A Community Devoted to the Preservation and Practice of Celestial Navigation and Other Methods of Traditional Wayfinding
From: Tony Oz
Date: 2021 Nov 16, 09:15 -0800
To refresh my long-forgotten ability to use logarithm tables I recreated the four-place one I used at school - by W.M.Bradis. (beware of typos in the scanned edition! my recreation is more reliable :D )
Then I tried to solve the division step of some all-Haversine azimuth calculation - which involved the divident == 0,2410 and the divisor == 0,3271.
0,2410 = 2,410·10-1; 0,3271 = 3,271·10-1
These "minus one" powers of ten mean that the logarithms of both numbers will have the "bar one" prefixes further denoted as "/1".
log(0,2410) = /1,3820; log(0,3271) = /1,5146
Now the subtraction step:
The resulting "bar one" prefix is born when I borrow from the minuend's characterisic digit - the "minus one", making it "minus two", then subtrahend's "bar one" changes its' sign and becomes "plus one": -2 + 1= -1, thus the "bar one" in the result.
Note that the fractional part is exactly the mantissa of the result, ready to be reverse-looked-up in the table - for exponentiation.
Now exponentiating the resulting mantissa (the 0,8674) I get 7,369. The "bar one" characteristic means that the result actually is 7,369·10-1, or 0,7369.
This is how I did calculations until electronic calculators and computers made me lazy and stupid...
I wonder how the "10+" logarithms would work in this example?