The Gates to the Food Forest

IMAG0560

Well, here they are. Two nice garden gates. Just need a bit of cutting big sticks into little sticks and putting them together. I decided to go with Douglas Fir because, well, its very cheap. But if you get into a fresh bunk of it there are some really good pieces.

Its been drying for a couple months now. It always amazes me how wet framing lumber is these days. As if trucking it by a kiln is the same as kiln dried. So it sits and dries and warps up a bit. Using 2×4 is intentional as well. I know you can get some good pieces of vertical grain from either side of your average 2×12, but 2×4 almost always have tighter grain these days. No more old growth big timber in the construction lumber market.

Each piece is marked at the end for what I will use it for. If I’ve learned one thing about what craftsmanship means as a woodworker it is orienting the grain. Or rather, a sensitivity to the material that allows each piece to find its best potential. You work with what you have and make the best of it, literally.

IMAG0564

It was very hot today and I really enjoyed working barefoot with my feet on the cool concrete floor of my shop. In permaculture there is a saying, “the problem is the solution”. The problem is that you want to punch yourself in the face when you drop a tool edge on concrete. The solution is sitting on the cool concrete while you work, at the planing board, staying cool through the summer, and not dropping your tools.

You know, I’ve never heard of Japanese shokunin using winding sticks. I wouldn’t be surprised if that’s just one more thing I don’t know of yet, but they sure are handy. My warpy sticks become straight, with the help of a 24″ Spiers infill jointer. Would you believe somebody I didn’t even know  just handed me this plane? But that’s another story. Oh, it came to me decked out in the obligatory white paint flecks, a universal constant for antique tools in the western world. I hear in Japan old tools come with red paint flecks.

IMAG0565

And then I became distracted thinking about cutting angles on scrapers. My test scraper is an on Nicholson file, with the end re-quenched to get it glass hard through at the end. I sharpened it on my 220 grit white grinding wheel. The saw plate is probably around RC 44-46 so take it for what its worth when it comes to Japanese saw plate. An 86 degree angle was pretty similar to what I first tried with my carbide scraper. The little curls of metal you see resulted from the corners of the scraper digging in, not good.

IMAG0568

At seventy degrees the scraper finally started to feel like it had bite. The volume of material I was removing increased two fold, and almost all of it was nice curly shavings and not small flake. The cutting edge works smoothly at about 20 degrees from the saw plate, no galling at the edge like the carbide scraper. Much different than the high angle I am used to when scraping cast iron.

IMAG0569

At sixty degrees the cutting edge was very nice for the the first few strokes, and then promptly dulled. I’m definitely thinking of making a few scrapers.

attachment

This is the goal. It was obvious I needed the control and leverage of the drawknife style handles. I’m deliberating now between making a handle from flat stock and a cutting blade from OF steel that can be screwed on to the handle, or going the full way and trying my hand at forge lamination.  I’d rather have a usable tool than an over-heated piece of badly laminated tool steel.  Either way I’ll need to make a Japanese bellows  and pick up some fire brick to get be able to heat treat a chunk of steel that size. I’ve used an prop-oxy torch for smaller heat treating, but its quite limiting if you want to heat something larger than a small carving knife, and not at all good for forging work.

IMAG0572

The work continues. Tomorrow is cutting and planing the kumiko, probably get a good start into the rail/style layout, and then into the kumiko joinery.  These gates are going to be big, 4’x8′ side by side.

I can’t think of a better way to enter the new food forest than a beautiful hand-crafted set of gates.

Diagonal Chomasaru Western Panel Saw

IMAG0554

This is my homage a Sebastian! Reading through his blog I came across an intriguing tooth pattern for cutting diagonal grain:

http://laborlimaetoolworks.blogspot.com/2014/11/diagonal-chumasaru-small-tutorial.html

It immediately reminded me of the Japanese crosscut timber saw that now hangs on my shop wall, with one rip tooth and a deep window gullet for every two crosscut teeth. It has the potential to be the saw that can cut acceptably in any orientation to the grain.  If I could only carry one saw, it would have this tooth pattern.

IMAG0543

I started by trying to scrape a bit of taper into the plate with my carbide scraper. I sharpen the edge at a 90 degree angle on a 220 grit diamond plate. This tool is used for hand scraping precision tool surfaces, usually cast iron, with a granite surface plate as reference. An hour of work yielded only .002″ taper on the .035″ plate and I stopped there. Japanese scrapers, sen, are truly rapid material removal tools, and I could tell that the cutting edge geometry on my scraper was all wrong for this job. Interesting idea though, using carbide. The edge holding on hardened steel is much better than even HSS, and a small piece can be brazed on to a mild steel backing.

IMAG0544

I’m using four crosscut teeth to one rip tooth, breaking off every sixth tooth and leaving the last twenty teeth at the heel intact. Its quite a satisfying feeling to break the teeth on a saw.

IMAG0545

I used a mill file that was just under the width of the window to file down the rip tooth gullets, and a small spiral cut chain saw file to round out the gullet at the bottom.

IMAG0550

With all the window gullets cut I filed the leading tooth to a rip tooth.  Getting the uwa-ba to a flat enough angle left me cutting into the adjacent cross-cut shita-ba. I figured all of the gullets should be a consistent height, so all of the cross-cut teeth would have to move over a bit by filing more heavily on the shita-ba.

IMAG0551

I then jointed down the remaining teeth to the rip tooth height, and  cut down the cross-cut teeth to shape.  I’ve spent a good bit of time with Japanese rip tooth geometry, but have had little practice cutting down cross cut teeth. The initial results were ugly, short, and un-graceful, but sharp. The next day after a good nights rest I went back with fresh eyes and evened up the cross-cut teeth, making them taller, and in general holding more consistent shape. It was really excellent practice, but I can tell I have a long way to go towards elegance. I found that it helped to focus more on making the tooth shape look right than how consistently I was holding the file angles. If the tooth looked right, the yasuri took care of itself.

IMAG0563

Here is the finished tooth line. The gullet between the rip tooth and the first cross-cut is still too high. It will require jointing down a good bit to move the cross-cut teeth back a little farther, but I’ll do it a little at a time as the saw needs sharpening. I left the neutral to negative tooth angles  because I don’t have a lot of faith in the steel this saw is made from to hold an edge well. Good for practice though, to see how far I can push the steel with the tooth shape.

IMAG0553

As per Sebastian’s suggestion, I thinned out the handle on my setting hammer. I removed even more wood from the handle the next day after taking this photo. It helped a good deal, keeping me from gripping the hammer with any force, instead allowing the weight of the hammer to be very clear and consistent from tap to tap while setting. After the initial setting I went back and re-jointed and filed the teeth. The height of the rip teeth were set by laying a small wooden straight edge on the tooth line and filing them down until a fine line of light, a hairs width, showed. It placed the rip tooth edge about halfway down the valley formed by the cross-cut uwa-me. Moving the rip tooth edge lower would probably improve cross-cut performance, but I’m happy with where they’re at for now. Then, after cutting through a bit of wood to remove the bur from filing checked for any last teeth that were still high; there are always a few. Getting those last few teeth sharp that I always seem to miss makes a good deal of difference in the final performance of the saw.

IMAG0557

My shokunin in training, Lilly. Ok, she’s a bit young, but has noticed that I’m a bit obsessed with saws. Mickey mouse tooth pattern anyone?

IMAG0556

Overall, for such a cheap (and I mean cheap!) hardware store saw this tooth pattern has done wonders. Its a bit too low of a tooth count for really smooth cross-cut in framing lumber. This saw, even with its small size, wants to be neck deep in big lumber to cut sweetly. Like, 6″ barn poles and the like. Or bucking smaller pieces of firewood. It excels at all diagonal grain cuts, with equal efficiency rip cutting. The biggest problem I had making it cut well was actually a warp in the plate caused by the warp of the flat-grain handle. I cut off part of the handle to take out part of the warp, and it had the added benefit of allowing fuller use of the tooth line. The handle is still cupping the plate a bit towards the heel, but there is enough set at the moment to follow a line. Short of a new handle I don’t know how to fix the problem.

As a carpenter I carry a similar size panel saw in my tool bag and I may have to switch it out for this one. The age of the saw has returned!

Viewing Saw Plate Distortion

Ok, not the sexiest of topics, but visualizing saw plate distortion is the first step in the path to removing said distortion. I should point out that using fluorescent lights like this only works well on shiny shiny saw plates. For the most saw work I still use diffuse light from the front with the saw plate viewed at a low angle to the incoming light.

IMAG0534

This is one of the thinnest and most flexible of my saws, a Nakaya Eaks 210mm kumiko saw. T-5 fluorescent bulbs put out a nice sharp light and the bulbs are relatively straight due to their process of manufacture. So, above is a corner deformation away from the viewer.

IMAG0535

Corner deformation towards the viewer. These are gross deformations, one should hope never to have distortions this bad in a dozuki.

IMAG0537

This would be what a very bad snag and jerk could do to a blade.

IMAG0539

Pressing up from the bottom to simulate a very large dent. Imagine my finger represented the force-area of a hammer strike. Look at how far out the deformation travels. This is exaggerated, but the point is that dents are almost never as local to the point of impact as you would think. If I were only to work the center area of compression from the opposite side of impact you would end up with an over stretched area in the middle of the blade that would want to pop back and forth like the top of a canning jar. Not good.

IMAG0533

So, can we use what we’ve learned to take a look at the saw without my heavy hand on it? In the above photo I can tell that this saw is far from perfect. The light bends down towards the toe of the saw, showing an overall smooth bend in the folded steel back. There’s a local bend upwards of the very toe of the saw from it hitting something too hard at the end of a cut. The line of light right below the steel back shows ripples, where the back is holding the blade with uneven pressure. Additionally, the reflection of the light tells me that I’ve snagged several teeth out of line, especially in the middle of the tooth line. Now, this is a disposable blade saw, so I don’t get in a huff about having damaged it. I know that when I can use one of these saws without creating so many problems that I am ready for a better hand made saw.

A stone thrown into a pond returns to stillness, but for the ripples of the wind.

Just Can’t Stop Sawing

IMAG0509

My maebiki-oga, showing its back. The line that runs parallel to the tooth edge puzzles me a bit. If I had to guess? Perhaps its the hardening line. I’ve read that only the teeth were hardened on these saws. If you filed past it from use it would require taking it back to a blacksmith to re-harden the tooth edge. But this line isn’t visible from the front side of the saw.

IMAG0496

The welding line for the tang? Honestly it looks a bit rough to me, but I’m not a blacksmith.

IMAG0497

Here, a shot of the front of the saw. When I look at this I see bands of hammer marks, three rows, parallel to the tooth edge. Perhaps the tensioning of the plate? Or correcting a warp from hardening the tooth edge.

IMAG0510

The tip of the saw slipped out of the packaging when it was shipped to me and was bent over very badly. I figured it would snap off when I bent it back, but its held on so far. The crack will grow with use of the saw and I’ll eventually lose it. This tooth also shows the original hammer setting marks, all on the top half/third of the tooth.

IMAG0512

I know most of us have read that too much set on one side can cause a saw to pull out of line, but have you ever tried to purposefully create the problem? I put a ridiculous amount of set in one side of a cheap disposable rip kataba and sure enough, it pulls out of line despite my best efforts to saw straight. Hammer setting teeth have to be judged by eye for evenness, and its something that I’m still a little unsure of my ability to do well. I’ve been sighting down the flat of the saw plate at a very low angle that allows me to judge the projection of the teeth. Its easy enough to see teeth that are over or under set with respect to the ones next to it, but from the back to the front I’m not convinced I can judge it accurately. Ultimately its down to the performance of the saw in the cut, but there are other factors at play that confuse the issue: sawing technique, poorly snapped lines, twisting logs that move as they dry in the process of being cut.

IMAG0520

Both of my yasuri are cutting-down files, and I use the diamond grit one as my Hatsuke-Yasuri. Not sure what brand they are, I bought them at my local Woodcraft store, and spent way too much because they come with a handle. I assume they came from JapanWoodworker, because Woodcraft recently bought them. My other maebiki-oga is the one with the tremendously hard couple of teeth in the middle of the tooth line. I haven’t worked with it yet, but I’ll attempt some audio if I can find an external mic for my phone.

IMAG0518

Here’s another shot of the “chone-gake” chipbreaker teeth. I use a square file to maintain the chipbreaker at about 1mm down from the cutting edge, with a length of the chipbreaker bevel also 1mm long and 45 degrees to the Shita-ba that forms the tooth edge.

IMAG0519

This is a shot from Des King’s book on shoji and kumiko work, which has a great section on maintaining kanna. I’m basically assuming that the maebiki-oga is taking a rough planing cut, and have shaped the chipbreaker accordingly. Des has this great blog we need to bug him into updating more often. His kumiko-zaiku work is just stunning.

IMAG0524

When sawing with the grain you get these great curly shavings, nicely curled to fit in the tooth gullets. When sawing against the grain you don’t get nice shavings like this until your cutting angle is quite low, around 30 degrees to the log or less. The sawdust from sawing down end grain is much more accordion like, also compressed by the chipbreakers.

IMAG0522

Ahh! Terrible loss of tracking to the line on my last log. What happened? I wanted to blame the saw, blame the lines, blame the log for moving on me, but its just my poor technique. Trying to saw from only one side of the log and not using a low enough angle to track well on the line. This is actually something that still I am wondering about.

Kobiki Sawing

The sawyer on the bottom of the log is leading in his cut with the toe of the saw. The cut from doing this is smooth and fast, but I’ve had all my troubles keeping on a line originate from sawing at an angle greater than perpendicular to the long axis of the log. The sawyer on the top is using a low angle cut, which feels very natural because the more of the weight of the saw is helping in the cut. They are both holding the handle of their saws towards the top, suggesting less pressure in the cut than the saw is capable of delivering if you pull from the bottom of the handle.

IMAG0523

Pulling back over onto the line required sawing at a very low angle. This is something that I could never do with my 350mm rip kataba noko. Supposing you are off the line on the same side both front and back of the log. Lowering the sawing angle and pulling over hard like this creates a concavity on the board right of the cut and a convex surface on the left. The thin flexible saw plate of my 350mm saw would then start to follow the cup and I know from experience that it would in fact become more exaggerated unless enough wedge is used. The plate of the maebiki-oga is much stiffer and while this correction does produce the curves I described, the saw doesn’t bend enough to follow the curve. Once I was more than the plate depth past the correction the saw would run without binding while perpendicular in the cut.

IMAG0528

The finished straight fletch cut, live edge boards. I learned much more about my technique from sawing this log without canting off the sides. Basically, I’d rather have accurate lumber than fast lumber. I’m humbled a bit, having to slow down and re-evaluate. The larger the log, the more important it is for accuracy to lead in the cut on the line that you can see, switching sides before over-cutting your line.  The saw really prefers to have at least 10 teeth in the cut. With all the switching back and for it felt a bit hoppy, even with the teeth freshly sharpened. Some of that tendency was overcome by sawing very flat, using a bit of speed on the pull stroke, and holding the saw higher up the handle. Smooth cuts remove the most material, and slow is smooth. Thus slow is fast.

IMAG0530

The payoff for my week of work! Imagine how much I could mill if I knew what I was doing.  The live edged stuff is my favorite. Once I get my sumitsubo carved I’ll get another log or two to work on. I’d like to try some quarter-sawing techniques if I can get the right tree.