In a previous post, I drew an analogy to practicing basketball; today I want to focus in on accuracy, and I choose to draw a parallel with the game of darts. While in the US and Canada, darts is largely a rec. room activity, in England it is a professional sport. Now, I’m not sure I would want to describe the dart throwers as ‘athletes’, but some have attained very high levels of skill indeed.
Darts is a repetitive activity done from a set distance, where the object is to throw the dart into particular scoring sections on the dart board. The highest scoring section is the triple-20 bar, which in the picture to the left is the lower red bar within the 20 wedge on the board. To get a maximum score of 180, one has to place all three darts (which comprises one player turn) within the triple 20 bar. Thus the accuracy required is to put the dart into a space approximately 0.25″ in diameter at maximum.
To throw the dart with that kind of accuracy, one doesn’t simply throw the dart randomly at the board. While random throws at the board may well give the bulls-eye occasionally, just like theorem of an infinite number of monkeys typing in a room will eventually produce a work of Shakespeare, it cannot be said that random throwing, hoping for the best, will produce success in terms of hitting the numbers you want. Similarly, one could be content with managing to hit a sector of the board, or a given number wedge, but really, if you want to get to the level where you can reliably hit the number you want you must strive in practice to hit the numbers you want exactly.
It’s the same in joinery work, and what that boils down to is: what level of accuracy do you aim for and how well are you able to achieve it? This relates to the subject of metrology as well, the development of which was a precursor to the industrial revolution. If you doubt whether metrology is of any importance to you, ask yourself if you would care if the local gas station pump was calibrated accurately when you put gas in your rig. Obviously, with fuel comparatively cheap in the US at the moment, the average person may feel he could afford to spill a few drops – when gas returns to the prices it reached last summer, I suspect this attitude would be different.
Of course, in carpentry and furniture-making, measuring to the sub-atomic level is clearly ludicrous, given the inability to machine wood to that degree of refinement, but the question is, to what degree of refinement do we take measurement? If your work is rough and ready, rustic, etc, or involves the construction of, say, a dirt mound covering many acres, then an accuracy of +/- the thickness of your thumb may well be adequate. Some timber framer’s work to the nearest 0.125″, or nearest 0.0625″, etc, – again, where do you stop? Do you stop where you can’t visually distinguish between one measurement and another?
Here’s my take on it: I strive to work to the highest level of refinement my cutting tools can produce. If my tool of choice, for example, is an axe, I am unlikely to produce a surface, reliably, with an accuracy of greater than +/- 0.125″. And that’s with a sharp tool – with a dull axe, I suspect the accuracy would plunge to +/- 0.25″. With a paring chisel, sharp, I could probably achieve a reliable accuracy of 0.01″ on a surface. With a well-tuned handplane, the accuracy can get down to the micron level, as is often seen at Japanese planing meets.
The router can be a most accurate cutting tool. When I first got into woodworking I looked a bit askance at routers, associating them to the ”California Roundover” phenomenon of the 1970’s (on-going perhaps) and as a piece of equipment that people without any hand tool skills would choose to use. I have, fortunately, completely reversed my thinking on this and dropped those prejudices.
It remains true however that one can use a router in a completely inappropriate way, even a dangerous way, and indeed, my experience working in lots of different shops is that most people do not know how to use a router properly. The most egregious examples of poor router use I have seen in timber framing shops, even very high end ones. Now, this might seem a little harsh on my part, after all, the router looks simple enough, how could anyone use it wrongly?
The main mistake: using the router to hog material out. The router excels as a fine trimming tool. When used to take heavy passes of material, the cutter rapidly heats and dulls, the motor strains, the work quality suffers as does the increase in likelihood of a mishap that will spoil the work or cause an injury. One has only to look at the cutter on the router- if it is burnt, that is a sign of using the router the wrong way. Router cutters are disposable tooling, though not exactly dirt cheap. To prolong the life of the tooling, it needs to operate at the right rpm and feed rate so as not to overheat the cutter. Other tools should be used to remove the bulk of the material, and leave the router for the light trimming to final depth or shape. With this approach, the cutters do not overheat, thus they last a long time which saves money, they leave clean surfaces behind, and can do highly accurate work
I didn’t come to these conclusions independently. I also mis-used the router for a good while. But then I found enlightenment when I discovered the router ‘guru’, a man named Pat Warner. This guy is an industry expert, the person that Dewalt and other companies call upon to test their tools and offer technical advice. He’s uses the router to produce accurate aluminum try squares for instance. His books are available readily – one’s listed in the sidebar – however despite title phrases like “Fast and Easy” I would suggest his books are not for beginners.
When you realize how accurate the router can be, a curious thing happens as you plunge down that rabbit hole: the router reveals where the inaccuracies are in other aspects of your work. If your straightedge is not straight, or your square not square, or your router base not flat, or if your router does not plunge exactly 90˚ in relation to its base, the surface left behind will be showing you that. This is similar it would seem to the handplane, where shortfalls in sharpening, fit of blade to block, tuning of the sole of the plane, and so forth, will be revealed by the thinness and quality of shaving pulled from the mouth. The shaving tells the story – which doesn’t mean it is necessarily an easy read however!
At first, all the variables to consider in the pursuit of accuracy can seem overwhelmingly diverse and might cause some to simply give up, assuming that such accuracy simply isn’t possible. However, like a flock of sheep that have escaped from the pen, in time, patience will win out, you will corral those sheep, and you can thus narrow down the variables so as to produce very accurate work. As a result, one’s approach to work will change. I find it a real pleasure to be able to predictably achieve accurate results, like a tenon that is +/- 0.005″ of target dimension, and sometimes I get even more accurate than that. It’s possible to work to 0.001″ with the router, and I haven’t got there yet. Aluminum jigs are the way forward in that regard.
Another area where errors crop up in is simply in measuring. One thing I got rid of long ago in terms of measurement practice is the use of fractions, like 1/8″, 3/32″, 1′-2-3/4″, and so forth. I have found the likelihood of transposing numbers in one’s mind in moving from a place where the measurement is noted to the place where it is applied is too high. 5/16″ is noted, then 5/32″ might be applied, for instance. Fractions confuse -I prefer to use whole numbers and decimals. instead of 10′-6″, I use 126″, instead of 3-5/8″ I use 3.625″. It takes a little while to become familiar with the decimal values, but the trouble is worth it in my view. Also, I use a calculator a great deal, a $12 jobbie picked up at the local pharmacy, and it gives numbers in decimals, so to use 16ths or 32nds I must always convert. It’s easier and one less source of potential error if one can reduce such conversion operations.
In light of this fact, when the opportunity presented itself a few years back to design a new inch-scale Japanese framing square (sashigane) for Shinwa of Japan, I set it for base-10 inches, subdivided into 10ths and 20ths. Thus one can go more readily from the calculator, and for Japanese roof carpentry base-10 is used, not base-12 as in the English imperial system. That square is in production and available here.
Tape measures are a source of inaccuracy. The latest fad in tape measures are the ultra-wide tapes, like ‘Fat Max’ which proudly advertise their ‘stand-out length’. It seems to me that this is might well be a clever marketing trick, using a phallic allusion, and thus having a such a ‘massive tool’ suggests, subliminally perhaps, having a big ‘stand-out’ of your own(!). Ahem! Perhaps I stretch the truth, maybe Stanley created the name for very different reasons.
While being able to extend a tape 8′ or 10′ without it folding over from it’s own weight is handy sometimes, the blade of this tape is very wide and quite cambered across the width (that’s how the stiffness is achieved). Besides making for a heavy tape that, when hooked onto the belt, could pull your pants down, the wide crowned blade must be rotated down to the surface to take or make a measurement, and this is a potential source of error both in the twisting of the tape and in the potential twisting of the tape’s hook.
Another source of error with tapes relates to the hook on the tape end, a part easily damaged if the tape takes a tumble to the floor. If the hook is bent, then your measurements will be off if you reference from that hook. As a result, I no longer trust the hook at all and take my measurements, again using whole inches instead of feet and inches, from the 1″ or 10″ mark on the tape.
Further, if you take several brands of tape measure and play them out side by side, you will find it is often the case they don’t agree exactly. Thus, for utmost accuracy, it is smart to always use the same tape on a given project, and if there are other people working on the same project, they should have the same tape too. My favorite tapes are made by Tajima, and Starrett makes some nice ones as well. I like a tape that measures 25′ with a 1″ wide blade.
Yet another source of mis-measurement can crop up when using thick rulers: parallax error. To avoid this problem, one must be fastidious to view from the ruler to the surface below in an absolutely perpendicular alignment with your eye, OR, (here’s what I do), use a thin ruler which will minimize parallax.
As for measuring in general, my tool of choice for joinery is a Mitutoyo Digimatic caliper. I have no interest in squinting at marks on a rule if I can avoid it, or trying to decide what the measurement is when it falls between divisions on my tape. The Digimatic gives me the answer right away, and since I work in decimal inches, I don’t worry about making conversions very often. Of all the measurement tools I own, my calipers, and I have sizes from 6″ to 18″, are my most valued instruments. That said, they are also surprisingly robust, as my 12″ unit has taken, unfortunately, more than one trip to the ground. It’s still more than accurate enough for measuring wood, though I would leave it out of the machine shop.
And as far as the idea that, “what’s the point, wood moves?” is concerned, I say, it sure does, and you know what, I can know exactly how much. I can mill a piece of wood on Tuesday to 1.505″ out of the planer and check it the next day and find it is 1.518″, and that tells me something useful, especially when looking at the dimensional change in light of the piece’s grain orientation, how I might choose to cut joinery on that piece, and so forth. I’ve taken scraps and cut mortises and tenons in them, at various moisture contents, then dried the pieces in the microwave and measured the changes – this is a great way to learn more about the material and disabuse oneself of various myths concerning wood movement. Wood does move, but it’s not some sort of arcane mystery. Wood, I have learned, is not a material to be mastered so much as it is one to be corraled, temporarily, like those sheep in a pen. That’s what I like about wood, it lives it moves it breathes – it has a fantastic diversity of qualities, and working such a material is an evolving and deepening path of study.
Finally, one of the best ways of getting in touch with wood movement on a daily basis, requires only one, decidedly primitive tool: the wooden handplane. I’m specifically referring to the Japanese wooden plane, or kanna. The wooden block holding the blade, called the dai, moves all the time, not just day to day, but even during the working day. These changes are apparent when one is trying to get thin shavings, where the slightest differences, just like a musical instrument being in or out of perfect tune, make all the difference. some would say these antiquated tools aren’t worth the trouble, but dealing with a wooden plane puts a woodworker right in touch with the daily vicissitudes of wood movement, and as you can see by the video linked above in this post, it seems like it’s worth the trouble considering the marvelous results achieved.