There’s No Such Thing as a Simple Component – Take the Resistor for Example

There’s No Such Thing as a Simple Component – Take the Resistor for Example

In any electrical engineering curriculum, one of the first things a student learns are the basic laws related to passive components (resistors, capacitors, and inductors), usually starting with Ohm’s Law (voltage equals current times resistance, or V = I × R). Along with the equations come the schematic drawings, with their North American standard representations. (Other regions of the world use different symbols, but that’s a story for another time.)

Looking at that zig-zag line representing the resistor and knowing its function, you may think, “What could be simpler?” The resistor is defined primarily by its resistance value in ohms, and then perhaps by a few other parameters such as power rating, and that’s pretty much all the student sees. Even in the hands-on lab, nearly all the projects are low power and low voltage, so their resistors take one of two forms: the “leaded” (also known as “through-hole”) version that is handy for breadboarding. An example of one such device is the carbon film resistors.

The other form is the basic “chip” surface mount device (SMD) package, such as the ROHM MCR03EZPD302. The SMD sits atop a pc board but is much more difficult to handle or probe.

What happens next is that this engineering student gets a job, encounters various real-world circuits, and may have to help fill out the bill of materials (BOM). That’s when reality hits, and it can hit hard.

Why so? If you enter the basic search term “resistor” into the Tmartis search box, you’ll get five major categories of fixed value resistors:

Chip Resistor - Surface Mount

Through Hole Resistors

Chassis Mount Resistors

Resistor Networks, Arrays

Specialized Resistors

That’s just the start: if you drill down, there are even more sub-divisions. For example there are high-power, low-inductance, milliohm-range ones used for current sensing, as well as power handling, highly inductive wire wound ones in the tens of kilohms.

Picking the right resistor

Which is the “right” resistor for the project? Sometimes the decision is relatively easy. If it’s a basic, low-power, low-voltage design using a standard pc board, a chip resistor is probably the place to start. Even so, there may be issues to consider:

What initial tolerance is acceptable: ±20%, ±1%, or an in-between value?

What happens when the I2R power dissipation is beyond a small value, less than a watt?

What about temperature coefficient of resistance (TCR), which can be as high as 1000 parts per million per degree Celsius (ppm/⁰C) down to a few ppm/⁰C?

How about self-inductance, which may be a non-issue in a DC circuit, but is a very big deal in a circuit operating at frequencies in the tens or hundreds of kilohertz and higher?

Should you use individual resistors or a resistor array that saves space and provides tracking of temperature coefficient, but may require more complicated pc board trace routing?

There are also issues of reliability, ruggedness, and stresses which are subtle.

What internal and external operating conditions will this resistor face in normal and perhaps somewhat abnormal use?

Is the resistor going to be used in an automotive application where it must meet the AEC-Q200 specification “Stress Test Qualification for Passive Components”, and if so, which of its five temperature grades, 0 through 4, is appropriate?

How about the many military-related reliability standards such as those called out in “Military Directives, Handbooks and Standards Related to Reliability”?

Many larger companies have specialists called “component engineers” whose expertise lies in evaluating the suitability of a selected component for the application, going beyond the top-tier specifications. Often, these engineers don’t get a lot of respect since they are not involved in the “creative” part of design and debug. However, if you don’t work with them early in the design phase, you may find yourself wishing you had. They can alert you to possible traps by asking, “Have you taken such-and-such into account for when the product is in the field, where it will be subject to (pick one or more) extreme temperature, humidity, vibration, salt spray, ESD, EMI/RF…” and the list goes on.

For example, there’s the KOA Speer Electronics RK73H1JTTD0000F family of flat chip resistors. Not only do they meet AEC-Q200 standards, but they also offer anti-sulfuration characteristics due to their use of a high sulfuration-proof inner top electrode material; excellent heat and weather resistance due to the metal glaze thick film, and high stability and high reliability with their triple-layer electrode structure. Note that this is not an “oddball” resistor: due to its AEC-Q200 rating and other factors, they are in big demand.

It’s easy to oversimplify component selection due to a combination of ignorance and even some arrogance. Years ago, at a company that made large materials testing systems with major mechanical design aspects, a senior mechanical engineer (ME) expressed his concern about a potential long-term problem in one of the structural beams. In response, one of our electrical engineers (EE) quipped, “What’s the big deal? Just get an aluminum extrusion to support it.” The ME went into his office and came back with a fat book listing every industry-standard aluminum extrusion along with its profile, tensile strength, brittleness, corrosion resistance, and other factors. He tossed the book on the table and said to the EE, “Go ahead! If it’s so simple, you pick one.”

Conclusion

The lesson here is clear: behind every “seemingly simple” component – even the basic resistor – there’s a lot to consider. For some designs, the key specification is relevantly modest with respect to tolerance, power, size, and operating conditions, where a basic, fairly plain resistor should be adequate. But outside of those cases, there are many second-and third-tier specifications that can make or break a design in evaluation, or even worse, in the field.

Education is a good way to start dealing with this issue. A search will bring up many articles and application notes on these topics. Vendor application notes are also a valuable resource, and while some may be biased towards the vendor’s offerings, a good engineer should be able to sort through the claims and extract what makes sense. Distributor application engineers are also a very good resource, as they have a broad view and provide big-picture perspectives, as well as experience from working with a diverse customer base. Stop, ask, look, and listen, and you will hopefully avoid problems.