Friday, 20 September 2013

Standing Tall...Not Me - The Plane...

It took a bit longer than I expected, but the 10½ now has a mount, or at least one that is as far as I can take it for now.

Here is where the mount stands right now...




There is a cap missing at the left end of the base which has to be added to lock in the sliding dovetail of the heel mount and extend the base to its full length. I can't add that piece until the final glue-up because, for strength, it has to be glued to the finished bracket below it, rather than to the end grain of the base. The right bracket is also unfinished as it still has to be altered to accommodate whatever tool I decide to mount below it. Until that happens, I have to be able to disassemble the entire mount.

While I stated I wanted to keep the mounts for each tool as separate as possible, I just couldn't resist including the left bracket that holds up the 10½ with the upper mount for the No.7 mounted below. This resulted in remaking the No.7's upper mount, but I think it was worth the effort and wasted material, despite the maze appearance. I thought I left enough waste on all the upper mounts, but as it turns out, I didn't.

Yes, I know, if I had drawn up plans beforehand, this problem wouldn't exist, but realize that drawing up plans for a layout like this would probably take as long to do as it would to make the layout. Also, in a case like this; mounting a tool is one thing, but getting it in and out of that mount is quite another. Because of that, it is almost impossible to create a good set of working-plans in a 2-dimensional space and less than fool-proof creating them in 3-dimensions. You just can't take into account all the variables that come into play when it is time to place the tool into its mount and get it out again. I know, I tried as, remember, this is my fourth attempt at fitting-out this cabinet. For me, the challenge of a job like this is far more enjoyable working from a idea that is liquid in my head then it is working from a fixed commitment on paper. If I drew up a plan, each part of the design would have to be followed because all the other parts depend on it being made as it was drawn. If you change one part, you throw all the other pieces out of whack. You also can't think of everything. While I commit to creating the piece following the idea I have for it in my head, once I get into the actually making of it, designs and concepts come to mind that I never considered previously. If I was working from a plan, I would either have to ignore those new designs and concepts, or commit to them and risk throwing all the other pieces in the plan out the window. For me, not only is working with liquid ideas is more enjoyable, I think it makes the most sense in the long run.

Ok, so this mount is like the rest; made from solid walnut stock. Overall, it is, or will be when the toe cap goes on, 11" long and 2½" wide.

The sliding heel mount is made from two pieces; a 2" by 2¼" by 1" block for the top piece and a 1" wide by ³⁄₈" thick by 1½" long piece for its dovetailed slider. These two pieces are joined by glue and two small screws from the underside so replacing them will be possible if one or the other breaks down the line.

The toe mount was also made from two pieces; the top piece having its grain running with the base and the second piece, the spacer, glued under it with its grain running across the base. These three pieces were cut to size individually with the only fitting done to the spacer, shaping one edge so it fit tightly against the toe of the plane. Once all three were to size, they were glued together and then shaped as one.

The dovetail pathway was cut in the base using a fine Disston dovetail saw, then cleaned up with a chisel and smoothed with sandpaper. The dovetail slider was shaped using a plane then, like its pathway mate, had its edges smoothed with paper.

The two brackets that support the mount were made together from one piece of stock 2" by 2" by 8" long. I created a step-down in two of its corners on both sides of the stock using three sizes of Forstner bits. I then rough-shaped these and the stock between them using gouges and chisels, smoothing the results using blocks and sandpaper.

Once I had the different pieces shaped and set together in their final configuration, I then decided which edges and corners would be rounded over and by how much. I used a chisel to rough them in to define their radius, then smoothed them out with sandpaper.

Sandpaper is frowned upon by many, and while it is true that a good plane and scrapper will work rings around sandpaper on larger stock, with these small, fiddly, little pieces I didn't have much choice. Besides, I like working with sandpaper as it allows me more control when it comes to finer details. Maybe this comes from all the work I have done on car bodies over the years, I don't know, but it works for me.

Because discussions of sanding are not covered as much as other tools on the web, I have assembled a few rules for using it below. These rules are based on what I learned decades ago while working as an apprentice auto-body man. My love of cars got me into that apprenticeship, and while I'm thankful for it teaching me that I didn't want to grow up to be a body man, it also taught me what I did want to know, like how to shape material using different tools, including sandpaper.
  • When you start the sanding process, get comfortable, your going to be at it for a while. To sand properly, it takes much more patience than you think. With this job, I have gotten into the habit of hauling the stuff I need, including a drink and snack, out onto our front porch and settling down on one of its steps for the long-haul. The open air disperses the dust and the neighbourhood traffic offers me enough distractions to keep me interested in continuing on until the job at hand is done.
  • Whenever possible, use a block, no matter how uncomfortable using one turns out to be.
  • If you are working a flat surface, use a flat block. The longer the stock, the longer the block. This may mean buying your paper in rolls, rather than sheets, if you are working longer boards. There is a point, however, where even a dedicated sander like myself will turn to a long plane with a sharp blade and a scraper with a good hook over an array of sandpaper in assorted grits. There is no denying that, on longer boards, the former is much better and faster than the latter.
  • When working a rounded surface, make or find something hard that fits the contour you are working on to use as a block. As my wife never reads these posts, I can be honest here and tell you that her stainless steel container that she loves having on our kitchen counter to hold all her fancy cooking utensils has a radius that is an exact match to the curvature of the toe on a Stanley plane. I think you can imagine how I came to know this.
  • Use course grits to waste stock and rough-shape the larger areas of the piece. I never go rougher than 80-grit. The exception to this is when I'm working harder wood, which can force me into using 60-grit to help get the job done in a reasonable amount of time.
  • To get the initial shape true and flat, whether a contour or a flat, wrap an appropriate block with the sandpaper and work in a 45° angle to the grain, first in one direction, then at 90° to that first angle. Working in cross-grain directions will cut away the high spots and not allow the different types of veins in the wood to affect the block's travel. Once you have the area flat or evenly contoured, sand it again, still using the block, in the same direction as the grain to remove the scratches.
  • Change your approach to the piece every so often to "see" the surface. This sounds weird, but when you sand, the pressure you exert on the paper is stronger on the forward stroke than it is on the return. Changing the approach, or the direction you are working from, results in a change in the direction of the paper's cut, raising the grain in an opposing direction. When you look at the wood under properly angled light not long after you started working from the new direction, you can quickly see areas that are slightly different in colour. Usually, the darker areas are the high spots and the lighter areas are the low ones, giving you a guide to tell you how you are progressing.
  • To keep edges square, never allow more than a quarter of the block's length to pass beyond the end of the surface. If that requires shortening your strokes considerably, then so be it.
  • Change your grit according to the area you are shaping. The finer the detail - the finer the grit.
  • Once shaped and you begin to work a finished surface, always, always, always work the paper in the same direction as the grain, no matter what grit you are working with and no matter how uncomfortable or short your strokes have to be.
  • Don't leapfrog over grits to finish the piece quicker. It doesn't work. The pieces I have made so far are NOT ready for finishing with varnish. They are only shaped. When I am ready to finish them, I will return to the 80-grit wrapped blocks to ensure the larger surfaces are true, then go to 120, then 180 and finish with 220. I will then wipe each piece with a damp towel. One purpose for this is to raise the grain, which I take down with 220-grit. The other important reason to do this step is so you can "see" scratches and flaws. Any scratches and flaws that show on a damp piece will also show even worse on a varnished piece. 
  • To explain how a finished surface works; when light hits a surface with imperfections, it bounces off at different angles, which you see as a dull finish. When light bounces off a truer surface, the rays are reflected in a more uniformed direction and you see it as being shiny. Varnishes even out the surface and make it truer, causing the light to reflect uniformly. Special "flattening" additives are added, the amount dependant upon whether or not you want a flat, sheen, semi-gloss or high gloss finish. A flat varnish has a great deal of these additives while a gloss varnish hasn't any at all. They all produce a truer surface, but those with the flattening additives cause the light to bounce in a non-uniform manner, not only off the surface imperfections, but off the varnish's additives as well, masking the surface imperfections in the process. With gloss varnish, the truer surface reflects the light uniformly while the surface imperfections don't, making them stand out like a sore thumb.
  • If a piece is going to be stained and/or varnish, I wouldn't go finer than 220-grit. If a surface is any smoother than that, the adhesion ability of the varnish will be compromised. The Varnish needs a "tooth" to hold onto, seeping into their nooks and crevices. Make the surface to stick and the thick varnish won't have anything to seep into. Wax raw wood, however, is a different story. When waxing, the smoother the surface the better. If a piece I am working on is to be only waxed, I will continue on from the 220-grit with 280, then 320, doing the final sanding with 400-grit. It doesn't stop there, though, as I then apply multiple coats of wax using a progression of ever-finer steel wool, the steel wool taking over where the sandpaper left off. When working very hard woods, I'll follow the same process but instead of stopping the sanding at 400-grit, I'll do one more with 600-grit, then apply the wax as explained before. The steel wool isn't as effective on hard woods as it is on soft, so I find that extra grit helpful in ensuring a good finish.
As for the cabinet, here is an over-all shot of its slow progression.


I'll see you again in a few days to show you what I end up mounting beneath it.

Peace,

Mitchell