Day 7:

Lots of sitting around today.  I am glad I grabbed a bag of pretzels on the way in and had a long book to read.  When I arrived, Mike was almost finished with the first coat of UV filler.  It’s wiped on, allowed to dry for about 10 minutes, then the magic UV wand is waved over it.  After that, it’s sanded with (I think)1000 grit sandpaper.  Rinse and repeat, kinda like shampoo.

Finally, clear coat!!!  The clear coat is a UV cured Polyester Resin.  It’s freakin’ amazing!  2-3 coats in a day and it’s harder than nails when done.  It stinks like hell, but it’s certainly no worse than lacquer and from what I’ve seen a lot less hassle.  I’ve never sprayed an instrument, but I have more experience with prepping and spraying laquer than I ever care to admit.  The polyester resin is sprayed onto the uke then allowed to flash for 30 minutes or so.

Then the magic UV wand gets waved over it and voilà! it’s time to wet sand.  After wet sanding, the uke is wiped clean and resprayed and UV-ed.  Once that is done, the uke is gently wetsanded one last time and then buffed out with buffing compound to a nice glossy sheen.

Once buffing is finished, it’s time to glue up the bridge.  The bridge is temporarily installed and a razor knife is used to cut thought the resin and liberate the lacquer tape.  The bridge is then dry fit and clamped to check for fit and any possible problems.

After that it’s popped off and the bottom is cross-hatched with a razor knife to increase the gluing surface area.  White glue is applied and the bridge is then clamped.
While the bridge glue is drying, the fretboard is liberated from the lacquer tape used to mask it off.  The edge of the fretboard is lightly filed to feather out the resin/ fretboard surface corner and the frets are redressed where necessary.

Next the tuning pegs are reamed out and the tuners fitted.  Following that the outer string positions are marked on the nut and rough cut with a razor saw.  Mike uses a spider divider to mark his inner nut slots.  I’ll use the Stew-Mac string guide as I happen to have one.

Finally, time to string it up and hear the moment of truth!  As it turned out, we needed to tweak a couple frets with a quick fret level, but all in all I think it could have been a lot worse.

(Some) Final thoughts:
Building a uke wasn’t near as difficult as I thought it might be.  I can’t wait to build my next one.  There were some steps where it was obvious one needs to pay attention, but there is this whole aspect to it that reminded me of cooking.  Respect the ingredients and just be zen with it and it will come out to your liking.  Thanks to Mike for taking the time to teach me to build and special thanks to our friends with whom I stayed while in the Bay area!

Day 6:

A nursery on the way to Mike’s that makes me think of Mr. Miyagi every time I passed it.  Well, that and I figured my wife would have loved the place…

We started out by drilling holes for the side dots with the pin vise. Mike had some brass rod that I thought would look great as position markers. In the lighting in his shop they looked like they might be subtle, but in actual use I’ve found them to be perfect. Dots were drilled at frets 3, 5, 7, 10, 12 and 14. I see no point in having fret markers on the fretboard itself. Why detract from a beautiful piece of wood with a marker you shouldn’t be using while you’re playing anyway? But I digress…

The rod end is dipped in a little super glue then stuffed in the hole and nipped off with the end nippers. Once they’re all in place, we filed them down flush with the edge of the fretboard. At first the thought of digging up the edge of the fretboard while filing down the brass scared the crap outta me, but in reality the brass gets filed down while the wood goes untouched as long as you hold the file level.

Fretting time! I didn’t actually think to ask Mike which way he prefers to fret necks, but he has a fret hammer (which based on the condition of the plastic head I will assume he prefers) and a fret press caul in a drill press. I tried both methods and I was infintely more comfortable with the hammer than the caul. The caul seemd more efficient, but I was worried about crushing the backside of the neck. (Turns out I did dent the neck slightly and had to sand it out later.) Although slower, I felt I had more control with the hammer.

Fretting was started at the 14th fret (where the the neck shaft ends) then worked up towards the first fret. This way the neck shaft together with the fretboard takes the blow of the hammer and you don’t need to worry so much about snapping the end of the fretboard off with a poorly aimed blow. Mike has a caul that the tenon of the neck slips into and supports the fretboard while you fret the high frets. It’s not a solid block like the 1st-14th frets so it’s a little wonky striking it with the hammer, but on the upside, the fretboard can flex a bit so it actually arcs down slightly as the frets get pressed into the slots making it easier for the frets to go in.

Once the frets are in, the ends are clipped down close to the body, and then filed flush with the fretboard edge. Using a mill file embedded into a piece of wood at a 35 degree angle the frets are beveled in towards the centerline of the neck. After that is the fret dress and leveling.

Next up was glue up of the headstock veneer. I chose tulipwood for this as I thought it nicely highlighted both the bloodwood and the fir. After the glue dried, it was trimmed to shape on the belt sander. As an aside, I never realised how handy a huge-ass belt sander could be. I am sold!

The nut was next. We chose corian for the nut and saddle, mainly cause in my opinion bone is a pain in the ass to work with. For me it’s the smell. I am not a fan of the smell of burning flesh and anyone who has worked with bone will know of what I am talking about. The nut is 1/8″ thick and placed at the end of the fretboard and it’s location marked with a razor knife. Still holding it in place, a razor saw is used to cut through the headstock veneer. The veneer is chipped away with a small (and I mean very small) chisel and then filed flush with a nut file.

The top of the nut should be 0.025″ above the top of the first fret. As it turns out, Mike has a scraper that is 0.025″ thick, so that is laid across the frets and used to mark the nut top. The nut is then trimmed to size on the belt sander. Did I not say it was a handy tool? After that, the nut is glued and clamped. While the glue is drying, the tuner holes are marked, punched and drilled.

The neck is then dry fit and clamped to make sure there are no problems before glue up is started. Once everything is ready to go, it’s unclamped and the neck end, tenon and exposed under surface of the fretboard are covered with glue. The big trick is getting the glue spread and the neck fitted before the tenon starts to swell with glue making it difficult if not impossible to get the neck to fit on correctly. After that, clamp it up and let ‘er dry!

While the glue dried it was time to pick a bridge. I prefer a tie bridge, Mike generally likes the Martin-style bridge. Either way, he has a supply of various styles in different woods to choose from. I didn’t ask, but the ebony bridge stock he had looked like Gaboon to me which was too black and did not go well with the Macassar ebony fretboard. In the end I picked Rosewood, as it complemented the lighter browns in the fretboard very well. Rough shaping was done on the bandsaw, then final shaping was done with sandpaper. The tie holes were drilled using an end mill or small routing bit.

The bridge was then lined up in its proper location, and marked with 2 pieces of laquer tape stacked on each other to create a lip. The perimeter of the bridge was then masked off. Finally, the bridge was removed and a piece of laquer tape was placed in the bridge location and trimmed to fit with a razor knife. After that, the excess tape was removed. Next a piece of 120 sandpaper was laid over the bridge position and the bridge sanded side to side to match the radius of the soundboard.

Two small holes were drilled in but not through the bottom of the bridge and pins were installed to act as guides for later bridge alignment. The bridge was again lined up over the tape and the pins pushed into the wood to mark their location. The pin vise was then used to drill pilot holes for the pins and the bridge placement was checked for accuracy.

The first order of the day was to draw a heel cap shape. (The Bloodwood heelcap was glued on yesterday, which I believe I forgot to mention.)

The trick of course is to make sure the heel cap shape is lined up with the center line of the neck. The rough bevel of the heel was cut very asymmetrically which gave me no end of trouble. Using the rasp I cut a 45 degree bevel just outside the line of the heelcap and used that as my basis for transitioning the heelcap into the neck.

Lots of working with the rasp and then light passes with the half round file to do clean up eventually resulted in a beautifully shaped (in my esteem) heel. At the time, I thought shaping the heel would never end and was a total pain in my ass, but due to its difficulty (and my eventual triumph), I believe (blindly or not) the next neck should go insanely faster.

After I decided I was happy with the neck shape, we went over to Mike’s marble sanding table and ran the taper across it a few times to even up the line.

Next to get sanded on the table was the fretboard face to ready it to accept the fretboard.

With the fretboard thicknessed on the drum sander, we lined the fretboard to the neck center line and clamped them together with the vise. Using a pin vise and a small drill bit, Mike drilled 2 small holes in the fret slots to pin the fretboard to the neck. The pins used were simply nipped from a couple pushpins.

Glue was applied to the neck surface and the fretboard pins lined up and inserted. Using several shaped cauls, the fretboard/neck assembly was then clamped to a table.

When the glue dried, the fretboard was trimmed to rough shape using the bandsaw. The “tongue” of the fretboard was then shaped on the belt sander.

A quick test fit of the neck and then a quick planing to cleanup the edges. After the edges were cleaned up, back over to the marble sanding table to do a final sand on the fretboard edge. Part of the heel near the fretboard needed to get a quick file to bring it back to shape

The radius at the base of the fretboard was done freehand using the roller on the belt sander.

I of course forgot to take any pictures, but today was Wednesday and a meeting night for the Berkeley Ukulele Club. I got to meet several people, all of whom have names but which I’ve forgotten. Well, all except Pam, but that’s because I sat next to her all night and had a pleasant time talking to her. Next time I am in the Bay area, I will have to try to time it on a meeting night again…

Ahhhh, the neck. The neck was the one thing I feared probably more than anything else. How was I with no skill with a file or rasp going to ever be able to shape a proper neck. Have you looked at a uke or guitar neck? Compound radii does not give proper expression of what goes into neck shaping. But I am getting ahead of myself…

The first thing to do is actually get a rough shape cut into it. To that end, Mike has an acrylic template he uses for neck/headstock profiling. Then with a marking gauge (I always wondered what they were for) we marked the center line, then filled it in with a pencil line to make it more visible. The headstock thickness gets cut on the scarf joint jig to keep the line straight. After that it is pulled off the jig and freehanded on the bandsaw.  That takes care of the “thumb” side of the neck. Mike has a 1/2″ thick neck template he screws to the fingerboard of the neck blank and uses it to cut the rough taper and headstock shape of the neck. Then the neck is placed fretboard side down onto the router table and the remaining wood is removed with a flush cut bit to give the neck it’s proper taper.   Then Back to the bandsaw to cut a rough approximate angle for the heel.

Now for the fun part. Neck shaping was broken down into three relatively discrete steps. One, shape the profile and taper of the neck. Two, shape the transition from the neck to the headstock. Three, shape the heel.

To start, visualise (or as I did) draw a centerline down the back of the neck. Symmetry is obviously key here. (unless you’re going for an asym shape, but that’s your problem) The idea is to get the rough shape across just an inch or so in the middle of the neck and then push it out to the edges and then bring it down to thin out the neck when you have a proper profile carved. Carving (in a way) was similar to the braces. Start with a 45 degree angle and then facet again and again. Work one side, then work the other to keep things symmetrical. The rasp was odd to get used to. I really have no experience with a rasp or much with files either.  Well, other than using them in ways I am sure would make most knowledgeable woodworkers cringe in anxiety. That aside, I think I can get my next neck down to half the time this one took. This one seemed to go on and on forever, but I digress…

After the profile and taper is pretty much defined, Mike had me draw in a sort of rough transition shape from the neck to the headstock. I was pretty bloody sure I’d fornicate this up in a major way, but somehow didn’t. Really, it was all about keeping off the taper line and carving down to the transition line, then just morphing it together.  It felt like I was taking for bloody ever to get the neck shaped and in the end it is probably a little thick, but still a nice feeling neck. Certainly not the thickest I’ve played by a long shot, though…

In the meantime, body work continues as it’s ready while I shape the neck.  We removed the clamps from the body and sanded the edges smooth.  The sides tend to facet aroundthe “flame” of the koa instead of bending nice and smooth like plastic would.  A little sanding with the random orbital sander takes care of that.  If it is left, the faceting telescopes into the binding/purfling and you end up with a very wonky line.

The binding/purfling ledge is routed with a laminate trimmer and a 1″ rabbeting bit with a properly sized bearing to get the depth of cut required.  As I was looking for a simpler, cleaner look with this uke, I decided to forgo any purfling.  If purfling is used, it is routed first then the binding ledge is cut.  Mike has a little router carriage set up and a cradle for the uke.  The uke sits in the carriage and you essentially spin the uke around to cut the ledge.  All in all, a fairly simple process.

I decided on a Bloodwood binding only on the soundboard side, so we  took the Bloodwood pieces and stuck them in the bender and bent them much like the sides were bent.  We bent two sets in order to have a backup and picked the best ones to use.  The binding pieces were taped together flush, then inserted into the bender.

To install the binding we clamped the first piece tight at the waist, then added another clamp the depth of the uke.  From there we worked toward each end of the uke spot tacking with cyanoacrylate as we went along, using the accelerant to speed things up.  After the side was trimmed into place, we saturated the binding with the glue and sprayed it with accelerant.  The butt joint is important to get clean, but the neck side joint is less so as the neck will (obviously) hide it.  Repeat for the other side.

We nailed the thickness of the binding pretty damn well, so it was only minimum sanding to get it flush to the body.  I then rounded the edges by hand with 220 paper on a block.  If you look at Mike’s Gallery, you will notice he uses the faux tortoise shell binding a lot.  Having seen too many bloody ugly pick guards on guitars made from faux tortoise shell, I have an extreme distaste for tortoise shell.  However, Mike uses it to great effect as a binding and part of it I feel is because is the binding is just slightly smaller than a guitar’s binding so that it’s not as obnoxious.  Mike tells me it’s because he rounds it over enough that the light shines through it giving it a very nice look.  Nevertheless, I never thought I would admit to liking faux tortoise shell.  Not that I about to use it on this uke, mind you…

Dovetail time!  The uke is placed in the router table sled and lined up with the centerline in the acrylic fence.  Once it is lined up, it is clamped to the fence and the dovetail mortise is cut.  Using the hand router, a 1/2″ plate aluminum jig is taped to the neck with double-sided tape and the 10″ radius is cut into the heel to match the body profile.  60 grit sandpaper is then taped with double-sided tape to the body and the neck is clamped into a vise and then sanded to fit using the body as a template.  The neck is centered and clamped to the router table sled and a stepped mortise is cut into the heel block to accept the dovetailed tenon blank.  The tenon block is cut to fit with the bandsaw, then shaped with a chisel.

We also slotted the fretboard today.  First thing to do is to use the jointer to create a reference edge to assure the slots are perpendicular to the fretboard’s length.  Slotting is done on the table saw with a pinned slotting template.  The saw blade appeared to be a Stew-Mac slotting blade, although I didn’t actually ask.  The acrylic template is attached to the back of the fretboard with double-sided tape and a pin is inserted in the template at the first hole and a cut is made using a sled.  Each successive cut is made by lifting the clamp off the fretboard, moving one pin hole over and making another cut.  Slotting was almost
anti-climactic in its ease.

I decided on a tulipwood headplate.  It looks great with the bloodwood binding!

P.S.   As long time listeners will have undoubtedly noticed, I have finally spent 30 seconds and worked out how to attach pictures to posts, so that should be more common as I finish posting Days 5-7.   In other news, I bought my first uke building tool the other day.  A dead table saw.  Exciting, no!?  I’ll use it to build myself a router table like Mike’s.  It was cheap, how could I say no?

We started the day by pulling the top from the go-bar deck and sanding the sound hole reinforcer and the bridge plate down to a small thickness. After that we cut the soundhole with the flywheel cutter. Total moment of truth. The good news is we didn’t fornicate up the sound hole. It turned out great!

Next up were the transverse braces. First they get thrown on the shallow dish and dished out to arch the top slightly. The soundhole reinforcer gets marked, notched, and chiseled so the braces sit flush against the sound board. From there it’s glue and clamp the braces onto the top. We then glued the center strip on the back. While those dried, we got neck blanks ready.

Mike uses Honduran Mahogany for his necks. High quality wood, I might add. He starts with a 4×3 chunk and cuts it into 4 pieces for, oddly enough, 4 necks. After cutting them on the bandsaw, they are jointed to make the faces parallel. Back to the bandsaw to cut the scarf joint.

After that the end is trimmed twice to get 2 blocks to be used for the heel. He has a swank jig (3 actually) for gluing up the headstocks and heels at the same time. While that glue dries, it was time to plane and scallop the transverse braces.

Planing was not so bad as not much needs to be planed before you go after it with a chisel. The chisel, however, took a bit more coordination and skill than I have, but slowly I increased my competence. Once the transverse braces were shaped it was time to cut and glue the fan braces. The fan braces went on similar to the transverse braces with the exception that the braces were ground out on one side to make room for the bridge plate.

Then it was time to work on the back. We sanded the back strip down then arched the braces. There are 2 back braces, one between the waist and one between the lower bout. Glue and clamp them and then it’s off to shape and scallop the fan braces. Things seem to move fast at this stage…

Once the fan braces are scalloped, it’s time for any touch ups, glue squeeze out cleanups, etc. Then it’s time to glue and clamp the top to the sides. We started with the head block, first with a small clamp to center, then 2 larger cam clamps to actually secure the whole of the top of the head block. Then the same with the tail block. After that it was simply a matter of following the curvature of the body without distorting the sides or top. While I can see some definite potential for some serious fornicating merde up, it went quite smoothly and easily. I admit, having someone who knows what they’re doing helps immensely.

While I waited for the top to dry, I shaped and scalloped the back braces. Once the top was dry, we unclamped it and threw it on the router table and trimmed the top down to size with a flush cut bit. I accidently let it get away from me near the top and check out some of the top. Fortunately for me it didn’t dig in or tear out into the body, so I was saved much heartache. Next time I know to bloody hang on tighter!

Mike then drew the taper in the sides to give the bottom of the uke its shape. Once done, he sanded it down with the edge sander.

Then it’s off to the deep dish and sand the sides where the bottom meets to level them. Then Mike used the random orbital to sand the top and smooth the sides out. When the sides are bent, the curl in the koa “facets” for lack of a better term. Smoothing the wood out ensures a nice feel, but mechanically more important, it gives a clean edge for the binding cutter to follow. Finally, we glued the lining on for the back to attach and cut the headstock down to a straight and even thickness.

It’s bed time. I’ll try to get Day 4 posted tomorrow, although I haven’t written any notes yet, either. The body got finished today and and the neck rough shaped. More later. Hasta!

Too tired yesterday to post so here goes Day 2.

I rough scraped the top and back to remove the bulk of the glue with a cabinet scraper. Then ran the top through the drum sander to do a quick clean up

Then I marked the center line, and used the template to draw the rough outline of the uke using a bearing as a spacer to get some excess wiggle room.

Off to the bandsaw to rough cut the shape. Then the sides are placed in the mold and the centerlines are marked for trimming. Trimming is achieved by eyeballing the line on the bandsaw blade and placing the top face down while making a cut.

It worked perfect since the sides are (in theory anyway) straight as referenced to the top and the blade is 90 degrees to the table. I thought it worked out quite slick. Mike has definitely impressed me in the “simple but elegant” solution area of many aspects of uke building

From there the sides go back in the mold with the bottom of the sides/uke flush with the bottom of the mold. Clamp the waist using the turnbuckle clamp and the head and tail blocks are ready to be clamped in. They get clamped flush to the top as the head block is shorter than the tail block.

The Rosette. I decided on a (I don’t know maybe 1/16th”-3/32nd” wide, maybe a touch more) band of bloodwood surrounded on either side by 2 thin lines of black. Mike has this cool modified/hybrid flywheel cutter he’s machined that has a 1″ clear acrylic disc about 4″ in diameter to keep the cutter from walking. Unfortunately, we measured incorrectly and blew the rosette pattern on my top. Fortunately, we had a back up Fir top. While I have to admit I preferred the colouring of the original top better, the new top has grown on me, but I digress… After the rosette channels were cut we installed the black plastic lines and glued them in. After that they were slowly trimmed flush with the drum sander. To get the bloodwood to bend, Mike boiled some water and essentially made spaghetti. It went in without too much fuss after a couple of initial dry fittings and some minor sanding of the rosette material. (before boiling, obviously) After a couple quick passes of the drum sander to smooth things out, we thicknessed the top with the drum sander down to 0.070″, although it may have been 0.075″. I forget which.  (Did I get pictures of any of this?  No…)

That done, it was time to mark the lines for the transverse braces, fan braces, and the bridge plate. Then over to the Go-Bar deck to glue up the sound hole reinforcer and the bridge plate. After that, I took the glued sides still in the mold and sanded the top edge on the wide radius dish to prep the sides for lining. Gluing in the lining strips went surprisingly easy with the help of many notched clothes pins augmented by rubber bands.

By that point it was too bloody late to head back to San Bruno for dinner with friends, so I had Mike drop me off in downtown Berkeley where I had an overpriced, but delicious Spinach-Mushroom Ravioli for dinner.

Ok, I have notes from Day 3, but am too tired to type them up, so maybe I can catch up tomorrow. Mike is hoping that at the rate we’re going, we can wrap up by Friday. That would certainly make for a mellow Saturday before I have to head home.

Bloody hell did I learn a lot. Hopefully, I’ll remember some of it…

Forrest was kind enough to give me a lift down to the BART station. The sound of the BART is freakin’ *AMAZING*!!!! It’s like bein’ in Star Wars. Every time you go into a tunnel it makes the Tie Fighter scream. I’ll have to record it before I am done with this adventure…

Mike’s 1 hour discourse on tonewoods turned into an almost 3 hour discourse. I am still not sure if it’s because I am that slow at making a decision or he was trying to milk out of me a better comprehension of what I am looking for in a uke. I think he was having some fun because he did smile through most of it, and I actually understood a reasonable amount of what he was trying to explain to me about the qualities of the different woods in the ukes we were comparing. Well, at least from a theoretical perspective. Actually verbalizing it was another matter entirely. Finally, I was able to discern somewhat between the Fir topped, Primavera body uke, a Koa Kamaka he had on hand, and my little Koa topped, Mahogany body uke. If nothing else, I enjoyed the conversation. Even if it was longer than he had anticipated. I also got to look at the Left-Brained Lutherie book, which while I certainly wouldn’t admit to understanding a lot of the book, I certainly understood more than I expected.

I had been toying with the idea of the all Koa uke, because it’s *THE* traditional tonewood, it’s got a nice sound, blah, blah, blah… but ever the one to change my mind at the last second I opted for a Koa back and sides with a Fir top. It certainly would never have occurred to me to use Fir for the top on my own, but his concert sized fish uke has a great sound, (plus it’s cute as all get-out) and I thought it would be fun to live dangerously as it were…

So, as far as actually building the thing today; we glued up the top and back using the most ingenious system I’ve ever seen devised. It involves tape, of course, because what can’t be improved upon with a little judicious use of tape? The hardest part for me was using the planer on the shooting board to square up the edges. Considering my experience with a planer is limited at best, (mmm, make that nil) it went remarkably well. I could see how after a few ukes I could cut my time down to about a tenth, though. Well, as long as I stayed focused for the duration.

We also thicknessed and bent the sides, which believe it or not is much easier than I have been led to understand. Thicknessing was a pretty nifty operation. I got to learn how to read a barrel micrometer, it’s a nifty tool designed by people with a fetish for thousandths of an inch. It’s not 0.07″ (seven hundredths of an inch) it’s 0.070″ (seventy thousandths of an inch), yeah, because there really *IS* a difference, right? It took a few minutes to wrap my head around that one as I am used to shortening things up. I could see if you were working in an industry that relies on a margin of error in the thousandths of an inch, keeping your measurements on the same scale would be copasetic. The thickness sander is pretty cool. I couldn’t imagine trying to achieve proper thickness with a set of planers

As far as bending goes, wet it with a spray bottle then slip it between a couple stainless steel sheets over a preheated thermal blanket an’ gently start squeezing the center caul down onto the waist. From there slowly crank down the bouts always listening for the ping that tells you you’re starting to crack, which is, of course, not good. I haven’t seen the results yet as it was time to go before the sides had cooled, so there’s still time to have fornicated it all up. Still, all in all, a lot smoother than the reports I’ve seen on bending using the Fox style bender.

No work tomorrow as Mike’s busy, but I’ve been tasked with picking out a rosette design. I’ll have to dig through his photos to get an idea. I was thinking maybe just a simple loop of wood around the inside of the soundhole. Simple might look best. No gaudy abalone on this baby…

I took some photos, but haven’t yet worked out how to upload them. Ok, so I haven’t really tried, either. But that’s not really the point…  (Update: 10/03/08.  Photo uploads have begun, yeah!)

On the upside, I scored a ride to the BART station tomorrow, so I won’t start the day off with a 2.5 mile hike.  Tomorrow I should know much more about ‘ukulele’s.  We shall see, we shall see…

So, on Thursday, Sept 4th I am bound for SFO to meet up with Mike DaSilva in Berkeley to build me an ‘ukulele. This blog is predominately for my own gratification and memory, so updates may be sporadic or simply done after the fact. Only the future will tell how lame my documentation will be. However, the experience should be fantastic!