Laboratory Experiment

Resistor Accuracy - Lab Problem

Are you a skeptical, maybe cynical, person? Then consider this idea.

Imagine that you are a resistor manufacturer. As you manufacture resistors, measure them as the come off the line. Say you're trying to make 1000W resistors. You know that when you're done you are going to have a whole bunch of resistors and that they are going to range from 800W to 1200W. That means that they run from 20% low to 20% high. While that's not bad, you also realize that you can sell 1% resistors for a lot more money.

So, here's the idea. As you measure the resistors, select out the ones that are within 1% of the value you want. Leave the rest in the bin. Maybe while you're at it you put the ones that don't make 1%, but are still within 5% into another bin. The result of this process is that you are left with a bunch of resistors that don't include any that are really accurate. You sell those as ordinary 20% resistors.

Next, you package the 1% resistors and 5% resistors separately and sell them at a higher price.

Now, the question is "Is the above process what actually happens?". In this laboratory problem you will test that idea and gain some practice measuring resistors in the process. (You can find more discussion of this idea in "Electrical Engineering Uncovered" by White & Doering (Prentice Hall, 1997) in Chapter 4.)

Here's what you should do.

Go to your resistor supply and get twenty (20) resistors of the same size.
Choose something from 100W to 10,000W.
Measure and record the values of those resistors and from your measured values come to a conclusion about the claim above.
If you need to, you can click here to go to the lesson on measuring resistance with an ohmmeter.
Your report should evaluate the claim that resistors are sold by selecting out the accurate values and selling them separately.
Your reasoning should be based on the experimental values you obtained from your measurements seasoned with your knowledge of statistics.

Resistor Accuracy

Here's what you're trying to get from this lab problem.

Given that you need a resistor of some nonstandard size,
To know what to expect when you use a resistor from a bin of standard resistors and use them in series and parallel combinations.

Resistors in series should have an equivalent that has a resistance that is the sum of the values of the resistors in series. In this lab problem you will check that theoretical claim, and you will get some practice wiring series elements. Here are the goals for this problem

Given that you need to put two resistors in series or parallel,
Be able to predict the equivalent value with confidence,
Be able to make the physical connection in series on a circuit board.

Here is what you should do.

Get some resistors - 10kW, 22kW, 47kW, and connect all series combinations of two resistors (including two the same) and check individual values and series values.
To connect two resistors in series, use two resistors in a circuit board as shown below. Notice that the point in common to the two resistors - the node to which they are both connected - is made by connecting to a group of five points in a single row. Make sure you understand the correspondence between the circuit diagram and the actual circuit.

Your report should answer the following questions.

Is the claim true?
Can you predict the series resistance more accurately if you take the "stripe value" for the resistance, or is it better if you actually measure the individual resistors?
Does the order of the resistances matter? If you interchange the two resistors does the series resistance change?

Resistors in parallel should have an equivalent that has a resistance that is computed using a standard formula. In this lab problem you will check that theoretical claim, and you will get some practice wiring parallel elements.

Here is what you should do.

Get some resistors - 10kW, 22kW, 47kW, and connect all parallel combinations of two resistors (including two the same) and check individual values and parallel values.
To connect two resistors in parallel, use two resistors in a circuit board as shown below. Notice that points in common to the two resistors - the nodes to which they are both connected - is made by connecting to a group of five points in a single row. (Those five points constitute a node.) Make sure you understand the correspondence between the circuit diagram and the actual circuit.

Your report should answer the following questions.

Is the claim true?
Can you predict the parallel resistance more accurately if you take the "stripe value" for the resistance, or is it better if you actually measure the individual resistors?
Does the order of the resistances matter? If you interchange the two resistors does the parallel resistance change?