluted in alcohol - as a flux for soft soldering metals. I have never tried it as a varnish so out of curiosity will now run a test to see
how it turns out.
|
|
|
jdowning
Oud Junkie
Posts: 3485
Registered: 8-2-2006
Location: Ontario, Canada
Member Is Offline
Mood: No Mood
|
|
On further checking it would appear that primitive 'varnishes' made from raw pine resin or refined rosin dissolved in either alcohol or turpentine
(made from the distillation of pine resin) were only used in the past on cheap, utility wooden items as the coating is not durable and is subject to
scratching. This type of 'varnish' dries by the evaporation of the solvent leaving the resin on the surface of the wood whereas a traditional oil
varnish dries by chemical action to form a stable durable coating.
The test piece 'varnished, with Rosin dissolved in alcohol appears to have dried overnight but further tests with this material will not be
continued.
Although it is generally thought that the polished surface produced by burnishing a pine or spruce wood is caused by the heat of friction melting and
distributing any resin that might be present near the surface of the wood I think that this is something of a 'red herring' and probably incorrect.
There are in fact few cells containing free resin in Spruce (but more in Pine) and in any case wood containing visible pockets of resin would not be
selected for sound board material (and otherwise avoided even for general woodworking where possible)
The dark areas visible in the end grain macro images of Sitka Spruce previously posted are minute latewood cells not resin deposits. The lighter but
still barely visible cells in between are the larger earlywood cells.
The burnishing process would appear to work by compressing cells on the surface of the wood - the smaller the cells the greater degree of polish,
Hence the greater shine apparent in the burnished surface latewood cells compared to earlywood cells seen in the previously posted images. The finer
the grain of a wood the more uniform and higher degree of polish produced by burnishing - very fine grained sound board woods often being
preferred.
To further test this idea a piece of boxwood (hard and extremely fine grained with no visible cell structure) was easily burnished to a near mirror
finish with a piece of burlap cloth. Boxwood contains no resins or oils.
The other evidence pointing to cell compression as the mechanism is that polished burnished wood surfaces do not make good surfaces for varnish or
glue - the closed cells presumably preventing absorption of the varnish and glue leading to poor adhesion and weak joints.
|
|
|
fernandraynaud
Oud Junkie
Posts: 1865
Registered: 7-25-2009
Location: San Francisco, California
Member Is Offline
Mood: m'Oudy
|
|
Sounds logical enough, and compression can account for reduced penetration of blood, drool, ketchup and other stains.
|
|
|
bulerias1981
Oud Junkie
Posts: 771
Registered: 4-26-2009
Location: Beacon, NY
Member Is Offline
Mood: John Vergara Luthier Lord of the Strings instrument making and repair
|
|
interesting thread. I look forward to further conclusions. Perhaps I'll try this method on my next instrument if all goes well.
|
|
|
fernandraynaud
Oud Junkie
Posts: 1865
Registered: 7-25-2009
Location: San Francisco, California
Member Is Offline
Mood: m'Oudy
|
|
I was going to apply a bit of shellac to a soundboard that was saturated with plain egg white. Its (only?) flaw is that it's not very effective at
repulsing stains and liquids. Maybe burnishing THAT would be worth trying? How exactly would you do it, how fast/slowly, what kind of "pattern".
etc?
|
|
|
jdowning
Oud Junkie
Posts: 3485
Registered: 8-2-2006
Location: Ontario, Canada
Member Is Offline
Mood: No Mood
|
|
That's a good question fernandraynaud and one that I have been considering with respect to my experiments with egg white foam distillate recently
reported on this forum. Egg white foam distillate is not the same stuff as raw egg white so I am not sure how the two might compare.
Anyway my thoughts went back to the famous and innovative 19th C Spanish luthier Antonio de Torres who is said to have applied egg white (or was it
egg white foam distillate?) to his guitar sound boards. So if the earlier Spanish luthier tradition was to burnish the sound boards perhaps Torres may
have applied egg white (or distillate) and then burnished the sound board? Why he would want to do this (if he did) is another question - was it to
protect the sound board against staining or to gain some acoustic advantage or a bit of both?
Yesterday I found the Sitka Spruce samples used for the egg white foam distillate trials and burnished the coated surface to a shiny finish using a
boxwood burnisher. The cloth burnisher seemed to work as well.
A drop of water was applied to the surface, allowed to stand for 30 minutes, and was then 'blotted' with a paper towel. The damp surface after air
drying showed no sign of 'grain raising' or residue or staining. Another test with a drop of mineral oil showed staining after the oil was wiped off
indicating absorption of the oil into the wood cells.
An enlarged image of the water droplet shows air bubbles in the exposed but compressed early wood cells - a typical feature found in the earlier
experiments with egg white foam distillate. It is thought that the distillate retained in the wood cells forms a molecular film around air trapped in
the cells by the water preventing further absorption of the water into the wood.
In effect the properties of the egg white foam distillate to repel water seems to have been reinforced by the burnishing operation? However, this only
seems to work with water not oils or greases which immediately breaks down the protective film of the egg white distillate.
Although the egg white distillate is somewhat effective as a sealer against moisture it is not - as expected from the earlier experiments - effective
against oil absorption. Burnishing the surface appears to improve the moisture resistance to some extent but, again, does not improve resistance
against oil absorption.
So if this treatment is not completely effective as an impervious coating (like a varnish) does treatment with raw egg white (or egg white foam
distillate) followed by burnishing have a significant acoustic benefit - perhaps by creating a very thin (elastic?) skin of compressed wood cells
protected against swelling/deterioration (due to moisture absorption) by the egg albumin protein molecules?
The best way to find out practically is to run some 'before and after' subjective tests.
fernandraynaud already has an oud with the sound board treated with raw egg white. If the sound board is now burnished to a polished finish it would
be interesting to know if there is any perceived acoustic 'improvement' - sound quality, volume etc.
On my part I am currently working on a refurbishment renaissance guitar project on the forum. The sound board of this instrument is currently without
any kind of finish. The guitar will, therefore, be first tested acoustically with the sound board unfinished, then with egg white foam distillate
coating applied and then with the sound board burnished. The results will be digitally recorded and analysed using free 'Audacity' software for
comparison. Should be interesting.
Burnishing may be undertaken either 'with the grain' of the wood or 'across the grain'. Hard burnishers should be polished smooth without any sharp
edges or corners that might score or mark the surface of the soundboard and soft (cloth) burnishers should be free of any grit or hard matter that
might also scratch the surface. Apart from that it is a matter of learning from experience!
|
|
|
jdowning
Oud Junkie
Posts: 3485
Registered: 8-2-2006
Location: Ontario, Canada
Member Is Offline
Mood: No Mood
|
|
Just to capture another passing thought.
Oud makers working in the Turkish tradition believed that any oil or grease contaminating the sound board surface destroyed the acoustic properties.
It is said that the Turkish master luthiers of old themselves only would undertake the final preparation of a sound board allowing no one else to
touch it for fear of oil contamination.
Could it be that the 'final preparation' by the Turkish master luthiers included (secretly) treating the sound board surface with an invisible egg
white or distillate coating - then burnished for optimum acoustic properties - a coating that would be broken down and destroyed immediately on
contact with any kind of oil?
|
|
|
jdowning
Oud Junkie
Posts: 3485
Registered: 8-2-2006
Location: Ontario, Canada
Member Is Offline
Mood: No Mood
|
|
Note also that - in an attempt to speed things up - burnishing using a Dremel type tool fitted with a hard felt buffing wheel was tried. This was not
successful as the operation generated a white dust and tended to tear out the wood grain. Further suggesting that it is localised pressure of the
burnishing tool and cell compression not generated heat that does the polishing of the cells.
When burnishing heavy pressure on the sound board should not be necessary as the contact surface of the burnishing tool is small and so the area of
pressure is very localised. It just takes some time to properly burnish a large sound board area by hand.
|
|
|
jdowning
Oud Junkie
Posts: 3485
Registered: 8-2-2006
Location: Ontario, Canada
Member Is Offline
Mood: No Mood
|
|
Searching through my wood scrap box turned up another test piece from the egg white foam distillate trials - this time a piece of fine grained
(quarter sawn, 20 grains/inch) Silver Spruce with one part coated with raw egg white and the other with egg white foam distillate. Both surfaces were
burnished to a shine with a boxwood burnisher and tested with a water droplet for absorbancy. The attached images show the water droplets after 30
minutes. The droplets were then removed with a paper towel. After drying the locations of the droplets were still visible as dull marks. These dull
areas were restored to a full shine by further burnishing.
Now for a small diversion.
Browsing the web site of luthier Alexander Batov there is an image of the underside of the sound board of a Baroque guitar by Francisco Sanguino that
has been 'scorched' with a hot iron. Batov comments that this kind of sound board scorching can be found on other early plucked and bowed
instruments.
Testing a piece of Sitka Spruce with my mini electric iron (used for hide gluing operations) set at maximum heat and rubbed over the wood surface
caused minute areas of free resins contained in the wood cells to be drawn to the surface. The heat of the iron was not quite enough to scorch the
wood but - if it had been - those traces of resin would likely have been burnt off (i.e. removed).
Was this the reason for treating the sound board in this manner - to remove any traces of resin that might otherwise adversely affect the acoustic
response of the sound board?
Only one way to find out!
|
|
|
jdowning
Oud Junkie
Posts: 3485
Registered: 8-2-2006
Location: Ontario, Canada
Member Is Offline
Mood: No Mood
|
|
To test the effect that mechanical surface treatment (burnishing or scorching) might have on sound board material a basic test rig has been set up -
sample strips of quarter sawn Sitka Spruce sound board material 25.4 mm wide (1 inch) and 2.6 mm thick clamped at one end and a 100 gram load applied
at the free end the deflection at the free end being measured against a scale (5 mm spacing) marked on the side of the test rig. The overhang between
the clamped end and the load is 85 mm.
This arrangement - for the technically minded - is a simple cantilevered beam (i.e. the test piece) where the deflection of the free end is a function
of the load, the area and geometrical section of the beam, the length of overhang, and the modulus of elasticity or 'Young's Modulus'. Young's Modulus
is a measure of the stiffness property of a material so that a low value means that a material is less stiff than a material with a higher value. Low
stiffness equates with a greater beam deflection.
The environments tested were:
1) no treatment
2) burnished upper surface
3) lower surface scorched at 150° F (65°C)
4) lower surface scorched at about 450°F (232°C)
Measured deflections after 24 hours (to achieve stability) were:
1) 2.5 units deflection.
2) 3.0 units deflection
3) 3.75 units deflection
4) 5.25 units deflection
As all else remains constant except the stiffness of the wood it can be concluded from these results that any form of mechanical surface treatment
results in increased flexibility (or reduced stiffness) across the grain (i.e. across the sound board width) the maximum effect being achieved by
scorching the underside of a sound board - producing up to a 50% reduction in stiffness.
It may be that this was a technique used by luthiers to fine tune their sound board acoustic properties?
|
|
|
fernandraynaud
Oud Junkie
Posts: 1865
Registered: 7-25-2009
Location: San Francisco, California
Member Is Offline
Mood: m'Oudy
|
|
I dunno, John. Why would more flexibility make a soundboard more effective? on the contrary, floppier seems less resonant! and aren't there more
parameters than just elasticity?
|
|
|
jdowning
Oud Junkie
Posts: 3485
Registered: 8-2-2006
Location: Ontario, Canada
Member Is Offline
Mood: No Mood
|
|
I don't know either.
Perhaps 'floppier' across the grain (but not along the grain) equates to a sound board that responds faster and with greater amplitude to string
vibration? It might also favour a better bass response? On the other hand, I can't imagine why a luthier would deliberately scorch the underside of a
sound board (as they once did) if this action resulted in an inferior acoustic performance. I am sure that there is much about sound board acoustics
that is less than obvious or well understood.
The results are opposite to what I had anticipated as I imagined that scorching or burnishing the wood surfaces would have resulted in a harder,
stiffer sound board.
I should mention that all the test samples recovered slowly (after 24 hours +) to their original flat state after the load was removed. All except for
sample 4 that has recovered to a deflection of 1.5 units so has been loaded beyond its elastic limit.
|
|
|
fernandraynaud
Oud Junkie
Posts: 1865
Registered: 7-25-2009
Location: San Francisco, California
Member Is Offline
Mood: m'Oudy
|
|
Seems like sample 4 was also scorched beyond its limit. It's what is known as burning the toast 
|
|
|
jdowning
Oud Junkie
Posts: 3485
Registered: 8-2-2006
Location: Ontario, Canada
Member Is Offline
Mood: No Mood
|
|
.... and to think of the many guitars, ouds, lutes, violins etc out there that have similarly scorched rib surfaces - a consequence of the hot bending
process!
|
|
|
jdowning
Oud Junkie
Posts: 3485
Registered: 8-2-2006
Location: Ontario, Canada
Member Is Offline
Mood: No Mood
|
|
I have had confirmation from Françoise de Ridder - http://www.sinier-de-ridder.com - that in their experience sound board 'scorching' has been found on both early guitars and mandolins particularly
from Naples.
Note that until the rebellion of 1647, Naples was part of the Spanish Empire and so luthier traditions may also have been those of the Spanish guitar
and vihuela makers of the 16th C.
|
|
|
jdowning
Oud Junkie
Posts: 3485
Registered: 8-2-2006
Location: Ontario, Canada
Member Is Offline
Mood: No Mood
|
|
There appear to be a number of theories about why Baroque guitar sound boards were scorched with a hot iron (on the underside) - ranging from
modifying the acoustic properties of the sound board - to artificially 'aging' the wood - to discouraging wood beetle attack - to preheating the wood
prior to gluing.
The most prevalent opinion among luthiers seems to be that scorching the wood surface makes the wood stiffer and so is a technique applicable to early
guitar sound boards where there is no cross bracing (below the sound hole braces). This technique would therefore influence sound board response.
However, the tests previously reported in this thread indicate that application of heat - even below scorching temperatures - significantly and
permanently softens the wood and reduces its elasticity - quite the opposite effect to that which might intuitively be expected. After all application
of heat would dessicate or dry out the wood and make it harder, more brittle and hence stiffer wouldn't it? Unless I am missing something in these
admittedly very rough and ready tests this does not appear to be the case.
Not that making the sound board less stiff across the grain (the affect along the grain has yet to be determined) would necessarily have an adverse
effect on sound board response and acoustic properties - it could be quite the opposite. Luthiers in the past may have discovered that scorching does
improve sound board response of guitars (better bass response?) but for a different reason than they imagined.
One image recently seen of a Baroque guitar sound board that has been scorched shows that the hot iron has been applied quickly and in a random
fashion - rather like a pencil 'scribble' drawn across the grain over the sound board area. This might have been to ensure that the applied heat was
not applied for a long enough time to cause separation of any glued joints.
These trials measured the effect on wood samples that were uniformly heat saturated to the extent that minute pockets of resin in the wood were melted
and drawn to the surface. So the next question is if the heat is very quickly applied in a random, non-uniform manner does this result in stiffening
of the wood? The answer is no it does not. A further test with a 5 mm wide light scorch mark down the centre of the test piece gave a deflection
under a 100 gram load on the test rig of about 4.25 units after 24 hours. This is a greater deflection than test piece #3, previously reported, that
was uniformly 'scorched' to a temperature of 150°C.
Clearly there may be some variables in the methodology that might affect the results to some degree - particularly Relative Humidity (greater RH =
greater deflection?). However, the measured RH during these trials has remained in the range 72% to 77% so this should not have greatly affected the
outcome if at all.
It is anticipated that a similar but reduced permanent softening will also occur along the grain (i.e. along the length of a sound board) so will test
that theory next.
|
|
|
jdowning
Oud Junkie
Posts: 3485
Registered: 8-2-2006
Location: Ontario, Canada
Member Is Offline
Mood: No Mood
|
|
A retest has been carried out on the 5 samples by applying the 100 gram load and measuring deflection after 24 hours and the recovered no load
deflection.
Interestingly the heat treated and scorched samples that were softened to the point that they permanently yielded under load (and so did not return to
their original zero deflection unloaded) when subjected to another loading cycle only deflected the same amount as the untreated sample.
So, for example, the deflection of the untreated sample was 2.5 units under load - returning to zero deflection unloaded.
The original 'scorched' sample #4 after loading had a permanent set unloaded of 1.5 units. When this sample was cycled with another loading of 100
grams, the sample deflected only 4 units after 24 hours and then returned to its unloaded deflection of 1.5 units - total deflection 2.5 units, the
same as the untreated sample.
It might be concluded from these limited test that if a softened 'scorched' sample is loaded beyond its yield point it then becomes stiffer - at
least, returning to its original untreated stiffness.
So, perhaps the reason that luthiers scorched the sound boards of their 'Baroque' guitars is analogous to what is known in engineering terms as
'Stress Relieving' of a material - a heating process that removes local (and potentially adverse) stresses in a material? A stress free sound board
being a better sound board acoustically?
Just a thought.
|
|
|
SamirCanada
Moderator
Posts: 3405
Registered: 6-4-2004
Member Is Offline
|
|
Interesting... I will also try this next time I make a top.
@samiroud Instagram
samiroudmaker@gmail.com
|
|
|
jdowning
Oud Junkie
Posts: 3485
Registered: 8-2-2006
Location: Ontario, Canada
Member Is Offline
Mood: No Mood
|
|
Note that evidence of this technique has so far been found on 'Baroque' guitars and mandolins.
Not sure if it has ever been found on old oud or lute sound boards.
|
|
|
jdowning
Oud Junkie
Posts: 3485
Registered: 8-2-2006
Location: Ontario, Canada
Member Is Offline
Mood: No Mood
|
|
To complete these brief trials a sample (same dimensions as samples previously tested) was prepared from the the same piece of Sitka Spruce sound
board material - this time with the grain running along the length of the sample rather than across. This represents a measure of stiffness along the
length of a sound board.
Under a load of 100 grams the untreated sample deflected about 0.25 unit compared to about 0.5 unit deflection when the sample was scorched. So - not
unexpectedly, despite the small size of the samples - this indicates, from the deflection measurements, that the stiffness along the grain is
significantly greater than stiffness across the grain for both untreated and scorched samples.
|
|
|
![[*]](images/xmbprobrown/default_icon.gif){kind=icon}
