How To Choose A Resistor

How do we choose the right resistor when designing and building an electronic circuit? Here are the major parameters that should be kept into account.

A resistor is a component made out of a poor conducting material, so that it can offer a resistance to the flow of the current.

You can think to resistance in terms of the obstacles that charges encounter when moving from one end to the other of a conductor. The more obstacles, the higher the resistance. In a metallic wire, for example, the charges are the electrons of the conduction band (see this post and this other one for further details).

In today’s post I would like to address an issue that sometimes is underestimated when designing an electronic circuit: how to choose the right resistor for the job.

Resistors are not all the same. Besides the resistance value that distinguishes one from the other, there are other factors that are important as well.

Here is a list of all the important factors, why they are important, and what are the consequences of not choosing a resistor based on each specific factor.

• The first thing that comes to mind is the tolerance, which is usually provided on the body of the resistor itself, along with its resistance value.

In color coded resistors, the tolerance is defined by the band that is far away from all the others. In the above picture, for example, it is the gold band, which means that the tolerance is of 5%. In other resistors, where the resistance is explicitly written on the body of the resistor, the tolerance is usually written in clear along with the resistance. More in general, you’ll have to refer to the data sheet provided by the constructor to figure out its tolerance.
Tolerance is an important factor for those circuits that require very precise resistors, like measuring instruments and the like. It is also important when the resistor is used for the polarization of a critical component. If the resistors used in the project have a tolerance that is too high, the whole circuit may not function properly because the actual value of the resistor is too different from the one that was required.

• Operating Temperature. This depends both from the ambient conditions and by the temperature raise produced by the power dissipation. There are two reasons to keep the temperature range into account. First, resistors slightly change their resistance with the change of the temperature. Using the resistor outside its temperature range would cause a variation greater than the one considered by the tolerance. Second, but not last, when the resistor is traversed by current it heats up. As long as the current stays within a range for which the power dissipation is not exceeded, everything is fine. Otherwise, the resistor can easily overheat and burn.

• Maximum Voltage. Operating a resistor above its maximum voltage rating may cause sparks that would destroy the resistor.

Resistors used in low power circuits usually have a maximum voltage in the order of at least 100V, and that’s why people usually don’t care or it doesn’t even know that there is such a parameter. In fact, low voltage circuits will normally never exceed the maximum voltage of any resistor. However, there are specific applications where voltages in the circuits can be above the 100V threshold. In such cases, it is important to verify that the resistors used in the circuit can withstand those voltages.

• Temperature coefficient. This is the parameter that tells us how much the resistance changes per degree Celsius. It depends on the material the resistor is made of, but also on the heat dissipation capability of the component. Some resistors are built with an embedded heat sink to reduce the value of this factor.

This information becomes important in those cases where it is known that the resistor is going to dissipate a considerable amount of power. Based on that, it is possible to figure out if the resistor needs an external heat sink and, eventually, the heat sink thermal resistance.

• Parasitic Capacitance and Inductance. A real resistor does not have only a resistance but also a very low value of capacity and inductance that may affect its functionality at high frequencies.

These parasitic capacitance and inductance are caused by the physical dimensions and shape of the component and cannot be avoided. When working at high frequencies, these values need to be taken into account, since they will generate both capacitive and inductive reactance that will affect the value of the resistor at the particular frequency it is going to be used.

• Packaging. This keeps into account where and how the resistor is going to be mounted. It can be a through holes resistor, which is provided with two leads to make the connections. The leads are usually inserted in the holes of a perforated board or of a Printed Circuit Board (PCB). Or, the resistor can be a Surface Mounted one. This has no wires, just two pads that can be directly soldered on a Surface Mounted technology (SMT) PCB. Other factors affecting the packaging include the possibility of attaching it to an external heat sink, and/or the necessity to properly ventilate it, to guarantee enough heat dissipation.

An electronic load to test DC power supply devices up to 100W.

What do you use when you have to test a new power supply that you just built, or one that you bought and want to know if the declared specs are true?

One thing you’ll need is a passive load that you attach to the power supply output to drain a certain amount of current, both to verify that the power supply is capable of providing that amount of current, and to verify the amount of ripple that was not filtered away by the power supply itself.

A classic method for doing so is to use a rheostat, which is essentially a potentiometer capable of dissipating the amount of power produced by the power supply. The resistance of the rheostat can be changed and therefore different amount of currents can be used to test the power supply. However, rheostats are big, heavy and cumbersome.

An alternative to rheostats is to have a so-called Electronic Load. These are electronic circuits that are capable to emulate the functionality of a rheostat.

I’m proposing here two simple versions of an Electronic Load, functioning in DC, and capable of dissipating up to 100W, all of this in a very condensed space, and very light in weight.

The first version is a very simplistic one.

It uses a cascade of three transistors, in a configuration called Darlington. This configuration is effectively equivalent to a single transistor with a gain (hfe) that is the product of the gain of all the transistors in the configuration. This allows for a very little control current flowing into the potentiometer used to regulate the base current, and for a high current available between the collector and the emitter of the transistor Q3, in the above schematics.

This circuit does not need its own power supply, since it gets what it needs directly from the power supply under test. The resistor R1 is calculated based on the highest voltage that the circuit will be able to handle, in this case 50V. It is important to note, however, that using this load at lower voltages will prevent the possibility to use the full excursion of the potentiometer, thus limiting the sensitivity of the circuit for the control of the current.

The next circuit eliminates the sensitivity problem by providing it own power supply to control the base current of the Darlington, thus eliminating the dependency from the external power supply voltage.

In fact, in this case, the base current can be adjusted with the potentiometer RV1, which is polarized through the resistor R1 and the trimpot TR1, which can be adjusted to maximize the useful range of motion of the potentiometer, regardless of the voltage applied at the input terminals. This way the electronic load can have the same sensitivity for any input voltage. A digital Volt/Am-meter, powered by the same internal battery, provides a visualization of the voltage of the system under test and the current being drained from it.

I will soon publish a video on my YouTube channel that shows the Electronic Load I built for myself. Please watch for that video to come out. And, in the mean time, can you think which one of the above schematics I used? Did I use the simple one, because less expensive? O did I sacrifice a few extra bucks to gain more sensitivity on the regulation of the load?

I strongly suggest to subscribe for free to my YouTube channel, and also to click on the bell icon that appears after the subscription is done. This will allow you to automatically receive an e-mail whenever I publish a new video. This way you won’t have to go periodically to my channel to check for new posts.

Happy Experiments!