Re: The amount of times...
Hmmm, 100C is where the vapor pressure of pure water is the standard atmospheric pressure at sea level. At 1 to 1.5km of elevation, the drop in temperature at where the vapor pressure of water is equal to the ambient pressure is enough to require adjustments to recipes when baking. The more natural point for 0C would be the triple point in water. Fahrenheit's scale was 0F being the coldest achievable temperature with water ice and NaCl, with 100F being core body temperature. A real SI scale fr temperature would be eV...
For doing thermodynamic calculations, the appropriate scales are Kelvin and Rankine, and there really isn't much difference in usability between K and R as all sorts of conversions need to be done to get answers in Joules or MWHr. Another "fun" problem is dealing with speed involves Joules being watt-seconds, while vehicle speeds are usually given in statute miles, nautical miles or kilometers per hour. A fun factoid is that 1 pound of force at one statute mile per hour is equal to 2.0W (1.99W is a closer approximation).
As for feet, a fair approximation is that light travels 1 ft/nsec, too bad the foot wasn't ~1.6% shorter as a light nano-second would be the ultimate SI unit of length. The current definition of an inch, 25.4mm, was chosen in the 1920's to allow machine tools to handle inches by having a 127 tooth gear instead of a 100 tooth gear.
FWIW, Jefferson wanted to base his unit of length on a "second's" rod, i.e. e pendulum whose length would have exactly one second period when measured at seal level and 45º latitude.
Don't get me started on kilograms of thrust.
Friday 27th January 2023 06:22 GMT
IvyKing
Reply Icon
Re: The amount of times...
From somewhere in the later half of the 19th century to ~1920, the US inch was defined as 39.37 inches equals 1m. According a ca 1920 issue of Railway Mechanical Engineer, the machine tool industry was making a push to defining the inch 25.4mm so that by using a 127 tooth gear to replace a 100 tooth gear a lathe could be set up to produce metric and imperial threads.
One problem with converting the US to pure metric is that almost all land titles use feet, not meters. The US legal definition of a foot was 1/66 of a chain, a mile was 80 chains (66x80=5280), a section of land under the Northwest Ordnance of 1787 (passed under the Articles of Confederation, NOT the Constitution), which was 6400 square chains and the acre being 10 square chains (640 acres per square mile). The surveys for the Townships (36 sections) didn't really start until ca 1796, so if the arrival of the metric standards had not been delayed by the storm and the English, the US might have re-written the 1787 law to use metric measurements.
Another problem with the US converting to metric was Herbert Hoover's success as Secretary of Commerce in setting national standards for pipes and other hardware.
One final note about metric versus imperial is that a nautical mile is defined as 1 minute of longitude at the equator, so works well with the degrees, minutes and seconds customarily used for angles. Metric navigation would favor a decimal system for expressing angles, i.e. the gradians.
doublelayerSilver badge
Reply Icon
Re: The amount of times...
"Fahrenheit's scale was 0F being the coldest achievable temperature with water ice and NaCl, with 100F being core body temperature."
Wrong on both counts. On Fahrenheit's original scale, 0 was the freezing point of a solution of ammonium chloride (NH4Cl), not table salt (NaCl). As neither compound is used directly on roads, the point at which it is not useful depends on which specific salt is being used in the area, and more importantly on where the compound has been applied and whether it has been moved or not. The temperature of the human body was not 100. It was 96. Of course, neither value is considered average for body temperature (and body temperature is incredibly variable in any case, whereas boiling points of things at a specific fixed pressure is stable). This is because the modern scale abandoned both limits by instead fixing 32 and 212 as the values for water freezing and boiling, moving both of the original bounds slightly and making use of the original scale inaccurate to modern users.
Wednesday 25th January 2023 01:55 GMT
-tim
Reply Icon
Coat
Re: The amount of times...
In 1700 it was much easier for a scientist to calibrate a home made thermometer using ammonium chloride cooling bath and a docile dog. The temperature of boiling water required a barometer at higher altitudes and calibration tables. The human armpit temperature of about 96 allows hand drawn hash marks in repeated halves. Many very early Fahrenheit thermometer are often marked every 3 degrees.
Monday 23rd January 2023 15:59 GMT
Michael Wojcik
Reply Icon
Re: The amount of times...
More importantly, for Fahrenheit the reference temperatures aren't 0 and 212; they're 32 and 96. 96 minus 32 is 64. And 32 and 64 are ... stay with me here ... powers of 2.
Fahrenheit based his scale on powers of 2 so that thermometers could be graduated by successive bisection (and then reflected to extrapolate outside that range, on the assumption that the mechanism was sufficiently linear within the desired range). That's an actual engineering reason, unlike "duh humans like powers of 10". There really isn't much reason to favor Celsius.
Kelvin, of course, is the one that matters. (Yes, Rankine works too, but for some SI operations Kelvin is more convenient.)
Celsius is today as much flavor-of-the-month as Fahrenheit is. The original justifications for them are no longer relevant; they're just a matter of taste.
