I had a friend ask me a question concerning his engineering homework some time back and it got me thinking. At first glance, I looked at it as an engineering issue, being such is the world where in I live… and all of a sudden, its like hey wait a second, this is just homework.
This isn’t a real circuit, where in Mr Murphy is lurking around every corner, just waiting to holler out gotcha should you overlook some seemingly minor detail. This isn’t a production issue, where in an error can equate to hundreds of man hours in rework, or an analysis to determine whether rework, or throwing the entire batch away and starting over is a more cost effective approach.
I thought back to my days at uni in circuit analysis class, where in we’d pound on complex resistor networks, using norton and thevenin to solve equations. We’d look at short cuts where in we could throw this and that sort of characteristics away, being their contribution to the circuit of a whole was minuscule. The classes back then served me well for years, and continue to do so.
And yet, there are many times where in that resistor, isn’t really a resistor, its a complex device, with potentially a capacitive or inductive component or both. It may have some weird impedance curves as a function of frequency. It will have thermal dependencies due to ambient temperatures, but it may also have issues with self heating, or heating from nearby components. And then to add insult to injury, it may have its own internal voltage sources with both ac and dc components as well. All of the above could change throughout its life cycle, or could even change due to mechanical stresses, either as one time things due to assembly issues, or as life long concerns due to stresses on the pcb itself. It could also be influenced by environmental factors such as circuit board pollution, crystalline growth on the terminations, or even water or chemical ingress through its packaging.
Fortunately, most resistors are applied in such a way that they are merely resistors, the factors above end up falling into the miniscule category and can be discarded as too small too have much of an effect on the circuit as a whole.
As such, the primary drivers of resistor selection end up being the idealized value and its associated tolerance, the power and voltage rating, its physical size/foot print, and whether it currently exists in inventory. Application specific concerns may require a specific material spec, carbon comp, metal film, and wirewound to name a few. Temperature coefficient, and/or reliability specification may also be in driving factor in some designs.
Alas, for my friends homework, a resistor was just a resistor.