Skip to content

Joining the wood for the back and front plates

June 20, 2012

While I was gluing the blocks into the mold, I didn’t want to waste a bunch of the glue, so I decided to join the two quarter-sawn pieces of curly maple for the back plate of the violin. First, I tried planing the edges for the joint.  But my plane blade kept slipping (I usually bumped the mechanisms that hold it in).


So in the end, I ended up just sanding them flat on my sanding board and then finishing it with a scraper.

I did this for the front plate a few days later, when I needed to re-glue a block that came loose from the mold (I wasn’t surprised –I had had troubles gluing that one in, to start with).

Overall, I’m much happier with how the front plate turned out, than with how the back plate turned out.  The joint of the back plate isn’t as tight as I would like.   I’m trying to decide if it’s worth attempting to unglue it and try it again…  I’m not sure how possible that is.

I made them both “rubbed joints”.


In the meantime, though, I think I’m getting better at resetting the blade of my plane so that it’s more square and makes a really shallow cut.  (I find/make some saw dust on a very flat surface and then set the blade to a depth where it just scrapes up the saw dust…)  This will presumably be very helpful later on when I need to plane anything that’s curly maple –like the back.  (Or, if I break one of the ribs while I attempt to bend it, I’ll need to start over with thicknessing a new rib…  Lets hope that doesn’t happen!)


Shaping the blocks

June 17, 2012

I did some more work on the blocks, this weekend.  Here, I’m drawing the outline of the violin onto the blocks.  I had to anchor the template down so that it wouldn’t move.


I already had the holes for anchoring the inside part of my template. So I anchored it down and then placed the outside section snugly around it and drilled tiny holes from the outside template into the waste-wood of my corner blocks to keep it in place. Then I removed the inside template and traced a “reduced” pattern onto each block with a washer and a small piece of cardboard.  These lines show where the ribs will go (and therefore, what wood I need to remove from the blocks so that the the ribs can actually go there).


Finally, I got brave enouth to start cutting away the extra wood from the blocks. The spruce had a pleasant smell.  I used some gouges, at first, and then a half-round rasp, and finally some sandpaper wrapped around the handle of some tool I found in my toolbox.


I just had to eyeball it, when determining if my work was square… I might need to touch it up, before I glue anything on.  (I rather wish I had a drill-press sometimes… But then again, I’m marginally scared of power tools anyways, so maybe it’s a useless wish.  ;)  Haha. Well, ok, that’s a bit of an exaggeration.)

Conquering the fear of hot hide glue.

June 10, 2012

I finally got around to drilling the holes for clamping the ribs, later…  Hopefully I’ll be happy with where I put them.


This was with a borrowed hand drill from my dad.  Fun to use.  ;)  And I loved getting to smell the aromatic cedar of my mold, again.  (Incedentally, I kept trying to google for pictures of where to place the holes when you use the tangential style of mold…  And I kept getting pictures from my own blog.  Haha,  weird!  So the answer is: I don’t know. ;) )

I also finally tackled the daunting task of using hot hide glue!  For the past year and a half, I’ve been dreading this task.  But I read online about someone’s cheap way of making a hot glue pot:
1)  Buy a $20-30 hotpot at Target (must have adjustable heat)
2)  Buy a glass jar of something small at the grocery store.   I bought a 2oz jar of pimentos.  I tried to find something that could withstand heat. So look for something that has already been sealed by heat and has the little popup indicator that tells you if the can has been opened or not.
3)  Manufacture an aluminum stand to set the jar on to [reliably] hold it up away from the main heating element of the hotpot.  I started with a rain-gutter leaf filter thing (it’s aluminum mesh) from the hardware store, and shaped/cut it how I needed.  I’m sure there are many ways to do this.
4)  Buy a cooking thermometer.
5)  Buy a 1 inch paint brush.  I’ve heard that you don’t want it to have a metal ferrule…  I don’t know why.  Perhaps to keep it from rusting? Or, I heard someone suggest that it would affect the color of the glue.

Here’s my end result:


As it turns out (after some patient temperature experiments), it really works!  :)
Be careful to check the temperature of the glue inside of the small jar often.  I find that it tended to vary by 10 or 15 degrees between the times when the heating element kicked on.  The glue granules melt into the water in your small jar at about 140°F.  I read somewhere that you don’t want to let the glue get more than around 160°F (supposedly it won’t form a very strong bond, if you do).   And I found that it started noticeably gelling at around 120°F.  So…  I would heat the new glue up to 140 to melt it and then turn the heat down a bit to keep it from getting too hot until I needed to use it again.  Then I’d decide whether to heat it up again.

Finally… I glued the blocks into the mold:

First, I sprayed the mold with a couple of layers of varnish. Then I inserted three “spacer screws” to raise the mold up off the table so that the mold would be mid-way up the blocks. I did this experimentally, screwing or unscrewing the screws and drawing pencil lines on the blocks and measuring how much space was above and below the lines.

Once they were glued in, I improvised a sanding board by rubber-cementing some sandpaper to a piece of glass. Then I carefully sanded the blocks down to the suggested heights: 32mm for the bottom block, 30mm for the top block, and the others somewhere in between.


Whew! It was a little stressful, because the blocks are really part of the final instrument. So my perfectionist worries started to kick back in. (I like to think that I’m a recovering perfectionist… But building a violin tends to cause a number of relapses… ;) )

Sizing the blocks

May 16, 2012

I accomplished my 2012 goal: to bring out my fiddle-making stuff again. Yay! It’s been a year and a half since I last worked on it… So I figured that just getting the stuff out from the corner it’s been “hiding” in would be a big enough accomplishment. (And then I even started working on it again. :) Which wasn’t technically part of the goal, but I figured that would be the easier part.) So… Today I finally shaped the blocks into the right widths to fit in the mold, and the right heights to come close to the final height of the ribs. The upper block is the shortest, and the lower block is the tallest. (I shaped both of them a year and a half ago.) The upper and lower corner blocks range in between. In the following picture, I’ve arranged them as shortest (near) to tallest (far). If you look very closely, you might be able to see the differences.


Also pictured here are some of the tools I used in the process. Sizing the blocks was interesting… And not always in a fun way. It was really tricky to convince the blocks to be perfectly square. The disk sander I used was somewhat flexible behind the sandpaper, so I always had to touch up my work with a block plane, lying on its side to get a perfect right angle. But it was always a trick to figure out which side of the block to face downward, because few of the sides were reliably flat, to begin with. The verier calipers were again needed to check the final sizes. I was aiming to get the blocks to be 1mm longer in height and width, so that I will only need to do a little sanding in the end. Whew! I’m glad to be done with this particular task. And I’m glad to be working on building my fiddle again! :) Tip: Close your soda bottle lid before you start sanding. ;)

“Thicknessing” the Ribs

December 4, 2010

I wouldn’t have ever thought that “thicknessing” was really a word… but I’ve run across it several times now in books and online articles when referring to planing or scraping wood so that it’s a specific thickness.  So I’ll just go with it.  ;)

The goal was to take the rib pieces from their original thickness (around 1.8mm) down to as close to 1.00 mm as I could manage. In reality, though, the digital vernier calipers that I used measured in increments of .05mm. So I typically saw numbers like 0.95, 1.00, and 1.05mm.

It was very tricky to thickness curly maple!  Not so much because the numbers are so precise and small (although that has its own difficulties), but more because of the “figuring” of the wood.

Since the wood is “curly” (also called “figured”), that means that the fibers of wood wave back and forth (or you might say “up and down”, if you lay the rib piece flat on a table), instead of being straight/flat. (It’s a lot like curly hair vs. straight hair.)  If you’re not careful enough, you get “tear-out” of the wood, where the blade from the plane you’re using catches the fibers (instead of cutting them), and then where the fibers angle down, the wood separates along the downward angle.  Since it’s “curly”, though, the fibers change direction and start angling up.  So they tear out.  Basically you end up with little dips in your wood where the tear-out occurs.  If you have a sharp enough blade and if you use the right angle when you plane, the blade should cut through the grain, instead of grabbing it and pulling it along.

As an aside, I’m not sure exactly what causes the figuring.  My personal speculation is that it’s somehow caused by the twisting rotation of the tree as it grows (perhaps following the sunlight throughout the day?).  But that’s just a guess!  I really have no idea, but I’d love to know!  :)


I began thicknessing a set of ribs (three pieces) with a block plane (turned at an angle, up to 45 degrees from “forward”) and with the wood clamped to my work bench.  Right away, I was getting too much tear-out.  I read somewhere that one way to prevent tear-out is to get the wood wet.  So I got a bowl of water and a washcloth, and sogged the wood a bit (evenly, front and back, to prevent warping), and then started planing.  It seemed to work.  It prevented most of the tear-out, anyways.  When the wood fibers seemed to start catching on the plane again, I would get it wet again.  (But by then (when I was first figuring out how to plane the wood with this method) it was usually too late and the tear-out had already occurred.)

This wet-planing went really quickly.  But a couple of warnings:

  1. Don’t forget to dry your plane, when you’re done!  Else it might rust (even if you think it won’t!).
  2. The wood will actually change size when it’s wet.  The ribs, when they were wet, were wider by at least a few millimeters (which is a lot, in the realm of violin making).  I imagine they were also thicker when wet, as well, so you need to take that into account when measuring with the vernier calipers.
  3. The end result is kind of rough.  You’d still want to scrape the wood when the ribs dry.  It’s no where as neat and shiny as if you planed it dry.

Also, I still got some tear-out and the ribs in those places ended up much too thin.  So having a sharp blade is still important.

But in the end, though, I’m not sure you want to get the ribs wet like that.  I’m not sure what affect the water (and the wetting/drying process) would have on the final result, but I’m a little skeptical.


So I basically chalked up this set of ribs as my practice ribs, and I test things on these ribs, now (which have become really, really thin, as a result, and would never be usable.)

One of the things I tested, at someone’s recommendation, is setting the blade of the plane very, very shallow and making it very very sharp.  I tried that, and  I also tried NOT clamping the wood down, but holding it with my other hand (so that I could feel when it was about to tear out, better, and so that the rib would slip out of my hand rather than the plane tearing the wood fibers out).  It’s much, much slower that way, but the end result was really close to being amazing!  The only problem was that I still got some tear-out, even after that.  But the wood, where it wasn’t torn out, turned out to be as smooth as silk!  It’s really cool!  Unfortunately, though, part of the deal is that the blade of the plane needs to be really, really sharp.  So I needed to sharpen it again, before testing it on the next practice rib.  And then I needed to reset it back in the plane.  I was never able to get my other ribs as beautifully planed, after that.  :(  bummer.  I suppose with a lot more practice of sharpening and resetting the plane…..


Anyways, since that set of ribs was ruined (but not without benefit! I learned a lot!), I decided to try thicknessing a new set of ribs.  This time without water, and without using the block plane at all, and without clamping it down.  Things went much better once I figured out what would work.  The picture below shows the various tools I finally settled on using and their resulting wood shavings.


The below picture shows the tiny Ibex “finger plane” I bought.  (2 centimeters long, if I remember correctly.)  I also bought a “toothed” blade to put in it.  If you look closely at the inside part of the plane, you can see the “teeth” in this picture.  The teeth help minimize tear-out (but I’m not sure how).  They really do help!  But it doesn’t work perfectly.  You still need to be careful to avoid tear-out, and you still need to set the depth of the blade just right.  Trial and error…

The wood shavings are long and skinny and curly, after the first pass.  Kinda fun.  :)


I knew that often people use “scrapers” to remove wood when making a violin, so I’ve been trying to figure out what will actually work for me and what doesn’t.

In the next picture is a knife I found somewhere in my kitchen.  I think I got it from someone.  It’s not exactly a paring knife… maybe it’s a sharpened butter knife or something.  I’m not sure.  But it’s stainless steal, so it’s hard enough to use as a scraper (I’ve tried using aluminum.  no good).  I’m right-handed, and the bevel is on the “left”, while I’m scraping towards the right.  I alternate the angle that I hold the knife.  Sometimes I hold it so that it leans to the right and I scrape towards the right, but sometimes I hold it so that it leans to the left and I scrape towards the left.  The bevel is not equal everywhere.  It’s more blunt at the tip.  Again, trial and error.  I also found that turning the blade to an angle would sort of drive the shavings away to the side of the blade as I went, so I could go farther without having too many shavings in the way to see.  Plus, there was less possibility of tear out, I think.  And I had less of a “rippled/washboard” effect on the final product.

The wood shavings look longer than they really are.  The shavings compress into one another and result in a longer-looking, fragile, curly shape.

Interestingly, I think I was able to remove more wood in the same amount of time as the toothed finger plane.  But maybe if the toothed plane were the normal size of a block plane, that would work the fastest…  I’m kind of cheap, though, so I’m not sure I’ll buy such a thing to try it.


The “scraper” in the next photo is actually just a “painter’s knife”, which I found at an art store.  (I think it’s used for mixing oil paints…?)  I start by using it similarly to how I use the kitchen knife, above.  (although it’s more awkward to hold.)  It has no bevel (just a 90 degree edge), but is really thin and flexible stainless steal.  (A little less flexible towards the handle.)  You can see that the shavings are really small.

Sometimes I angle it towards the right, and then move leftwards with it.  I do that especially on places where I think the kitchen knife was too harsh on the fibers and started bending them instead of cutting them.  This method sort of seemed to undo the damage by cutting them from the opposite direction and possibly making them lay flat again.  Anyways, it brings the “shine” back again, where the wood looked a little rough.

Sometimes I brush it back and forth like a paintbrush to do some final scraping.  The result is more dust-like.


From what I hear, scraping the wood is always better than sanding the wood on the parts of the violin that go towards the outside (the parts that get varnished).  I suppose you don’t get the scratches left over from the grit/abrasive of the sandpaper.

I still wish I could have perfected the planing method that left the wood feeling as smooth and cool as silk, but this will have to do, for my first violin.  The thickness is really quite decent, so I don’t want to mess with it much more or else it’ll get too thin.

Mold and upper/lower/corner blocks

November 24, 2010


I cut the wood for the corner and end blocks, as well as the recesses in the mold for the blocks.

I decided to use the “tangential” method of setting the corner blocks in the mold because I wanted to make sure the grain was facing the points of the corners. (supposedly that’s the “right” way to do it. I think it makes the corners more resistant to damage from a side impact… or maybe it’s just easier to carve the blocks when they’re set that way.) So then I only had enough wood thickness to set them tangentially.  I think it’s far more common to set the blocks “square with the world”, but it means that you have to have “deeper” blocks. (Google for images of inside violin molds to see what I mean. There are very few pictures of the tangential method, and lots with the perpendicular/parallel method. Not-coincidentally, that means I’m not sure what I’m doing. ;) I’ll have to just make it up as I go, I guess. )

You can see the rough-cut blocks in the picture. The widest block (on the right) will be the neck block (aka upper block). Second-widest (moving left, in the picture) will be the lower block (aka bottom block). Then the two lower corner blocks are narrower still, and then the two upper corner blocks are the narrowest.

In the picture, the east-west direction of the blocks will become the up-down (top-plate to bottom-plate) direction in the mold (the “height”). I made them all around 35mm “tall”, to start with. In the end, the idea is to have the lower block be about 32mm tall, and the upper block about 30mm tall (I think… I need to verify that, though…) and the corner blocks are some height in between. (I’m not sure what the point of that is, though. I’ve read somewhere that it makes a difference in the sound, but I’m not sure what or why, at the moment…)

The up-down direction (table-to-sky) of the blocks in the picture will become the inside-to-outside direction of the blocks in the mold. This “depth” will vary… For the corner blocks, I will need every millimeter I have (I can’t remember at the moment, but approx 20mm?). The upper and lower blocks don’t need to be as deep. (Maybe around 15mm?)

And finally, the north-south direction of the blocks in the image will become the “widths” of the blocks around the circumferance of the mold. This varies, too. Strobel gives suggestions on these “widths” for the upper and lower blocks. The corner blocks are up to you, though, I guess. I measured the distances on my geometric template pattern between where two curves of different radii meet on one side of the corner to the next curve-change on the other side of the corner. I think, though, that I could have made them smaller. Oh well. If I’m careful about where/how I glue the ribs to the blocks, I can probably still make the corner blocks narrower prior to cutting/bending/gluing the linings. Or I could fashon some sort of inset to shrink the mold recesses down and then resize the blocks prior to bending/gluing the ribs.

The remaining wood that the blocks were cut from will be cut into the lining pieces. More about that when I get that far.

“Strad Poster”

November 21, 2010


Yay! My Strad poster finally made it.  As you can see, my kitty is excited too.  (He wanted to walk on it…  naturally.)

Anyways, this particular poster is of a violin known as “Titian” which was made by Stradiveri in 1715.  The front side of the poster shows photographs of the violin from different views. As well as a CT scan of the top. (Yes, the same kind of CT scan that a doctor uses. Technology comes in very handy for many reasons!)

The back shows the measurement details (lengths, distances, thicknesses, etc.) And more CT scans for additional thickness and arching detail. It also shows a thickness map of the front and pack plates, derived from the CT scans.


Being new to this, I’m not quite sure, but I think a lot of the stuff is meant to be to scale (although clearly not all of it). I don’t have time right now, but I’m sure I’ll analize it more, later.