Greg and Karl are correct--virtually nothing in building is measured on the
job to millimeter precision. One of the hallmarks of inexperienced workers
is that they try to measure too precisely. That's a waste of time, and the
foreman will soon chastise them, because buildings have loose tolerances.
Buildings CANNOT be designed, measured, or constructed to millimeter
precision. Many of the components (block, brick, dimension lumber) are not
even manufactured to millimeter precision (although plywood is). The
dimensions on wombat architectural drawings are all whole inches. I've never
seen any fractions. So drawing dimensions to the millimeter are about 25
times more precise than needed. It is false precision, because the last
digit is incorrect, whereas on an inch-foot drawing all digits are
significant. (Note: There may be inch-fractions in the written
specifications, e.g., 1/2" pipe, but not on the dimensions labeled on the
drawing.)
>From my experience, I'd estimate that 95% of on-the-job measurements in most
trades are whole centimeters (3/8"). Modular parts are whole decimeters (4
inch multiples). Occasionally, on a custom fit, you may measure to the
nearest 0.5 cm, very rarely to 3 mm or so, but virtually never to 1 mm. So
it seems to me a whole lot easier to use centimeters, occasionally adding a
"point five" or something, than it is to add a falsely-precise unnecessary
digit or useless zero all the time. Millimeters would still be used in the
NAMES of small manufactured parts (e.g., pipes and plywood thickness),
although workers would not actually be measuring them.
HOWEVER, when building a soft-metric building with inch-foot parts, you DO
have to bother with extreme precision--even tenths of millimeters--when
calculating the spacing of parts. Otherwise, the error would accumulate and
the parts wouldn't fit. For example, if you ignored the 0.4 mm when
calculating the horizontal spacing of 16" (406.4 mm) concrete blocks, in a
30 m wall that error would be 3 cm. Since the mortar joint between the
blocks is 1 cm, that would mean that there wouldn't be any room for mortar
on several of your blocks (or some of them would have grossly wide joints).
Your blocks would be noticeably out of whack with each other and with other
parts of the building. This is why builders are so unhappy about
soft-metrication. It's a big headache for them.
Note to Greg: In the USA, auto-retractable wombat tapes, which are
available up to 33' (10 m), have ticks every 1/16" (1.6 mm), with 1/32"
ticks (0.8 mm) for the first few inches only. But the 16ths are rarely used,
and the 32nds almost never. Longer 100' (30 m) and 300' (100 m) reel-type
tapes are marked either every 1/8" or every 0.01 ft (about 3 mm), as you
note. Most long tapes have feet-and-inches on one side of the tape, feet and
decimals on the other. Short, 5 m all-metric or dual tapes are readily
available in hardware stores here, but longer, professional metric tapes are
usually special-order only.
> -----Original Message-----
> From: Gregory Peterson
> >
> I bought a house last year and I have been doing some minor work
> on it. I share your observations, Karl. One does have to get over
> natural tendency to use 1 mm precision. Heck, the standard
> Imperial tape measure is only resolves to the nearest 3 mm
> (1/8-in). I got over the 1 mm precision when I realized that my
> hand held circular saw took a 3 to 4 mm bite!
>
> I agree that writing down a 167.5 cm is quicker and easier for me
> than having to count off inches and fractions of an inch just to
> write down 5'-6". I've never worked in the Imperial format, nor
> will I ever.
>
> For the work that I'm doing, if I'm within 2 mm then I'm happy.
> If it's level then I'm ecstatic!
>
> greg :)
>
>
> >>> "Karl G. Ruling" <[EMAIL PROTECTED]> 2000-10-26 14:01:16 >>>
> I've been building things around my house using SI, and it's not that
> tough. Yes, it would be better if all the lumber were in rational metric
> sizes, but it doesn't really matter much. No piece of lumber is exactly
> what it's supposed to be, so I've never been able to assume that
> a piece of
> nominal 1x3 would actually be 2 1/2" wide. I've always had to measure it.
> If I measure it as 2 7/16" or 62mm, it's the same width, except
> that "62mm"
> is easier for me to scribble on my construction notes.
>
> As for the difficulties of dividing a 4' x 8' panel into three for the
> placement of studs (nominal 16" centers) using SI, this is also
> largely an
> illusory problem. A nominally 4' panel -- if it is truly 4' wide -- would
> be 1219.2mm wide. The 0.2mm is completely inconsequential; if I drag the
> panel across the garage floor I'm likely to rub off that much on the
> concrete, so it's not worth worrying about. So that leaves the
> panel width
> as 1219mm, which looks like a pretty ugly number to divide by 3, but it's
> not. Dividing it by 3 gives 46mm with a remainder of 1mm. I can
> ignore the
> remainder because it's only about the thickness of a US dime. There's not
> much house construction that is built to the precision of the
> thickness of
> a dime! I'm very lucky if I can achieve millimeter-precision in
> my woodworking.
>
> I find that a lot of the problems of working with building
> products made to
> hard-inch dimensions while using SI vanish when I keep in mind
> the level of
> precision required or possible. A millimeter here or there simply doesn't
> matter in most cases.
>
> I do appreciate it that having inch-sized building products does make
> building in SI a mental challenge (you have to get over worrying about a
> millimeter here and there), but I'm puzzled that this was an
> insurmountable
> problem for the Michigan road construction industry. What's the level of
> precision used in pouring concrete slabs or laying asphalt? What's the
> level of precision involved in bending reinforcing bar, binding
> it together
> with rebar tie-wire, and imbedding it inconcrete? Who's using telescopic
> rifle sights to lob water balloons?
>
>
