Archive for July, 2016
Varnishing Procedure; Part Two
Leveling the early coats of Varnish
In the last post, we were looking at the third coat of varnish. It looked pretty raw, still. You can imagine how much brighter the white wood might have looked under the varnish, without having stained it first. Some people hang their newly made instruments out in the sun, so that the ultraviolet light oxidizes the wood somewhat, and darkens it quite a bit. (Especially those who live where the sun actually shines: I, on the other hand, am in western Oregon, where the sun is a novelty… I do have a U.V. cabinet, but have not been very impressed with the results, so I haven’t been using it.)
Something I have occasionally done is to apply a few coats of varnish, then strip it all back off, down to the wood. It does not affect the mineral ground, nor the sealer, but it does affect the foundation color of the wood, much improving the finished look. But it is a lot of extra work…so I have only done it a few times.
Regardless of how I accomplish the color I want, the next step is to level the varnish.
I use fine abrasive paper (400-grit), to gently remove any bumps, and to leave a matte-finish all over the violin. I am very careful to not fully remove the texture of the raised grain, left from the coffee stain, but only to scuff-up and smooth the varnish itself. The wood should have been the way I wanted it before I began varnishing. I prefer that my violins still have some “wood-texture”. Some people like the mirror-bright plastic look, but I don’t. On a guitar, that is one thing. These are supposed to be concert violins… another thing entirely.
The things I am especially vigilant about are sags, drips, runs of any sort, and brush-hairs, etc. If there is an actual flaw in the varnish, I want to remove it before I move on. But, many times, I will deliberately (later) fill any “texture” with a darker substance, again emulating the “wear and dirt” of old instruments. I limit this to pretty minor stuff, usually. I have known fellows who took “antiquing” much further; as in, breaking wood off corners, etc, and abrading away varnish right into bare wood, then rubbing dirt into the wood. Sorry…I am not going to do that. 🙂
The important thing is to realize that any anomalies in texture will be highlighted by the varnish, not hidden. Something as small as a brush hair or a speck of sanding dust will become terribly obvious if one continues to varnish over it. (I learned this the hard way….) The varnish piles up around it and literally makes a mountain of a molehill. It pays to take the trouble to get things smooth before you move on.
Completing the Varnishing Sequence
I try to make sure that each layer of varnish actually contributes what I want; building toward the whole effect. So, I have to decide when to do any shading or antiquing. I have to decide how much extra color to apply in the darker areas, and what the overall color impression is to be when a person first sees the violin: Dark? (How dark?) Antiqued? (How much?). So, there usually comes a point where I decide the violin is dark enough, and I add two or more additional coats of yellow or clear varnish to increase the feeling of depth, and to shift the overall color impression back toward golden. Sometimes that works well; sometimes it does not. Occasionally I have had to give it up, strip the fiddle, and start over. 🙂
So, here is the rest of the varnishing, in the order it actually happened:
I neglected to take a picture of the fourth coat of varnish, but it wasn’t all that different from the third…same varnish, etc. So, here is the
After the varnish dried overnight, I gently sanded it again, then added a sixth and seventh coat. Here are the photos:
This one was mostly a case of adding strong color in the areas I wanted darker, and virtually none in the areas I want lighter. There was a tendency to streak, so I would feather the transitions out, using a fine sable brush, moistened (not wet) with alcohol. It worked well, but, if you look closely, you can still see streaks. These will be smoothed before the seventh coat, either by the alcohol method or by sanding. It is important to realize that, on the flamed maple, your brush-strokes should follow the flame, and on the spruce, follow the grain. In both cases, the natural lines in the wood tend to disguise any leftover brush-marks.
I allowed the sixth coat to thoroughly dry, then sanded it with 400-grit, well-worn abrasive paper, and applied a final “color-coat”: after this one, I will only add yellow or clear varnish, to accentuate depth or shift the color range toward gold.
I doubt you would really see a great deal of difference in this coat. The photographs are not very accurate in terms of color– I guess the flash does that. They are too red, and too bright-colored. It is not really that bright, but the color is shifting back toward gold, as I apply the last two or three coats in yellow varnish.
So: I think that is about as far as I will take the color, for right now. I will very likely decide later that I want it darker, and, if I do, I can always add more color. But I think I will go ahead with the saddle, fingerboard, nut, etc. and see how I feel about it. If I decide it is fine the way it is, then I will go forward with set-up. If I want it darker, I can add color at that point, and another yellow or clear coat, and then do set-up.
So, that is as far as it is going tonight.
Thanks for looking.
Varnishing a New Violin
Finishing Sequence: Sealer coat first
In the last post, I showed the photo of the violin with just the turpentine/pitch sealer coat in place, and not totally dry. Remember that this was applied over a coffee stain, and a mineral ground that I had rubbed into the wood.
When that coat was finally dry, I checked for any distortions (from the coffee stain, I guess) and corrected them with plain water, just moistening any low areas with a damp rag, and watching them come back to normal. Since I just barely moistened those areas, they stayed in the correct position after drying the second time. This has been a rare occurrence in my experience, but I was grateful that it turned out to be a relatively easy fix.
Then I sanded lightly with worn 400-grit, to remove any bits of loose debris and/or any little fibers of wood that had lifted above the smooth surface. I had already done this after staining (both times), but it always pays to go over things again.
First Coat of Varnish
The first coat on this instrument was a very blond spirit varnish; not my usual. I am not entirely pleased with the result, but it is acceptable.
Scond Coat: Darker Yellow
I sanded it lightly, again, and then applied a darker yellow varnish. Fortunately, spirit varnish dries very rapidly, so I can sometimes get two or three coats in one day, early in the sequence. As the varnish gets thicker, it dries more slowly. I assume that this is because it can no longer soak into the wood at all, so every bit of the drying has to happen from one side of the varnish film; but perhaps there is more to it than that. At any rate, as the instrument nears completion, I have to allow longer time for drying. The other side of the “fast-drying” coin (or two-edged sword) is that it is extremely sensitive to the next coat of varnish, as the solvent in the new coat can easily lift the previous coats, forcing me to completely start over, in some cases. I really need to be patient, and work carefully, applying many thinner coats, rather than fewer thick coats.
Third Coat: Red-Brown varnish
After the yellow varnish dried I began adding (several coats of) a darker red-brown varnish, allowing each coat to dry, and sanding lightly between coats, to make sure the finished result is good. This is the first coat of the red-brown varnish, so, actually the third coat, overall. It will get at least five or six more coats of varnish before it is done, but the differences become less and less obvious, as the varnishing nears completion. I am enjoying looking at the beautiful European maple and spruce. I ordered this wood from International Violin Company, in Baltimore.
As I continue to add coats of varnish, I am keeping an eye on the general “flavor” of the instrument. I may skip certain areas for several coats, to leave the varnish thin in those areas. I deliberately try to emulate the look of some of the more gently-used “Old Master” instruments. I am not attempting to “fake age”, so much as attempting to capture some of the charm and appeal of those intruments. If anyone has a question about my motives, all one has to do is check the label: every instrument is signed, numbered, and dated. The date on the label is the day I actually closed the corpus, so, perhaps a few weeks prior to final completion, but no more than that.
I also may switch back to a yellow varnish at some point, to shift the color back toward gold, rather than just a red-brown. And, occasionally, I have stripped everything back off and started over. As the original maker, I have that option. and, invariably, the result the second time was far better.
I will post more varnish photos as the violin nears completion.
Once the varnishing is complete, I will replace the Fingerboard and begin the final fittings and set-up of the violin.
- Tuning pegs
- End Pin
Hopefully, all of that will be covered in the next post.
Thanks for looking.
Finishing My Newest Violin
For every violin, viola, five-string fiddle, cello or double bass I build, the finishing process is nearly the same. After completing the last of the purfling, the order is as follows:
- Channel (the concave curve in which the purfling is nearly centered, but which must fair smoothly into all the other surfaces, concave or convex.)
- Final edgework (making the outer edge (and inside of the plate) curve together into the channel and arching)
- Varnish prep (final fairing-in of all curves and elimination of all bumps and unwanted ridges)
- Water-based stain (usually two or three applications)
- Mineral ground (one application)
- Seal-coat (one application)
- Varnish (numerous coats)
- All set-up issues: saddle, fingerboard, nut, pegs, soundpost, bridge, endpin, strings, tailpiece and (on the smaller instruments) chinrest.
The Channel, and Final Edgework:
I scribe a margin around the entire outer edge, usually about 1.6mm in from the outer edge. This will serve as the “limit” for the channel. I cut the channel to the desired depth, using gouges first, then small planes and scrapers, bringing the smooth curve up to match the scribed limit. Then I file the outer edge so that it curves in from the center of the edge of the plate up to meet the scribed margin of the channel. The other margin of the channels are scraped so that they fair into the curve of the plates themselves. I true-up all my lines using scrapers, files, and very fine abrasives, and then I am ready for final varnish preparations.
On my last post, I had just completed the back purfling. That seemed a good place to “call it a night”.
Then I began preparations for cutting the channel, by scribing in the boundary line:
I have a particular spot on my basement stairs, where I can get a very low-angle, not-too-bright light source, which casts shadows across the face of the instrument, and makes is easy to see irregularities. So, the last thing I do is stand there, holding the fiddle and rocking it very gradually, to see any shadows that indicate a rough spot or an irregular shape. When I find one, I gently scrape it away with a very sharp scraper. I want to move as little wood as possible at this point in the build: I already have the arching and graduation just about exactly where I want them to be. This is strictly fine-touch finish-scraping.
Some people are very particular about not using ANY abrasives. Sorry– I do use them, in a very limited way, but only at the very last part of the build. The edges of the spruce plate, for instance, are so soft that a scraper is simply too aggresive for final shaping, especially after I raise the grain with my water-based stain (strong coffee, in my case…yes, and I drink the leftover “stain.”) So I use a very fine (worn 400-grit) abrasive paper to make the edges smooth and pretty. I also use it later, between coats of varnish, to make sure that no bits of dust, splinters, etc. can spoil the next successive coat(s) of varnish.
I use a coffee stain because it imparts a gentle yellowish look to the wood, without doing any damage to the wood. Many people use strong tea, and I have done so. Some have a particular tea they use, for the color they want. That is fine. Some have alcohol-based, color-fast dyes to accomplish a similar result. That is OK too. I’m just telling you what I do. People have sometimes asked whether I use de-caf or regular. I love that question: I say, “Oh I always use regular, to wake up the tone!” 🙂 (Riiiight…)
This is something I got from Roger Hargraves. He is a master luthier living in Germany, and he wrote an online book as he built a magnificent double bass. In it he explained why he uses a mineral ground, and what it is (it is effectively exceedingly fine gypsum dust, introduced into the wood pores in a liquid suspension. He used a very thin varnish as I recall…I used coffee, again. It dries slowly, is easily applied, and cleans up with water.) So, I apply the mixture with a brush, one area at a time, rub it in with my fingers, and immediately rub back off as much as I possibly can, using a rag. I do NOT want a thick layer of stuff on top of the wood– I want the pores themselves to be filled with the particles. I coat all areas except the “handle”portion of the neck in this manner. Supposedly it leeps the varnish from saturating the wood, thus avoiding overloading the wood with varnish and deadening the sound. Does it help? He says it does, and I know for a fact that he did it…and I saw the bass in person, last year at the International Bassists Society convention, in Ft. Collins, CO.
The ground dries chalk-white, in spite of the fact that I used coffee for the liquid. I sand it gently to remove any excess gypsum (there always is a fair amount that comes off the surface at this point). Sorry, I didn’t take a picture at this point. You’ll have to take my word for the fact that it simply looked like a chalk-white violin, with the grain and purfling quite obsecured by the dry mineral ground.
Then I seal the violin, all except the “handle” portion of the neck. There are many things people use for sealer: Some use special commercially available preparations (I have done that, too), and some use very thin varnish, or shellac. For now, my sealer is a combination of Pine sap (yes, that sticky gunk that flows out of a cut in a pine-tree) and turpentine. The turpentine penetrates the wood very readily, taking the pine pitch with it. I allow the turpentine to evaporate off, leaving the pitch in the wood. I am of the opinion that the pitch is an OK thing in the wood, and that it seals the wood against further penetration by varnish. That’s my theory, anyway. 🙂 The mineral ground turns completely transparent under the saturation of the turpentine and pine pitch, and remains that way after the turpentine evaporates. Here is a photo of the violin with the mineral ground rendered transparent by the sealer. It will probably take a few days to get absolutely dry, (turpentine evaporates slowly) so this is the last photo for this post:
I have used a wide variety of varnishes, but currently I am using a spirit varnish concoction. If I need more color, I use the “Transtint” dyes available in woodworking stores and online. It does not take a great deal of dye…you want to go easy on this stuff. I have considered extracting dyes from plants, etc. Never have gotten around to it. 🙂
I’ll show the varnishing process in the next post.
Thanks for looking.
More Progress: The Plowden-model looks like a violin!
Removed the Mold and Installed the Back Linings
One nice thing about this (French-style) mold is that it is easy to remove– it usually takes less than five minutes to get it out, provided I have waxed the non-glued surfaces of the mold, so there is no chance of an accidental adhesion to the ribs. I simply use a flat chisel to crack the glue loose at each block, where it is attached to the mold, and then slip my parting knife (in reality, it is an old potato knife) between the block and the mold, to make sure nothing is still adhering to the mold. After that I can slide the mold out by lifting as shown, while wiggling each individual block to make sure they are all moving evenly. The only drawback is that I have to install the back linings after the mold is removed, and the rib garland is pretty fragile. In the past, I have used a two-part collapsible mold, in the italian style (centered between the front and back edges of the ribs); but it is more difficult to remove. The only real advantage is that I can install all the linings while the corpus is still on the mold, and still get the mold out without too much trouble. I believe it does have a slightly greater tendency to leave the ribs slightly convex in the center, as well, and some people find that attractive. (I do too…. Perhaps I will eventually go back to using the Italian-style molds.)
This time, the linings went in without a struggle, though, and the whole task took about 30-40 minutes. A few hours later, the glue had dried, and I was able to remove all the tiny spring-clamps, and get going on leveling the back of the garland, and cleaning up the interior.
Cleaning up the Interior
Cleaning the interior primarily means shaping the linings and blocks, after the mold is out of the way, but it also is a final opportunity to scrape any rough spots, and make any minuscule repairs. So, I put on my magnifying visor, and look things over very carefully. I use a knife to trim the linings so that they are triangular in cross-section, thickest at the edge, to suport the edge of the rib, but tapered so as to not restrict the vibrations of the middle of the ribs too much.
One thing I have to remember, at this point, is that it is also my “last shot” at the interior of the back plate, so, before I install it, I carefully check and scrape it, and, as a final “seal of approval”, I install my signed and numbered label, gluing it in place where it will be visible through the bass-side sound-hole. All my instruments are individually hand-crafted, and each ends up a signed and numbered original, not a mass-produced instrument. I’m really not interested in doing things the “factory” way.
Installing the Back Plate
I used to complete all the work on the back plate, including purfling and final edge-work before installing it on the garland. But I have recently altered my order of operations so that I purfle and do edgework after the plate is installed. This allows me to adjust the plate overhang boundaries, and make final decisions as to how I want things to look, before permanently locking the shape in place by installing the purfling.
I also used to “self-induce panic,” by spreading hot hide glue around the perimeter of the garland and plate, and then frantically trying to align the edges perfectly and adjust all the (25-30) spool-clamps, etc. before the glue gelled. That was a fool’s-errand, as the glue usually gels far more quickly than I could ever get everything clamped in place. The late Sam Compton shared how he applied the glue and allowed it to gel, then aligned everything at a leisurely pace, gently clamping everything in place. Afterward, he went around the perimeter with a hair-dryer, gradually warming the plates until the glue re-liquefied. Finally, he tightened the clamps just enough to completely close the joints. Pretty clever.
I ended up clamping everything dry, then loosening a few clamps at a time, and slipping the hot hide-glue into the seams, using a very thin palette knife. That seems to work pretty well, and definitely eliminates stress. (His method probably is even better, but this is how I do things, for now.)
Final Edge Shape and Purfling
Once the back is securely glued in place, I can trim the edges to exactly the shape I want, so that the overhang is matching the front plate, etc., and then begin purfling.
I use the purfling marker to scribe in the double lines (except for the corners) where the purfling will be inlaid. I then draw in the corners very carefully. I have occasionally used a template for the corners, but it seems to not always work, since the corners are not always exactly the same. So I end up just making some layout marks to get the limits delineated, and then I draw the corners in, freehand.
I incise the purfling slot with a small, sharp knife, in three or more steps:
- The first pass with the knife is very shallow and light, only serving to deepen all the scribe marks.
- The second and third passes bring the purfling slot to the correct depth, and
- The last step is to use a purfling pick to remove all the waste wood between the paired cuts.
I check the depth and width of the slot by inserting a small section of purfling into the slot, all the way around, until I am sure that the slot is correct. then I bend and fit the purfling strips, and make sure the fit is as exact as I can make it, terminating in the classic “bee-stings” at the corners, rather than a simple “mitered corner.”
Finally, when all is as close to perfect as I can manage, I lift each section of purfling partially out of the slot, one at a time, and slip hot, thin, hide glue into the slot, using the thin palette knife again. I press the purfling all the way down to the bottom of the slot, and allow the glue to dry before beginning the final edgework.
The Channel, and Final Edgework:
I scribe a margin around the entire outer edge, usually about 1.6mm in from the outer edge. This will serve as the “limit” for the channel. I cut the channel using gouges first,then scrapers, and bringing the smooth curve up to match the scribed limit. Then I file the outer edge to round in from the center of the edge of the plate up to meet the scribed margin. I true up lines using scrapers, files, and abrasives, and then I am ready for final varnish preparations.
We’ll talk about all that in the next post, and post pictures. I need to get this one wrapped up. I’ll show close-ups of the edges and corners next time. there is still a long way to go, but it is looking more encouraging. 🙂
Thanks for looking.
Progress on the “Plowden” Guarneri del Gesu-model violin
The Plowden is “in the lead” and pulling ahead!
One fiddle is running ahead of the others. That is how it usually turns out when I am building more than one instrument simultaneously. Something catches my attention in one instrument, and I bolt for the finish line with it. I don’t know why…happens every time, though. I think this will be one of my best efforts, and should turn out to be a concert-quality violin. We will see, though.
Scroll and neck complete– Fingerboard temporarily glued in place
There will still be fine-point tweaking and smoothing I will do, up until the day I begin varnishing, but the scroll was nearly enough complete that I went ahead and mounted a fingerboard on the neck, and began shaping the two of them together. The string “nut” (the tiny bar of ebony that will cross the top of the fingerboard as a support and anchor for the strings as they cross over into the pegbox) will be fitted and installed pretty much the last thing before the bridge, soundpost, and final set-up.
Setting the Neck
Once I had the entire neck shaped, with the exception of the final shaping of the heel (which is completed after the back plate is installed), I could begin setting the neck.
Now: I used to install the back plate and then install the neck, but that meant that when I set the neck, I had to continually worry about the angle of the heel as it presented itself to the button (the top “tongue” of the back plate that overlaps the heel of every violin-family instrument…critical to the strength of the joint.) If I install the neck first, making sure that the heel overlaps the neck block at least a little, so that I can plane it absolutely flat before installing the back plate, I eliminate the struggle to get that perfect heel-to-button joint. It becomes perfectly easy.
So…for all you luthiers reading this, yes, I am aware of the traditional way to fit this joint. I learned this particular option from reading the work of the late David Rubio. It took me a few years to recognize the value of the change, and I tried it for the first time when working on my first cello. It worked so well that I have gradually begun to do it on all my instruments.
Here is the back of the instrument, mold still in place:
Here is the side view:
And the front view:
The next step will be to level the back of the corpus, so that the blocks and the neck-heel are all in the same plane. I may have to wait until the mold has been removed before completing that task: the mold is pretty much level with the blocks, and I have no desire to sand the mold. So, I may end up removing the mold, installing the linings, and then levelling the back of the corpus, along with the heel of the neck. Then I can clean up the inside of the corpus, shaping the linings and blocks, and cleaning up any rough spots. Then I can install the back plate and get moving on the purfling. After that it will be time to do all the final preparations for varnishing.
So…that is the progress report for this week.
I will try to catch up on the 14-inch viola and the 3/4 size violin soon.
Thanks for looking.
Progress on the Small (14-inch) Viola and the “Plowden” Guarneri model violin
Here are some photos of what is happening with these two fiddles. I decided to add a third instrument to the bench, so to speak, a 3/4-size violin (separate notes on separate thread), so it is slowing me down just a little.
Both the viola and the violin are moving along:
- Arching is complete on the front plates of both instruments.
- F-holes are laid out on both instruments, cut out and complete on the violin.
- The bass bar has been fitted, installed and trimmed in the viola.
- Graduation is nearly complete on the viola, complete on the violin.
- The scrolls are partially carved…still a fair way to go.
- The back plates are arched, but there is still some work to be done on each before I would call them absolutely complete.
- The top plate has been installed on the violin, and purfling installed.
- The violin top plate and rib garland are nearly complete…the edgework is done, but some refining will still happen.
- You can see that I trimmed a couple of millimeters off the corners of the violin front plate. I will do the same on the other three plates as well.
Here are some photos:
So…you can see that progress is happening. Not at a very exciting pace, but I hope the wait will be worthwhile.
My goal is to produce three very good instruments this summer/fall:
- the 14-inch viola,
- the Guarneri-model violin, and
- the 3/4-size violin,
and then show them, along with a larger viola, to orchestra directors and teachers in the Greater Portland Area.
My rationale is that good small violas are hard to find, and so are good fractional sized instruments. If I can demonstrate to the teachers that I can produce very good instruments in smaller sizes, as well as the larger sizes, then perhaps they will recommend their students to me.
All I can do is try….
Thanks for looking.