Measurements
A harpsichord is made primarily out of wood. While I've never counted the number of individual parts that go into a harpsichord's construction, it must be well into the hundreds.
Most of these parts have three dimensions - length, width, and thickness, and our job is to fabricate these parts as accurately as possible. Unfortunately, this is not as easy
as it sounds. For example, let's consider the Instrument Workshop jacks that we will be using in our instrument. These jacks are said to be 5/32" thick. This measurement appears to be fairly easy to understand, but is
this dimension accurate, and does it matter if it's not? The jacks are actually .153" thick using decimal inches as our method of measurement. 5/32" is .156" using decimal inches. Using
the metric system, .156" translates to 3.96mm, and 3.96mm does translate back to .156", but using fractions (5/32") or metrics (3.96mm) both result in a measurement larger than the
actual thickness of our jacks (.153"). Of course, we could figure out the exact metric or fractional equivalents of .153", but doing so would require many decimal places (3.886200mm) or absurdly large
fractional denominators (627/4096), thereby making them useless.
At this point, some will say that this difference of 3 thousandths of an inch is negligable. After all, we're working with wood and wood will expand and contract more than .003" when the
humidity changes. True as that may be, if a part should be .153" thick and it expands, due to a rise in humidity, to .156", it is now .003" too thick. However, if we make the part 5/32"
(.156) to begin with, and it expands .003", we are now .006" thicker than it should be. In some cases, .006" will be the difference between a part that works and a part that doesn't. Our jacks
must slide up and down in slots cut into the registers (see Chapter 12). If these slots are more than a couple of thousandths of an inch wider
that the thickness of the jacks (.153"), the jacks will be too loose in the slots. If the slots are even .001" smaller than .153, the jacks will not slide at all, but will stick in the register.
For this reason, I try to make all parts as close as possible to their stated dimensions, confident that they will be as close to their stated measurement as possible regardless of any humidity related changes.
Of course, a little common sense is necessary. Worrying about thousandths of an inch over relatively long distances, like the length of the instrument, is ridiculous, but in areas where extreme accuracy matters,
a good micrometer or digital caliper can be your best friend.
My point in writing this techno-babble is not to scare you, but to impress upon you the importance of working accurately. As you will see, we will measure many critical parts by observation and feel
rather than by measurement. The measurements in this book are given in both Imperial and Metric. That said, I normally use neither, but prefer to use decimal inches as I feel it's the most
accurate way to represent a measurement. The jacks are not 5/32" thick, they are .153" thick. This is a measurement I want to know, and the only way to know
it is to use a digital caliper and measure it. It's a good habit to get into.