Friday, November 2, 2012

Basic: American customary units

In most of the world, the standards of measurement are chiefly SI (System International), which is also referred to as "metric". In the United States, these units are used in a scientific context, but in popular use they are not applied by common people.

The units used in the United States are an evolution of the earlier British "Imperial" or "Avoirdupois" standard. British units are still used in the UK, but because of that country's much greater proximity with Europe, it has adopted SI units for some facets of daily life. In the USA, these measurements are usually called "standard," while the international system is referred to as "metric" or "SI." SI units are almost never used casually in the United States, except for liters in some cases.

In a scientific sense, the American or old British standard units can be lumped into the FPS (Foot-pound-second) system, to correspond to the MKS (Meter-kilogram-second) system that is overwhelmingly used in the rest of the world.

The intent of this article is to provide an introduction to the differences between them. I am looking towards my foreign readers in intending to inform them how completely outside the mainstream SI is in the United States, so they do not get confused when Americans fail to understand kilograms or Celsius or kilometers or liters.

Mass versus weight


In SI, the unit of mass is the kilogram (kg). The SI system chooses to axiomatically define mass as the most important, pre-existing inertial quantity. They derive gravitational force from mass; hence force is a derived unit. The SI unit of force is the newton (N). 1 N is equal to 1 kg being accelerated at a rate of 1 meter per second per second.

This is scientifically sound, because in the modern era we know that mass is an inherent, unvarying property of matter anywhere in the universe. However, historically weight and mass were conflated. And in the real world, they often still are, because it's convenient to think that way. Earth's gravity, as it was in the olden days, is nearly homogeneous over the surface of the Earth.

Europeans and Africans and Asians and South Americans describe the masses of various objects in kilograms, and they are correct as far as that goes. But it is far more common to test for weight than mass; an object's mass is usually found on a scale or balance, and this is reliant on the Earth's gravity. If we attempted to find mass in kg by electronic scale on Mars, it would return an incorrect value of mass because this measurement depends on the force of gravity. Actually, one could get around this problem by using a triple-beam balance with known mass values for comparison.

In the FPS system, the pound (lb) is the basic unit of mass... and of force. Say what?

North Americans describe an object's mass in terms of pounds, which is also correct if it is rigorously used just to describe mass. But it's an even more confusing conflation because pound refers to both a mass and a weight. A pound-mass is actually a quantity of mass. It is no longer defined in terms of earthly or astronomical constants, and it is actually defined as 0.45359237 kg. This is exact; it is a defined and not measured quantity, so there is no rounding error. However, the US standard system also uses pounds as a force. The expression "lb" could refer to pound-mass or pound-force depending on the context.

Happily, for those of us who want to stick with standard units, there is an alternative. An invented unit called the "slug" exists, which is a unit of mass derived from the pound. A pound can be thought of as one slug of mass accelerated at one foot per second per second. If you want to hold onto lb for mass, 1 slug is about equal to 32 lb, because the constant of Earth's gravity (32 ft/s) is already taken out of the pound-mass figure. This alleviates ambiguity and puts the pound in the realm of force only.

Notice the philosophical difference in units. In the FPS system, the force is a fundamental unit and mass is derived from it. In the MKS system, mass is a fundamental unit and force is derived from it. There's nothing wrong with either approach. If the Imperial/US standards seem arbitrary, remember also that the kilogram is based off a single platinum-iridium bob that's held in a secure facility in France.

Decimals versus fractions


The decimal approach, which is the very basis of SI, is the best when used in a laboratory setting in conjunction with conscientious evaluation of significant figures. I have blessed scientific notation and the easy scalability of decimal numbers when you are taking precise measurements that might end up in an academic paper. However, it doesn't lend itself as well to rough-and-tumble estimation.

The first gripe I have with decimals is the ease with which proper significant figures can be ignored and abused. Let us presume we are high school students measuring the length of an object with a digital caliper. We have access to so many digits of precision that the urge is compelling to use them all even if they are meaningless. The caliper can be loosely placed around a pencap to get a value of 2.22 cm or pressed harder to get 2.18 cm. The real value shouldn't falsely claim itself to be 2.22 cm or 2.18 cm, but you will see this mistake made all over lab textbook pages. Realistically there is enough variation that you should truncate it to 2.2 cm and be happy with that; only two significant figures are valid. Realistically, if the speed of a Corvette ZR-1 as it crosses the mile mark on the Bonneville Salt Flat is 176.549 mph, best to just leave it off at 176.5 mph with practical tolerances of the equipment in mind.

To conclude this example, let me frankly say that most people who measure things are far too trusting of their equipment, and the notion of significant figures is not widespread enough. Too many experimenters just simply trust the data given at face value. A fractional system is not applicable for all situations, but when it can be used, it requires no background knowledge of significant figures; the smallest fractional value is simply the quantum of measurement, and can be set at whatever the practical requirements of the moment are.

In fractional-based mathematics, the real world is always "breathing down your neck," so to speak. If you are a carpenter in your home trying to build a bookcase, attempting to measure something more precisely than 1/32 inch is meaningless because your saw cannot cut any more precisely than that. Dividing a length of measurement into equally spaced parts inherently yields a fractional system. It is only when the gaps are specifically placed so as to have nine smaller markings between each larger marking, that a decimal system works. 

In the real world, measuring to the nearest 1/16" and leaving off there is actually faster than estimating how many millimeters are between the neighboring centimeter marks of a measurement. When it comes time to actually measure something, I instinctively prefer a fractional way for quick and suitably precise measurements.

Distance


If a distance is given in meters, we Americans can sort of figure it out by realizing that a meter is similar in magnitude to a yard (which is 3 feet). But talking of kilometers requires an estimation that can't be done mentally on the fly (1.6 km = 1 mile). 

I have heard foreigners speak with disdain about the confusing nature of American distance units. They are absolutely entitled to that opinion. But I think the charming variety of names in the old standard system ensure that they are likely to stay around for a while in this country. Every American child knows it's 12 inches (12") to the foot, 3 feet (3') to the yard, and 1760 yards ( 1760 yd) to the mile, or (more directly) 5280 feet in the mile. They do require memorization, but if you memorize it, you will simply never forget. Apart from that, miles are sometimes mentioned fractionally (as in the quarter-mile dragstrip), but more often as decimals (37.4 miles to destination).

Temperature


The SI unit of temperature is Kelvin (not degrees Kelvin, the unit itself is just called Kelvin, abbreviated K) but in common usage, degrees Celsius are instead used. The magnitude of a Celsius degree is the same as a Kelvin, but 0 K is set at absolute zero so that there can be no negative temperatures in Kelvin, while Celsius is offset by +273.15 so that 0 C is set at the freezing point of water at sea level.  Celsius degrees were made of such a magnitude that the boiling point of water at sea level is 100 C.

Scientifically speaking, the Kelvin/Celsius scale is faultless and easy to use. It makes sense even for American students. We are already happy with it in academic circles because of its solid reliance on absolute zero and the easy way to remember the boiling point and freezing point of water.

By comparison, the Fahrenheit system used in North America doesn't set 0 F at any particular value of interest. The freezing point of water is +32 F, and the boiling point of water is +212 F. These values are known by all young children in North America, but they do not make intuitive sense to SI users.

You are not incorrect if you state that 0 and 100 are easily remembered. But in the real world, if you find it extremely difficult to memorize 32 and 212, you aren't really trying at all. I'm positive I haven't forgot either a single time since very early childhood.

If we discount the fact that 0 F is not scientifically meaningful (which is a sign that the normalization of this temperature scale is somewhat arbitrary), then Fahrenheit is actually fantastic for human casual use, because it is more precise than Celsius without resorting to decimals. A Fahrenheit degree is rather close to the threshold of temperature detection by a human being; that is, a change in temperature of +1 F or -1 F can be noticed by a human, whereas Celsius degrees are larger and consequently more precise measurements of temperature often include a decimal point, which is irritating to me.

For human experience, the 0 and 100 points in Fahrenheit do not seem so arbitrary to me. If we aggregate the climates of the world, I think we can broadly say that 0 F is a very cold winter day, and 100 F is a very hot summer day. 50 F is a midpoint that might correspond to a cool day in spring or autumn. Whereas in Celsius, 0 C merely corresponds to a somewhat cold day, meaning winters often need to handle negative numbers, while there is no convenient threshold value where something becomes "hot" in Celsius.

That is, of course, just a personal opinion.

Volume


In a certain sense, liters have totally overtaken cubic inches in regard to engine displacement. There are now only a very few American vehicles (the newest Hemi 392 Challenger and a special-edition Corvette 427) which even provide their displacement in cubic inches, and of course their brochures will also list the size in liters.

On the other hand, liters do not retain familiarity for dry volume at all in America. You will never see calls for metric-based quantities in recipes.

I worked in a brake shop and I can tell you, at least as far as steering, braking, and suspension are concerned, that you have to go very far in the past (20+ years) to find wide use of non-metric bolts and fasteners. The only exception  I encountered is some lug nuts, typically on Dodge full-size trucks, which are 7/8", which is not really an exact fit with any metric socket. Apart from that, I never reached for an inch-sized wrench or socket. The US automakers never operated in a vacuum and keeping tooling in two different standards for different region was a costly luxury. Since we could hardly expect the rest of the world to go to inches, we went to millimeters as far as tools were concerned. There is no metrication needed in the auto shops of America: every decent mechanic has a full set of metric tools, and probably uses them far more often than inch-sized tools.

But in the kitchen, prepare to be bewildered by the variety of measurements in North America. Talking of dry volume measurements, we usually start at the teaspoon (tsp), three of which form a tablespoon (T or Tbsp). Liquid measurement starts at the ounce (oz), eight of which form a cup (c), two of which form a pint (pt), two of which form a quart (qt), four of which form a gallon (gal). Two dry gallons form a peck (pk) but this is rarely used. If you're a farmer or a restaurant manager, you'll also retain familiarity with the bushel (bu), which is equal to eight gallons of dry volume.

For some real world interpolation for the benefit of my foreign readers, let me start by saying that a teaspoon is 5 mL. Bottled beer typically comes in 12 oz portions, of course pint glasses contain 1 pt, individual servings of soda may range from 12-20 oz., and milk is usually sold in half-gallon or gallon containers. Petrol, which is always called gasoline or gas in the US and Canada, is sold by the gallon, which is roughly 3.79 L.

No comments:

Post a Comment