Constructing a lyre via hollowing isn’t a hard process on paper -
there aren’t a great lot of steps, broadly speaking. And it can be as
hard or easy of a process as you like, depending how intent you are on
good results.
I am very intent on good results.
With the large design questions settled beforehand, I went over to
my friend’s garage and got to work. The first step is cutting out the
general profile on a bandsaw. Since making fine turns with a bandsaw
wasn’t going to work, I radiused all of my inside corners with the drill
press. On the three string, I left the wood inside the yoke in place
for the time being - I wasn’t going to cut through the head of the
instrument, and no jigsaw was available. For the four-string, I meant to
do a tongue and groove joint anyway, so the bandsaw accomplished most
of the material removal, save for some remnant at the shoulder.
After that, reducing the depth of the arms would give me handholds
while doing everything else. As my boards were too wide for the bandsaw,
I did this with a forstner bit in the drill press, by drilling from the backside. I couldn’t
get exactly to size, because the bit has a spike in the center. That meant I had to stop just short of the spike leaving a hole in the arms when I was done, which left at least an eighth inch of extra material clinging on for the time being.
After this, I needed to define the slope on the back of the
instrument, so I re-traced the angle and removed the relevant material
with a belt grinder. Since hollowing was dependent on the drill press, having the exterior shape defined first meant that I could make the inside and outside parallel reliably.
The last and most time consuming step that I got done for both
instruments that weekend was the rough hollowing itself. I started with a
forstner bit, as with the back side of the arms. The table on the drill
press was set to roughly 1/4” down, so as to avoid the aforementioned
hole problem. The four-hole jouhikko was actually thicker than the
maximum throw of the drill press, so my initial drilling was simply as
far down as possible. The second drilling targeted roughly 3/16”, but
this time, the forstner bit had its tip ground off. This caused a fair
bit of wander until it found purchase, but it made a mostly flat bottom,
save for a faint ring around the outside and a raised dot where the
cutting surface was gone. Doing this, I averaged five gallons of wood
shavings per instrument, as measured by overflowing the scientific precision garbage can.
On our second day together, we went over to a table saw and engaged
in joinery cutting, where we cut the tongue and groove joint for the
four-string and broke down of a spare maple board. Parts from that would be useful for pegs later.
At home, the longer process of refinement could begin. I took the
jigsaw and cut out the last piece from the three-string, then looked to
the joinery on the four-string. Unbeknownst to me, a lot of tension had
been released from the wood, so the arms had moved together by roughly
1/4”. So, I trimmed the yoke to fit… and discovered my mistake.
Fortunately I only had to buy a little extra wood to cut a new yoke, and
my caliper and router work was precise enough that it fit perfectly.
Almost makes me feel like a real woodworker.
More work on a bench sander squared up the outside of the bodies
and removed the worst of the 60-grit texture. This handled the flat
areas, but the bottoms of the arms had jagged excess from the forstner
bit. This cleaned up with the router, as did the outside radius on the
yoke. It also cleaned up the inside of the body beautifully up to an
inch down. That’s as far as my 1/4” shank tool would go.
Special attention was required for the base and inside of the arms,
and the tailpiece hooks. The arms took a half-round file to flatten out the imprecise cut at the base of the arms and keep the curvature at the base of each arm consistent in all dimensions. The file was also the best way to keep
the insides of the arms straight.
I cut out the bottom (and on the four-string, the top) of the tail
hook with a hand saw and did the rest of my contouring with hand files.
The round file gave a nice profile for the actual hook portion with
minimal extra cleanup.
To get the back face down to thickness, I put a straightedge across
the face and measured down with a caliper. Both instruments were quite
thicker than I figured necessary at this point, so cleaning up the last
of the sides would go hand in hand with stock removal from the bottom. I
set my cleanup depth in several locations with a small drill bit with
tape around it - once the hole disappeared, I would be done sanding. Once again, the file sander was the best tool for the job, since
anything that removed material faster didn’t fit in the space.
My original bracing was made from maple, until I read that it was
best to match the wood that the soundboard would be made from. I ended
up with extra cedar left over, so I carved bass bars and soundposts from
the leftovers of the same boards on each instrument. The soundposts were hand-carved to rough circles about as thick as the soundboards were wide. The bass bars were whittled and sanded to a somewhat traditional shape, with a bump in the middle and a gradual taper to the ends, so the greatest stiffness would be right around the bridge.
My sound holes
weren’t especially scientifically laid out, save that I had roughly the
right clearance for where the bridge would go and space for the bracing.
I went with a layout vaguely resembling the original on the Trossingen lyre,
and did something a little different for the sake of difference on the
Sutton Hoo - no surviving reference on that one, so I assumed that it
would be roughly like its contemporaries.
Not all of my original maple bracing went to waste, as the maple jouhikko
ended up a little thin in the back. When I found I could flex it
slightly with my fingers, I was worried that it wouldn’t support the
soundpost. A maple bass bar added to the back fixed that.
I had a particularly new experiment for this project in that I
decided to use hide glue for assembly. Reports online said that regular
wood glue dampened sound and hide glue didn’t, and if worst came to
worst, I could open the instrument again. Rehydrating the glue to the
correct consistency was an experiment, but overall I didn’t notice any
of the smell that I had heard about, and the glue was easy to use.
I did cheat slightly and glued my soundpost to the top board after
experiments trying to wiggle it into place through the holes went
poorly. It also helped me set the length of the posts correctly. I found
out afterwards from Reddit that my bass bar placement was technically
backwards - it should be parallel to the strings. The way I installed
it, it would be… on a violin, where the strings come together on the fingerboard instead of fanning out between the arms. That’s something
to learn from on my next instrument, since I plan to do one more eventually. Meanwhile, both of my lyres repeated this mistake, since I glued them at functionally the same time.
Once the lid went on, I was able to file the edges flush and sand out all of my tool marks.
I copied my peg spacing from my existing instrument and drilled the
pilot holes on the drill press. My laurel had a spare set of tapered
peg tools, so I was able to ream the holes out with those. The taper
lets the friction pegs work, setting their tension by pressing them
manually into place. I carved the pegs from maple for both instruments. I
tried mahogany at first for the 3-string, since interchanging the woods
would be pretty, but the mahogany broke in half during the test fit, with no string attached. So
that was a bust, and I had to start over.
After sanding the instruments to 400 grit, this is the point where
the posts diverge, because I finished one instrument long before the
other! Stay tuned for Finishing Out the Jouhikko.
No comments:
Post a Comment