This is my original mold– my first five-string was built on this mold, as was the commissioned instrument from a year ago. I will not complete it before the end of this year, obviously, but it is on the way.
The neck and back are made of spalted, heavily flamed Big Leaf Maple, salvaged from the yard at the home where my wife grew up. Her mom and dad had the tree taken down a few years ago, and my son and I salvaged a little of it. The front plate is sitka spruce.
It will be a few weeks…and there is another coming right behind it, but on the Guarneri mold.
Wet down the wood,to raise the grain and accentuate “problem” areas
I wetted the whole bass down with coffee, mainly to raise the grain a bit, but partly to add a little color to the white wood of the Sitka spruce belly. The Oregon Big Leaf Maple is already pretty colorful.
Then I turned the lights off in the workshop and went over the whole bass, inch by inch, with a small flashlight, held at a low angle, to make shadows wherever there was a discontinuity in the surface. As I located them, I scraped or sanded away the problems, before moving on. It took quite a while, but I was pretty happy with the outcome.
The next step is to coat the wood with a mineral ground: a suspension of extremely fine particles of gypsum in water is what I use. I brush it on, one section at a time, rub it in vigorously, to get the tiny particles into the pores of the wood; then rub off any excess, with a rag. It always looks as though I took all of it back off, until it dries.
After the ground is fully dry, I sand all over with fine abrasive, to remove any dry patches of excess mineral. There will be very small discontinuities that have been filled by the mineral ground: this is desirable, and I am not trying to remove those places.
When the ground dries, the bass will be stark white, but when I apply the sealer, to lock the mineral particles in place, the mineral ground becomes completely transparent, permanently. It will never be visible again.
So, here is the bass, with the gypsum fully dry, mounted in my varnishing fixture, and ready for the sealer:
I am currently using rosin dissolved in turpentine and alcohol as a sealer. The liquid (alcohol first, then turpentine) evaporates, leaving the rosin in the pores of the wood. When dry, this helps prevent the varnish from soaking into the wood, so as to minimize the sound-dampening effect of excessive varnish penetration.
This is the part of finishing I like best: it seems almost magical to see the stark white of the mineral ground disappear instantly and permanently as the sealer permeates the gypsum and renders it transparent, so that the beauty of the wood is revealed.
After the sealer is dry, or just before it is completely dry, I rub down the surface of the wood with alcohol, to pick up any rosin that may have remained on the surface. When the sealer is fully dry, I go over the whole surface, lightly, with fine sandpaper, to pick off any bits of wood fiber that may have raised during the ground and sealer process.
I always begin with a yellow varnish: I like the way it shines through the darker pigmented varnish when all the finishing is complete. In this particular case the maple was dark enough that even the yellow varnish will end up looking pretty dark. So, though I will still begin with the yellow, I will have to add a good deal of darker varnish on the front, to balance the color with that of the back. This is just a type of “Judgment call” that the maker must always consider when finishing an instrument. As I add coats of varnish, I will pay attention to which areas need darker varnish, and which could use yellow or clear varnish.
When I last posted, I had flattened the back plate, using a plane, but the shape was still oversized.
So I traced out the plate shape using a small section of plastic pipe as a guide, and a ball-point pen inside the pipe to make the mark. Then I cut out the plate using my very old Craftsman “Auto-Scroller” saber saw.
My beloved wife, Ann, bought me this saw when we had been married for less than two years, and it has served me well for the last 38 years, but this may be the final plate it will cut out. It overheated rather badly during the cut. 🙁
Once the plate was cut out, I used my curved-sole scrub-plane to remove waste wood, and rapidly bring the plate to near the proper thickness around the edge. As the thickness gets close to the target dimension, I switch over to the Ibex Finger-plane with the toothed blade and the wooden handle, to complete the thicknessing of the plate edge. The Oregon Big Leaf Maple is much more difficult to carve than the Spruce was, both because it is harder, and because the grain is highly flamed, meaning that it changes directions every centimeter or so, resisting all efforts to smoothly plane off the wood. The toothed plane helps, but when I start getting close to the right thickness, I will have to switch over to a scraper before the tear-outs from planing are too deep to be removed.
You can see the longitudinal arching template in the above photo: it is just a thin piece of plywood with an 11′-3″ radius circle section cut out of it so as to leave the correct arching height in the center. I used that to help me establish the longitudinal arching. The Ibex plane is on the plate, and the scrub-plane is almost out of sight behind a small block-plane in the background. The small block-plane is helpful for smoothing the ridges left by the scrub-plane.
I am working to the rough sketch I made before beginning, with the plan for the back arching: (I did change the plan a little. I realized that I could extend the arching a little further “north,” as I have tapered the entire garland a little, so that the bend in the upper bouts will not be so severe, and the arching may be able to follow it a little way before flattening out to avoid the compound curve. It’s worth a try, anyway, and will not hurt anything.)
My hands and shoulders were getting too tired, so I went inside and used small finger-planes, files, and scrapers to refine the scroll. I am waiting on an order of carbon-fiber reinforcement materials to complete the neck, but other than that, I am pleased with how it is turning out.
I also completed the scraping of the Sitka Spruce belly, and it is pretty much ready to be glued to the garland.
I pretty much wore myself out on this stretch: I’m looking like a tired old man, here. And I thought I was smiling…
As I said in the post regarding tools, I built the little curved-sole scrub-plane with the specific intent of using it to carve out the inside of the Sitka Spruce front plate for this Five-string Double Bass.
As the depth approached the correct value, I began switching over to the palm plane, there in the foreground. But as it turned out, I actually had a long way to go before I was anywhere near too thin.
I used the bass caliper to register thicknesses all over the plate, and then began carving “dots” at each location, to the desired thickness.
As I found (or created) spots that were at the correct thickness, I wrote in the thickness, and highlighted them in yellow, to warn myself against going any deeper. Eventually, I had mapped out the entire plate at least approximately according to this diagram from Peter Chandler’s book “So you want to build a Double Bass”:
He had derived these measurements from a fine old master bass by Domenico Busan, which conveniently happened to be disassembled for repairs and restoration. He said that he had subsequently used these values on all his basses, and it always worked well. (Sounds good to me!)
I kept carving until I had “dots” all over the plate.
Connecting the Dots
Then I began “Connecting the Dots”:
As I planed away the excess wood, the “dots” got smaller and smaller, and, in some areas disappeared. By that point I had switched over to the palm plane which is less aggressive and makes a smoother surface.
But eventually, it was pretty much all done, and time to cut out the f-holes. However, I decided to install the purfling first, and then cut out the f-holes.
I did not take pictures while this step was in progress: I just got going and pressed on until the job was finished, then took a few pictures. Sorry. I don’t always think about pictures.
I used this old purfling marker to trace my lines, then a thin-bladed knife to slice along the lines to make a slot…then picked out the waste wood and inlaid the purfling.
Cutting the F-holes
I used a coping saw to cut out the f-holes. It was slow and laborious but it worked, and there was little chance of any catastrophic errors. The result was two f-holes cut within a millimeter of the line and no errors. It is starting to look like a double bass!
I use a very thin paper gauze tape for chalk-fitting bass-bars.
The trick is to press the bar into the chalked tape, and “wiggle it” slightly, to pick up chalk on the high spots. then plane off just the chalked places and do it again, until all of the bass-bar comes up with chalk on it. That achieves a perfect fit. When the tape is finally removed, it takes all the chalk with it.
Then I warm the wood using a heat gun, apply a liberal coating of hot hide glue to both surfaces and clamp the bar in place. I leave it overnight to dry, just to make certain it will not pop back off (I have had it happen.)
The properly-installed bass-bar still has to be carved to the appropriate shape. I use planes to accomplish the carving.
Back Plate Vision
There is still a good deal to be done, before I can install the Front plate, so I am stopping there for the time being.
But I really wanted to get a foretaste of what the Big Leaf maple of the back is going to look like; so I planed the inside and outside of the back plate flat, just to have a look at it:
It is pretty stuff! I am really looking forward to seeing it completed.
This is not really a new project, but rather one that was “tabled,” for lack of better term…work was suspended until a better set of circumstances emerged.
I built the mold for this bass in 2015, began bending ribs in 2017, with a woefully inadequate bending iron, and a great deal of frustration.
A commission came in, so I set aside the bass, to work on the cello, and never came back to it…so it sat in the corner of my workshop silently sneering at me every time I looked that way.
But! Since I was laid off from my job, where I had worked for 33-1/3 years, in January, I am catching up with some projects and able to face others with new eyes.
Here is the five-string 16-1/2″ viola I am just finishing up, balanced on top of the bass mold:
Once I had the bass mold up on my bench again, it was easier to confront the problems, rather than avoiding them.
The New Bending Iron
The first thing I needed was a new bending iron. A fellow I met online, John Koehler, a fellow bass maker, told me how he built his bending iron. So I followed his lead, and built a new bending iron:
It is a section of exhaust tube, welded to a piece of angle iron, so that I could clamp the apparatus in a vise. Heat is supplied by a 550-W electric charcoal briquette lighter, controlled by a 600-W dimmer switch. It took a little trial and error to get it set up correctly and to calibrate it, but it turned out to work very well! (What a relief!)
Bending the ribs
Bending the remaining two Big Leaf Maple ribs was nearly effortless, and took about ten minutes, tops, not counting waiting for the tube to heat up.
Installing the ribs and linings
Then I glued the ribs into the fir blocks on the mold with hot hide glue, one at a time, and affixed the willow linings in the same manner before moving to the next rib.
Once one rib was completely secure, trimmed and lined, I rolled the bass mold over and repeated the operation on the other side.
I planed the linings flush with the ribs and blocks, and the garland was essentially complete. It will require careful leveling before fitting the plates, but not much other than that.
In the coming weeks, I will complete the center-joins of front and back plates, then complete the carving of the plates and the neck and scroll, and start putting this bass together!
Just as a teaser, this is the wood for the front, back and neck:
(Notice that there is a fair chunk left over where the neck pattern does not use all the wood it is on: watch that space! )
When I last posted, I had only begun cutting the outer purfling slots (I planned double purfling plus a weave for the back), so the next thing was to complete those slots.
Inserting and Gluing the purfling
The next step is to insert and glue the purfling in place, using hot hide glue. First I bend the purfling, using a hot iron, then I cut the ends to match the joints where the various sections meet. I insert the pieces dry, to ascertain that they fit, then, one-by-one, I pick the pieces back out and insert hot hide glue into the underlying slot, and quickly re-insert the purfling, forcing it to the bottom of the slot.
So that was the completion of the purfling. Edgework was next, shaping the channel through which both purfling slots will travel, as well as the outer edge and how the channel fairs into the front and back plate curvature. I used gouges, small planes and files, to get the edges to the required shape of a finished instrument.
While all this was happening, I was also getting going on the scroll, pegbox, and neck, but I will save that story for another post.
When I last posted, I had completed the carving of both plates and the garland, but had not begun assembling the corpus.
The next step was to install the bass bar. The bass bar is the only fixed, interior brace in violins, violas or cellos. Flatback basses do have some other bracing, but they are a different “branch of the family,” so to speak. All members of the violin family have a bass bar– a spruce brace, which runs “north-south” at a slight angle, nearly parallel to the centerline of the front plate of the instrument, and just inboard of the bass-side f-hole, so that it supports the bass-side foot of the bridge. In a five-string instrument, this becomes an even more critical part as the instrument has a broader range and has to have good support on the bass side, as well as the ability to sing in the higher registers.
I first carve the bass bar bottom to exactly fit the inside curve of the front plate, along the correct location, and at the correct angle, then glue and clamp it in place, using hot hide glue and special clamps, padded with cork, so as not to damage the soft spruce of the front plate.
Five-string viola corpus assembly:
Now the plates are ready to be installed. Before doing so, I used a small finger plane and half-round files to shape the edge all the way around on the inner face of each plate, hoping to avoid having to shape it after installation. (I am aware that sometimes adjustments have to be made, so I may have to do some tight-clearance work later on, in spite of this precaution. That’s OK.)
The next step was to install the back plate. This is an older-model mold, or “form,” (my first, in fact, as I mentioned in an earlier post) so it has some peculiarities, compared to my newer ones: it is a two-part mold, made to collapse, thus easing removal of the mold after installing the first plate. But in later iterations, I moved toward installing the front plate first, and installing the neck before removing the mold.
In this model, originally, I had planned to install the back plate, then remove the mold, and finally install the front plate, after which I could install the neck whenever I was ready to do so. Nowadays I personally find it easier, however, to install the neck before the back plate is in place, because I don’t have to concern myself with the back side of the heel aligning with the back plate button. (Annnd, it would have been a simple matter of planning, to still do that with this mold, if I had been thinking ahead: just label the front side of the mold as being the side without the screws (which have to be accessible) and you can install the front plate first, then remove the mold after installing the neck; no problem.) However…I wasn’t thinking ahead, and I used the mold exactly as I had originally designed it, so I have no choice, now: I am forced to install the back plate first, remove the mold and then (after shaping the blocks and linings and cleaning the interior of the corpus) install the front plate. So that is what I did. (By the way, in case you are thinking that the shape of the front and back plates are mirror-image of one another, the fact is, they virtually never are exactly mirrored, and are nearly never bilaterally symmetrical even if they were. So the front plate will not fit the back of the mold, and vice-versa.) Ah, well…hindsight, etc.
Here is the back plate, glued in place: the mold is still inside, holding everything rigid. Notice the spalting and curl in the maple back. This is a striking look, and some people love it…others do not.
After the back plate glue was dry, I removed the mold, shaped the interior blocks and linings, and cleaned up the interior of the corpus, so that it was ready for the front plate to be installed. I also installed the signed and numbered label, marking this as one of my handmade instruments.
Then I clamped the front plate in place, dry, just as I had done with the back plate, removed a few clamps at a time, and used a thin palette-knife to insert hot hide glue between the plate and the blocks and linings. As soon as I had the glue in place, I quickly replaced whatever clamps I had removed, before the glue could gel.
Once the plate was glued and clamped all the way around, I went back around with a blade, and picked out any gelled, cooled hide-glue that had squeezed out of the joint, so as not to have to deal with it later, in the form of hard, jagged chunks of dry hide glue. Then I tightened the clamps a little, and brushed hot water all around the joint, so as to reconstitute any glue that had gelled too soon, and allow the joint to close even more tightly.
Here is the corpus, all glued together. The next step will be to adjust the overhangs as needed, and lay out the corners so as to begin purfling.
Beginning Purfling the Five-string viola:
I used to do my purfling before closing the corpus, but I frequently discovered that the rib garland had moved a little, during the removal of the mold…or in some other way, things had changed, and then my plates no longer fit the garland, and I could not change the plates, because I had already installed the purfling…which locks in the shape of the plates, irrevocably (sigh…). So, I began waiting until after the corpus is closed and whatever needed overhang adjustments have been made, and then begin purfling.
I use a two-blade purfling marker to sketch in the location of the twin, parallel cuts needed to make the purfling slot, but I have to sketch the corners in by hand, with a pencil, because the purfling marker will not correctly lay out the corners.
I went ahead and began both the front and the back plates, but got too tired to complete them last night. (Today was spent getting last-minute things done, as we have heard they are mandating that all Oregonians stay at home, due to the coronavirus scare. Went and bought flour and other groceries, filled the car with gas, and got the snow-tires removed, as that deadline is soon upon us as well.)
One thing about the maple and spruce plates: the spruce is very soft, compared to the maple, but it is tricky to carve, because of that. The winter grains (reeds, they are called) are so much harder than the summer reeds, that the blade has a definite tendency to swerve and follow the grain instead of the line you are trying to follow. The maple is much tougher to cut, because it is hard all over, but it is much easier to follow your lines without digressing.
So, here is what the little viola looks like, today:
In both cases, the plan is to cut the two incisions, pick out the wood between them, and then dry-fit the purfling strips, before removing them one-by-one and gluing them in place with hot hide glue.
That will be the next post, unless I take a break and carve the scroll. Either way, it is starting to look like a fiddle!
I had originally begun this little viola ten years ago, wondering what effect it would have to go extra-wide on the lower bouts without changing much else: I was sidetracked with other, more pressing projects, so the little viola sat, partially completed, for ten years. I was somewhat disinclined to complete it, now, because my skills have increased significantly over the intervening years, and I can’t undo some of the “marks of the amateur” which now glare at me accusingly. But, I also felt the need to get some 5-string violas going, and if it didn’t work then I could still toss it out, or whatever…at any rate, it was just an experiment.
I completed the little 5-string viola about the time I completed the recent commission, so I numbered it accordingly, and set it up. It has a great big voice, and is balanced and clear across all five strings! What a surprise!
Accidental Imitation of a famous instrument
I sent pictures to a violist friend, and her immediate response was “A 14” Tertis! Bet that would have made the old boy mad!”
I had heard of Lionel Tertis, and his famous contribution to the viola world…but had never actually seen one of his instruments. So I looked up photos on the internet, and, to my surprise, it really doeslook like my little fiddle! So, I inadvertently retraced the steps of Mr. Tertis, and produced a fairly powerful little “Tertis-style” 14″ five-string viola! He, of course, was on the other end of the spectrum, trying to achieve a very large, but still playable, viola. (My apologies to the Maestro! I wasn’t being deliberately irreverent!)
Here are some photos:
The back, ribs and neck are Big Leaf Maple, harvested near my home, and the belly is Sitka Spruce. I used spirit varnish, as usual.
I have been seeing increased interest in 5-string violas, lately, so I am working to “populate” my five-string viola stock. This is the viola design I began with over 20 years ago (The mold says 1999; it was my very first instrument.) and it makes a very nice small viola. So I decided it would probably make a great 5-string viola as well.
I began the work quite some time ago, but other projects took priority, so the little viola languished on the bench. The back is curly, spalted maple from a tree that had been taken down on my wife’s parents’ place, and the belly is Sitka spruce. Both the front and back plates are one-piece in this case…something I seldom do. (One-piece front plates are uncommon. One-piece backs are quite common, and I prefer them.)
The rib-garland had been completed and the plates traced out months ago; so, after delivering the last commissioned instrument, I finally felt free to get to work again on the viola.
Five-string viola arching complete: ready to graduate the plates
I failed to take any photos while arching…so we are beginning with that portion complete.
On the backs, especially, I make a practice of carving “dots” to specific thicknesses, following a plan in my mind. Once all the dots are correct, I “connect the dots” using small planes, until the entire interior is a smooth continuum, and all the correct thicknesses.
The Spruce is a lot easier to plane, so I tend to just measure and plane. Most people use gouges for all this work, but I like using the planes.
Cutting the f-holes on the Five-String viola
Once the plates are completed (which also involved laying out and incising the f-holes) I still have to actually cut out the f-holes. I used to do this using just a knife, but it was time-consuming, and I found it difficult to get the round parts “round.” My grown children bought this tool for me, a special tool for cutting just the upper and lower eyes of f-holes. It works beautifully!
Beginning the Scroll for the Five-String viola
While things were being sorted around, and different tasks became logical, I decided to get a start on the scroll and neck. Didn’t get very far, but here it is:
There is still a long way to go, but it is feeling more as if I was getting something done, at least.
I began this instrument as an experiment, in 2010, but did not complete it. In fact, the only reason I know when I began it is that I wrote the date on the mold when I made it. It was intended to be a 14″ Viola with a very wide lower bout.
The wood for the neck, sides and the one-piece back is plain-sawn, Big-leaf maple with a two-piece Sitka spruce top plate. The workmanship is not as good as I try to produce today, so I might have simply discarded it, but, (I reasoned with myself) “it is just an experiment, to see what happens if you add a lot of width in the lower bouts.” If it turns out to work well, I will still have the mold and can simply make better, more professional examples of the same model.
What about a five-string Experiment?
So, back in September of 2019, it occurred to me that really I needed more 5-string fiddles on my website (especially violas), and that, if it worked, this was one I could complete more quickly…so I jumped back on it!
It still needs a good rubdown, a soundpost adjustment, and some play-in time, but today it is playing, and it sounds and looks OK, for what it is.
The voice is still a little too soft for what I want, but that could be the fact that the varnish is still pretty soft, too; and also, the soundpost is pretty tight (which I know will need to change, but I also recognize that there is a tendency for the arching to relax a little and for the soundpost to be looser, after some time with strings on. Longitudinal compression of the top plate tends to try to shorten the top, resulting in a change in how the soundpost fits.)
All that to say, I am reticent to just jump in and trim the post, when it may actually need the extra length, in a few weeks, and with the varnish hardening off, and the playing-in beginning to take hold, I want to take my time about making changes.
The sound is (mostly) balanced across all five strings, but the C-string could use a little more focus. Probably adjusting the soundpost and giving the varnish more time to harden will take care of all those issues.
Finally, because the repetitive tensioning and relaxing of the tuning is hard on strings, I have deliberately used an older set of strings for the set-up and trial period, so as not to sacrifice a brand-new set in what I know will shorten the life of the strings.
So! There it hangs in the dining room, where it will be warm, and we will see what the next week or so brings.