Resistors 101: What to Consider When Designing With These Essential Passive Components
Don't be fooled by the resistor's apparent simplicity, there's more to a resistor than meets the eye, or ohmmeter. Design engineers should understand the other parameters that define this ancient, yet essential passive component.
Along with the inductor and capacitor, the resistor is a one of the trio of fundamental passive components in electronics. Each appears to provide just a simple electrical function; for resistors, it is to impede current flow, which produces an associated voltage drop. The obvious primary parameter of interest when choosing a resistor is the resistance value, which can range from a fraction of an ohm, such as in a shunt resistor used for current sensing in series with a load, to megohms needed in low-current, high-voltage situations, such as in X-ray sources.
In addition to the resistance value, there are many other parameters and associated specifications that should be considered. A brief look at some of these additional parameters will convince design engineers that thinking “it’s just a resistor — what’s the big deal?” is simplistic at best and risky at worst.
Start with fundamental factors
Initial tolerance is the variation from nominal value that can be expected when the resistor is measured before it has been used. For many applications, such as a current bleeder across a large capacitor, a standard tolerance of 5, 10 or even 20 percent is adequate. Loose tolerance may also be acceptable if the circuit will be calibrated or trimmed before use. Other situations, such as precision analog front ends or battery load-current sensing, may need tolerance as tight as 0.1 percent or better.
Next comes the power rating. Every resistor dissipates power as a result of the current flowing through it, by the well-known formula P = I2R. Resistors are available in fractional-watt ratings as low as 1/10 W, and with ratings up to tens and even hundreds of watts. While a higher rating may provide a comfort factor, the cost and physical size of the resistor increases with the power rating, so over-specifying is not “free.” At the same time, using a resistor close to its maximum specifications requires careful attention to placement, airflow, and possible heat sinking in order not to stress the part and unintentionally exceed the ratings.
Voltage ratings are another parameter to consider. Most resistors are used at low or modest voltage, in the single or tens of volts, but there are some applications in hundreds and thousands of volts. The resistor’s physical construction must be able to withstand this potential difference across its leads and also meet relevant clearance and creepage regulatory safety standards. If the lead spacing is too close, the high voltage may result in arcing or flashover between the leads, even if the resistor is within its power dissipation rating.
Temperature drift and long-term stability come next
Temperature coefficient and drift relate to the constant challenge faced by all components: the fact that their nominal specifications vary with temperature and will even change with time as the component is used. Resistor vendors specify the drift either as a percent of nominal value called out per degree Celsius, or for precision devices, in parts per million (ppm). Drift values can range from ±1 ppm to several hundred ppm/°C.
Long-term stability specifies how the nominal value changes over time, even as the resistor is not stressed by using it close to its maximum allowable ratings. It too is characterized in percent or ppm change per day, month or even year. This is a form of aging that occurs as the component and its constituent materials undergo the normal aging process, as all materials do. Both temperature and time changes are primarily of interest for analog circuits used in sensor front- ends or for precision measurement of current using a sensing resistor. The vendor may also call out the maximum amount that the nominal value will change even at standard temperature (typically 25°C) as a result of burn-in and aging. Note that before analog circuits were used with microcontrollers and practical techniques for software-based calibration were common, designers of precision instrumentation sometimes resorted to specifying thousand-hour burn-in and thermal cycling of the resistors, prior to when they were installed in the circuit.
Packaging affects DC and RF performance
Fig. 1: The Bourns CR series of molded, SMD thick-film resistors is rated at 1/10 W, ±100 ppm/°C, and is available from 10 Ω to 1 MΩ. (Source: Bourns)
Packaging and lead style are established by the product circuit board or physical configuration as well as the desired power rating. Low-wattage resistors are almost always in surface-mount device (SMD) packages, compatible with standard PC board assembly techniques (Fig. 1a). As these dissipation ratings increase, the available resistor will need to have more or larger leads, may be available only in a through-hole package instead of SMD style, or have an unusual body shape requiring special mounting and handling. Most low-wattage resistors are tiny and getting tinier: the "1206" body is about 0.12 × 0.06 in (3 × 1.5 mm), the smaller 0805 is about 0.08 x 0.05 in., and there are even smaller packages: 0603 (1.6 × 0.8 mm), 0402, 0201, and 01005.
During construction, there are many ways to fabricate a resistor (Fig. 2). The carbon-composition resistor that was standard in the 1950s and 1960s has largely vanished, due to its mediocre performance in terms of initial tolerance and drift, but it is still used in some applications because it can withstand current/voltage surges that might destroy or severely affect other types. Most resistors are now made using thin or thick film deposited on a ceramic substrate, or metal alloy, film, or foil, which combines favorable accuracy, available resistance range, precision, tolerance, dissipation, and drift performance. Another approach is based on the use of wire, one of the oldest and most obvious techniques, and is used for relatively low-value devices. These “wire-wound” devices are just what the name says, built from a measured length of precise-gauge wire wrapped around a form. There are also other specialized fabrication techniques employed for niche applications.
Fig. 2: Despite the apparent functional simplicity, actual construction requires a sophisticated, layered combination of advanced materials. (Source: Bourns)
The operating environment is yet another parameter for design engineers to consider. Most resistors operate in fairly benign environments except perhaps for temperature, but there are some applications where ambient humidity is high (approaching 80 to 90 percent) or the atmosphere is corrosive such as in mining atmospheres. For this reason, vendors offer specialty devices that perform well in these environments, through the use of specialized coatings and lead materials, but at a steep price. Also note that resistors are available as networks or arrays (Fig. 3), with multiple devices in a common package (usually between two and eight resistors) (Fig. 4). These resistors can all be of the same value, but vendors also offer arrays comprised of commonly used groupings, such as a four-resistor array of 10, 10, 100, and 100 kΩ.
Deciding whether to go with individual resistors or an array is a trade-off that the circuit and board designers must carefully evaluate on a case-by-case basis. For a precision analog circuit used in a sensor front end, however, the array version has a distinct advantage: its resistors usually have closely matched nominal values, even if the accuracy is only fair. Equally important, their drift coefficients usually match very closely. For this reason, they are often favored by designers of such circuits.
Fig. 3: This thick-film resistor array of the CRA06 family from Vishay/Dale (10 Ω to 1 MΩ, 5%, ±200 ppm/°C) is available in 4- and 8-pin packages, with two- or four-independent resistors of equal value, respectively. (Source: Vishay)
Fig. 4: The schematic shows how it increases functional density, thus saving board space, shortening the BOM, and perhaps simplifying (or complicating) PCB layout. (Source: Vishay)
Written By: Bill Schweber
Bill Schweber is an electronics engineer who has written three textbooks on electronic communications systems, as well as hundreds of technical articles, opinion columns, and product features. In past roles, he worked as a technical web-site manager for multiple topic-specific sites for EE Times, as well as both the Executive Editor and Analog Editor at EDN. He has an MSEE (Univ. of Mass) and BSEE (Columbia Univ.), is a Registered Professional Engineer, and holds an Advanced Class amateur radio license. Bill has also planned, written, and presented on-line courses on a variety of engineering topics, including MOSFET basics, ADC selection, and driving LEDs.