Monday, September 10, 2012

4 crude benches



Look what I did with my Saturday! These 4 crude benches were fun and mostly pretty easy to make, although not without their challenges.
4 crude benches being tested by my assistants.
 I've had this large maple limb sitting in my backyard for a few months now, mostly because it's too massive to move. Removing the limb was half the motivation for this project, having seating for an upcoming outdoor part the other. I'll walk you through my process really quickly, mostly with pictures.

First, I sawed the log into a couple of sections 4 or 5 feet long. I used a beefy Japanese-style crosscut saw made by vaughan for this. My arm was tired by the end, but it allowed me to cut it right on the ground, which was good since I couldn't lift the limb.
I think it's called a "golok machete". I like it because of the heavy weight balance. It is an amazingly useful tool, and has more or less replaced my little hatchet for yard-work.

I found a natural split starting and followed it by pounding my machete into the end of the log. Mostly because my froe is [intentionally] much too dull for the task.
started to follow the split with an iron wedge
Once the split was started, I just had to follow it with my meager selection of splitting tools., starting with my only iron wedge.

The machete came right out, so I jammed the froe in the end as a placeholder.
My favorite tool, the humble but versatile shim.

Now comes the fun part, because at this point some would say I'm out of wedges. Rather than try to use the head of my axe, I used ordinary wooden shims, like what you might use to level your dresser. I use them in pairs so that they're stronger, and thick enough to do the job.

Skinning bark with the machete. Essentially hewing.
I skin some bark off to get a good look at the crack, so I can find a spot where the shims will do some good without getting twisted or caught, and gradually pound them as far as I can get away with.

Now that the shims are in, I can wiggle out the iron wedge by hitting it sideways a couple of times with the maul.
Iron wedge finding a second purchase.
...second purchase found.

Once free, I look for a spot where the iron wedge will do the most good. At this point I'm just playing wedge-leapfrog.
I hear cracking while pounding shims. I guess I was gently tapping them with the back of this axe.
shims, buried

The second time I use the shims, I find a spot where I can pound them home and I hear cracking several times as I'm doing it. So, they're more than just a bookmark.

The iron wedge in what is to be it's final position

By the third time I position the iron wedge, the log looks almost ready to go.
Success!

A few over-the-head, full-power swings of the maul later, and I get what I was waiting for.
The froe finally gets in on the action.

The log busts open. There are still some fibers connecting the halves, so I used the froe to lever them apart.

Four proto-benches, shown with tools used.
I went through this whole splitting process twice, producing 4 half-round bench seats.

This is where I found the shims.
Incedentally, the maul I used to beat wedges into the log was made out of another branch from the same tree. I highly recommend making one, even if you don't have the perfect tap-root handy. Mine is indispensable for jobs like this.

I thought about propping the bench seats up on some notched log half-rounds, but quickly decided to just make legs out of a hefty dowel. I purchased 6 1-1/2" poplar dowels in 4' lengths from the local big-box and retrieved my 1-1/2"auger bit and brace.
This is as far as I got using the brace.

Boring holes turned out to be a real challenge. My brace just didn't have a large enough swing to generate the requisite torque. I was repositioning my arm 3 times for every revolution. I gave up on that after about 1 inch.
Looks weird, doesn't it?

Since the auger bit has a hexagonal shank, my next idea was a socket wrench. It actually worked pretty well! It solved the problem of repositioning my arm, but it still had a hard time delivering enough torque.
Even this approach got tiring.

At this point I broke down and got my power drill. Last time I used it with an auger it was a little hard to control, but I figured I was out of options. Turns out, all it could do was hum and make ozone--not enough torque.
Electricity didn't solve the problem.

So I found an old vacuum cleaner pipe and slid it over the socket wrench handle.

Physics to the rescue!
With the pipe in place, I finally had plenty of torque. I finished the first bench with this technique. It was still slow going though, swinging that long pipe around a full revolution to earn every 3/32" or so.

Isn't there a saying about the right tool for the job?
At this point I spotted my neighbour and asked him If I could borrow a beefy drill. He was generous enough to lend me a battle-scarred Dewalt that instantly solved my problem. I had to be very deliberate to avoid taking my kneecaps out with the handle, but I felt comfortable with it very quickly and I think I drilled the last 12 holes in about 10 minutes.
cutting legs
I drilled the holes about 4 inches deep and did it all by eyeball. I cut the dowels into 16" lengths and pounded them home. I trimmed only 5 of the 16 legs to make the benches roughly level. I'm really please with the result, although I think I will hew the tops a little so they don't catch on peoples clothes.

Summary?

Summary:

I made 4 benches in an afternoon. That doesn't happen often. These sort of rough building techniques a fun change of pace from the exacting practices necessary to make guitars. The results have their own crude charm (I think), and it's satisfying to get big things done quickly. I'll be doing this sort of work again, and also bringing some of these tools into the workshop more often.




















Tuesday, June 26, 2012

Adam's plant bench is done

Home at last

Completing this project was an excercise in patience. Finishing wood before it's ready is always tempting, and always a mistake, so Adam and I challenged ourselves to be patient.

first coat of polyurethane dries
During smoothing, I used a 50 degree blade in my bevel-up smoothing plane because of the tricky grain direction around the miters, and learned why smaller cutting angles are preferable for softwood. The result was like a much smaller version of the pilling you sometimes see on a sweater.
a revealing moment of truth
We planed, scraped, repaired, made some mistakes, made more repairs, and ultimately sanded it to get the surface we wanted. We spread this over at least two evenings, which was barely enough.

tiles spaced while underlying mud dries
We started with one thin coat of glossy clear polyurethane thinned with a little mineral spirits. I like this finish because it's easy to get an even coat by wiping on and wiping off. Also, I like the way it soaks into the wood, leaving the wood's own natural sheen on display.

spreading grout, a messy process
This was enough protection that we decided to lay the tiles. Having never done tiling before, I was learning from Adam. We did a dry run first to check the spacing, then spread a layer of corrugated mud and set the tiles in their final positions. A bag of tiny spacers and finger pressure was all that was needed to get the spacing right. I was impressed with how easily this went, although the tops of the tiles were noticeably out of flat.
excess grout was no fun to clean off

Spreading grout was entirely different. We spread it over the tiles, packed it in the gaps, and scraped it off. At this point, the tiles weren't exactly clean, but the bulk of the excess was gone. After a few days, it was dry enough to wipe the excess off with a damp cloth. The difficult part of this was finding out that the grout (powder & water) was still fairly easy to dissolve in water, and tended to seep under masking tape, leaving dark stains on the wooden frame. These were sneaky but easily removed with yet another damp cloth. Also, the grout didn't come off very easily. I made some scrapers out of small padouk scraps, which worked well, but there was still a lot of elbow grease involved. I left Adam to do the bulk of this part and cleaned the shop instead. I wonder if the grout cleanup would have been easier if we waited longer.

Clean tiles. Finally.
Back to finishing. We masked off the tiles and went over the parts lightly with steel wool before applying a second coat of finish. After it dried, we both agreed that any more would start to build up a visible glossy layer, so we left it at two coats.

Last-minute adjustments.
Finally, we could assemble and admire the product of our labors, or so we thought. As it turned out, we had forgotten to take wood movement into account for the legs. Fortunately it was one of the hottest and most humid days of the year, so we learned this lesson by finding that the parts wouldn't fit, not by having the bench explode later. We adjusted the mating parts of the legs and top, leaving a little extra room this time, having learned our lesson. One coat of finish on the freshly cut surfaces and we were finally done.

The frame fits together and is ready to accept the top
Now that it's over, I've learned a few things:

- Woodworking is more fun with friends, but more challenging too.
- How to lay tiles. Grout is not my friend, but I'm glad for the experience.
- Adam's idea to make the whole thing disassemble without fasteners is really cool. I may use it again.
- Lots of miscellaneous hand tool skills. This project was crucial early learning for me. I'm very glad Adam was OK with his project playing this role.

It works!
Despite numerous errors, I feel pretty happy with the end result, and looking forward to using the reclaimed shop space for whatever is next. Maybe a tool chest.










Thursday, May 31, 2012

Fiberglassing lesson from Anthony

I have some pictures from yesterday evening, when my friend Anthony gave me a beginner's lesson in fiberglass. (Thanks Anthony!) He has done this on a few wooden kayaks, so I wanted the benefit of his experience when trying it for the first time.
The project was to protect and stiffen the blades of this kayak paddle that I made, but never finished, last summer. I gave a final sanding/scraping , and then we got to fiberglassing.

wetting out the cloth
First we rough-cut the fiberglass cloth to size, with an extra inch or more to spare on all sides. The cloth stretches on the bias easily, so we laid it down on a table and tried to get it to lay flat first. We put our gloves on for this part because apparently fiberglass is not so nice to skin.

Once the work area was all set up and the cloth pieces were ready, we started mixing the epoxy. We had a tough time initially figuring out what to use to measure it, but ended up using some little pvc pipe caps that probably held less than 1 oz. each. The hardener seemed to react with the pvc, but it wasn't in there for long. Next time, I'll probably get some pumps that measure as they dispense, which Anthony has convinced me are the way to do it.

We laid the cloth down and painted the epoxy on with a couple of disposable chip brushes, taking care to avoid bubbles or other unevenness. When first "wetting out" the cloth with epoxy, it goes from opaque to invisible in the span of about 3 seconds, starting when a dollop of wet epoxy lands on it. That was worth seeing!
Epoxy looks the same whether wet or dry.

I took a look at it this morning and the epoxy has gotten harder, but looks exactly the same. I tried slicing a little bit of the excess cloth off just to try it, but that's all the time I had today. At the next opportunity, I'll apply another coat of epoxy before moving on to marine varnish.

Excess cloth slices off easily once dry.

It would seem that applying fiberglass is a straightforward task for anyone with a bit of finishing experience. I'm definitely not going to hesitate to try it again.







Tuesday, May 29, 2012

Cutting up a neck blank

Yesterday I made some more guitar building progress. Every time I work on them, I realize that it's not quite as tough as I sometimes make it out to be. I also decided that I like breaking the process into individual steps, and posting about each one. I think it helps pace myself, or process my thoughts, or something.
I drew out all the steps some time ago. Now I just have to stick to the plan.
The plan was to rip off a piece to use for the headstock. It comes from behind where the fingerboard will go. Ideally, the neck ends up about the right thickness, and I have a nice piece left over to use as the headstock. This way the headstock will end up matching the rest of the neck.

Unfortunately, I had set myself up for a challenge by making a really dumb mistake when I prepared the layers of the neck blank. The piece of cocobolo in the middle is not the full thickness. I tapered it off when I cut it out of a large piece, apparently attempting to save wood. This means that I have just barely enough wood for a headstock, and next to no room for error.
The middle layer of the sandwich doesn't reach the bench. D'oh!

In order to cut out a headstock piece without wasting wood, I needed accurate lines to follow, and before doing that, I had to bring the surfaces up to snuff, accuracy-wise.

There is foamy dried glue showing near the top, but it will get covered by the fretboard.
I planed the fingerboard-facing surface first, making sure to keep it perpendicular to the glue lines. Since the veneer layers were out of alignment, there was some dried glue where there should be wood. Wanting to conserve wood, I planed down far enough to get a completely clean surface only where it will show. There are a few glue bubbles visible in one of the narrow walnut stripes, but these will be hidden by the fingerboard. A relatively fine setting on my cambered jack plane was useful for focusing efforts on one end for a little while. This ebony is difficult stuff to plane, so anything that focuses energy in one area can make planing a lot easier. I finished up with my jointer plane, set as fine as possible because otherwise it's very difficult to push. A 90-degree fence on the jointer helped make planing the two sides easier. I didn't bother with pictures of planing because it's not unique to this project. I left the back and ends rough, and moved on to marking out some lines.

Before I actually marked anything, I wanted to sanity check the dimensions in my drawings. I made a full scale cardboard mock-up to make sure I liked the size. I didn't end up deciding to change anything significant, but I did end up deciding that I will need to beef up the headstock a bit. If the picture of the mock-up looks weird, that's because it's going to be a weird guitar. The whole thing will be tiny, and the body is not integral to the design.
The neck blank with hardware, tape & cardboard gives me an idea of what this thing will look like
The part of the neck blank that goes through the body will stay full-thickness. The rest needs to shrink down a bit to be playable. Since I won't be using a truss rod, all of the strength of the neck needs to come from the wood itself. This is why I've chosen very stiff species. Still, making it too thin now would be a mistake. I'm going to make a final decision on the thickness later, after I've seen it under string-tension. For now, I'm comfortable with 3/4" at the headstock end, and 1" where it meets the body. This leaves just enough excess to use as a headstock.

Chalkdust makes knife lines visible.
I used a marking gauge set for 1" and a 1/4" brass set-up block to mark both depth settings. A square and marking knife defined the other dimensions. These knife lines are great for accuracy, but they're almost invisible. Chalkdust fixes this. I scribbled over the marks with chalk, then wiped off the excess, leaving sharp white lines to follow.
This spacer might have been overkill, but I like not having to change a gauge setting until I'm sure I won't need it again.

I clamped down a ruler to connect the dots with a marking knife. This is important, because if I make two lines by accident, I will undoubtedly end up following the wrong one.
The chalked knife line is crisp and noticeable.
With the lines marked, all that remains is the straightforward task of sawing to the lines. I'm pretty sure I have Chris Schwarz to thank for putting the idea in my head that this is a very important skill. Being able to saw to a line is liberating, because you can draw a line almost anywhere.

I used my favorite saw for this, which also happens to be the only decent rip saw I own, an inexpensive ryoba made by Vaughan (no financial connection etc.). It was important for this step to use a saw that cuts a thin kerf, and to stay right on the line--for at least two reasons. For one, I need to avoid wasting wood. For another, because of the shape of the neck, it will be very difficult to plane the back. Better sawing means less work later.
Sawing to a line is the key.
The saw I was using has a pretty thin plate and only very minimal set. This means it's very hard to steer. This makes it hard to correct for errors, but also hard to make them in the first place. The key to doing a good job is simply vigilance. I watched the line every second I could, and when I got off track, I corrected immediately, before it got any worse. I flipped it a few times so I was always sawing the line I could see. This was easier than I thought it would be, because patience and vigilance make up for a lot of raw talent that I don't have.
Ripping exotics is a good way to work up a sweat. Good thing I had a sharp saw!
To fully liberate the headstock piece, I had to make a small compound-angle cut. This is where sawing to a line has such an advantage over say, a table-saw. I could have done this on my table saw, but it would have required quite a bit of error prone setup. I think it would have ended up taking 10 times as long, plus I probably would have made a stupid mistake along the way. Seeing the line as you cut is reassuring.
This begins to suggest what the back of the guitar will look like.
 After I got the pieces free, I was pleased to discover very little roughness, but still plenty of room for improvement. This is one of those things that just gets better with practice, I guess. The scrap I was left with does look big enough for a headstock, so I'll call this a success.
This scrap should be just big enough to make a matching headstock.
 Next time I have a chance, I will do the same steps on the other guitar. It's softer wood and without the missing chunk, so it should be significantly easier.




Tuesday, May 22, 2012

Gluing up a guitar neck blank

ebony, maple, walnut, maple, cocobolo, maple, walnut, maple, ebony
Guitars are the reason I'm interested in woodworking. Building them is also very intimidating for me. I've built 2 in the past, neither from scratch. I've been designing, working, and mostly procrastinating towards the next two for the last two years.

Chalk demistifies planing dark woods. The high spots are gone.
Fortunately, my other projects have increased my basic skill level, and made these guitars a little less scary.

Even though this is not the goal, these ebony shavings really are beatiful.
The reason I first got into building guitars is because I wanted to own crazy ones that couldn't be purchased. The first one I built included two pickups that each covered only 3 strings. My next one was an electric baritone fretless 7-string from Warmoth parts. And this time I wanted a short-scale, 7-string, neck-through acoustic body guitar on a stick. No wonder building them is intimidating. If that didn't make sense, just keep following the blog and it will become clear at some point.

tape lets you open the layers like a book, but one continuous strip didn't leave expanding gorilla glue anywhere to escape.
Anyway, I thought it might be wise to also build something more simple and mainstream. So I'm also working on a normal-scale 6-string version of the same design. Since I didn't want to use truss rods, I looked up young's modulus on a bunch of wood species to figure out how to make a very stiff neck. I already glued up a neck blank for the 6-stringer out of hickory, with some walnut veneer.

spread the layers
Hickory is kind of a sleeper. It's much harder and stiffer than just about any other North-American hardwood, but not extremely expensive or flashy. Yesterday, I finally swallowed my fear and glued up the other neck blank. This one, for the over-the-top 7-string version, is made from cocobolo (center) and ebony (sides) with walnut and maple veneer layers. This kind of combination of exotics would probably look awful and be ridiculously costly in furniture, but for a guitar, I find it to be neither. Guitars are supposed to be weird.

The Ebony pieces I had were still a little rought from when I hand-ripped them, so I jointed them. A freshly honed 50-degree iron in my bevel-up-jointer did the trick. Very thin shavings were necessary to keep it moving at a reasonable pace through this dense stuff. After the planing, I aligned the wood, taped the flat side, opened it like a book, and temporarily removed the veneer. Taping along one edge is a great way to save time in face gluing, and I needed all the time savings I could get. In retrospect, I should have used several strips across the layers instead of one lengthwise. I used gorilla glue, which expands. Leaving the glue nowhere to escape on one side meant it kept trying to push the veneer out the other side.

wiped off some color
Supposedly cocobolo isn't very good at accepting glue, and I wouldn't be surprised if the same is true of the ebony, as dense as it is. I tried something I read about and wiped the faces down with acetone to remove oils from the surface. It also wiped off some of the color.

Next up, more procrastination. But just a little. I was about to glue up 9 layers of wood in one session. That means spreading glue on 18 surfaces and getting everything aligned accurately, and doing it all really fast. I'm not really sure what the open time is for gorilla glue, but I don't think it's a lot. So I paused and asked myself if this is really what I wanted to do. Maybe it would be safer to break it up into a couple of gluing sessions? I decided to just go for it. What the heck, you only live once.
the right glue spreader is KEY.

I misted the surfaces with water, because apparently that makes gorilla glue work better. Then I got my glue spreader and started working fast.

spreading glue like the wind.
As I went, the veneer started escaping, being pushed by the expanding glue. Eventually I had to get my hands messy, rip some tape off the back and start clamping in a hurry before the veneer moved again. Then I thought about how I wanted the glue lines to look (not full of bubbles), and applied every clamp I possibly could.

clamps!
my first look at the results
A couple hours later, after the glue was dry but not petrified, I returned to clean up the excess foam and take a look at the results. Getting it all off took some effort, but luckily the glue lines were tight and invisible. Once I hacked off the large chunks with a knife, I used a block plane and a chisel to clean it up a little more.

It was all worth it. Well, maybe not the procrastination.
This part made it all worthwile. Seeing shavings with veneer stripes included was incredibly satisfying.

I think you can expect to see more on this project in the near future. This little success has got me jazzed about the project. Also, since I had the idea to build these guitars a couple of years ago, I've gotten a lot more comfortable with things like sawing to a line. Now that I imagine the next step, cutting off a piece to use for the headstock, I realize I'm not even too scared to do it. Hooray!

Step-by-step drawings
Speaking of procrastination and being too scared to work, one positive thing did come out it. While I was having trouble imagining the order of operations for building the neck, I solved the problem by drawing out the steps. Drawing helps me think and it's probably underrated. I'm talking pencil and paper, by the way. It's graph paper for me.








2X each: bridges, neck blanks, fingerboard blanks