How Digital Scales Work

Scales come in all shapes, sizes and configurations, but the basic component doing the measuring is nearly always a load cell.

A load cell is a kind of transducer, a device that converts one form of energy into another. Through load cells, digital scales change mechanical energy -- the smooshing or stretching caused by a sitting or hanging load -- into an electrical effect. The widely used strain gauge (you'll also see it as strain gage), for example, reads compression or tension as tiny changes in electrical resistance in a Wheatstone bridge [sources: Mashaney; Omega].

Let's break that down using a compression strain gauge as an example.

Compression occurs when an applied force reduces an object's volume, but it can also refer to a more general decrease in size along one or more dimensions. As it happens, squishing an electrically conductive material changes its electrical resistance, because longer and narrower wires are more resistant than shorter, wider ones [source: Craig]. Think of it like water pushing through a pipe: The longer and narrower the pipe, the harder it is to force water through it.

Various materials experience different resistance changes under deformation, a quality known as gauge factor. Gauge factor can also alter in response to temperature. Consequently, constantan alloy (55 percent copper and 45 percent nickel), which performs well at room temperature, has established itself as the go-to material for strain measurements [sources: Encyclopaedia Britannica; Craig; National Instruments; Pratt].

To pick up the change in resistance caused by weight compression, one or more strain gauges are placed within a Wheatstone bridge. A Wheatstone bridge is an electrical circuit that can detect an unknown electrical resistance by balancing it against known resistances elsewhere in the circuit. In a sense, it's like a balance scale for electrical resistance: The "weight" (resistance) on one side tells you the unknown "weight" (resistance) on the other [sources: Craig; National Instruments; Pratt].

A given bridge can contain 1-4 strain gauges. When multiple gauges are used, they're arranged in opposing directions to improve sensitivity and to mitigate temperature effects. Because the resistance change in a strain circuit can be minuscule, the signal often requires amplification [sources: Craig; National Instruments; Pratt].

As a load cell measures compressive resistance change, it transmits a signal to the CPU, which converts it into input for a display board, which then shows the result on a digital screen. This principle remains true whether you use a strain gauge or some other kind of measuring device.

Research, industry and commerce require the capacity to measure weights under a seemingly limitless variety of environmental conditions and space constraints, while also controlling for possible errors. So, although strain gauges are the most widely used type of load cell, they're far from the only design in use today [sources: Eilersen; Omega].

Industries that require greater safety and sterility often turn to pneumatic load cells, which derive the weight of an object by measuring the air pressure necessary to balance it. These blowhards work well in the food industry or within hazardous sites because they don't contain fluids that might seep, drip or spurt into the environment. Pneumatic cells can heft a wide range of weights with high accuracy, but they require a clean, dry atmosphere and tend to take their sweet time responding [source: Omega].

Hydraulic load cells, which measure load as a change in fluid pressure, are commonly found weighing tanks, bins and hoppers. Because they function sans electricity, hydraulic cells work well in out-of-the-way locales where power is an iffy prospect. Pricey and complicated but rugged, these fluid-filled gadgets don't flag or fail, even under million-pound loads [sources: Eilersen; Omega; Takhirov et al.].

The terms "strain gauge," "pneumatic" and "hydraulic" describe operating principles -- essentially, the physical laws governing each weighing technique. But a buyer looking for a scale might also consider a cell's size, shape, configuration, materials and other physical aspects based on the job requirements involved. For example, a customer who needs to take measurements in a wet environment might choose a hermetically sealed canister cell, whereas a facility concerned with clearance issues might opt for a thinner bending beam cell [sources: Mashaney; Omega].

Form factors are also chosen for their ability to minimize the effects of extraneous forces. Scales typically measure force along a single direction called the principal axis -- which, in the case of weight, means downward (in the direction of gravitational pull). Weighing errors arise mainly from off-axis forces, which act parallel to the load, and from sideloads, which act perpendicular to it [source: Novatech]. Thanks to their zigzag design, s-beam load cells excel at eliminating sideload [source: Omega].

Within each of these categories and subcategories, there are sizes, shapes and price ranges to fit just about every need imaginable. But there's more to scales than just weighing your options. Measuring force is as fraught with technical troubles as any precision measurement -- no minor matter, when one considers that human lives -- and piles of money -- can rest on the difference of a few grams.

An old saying goes that a man with one watch knows what time it is, but a man with two watches is never sure. Scales must contend with a similar conundrum: If you weigh the same object twice, do you get the same result? How about on different days, under changing weather conditions? If so, can the scale remain consistent despite wear and tear?

These factors -- respectively known as repeatability, reproducibility and adjustment -- are all important, but a prospective buyer must also consider more basic questions, such as the scale's weight range, including its minimum and maximum capacity [source: AWS].

Then again, those statistics matter little if a device lacks the necessary accuracy and precision for the job. Although the two terms are often misused as synonyms, they're not interchangeable: Accuracy describes how closely an instrument measures an actual or expected value, whereas precision refers to how granular or exact that measurement is. A speed gun that clocks a 90.4213 mph fastball at 90 mph is accurate but not precise; a speed gun that clocks the same fastball at 88.3246 mph is precise but not accurate.

Scales usually express accuracy as a percentage of full scale (FS) -- the total calibrated range a device can measure. For example, a 1 percent FS accuracy can mean +/- 5 pounds for a 500-pound scale, but +/- 1 pound on a 100-pound scale. Alternatively, error can be expressed as percentage of actual reading (AR): For an object weighing 100 pounds on such a scale, a 2 percent AR would mean an error of 2 pounds, whether the scale had a range of 10 pounds or 10,000 [sources: AWS; Busch; Plint].

The precision of a scale is similarly variable. By industrial standard, regardless of capacity, scales are split into a maximum of 10,000 divisions: Hence a 10,000-pound scale will divvy into one-pound increments, whereas a 500,000-pound scale will divide into 50-pound increments [sources: AWS; Crowley; Mashaney]. Weighing devices are grouped into categories based on these divisions (see sidebar).

With this in mind, always try to select a scale that is has a high enough capacity to handle what you are measuring, but not much higher. This will ensure you get the most precision possible, pound for pound.