Robin,
I was always taught that if a number is below zero, you put a zero
before the decimal point. This is just to avoid any confusion about
the decimal point.
If the number is less than zero, you use a minus sign (-) to designate
it. The convention is for metric values (or certain academic
applications) to prefix values between -1 and +1 with a zero. This does
not apply to inch-based values and serves to differentiate the units in
use in the mechanical engineering world.
I was also taught that you put as many decimal points as your
accuracy. This is to stop confusion within calculations and knowing
the accuracy of your data.
The ISO and ANSI specifications in engineering disallow "trailing zeros"
on metric values. The standards call for tolerances and allowances to
be established independently of the numeric value for metric usage. The
convention (dating to 1915) for inch-based dimensions is to allow
"general tolerances" to be assigned based on decimal places used in a
dimension.
If you have 3 decimal point accuracy, then you have to use 3 decimal
places even if the number reads 3.000 so any further calculations know
that you can work to at least three decimal place accuracy.
You are speaking in terms of general, conventional mathematics.
Engineering has different definitions.
I have not read this documentation (It isn't free), "ASME/ANSI Y14.5 -
Geometric Dimensioning and Tolerancing" but doing a search shows that
it is for engineering drawings. What I found on Wikipedia helps me
understand your comments. It does state that tolerances also need to
be indicated. Do you want the tolerances as well?
Unless you do engineering drawing (drafting), my reference is going to
be of little value. The generalized convention (i.e. for most
industries, but exceptions exist) call for an inch-based dimension given
to three decimal places to have a ±.005 (inch) tolerance, a dimension
given to two decimal places to have a ±.01 (inch) tolerance, and a
dimension give to a single decimal place to have a ±.05 (inch)
tolerance. The equivalent conventions for metric dimensions are to
apply "flag note" standard tolerances to dimensions. Tolerances varying
from the nominal standard must be applied individually.
The reason this cropped up for me is that I just wrote a tutorial on the
differences between inch-based and metric-based gearing. I spent more
than three hours "adjusting" values in the metric tables as I could find
no format for a "leading zeros with no trailing zeros" setting. I don't
normally worry about this in spreadsheet operations as they rarely
become "public" and I know what I mean when doing calculations that
way. It's tables in documents that get me slapped on the wrist if I
don't follow the conventions.
While I admit that this is a fairly obscure application, I just watched
a fairly large American company lose a NATO development contract because
they applied the inch-based convention of using trailing zeros to define
tolerances to a metric drawing. Most people will never see such
persnickety application of standards, but they do exist.
.
Lew
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
