darkening the violin wood

[robertdo]

Hi LinkMan,  ask and ye shall receive.  Here's the beef! (I remember those commercials,  she was cool )

Here we have a cooked varnish with no coloring materials added, on top of white (un-darkened) spruce.  The varnish layer is something near a tenth of a mm maybe slightly more.  Something interesting I notice in the closeup photo is the bright line of light that seems to glow just under the varnish layer.   As I recall I think there is a ground on it but that only seals and in maple it enhances chatoyance but it doesn't darken significantly. This pic was taken in sunlight.  Inside in room light it's less interesing, more of a brown but I think that pic is on my other computer.     Don't ask me if I can repeat the recipe...      I hope I can.

 

P.S.  Found the pic of it in normal room lighting.  Third pic below. 

I think on the contrary that the colour on the third picture is a much nicer, warmer, more interesting brown (it looks like the tickness is more like 0.5-1mm though... )

[MikeC]

I think on the contrary that the colour on the third picture is a much nicer, warmer, more interesting brown (it looks like the tickness is more like 0.5-1mm though... )

It is kind of a nice looking brown but the camera seems to make things look better,  In bright sun though it looks like it's on fire and I liked that. It kind of depends on the angle too, looking straight on you can see the orange tint but at an angle looking through more thickness it's more brown looking, brown after all is the same thing as dark orange,   and then there's the problem of getting accurate photos with a cheap digital camera.   I have to admit it would probably look better if the wood underneath were darker instead of so white.  Suntan would be good for sure.

Got my curiosity going now,  I'm going to slice a little off and measure it with a micrometer. 

 

 

wow you're right on the thickness,  my visual estimate was way off.   I mic'd it and it's .013 inch which is 3 tenths of a mm. 

I notice also after aging about a year that it tends to be kind of chippy 

[Christopher Jacoby]

Mike, that's good looking stuff.

[KGE]

Returning to the initial openning questions by Peter, I have read the mentioned book even though at the time I did not get so interested in the question of the groundcolor.

I used the "Magister ground " and thought the  proplem of getting the necessary dark color is solved.

Unfortunately I am back at the beginning also searching for the best way of darkening the wood before the actual varnish.

Does anybody know how to isolate the groundcolor from the varnish, I noticed that many substances react with the varnish and disappear. Also the Magister ground without the sealer  (also tried kaliumnitrit).

Does anybody know if for example a casein coat works as a sealer?

Can anybody point me in the right direction...

 

[joerobson]

There are things we know and things we don't know.

 

We don't know specifically what was used on the classic Cremonese instruments before they were varnished.

We do know how the appearance of modern techniques compares to the classic finishes.

 

We do know that the analytical methods used in the Stradivari Varnish book are excellent science.

We do not know if the interpretations of these findings are correct or not.

 

We do know a lot about what happens to the wood as it ages under the varnish.  Unless the instrument was never exposed to light, there are 3 changes that occur.  The initial change is one we are familiar with: the reaction to UV light.

The result is a deepening of the natural colors of the wood and an increase in the contrast of those colors.  The second is the reaction to visible light, specifically the blue/purple part of the spectrum.  Exposure to visible light, over time, results in a minor decomposition of the hemicellulose.  One of the by-products of this decomposition is a chemical compound that has a specific color.  This color, though defined as a yellow or a gold, is not the yellow often seen in photographs of violins.  When the surface of the instrument is worn, particularly if this wear is severe, the yellow color further decomposes into a gray/green.

 

Joe

[Michael_Molnar]

Well said, Joe, especially the issue about the yellow color. 

 

I think B&G did some good science. The only fly in their ointment (varnish?) is the suggestion of a sealant like casein. As we discussed, this is most likely nothing more than contamination from handling glue.

 

Mike

[pbelin]

To answer the question,yes,casein works very well as a sealer. Either as an emulsion with oil or alone, I've had good results with it.

[Don Noon]

I use casein/ammonia fairly often, mostly on necks. It seals fairly well... but only after 3 or 4 coats, with light sanding to knock down the raised grain.

As a single coat, it doesn't seal enough to prevent thin oil varnish from soaking in, so when I do use it on the body, I also use some other kind of ground before the varnish. Again, there's the raised grain to consider. Visually, it may darken the wood slightly (due to the ammonia), but not much. I think the appearance would be compromised if it was used heavily on the body; a single coat doesn't seem to do much either way.

[MikeC]

 When the surface of the instrument is worn, particularly if this wear is severe, the yellow color further decomposes into a gray/green.

 

Joe

Joe do you mean that in some instances the ground has been worn through and the wood turns grey green?   Meaning the ground is wear resistant but not immune to wear ?

 

 

Something that will turn wood greenish gray is anything base like a water baking soda solution for example,  it will on maple anyway, haven't tried it on spruce.    Someone once mentioned using iron sulfate to darken wood but I didn't like the results. 

[John Harte]

Does anybody know if for example a casein coat works as a sealer?

 

 

Yes it can work as a sealer but there are certain characteristics that have concerned me.

 

The casein glue sizes that I have tried seem quite aggressive in terms of what they do to the surface of the wood.  Minute cracks in the wood surface structure seem to result.  (I seem to recall that someone else posting on Maestronet has mentioned this - Maybe Michael Darnton???)

Attached is a photo that hopefully illustrates this, at least to some extent...

 

Casein glue, at least my samples that I have examined under scanning electron microscope, are quite porous.  This fits in with Don's comments in post #58.

(See attached SEM photo.)

A hide glue size looks significantly less porous.

 

I've also attached a photo of a casein glue and oil varnish emulsion.  This looked better under varnish than the straight casein glue size.

 

The casein glue featured in the attached photos could have been made from commercially available casein powder, cheese or milk curd.  I don't remember what I used.  Whatever was used would have been mixed with quicklime in an attempt to create a film that was more resistant to moisture attack.  (I have also used borax but not here.)  Quicklime, as opposed to the casein component, may be the culprit causing the wood surface damage.  Don or others may be able to comment on this.

 

[Michael_Molnar]

Hi John,

 

Could you better describe each photo?

 

Mike

[John Harte]

Hi John,

 

Could you better describe each photo?

 

Mike

 

Hi Mike,

 

The first two photos are of old samples from 20+ years ago.  Both are sections on the same piece of spruce.

These photos were taken through a cheap 15x illuminated magnifier.

The first photo features a casein glue size on spruce while the second features a casein glue and oil varnish emulsion size on spruce.  I would guess that the emulsion is an oil varnish in casein glue emulsion as opposed to a casein glue in oil varnish emulsion.  Unfortunately my pencil remarks on the back of the sample don't state which.

 

These photos were taken very quickly yesterday using a camera that I am unfamiliar with.  (A new purchase...)

I will take a couple more photos of these samples today and see if I can get better shots.  The light direction is significant in terms of what is captured and maybe light coming across the grain as opposed to along the grain will produce more interesting images.  If I can get anything better I will post these.

 

The third photo is a digital photo of a standard black and white photo.  It features a casein glue layer on maple viewed through a scanning electron microscope.  In preparing this particular series of samples I built up quite thick layers, in each case well over 10µm thick. 

(Following publication of the Barlow and Woodhouse particulate ground research, I prepared a series of varnish and ground samples that I was able to view through scanning electron microscope at the local university.  This was back in 1990 - 92.  Scanning electron microscopes are much better now!!!) 

 

Attached are another couple of SEM photos from the same series of samples.

One features casein glue on spruce and the other hide glue on spruce.  As you will see, the hide glue is significantly less porous than the casein glue.

 

John

[John Harte]

Mike, I have tried to take better photos but haven't been as successful as I would have liked.  For what it's worth, attached are the latest efforts.

 

You should hopefully be able to see the labelling associated with each photo.  The main addition to the earlier shots are those taken with the light source directed across the grain.

 

The casein glue sample lacks the visual clarity of the emulsion sample resulting in the appearance of the medullary being more muted or veiled.  This is best observed via comparing the two sample shots where the light has been directed along the grain.  Capturing the damage to the upper wood structure in the casein glue sample has not been overly successful but hopefully the two photos of this sample give some hint of the damage.

 

As I have mentioned, the damage could be due to my use of quicklime.  This may not be an issue where ammonia has been used.

 

John

[Peter White]

I like Joe Robson's comments that the Stradivari Varnish book has good science, but I can not be sure that the conclusions drawn by the authors are correct.

They see things --or detect substances and varnish methods--that no one else has seen or photographed, as far as I know.  They are very careful in their interpretations but they seem not to hesitate in their conclusion that Strad stained the wood after sealing it.  Assuming that this single stain layer is really there, and that it was perhaps colored either brown or red, and that it was applied probably as a liquid, what do you think it might have been?

The authors are pretty sure that a protein sealer was used.  Probably casein. Let's assume that's the case. Then the violin was sealed and one coat of clear and one coat of colored varnish was applied.

To me this sounds like the kind of efficient varnish method that a hard working, very productive, very smart man would have used.

 The part of the process that is a NEW idea to me is that the violin was stained after it was sealed.  And this staining made the next two steps very easy for a man who was carving, producing, experimenting with forms, making all kinds of instruments, cases and God know what else etc.  He had to have the varnish system down to basic, consistent, reliable, aesthetic steps--four relatively quick applications.    This was accomplished especially if the resin to oil ratio resulted in a quick drying and hard varnish. This theory makes logical sense for a man who wanted to produce a LOT of work.

Staining made this quick and attractive varnish process possible. Now, what do you think he sealed the stained violin with?

I would love to carry on the conversation about this stain layer. 

Thanks,   Peter

[Roger Hargrave]

I use casein/ammonia fairly often, mostly on necks. It seals fairly well... but only after 3 or 4 coats, with light sanding to knock down the raised grain.

As a single coat, it doesn't seal enough to prevent thin oil varnish from soaking in, so when I do use it on the body, I also use some other kind of ground before the varnish. Again, there's the raised grain to consider. Visually, it may darken the wood slightly (due to the ammonia), but not much. I think the appearance would be compromised if it was used heavily on the body; a single coat doesn't seem to do much either way.

 

Yes it can work as a sealer but there are certain characteristics that have concerned me.

 

The casein glue sizes that I have tried seem quite aggressive in terms of what they do to the surface of the wood.  Minute cracks in the wood surface structure seem to result.  (I seem to recall that someone else posting on Maestronet has mentioned this - Maybe Michael Darnton???)

Attached is a photo that hopefully illustrates this, at least to some extent...

 

Casein glue, at least my samples that I have examined under scanning electron microscope, are quite porous.  This fits in with Don's comments in post #58.

(See attached SEM photo.)

A hide glue size looks significantly less porous.

 

I've also attached a photo of a casein glue and oil varnish emulsion.  This looked better under varnish than the straight casein glue size.

 

The casein glue featured in the attached photos could have been made from commercially available casein powder, cheese or milk curd.  I don't remember what I used.  Whatever was used would have been mixed with quicklime in an attempt to create a film that was more resistant to moisture attack.  (I have also used borax but not here.)  Quicklime, as opposed to the casein component, may be the culprit causing the wood surface damage.  Don or others may be able to comment on this.

 

 

This is an interesting thread. I wrote about casein and casein emissions in the early 1980's. Making casein with quick lime was probably the method used in classical times. Amonia was not freely available. It has the advantage (as a glue [i believe it is technically a cement]) of being more waterproof. With quick lime it creates better emmusions than ammonia/borax mixes and it emulsifies with a lot of things other than just varnish and resins. 

 

Casein has a longer working time than gelatin based glues and as such can be adjusted. It can be used in fairly cold conditions without gelling. In hot dry weather there is no need for constant fire in the workshop. When dry it is also resistant to oils and most common solvents, which is why it was used to join the wooden panels used by painters. 

 

Today we talk about the reversible nature of gelatin glues as being advantageous, but who wants a center joint to pop? And opening rib to table joints glued with casein is not as difficult as it sounds. It was not analyzed at the time, but I opened a del Gesu (ribs to back) in the 1980's that had never been opened. Under the blocks the holding glue was whitish and dry and did not react to water. 

 

Also if anyone has ever removed classical ribs from a corner block they will know that the usual methods for releasing animal glue do not work. This is especially difficult with cellos. It explains why, in the past, so many blocks were replaced when repairs to the ribs were done. 

[Kev Chanot]

Has anyone been able to scientifically analyse,and accurately date classical Cremonese varnish.If so,have these procedures been carried out on The Messiah Strad,for example?

[joerobson]

Mike C.,

We need to carefully examine the source and position of the colors.  The gray/green color I am talking about is in, not on the wood.  As the varnish wears off, the protective layer is thinner, so the transition from the "gold" to the "gray/green" happens more quickly.  The ground is quite protective, but not invincible.  The most graphic example would be beard wear.

 

John,

Excellent photos.  Thank you.  This kind of micro-scale information is rare and so useful.

 

Kev,

Yes.  Brigette Brandmair's work in "The Stradivari Varnish Book".

Joe

[Kev Chanot]

Joe,

 

Many thanks ,I may invest in "The Stradivari Varnish Book".

[Don Noon]

My tests of casein/ammonia, at thicker levels, showed similar results to John Harte's: some clarity is lost. If I only use a light wash coat, where there is no appreciable thickness, the subsequent ground appears to be able to get into the wood and do its job visually.

Acoustically, my early sample testing of casein/ammonia looked pretty good; stiffness was slightly increased, and damping was slightly decreased. It wasn't a huge change, and I'd guess the effects would be overpowered by the oil varnish. However, I still use casein/ammonia occasionally on the inside surfaces, especially if I have low density, processed wood that might benefit most from surface reinforcement.

[LinkMan]

 

This is an interesting thread. I wrote about casein and casein emissions in the early 1980's. Making casein with quick lime was probably the method used in classical times.

The case for casein

Roger Hargrave proposes a casein link with Stradivari.

[Michael_Molnar]

John Harte,

 

Great photos and explanations.

 

Thank you!

[John Harte]

This is an interesting thread. I wrote about casein and casein emissions in the early 1980's. Making casein with quick lime was probably the method used in classical times. Amonia was not freely available. It has the advantage (as a glue [i believe it is technically a cement]) of being more waterproof. With quick lime it creates better emmusions than ammonia/borax mixes and it emulsifies with a lot of things other than just varnish and resins. 

 

 Casein has a longer working time than gelatin based glues and as such can be adjusted. It can be used in fairly cold conditions without gelling. In hot dry weather there is no need for constant fire in the workshop. When dry it is also resistant to oils and most common solvents, which is why it was used to join the wooden panels used by painters. 

 

Today we talk about the reversible nature of gelatin glues as being advantageous, but who wants a center joint to pop? And opening rib to table joints glued with casein is not as difficult as it sounds. It was not analyzed at the time, but I opened a del Gesu (ribs to back) in the 1980's that had never been opened. Under the blocks the holding glue was whitish and dry and did not react to water. 

 

Also if anyone has ever removed classical ribs from a corner block they will know that the usual methods for releasing animal glue do not work. This is especially difficult with cellos. It explains why, in the past, so many blocks were replaced when repairs to the ribs were done. 

 

 

Roger, thank you for your very interesting comments!

Did you reglue the opened del Gesu rib to back seam with hide glue?  I presume that hide glue will satisfactorily adhere to remnants of casein glue??

 

Your article in The Strad, along with Raymond White's comments, fired my interest in casein. 

Thank you for your ongoing generous sharing of knowledge and information!

 

John

[John Harte]

My tests of casein/ammonia, at thicker levels, showed similar results to John Harte's: some clarity is lost. If I only use a light wash coat, where there is no appreciable thickness, the subsequent ground appears to be able to get into the wood and do its job visually.

Acoustically, my early sample testing of casein/ammonia looked pretty good; stiffness was slightly increased, and damping was slightly decreased. It wasn't a huge change, and I'd guess the effects would be overpowered by the oil varnish. However, I still use casein/ammonia occasionally on the inside surfaces, especially if I have low density, processed wood that might benefit most from surface reinforcement.

 

Don, thank you for your comments on the acoustic properties of casein/ammonia.

 

I should mention that the casein glue layer in the SEM photos that I posted is far thicker than that featured in the photos in Post #63.  From what I recall these involved a single relatively thin application of casein glue and the casein glue and oil varnish emulsion.  It looks as if these may also have had a very thin layer of over varnish applied to bring out whatever visual effect each produced.  

 

I can imagine that your casein/ammonia option might produce a clearer result than casein and quicklime.

 

John

 

[Patrick KREIT]

For the casein to be good, you have to make it yourself from non-fat fromage blanc (fromage frais or curd) and add lime or borax.  

Casein with borax is perfectly translucid. Casein with lime can yield a slightly yellow-green tint but with time, the color disappears.

 

Traces of casein paint have been found in the tomb of Tutankhamen and on the frescoes of Pompeii. Artists used it during the Renaissance and still use it today in tempera painting. It was widely used between the 17^th and 19^th centuries, when painted furniture was highly popular. In the form of glue, casein can be combined with mineral pigments or with iron oxide, as in artists’ paint.

 

Casein is a glue obtained by mixing curd (fromage blanc or fromage frais) with lime or borax. It is composed of proteins (calcium, phosphoric acid) more complete than those found in gelatins; hence their different characteristics, though both are of animal origin. This binding glue has been known and appreciated for its reliability from Antiquity through the Middle Ages and the construction of cathedrals. It generally creates a very solid, hard film, though breakable on flexible supports.

 

It is ideal for hard, solid surfaces presenting low or high porosity, such as plaster, lime, and especially wood, because it is worm-resistant. Lime-casein lets the surface breathe if need be, but is sufficiently waterproof to resist sponge washing.

 

Borax is a monovalent base that leaves a reversible film as does bone glue. A divalent base such as lime allows the formation of an insoluble casein film. Lime-casein improves water-resistance. It must be applied within one or two hours after preparation. Lime is rot-proof, but powdery and friable, whereas casein is solid and worm-proof.

 

This glue can be used to protect both the inside and outside of the violin body. Casein, an emulsifier, can be combined with linseed oil, castor oil, glucose, or glycerin to make it even more flexible and less permeable, thus dramatically limiting moisture resumption. Once coated with casein, the wood’s moisture content seldom exceeds 9% in the presence of prolonged high ambient humidity. This sizing also has proved to be an acoustic filter, yielding purer sound.

 

Due to wood’s recurrent hygroscopic instability, Italian violin makers exposed their violins to the sun before varnishing them. This procedure stabilizes the frequency of the top and back plates by preventing resumption of moisture at the highest level.

The surface of the wood exposed to sunlight or UV light becomes oxidized. The chemical reaction between the lime-casein coating and the oxidized wood instantaneously gives the latter a fine honey color. It is impossible to imitate this reaction by using other substances. Translucent and practically inalterable, casein accentuates the grain of the wood. The consistency, plasticity, and color of this primer resemble those of the old Italian violins. The advantage of casein is that it is applied cold and is quick-setting.

 

Casein with lime or borax can be mixed with iron oxide (yellow, brown, or red) to give a more intense color to the prime coat before varnishing.

 

Below 12 °C (53.6 °F), casein forms lumps (instead of glue), so it is unsuitable for winter use outdoors or in an unheated workshop. Collective memory must therefore have understandably disregarded casein in favor of bone glue that was easier to use and transport once it appeared on the market.

 

If the lime is old, the reaction with the curd fails to occur. The lime must be heated in a pan so that water evaporates from it. Let the lime cool before using it. Careful: it is practically quick lime!

The curd/heated lime mixture must not be used immediately, or the reaction will be too strong upon application to the wood and will yield a dark brown color. Instead, it must be set aside for approximately 30 minutes to let the quick lime become slaked lime.

 

Test the sizing on maple and fir samples before using it to coat the wood of an instrument. Depending on the reaction time of the casein glue being prepared, the result will be a lovely color of old wood or absolutely translucent sizing. Lime is an aggressive substance, so gloves must be worn while using this sizing in order to reduce the risk of dermatitis on sensitive skin.

Add 5% castor oil, linseed oil, glycerin, or glucose (honey or grape sugar) to the casein to make the size less permeable.

 

The casein can be applied on the wood without previously filling the pores, for it adheres perfectly and uniformly to the surface, thus decreasing anisotropy in the materials, particularly when it is also applied to the interior of the sounding box (including the ribs).

The quality of the violin's sound is enhanced. Casein acts as an acoustic filter. The sound is purer. A silica (quartz) or colloidal silica filler can be added to the casein if prior sealing of the wood pores is not desired.

 

Application of a casein-oil emulsion on wood presents no problems. This size can be applied equally well with a brush or a cloth. Casein size made with hydraulic lime can be used for 1 to 2 hours, whereas with borax, it can be used for approximately 24 hours.

 

Formulation of the casein depends on its use:

 

-          Casein with 10% lime has a pH of 11: vegetal pigments change color.

-          Casein with 45/50% lime has a neutral pH (7): vegetal colors do not change.

-          Casein with 10% borax has a neutral pH (7): vegetal colors do not change.

 

Casein with 10% lime yields a prompt reaction on wood that has been exposed to the sun or to UV light. It gives the wood a lovely, perfectly even, inimitable honey color. Casein is plastic and dries quickly. Another advantage is that it is applied cold. To lower the pH, add a few drops of lime juice or balsamic vinegar.

 

Lime casein sizing should be sanded after 20 minutes’ drying time; borax casein, after 1 hour. The second coat, however, must not be sanded, for it is very thin. It is preferable to apply it with a cloth and rub it in to obtain even distribution without blisters or overlapping marks.

 

Dye can be applied on top of the casein. All dyes used in violin making adhere well to this sizing.

 

www.kreitpatrick.com

[actonern]

_{I was under the impression that Brandmair and company specifically looked for and only found trivial amounts of protein in their analysis of classical varnish films... not enough to suggest an intentional coat as part of a finishing system.}

 

_{Is that wrong?}