When did people start using Revolutions per Minute (RPM) to measure motors, engines, other devices and where did the term originate? Why do we continue to use it instead of an SI unit like Hz?
From car engines to disk drives and fans we measure RPM in the thousands instead of HZ, e.g. 3000 rpm or 50 Hz. I realize its a bit of a convention now but it seems odd.
We often forget that even the minute is not an SI unit, only the second and its decimal multiples and fractions are. It is a leftover of the sexagesimal system (base 60), whose use predates the decimals by many centuries, and goes back to the ancient Babylon. So are the angular degrees.
Heinrich Hertz was born in 1857, and the International Electrotechnical Commission adopted his name as an SI unit in 1930. But back in 1827, for example, we find revolutions per minute (RPM) throughout Farey's Treatise on the Steam Engine. On p. 620, for example, it shares the page with such still current non-SI units as feet, pounds and horse-powers:
"A cast iron neck of 1 inch diameter, and 1 1/5 inches long, is capable of sustaining a force of torsion, amounting to 88 lbs. acting at half a foot distance from the centre of motion, without any risk of injury, for it will require ten times that force, or 880 pounds to twist off the neck. If such a neck made one turn per minute, the mechanical power it would transmit, may be represented by a force of 88 lbs., acting uniformly through the circumference of a circle one foot diameter, or a space of 3.1416 feet per minute; which is equal to 276.46 lbs., acting through a space of one foot per minute. This, divided by 33 000 lbs. for a horse-power, gives .00838 of a horse-power, which may be transmitted by a cast iron neck one inch diameter, and 1 1/5 inc. long, when making one turn per minute. Or conversely, such a neck must make 119.37 revolutions per minute, in order to transmit one horse-power."
This is followed by detailed instructions on how "to find the dimensions for the cast iron neck of an axis, which is to transmit any given number of horse-power, when it makes a given number of revolutions per minute", with a reference to millwrights and Watt's steam engines. Indeed, revolutions per minute are at least as old as horse-powers, see The Origin of “Horsepower”. In 1760s, when Watt wanted to impress upon people how powerful his newly minted steam engines were, he calculated how many horses just one of them could replace. It was PLAN/33000, where P is the mean effective pressure in the cylinder, L is the length of stroke in feet, A is the area of the piston in inches, and N is... yes, the number of revolutions per minute. The 33,000, that we meet again, is just 1 horse-power, in foot-pounds per minute. In the SI, it is 745.7 Watts, and the former has the same advantage over the latter as RPM over Hz.
So why are degrees, minutes, feet, pounds, horse-powers, etc., still around? People grow up with them, they make their way into various designs and official documents, accumulate a tail of working intuitions and connotations in practical contexts, textbooks and manuals are written around them. Abstract arguments about advantages of unified measurement units, even supplemented by government decrees, are as effective in washing them out of general use as the efforts to prescribe the use of words through dictionaries generally, i.e. not very. And RPM have an extra advantage of intuitive clarity. It is somewhat different with the use of terms and units by the scientific community, which is much more receptive and disciplined, but even scientists are people too.
