After giving the first two lectures at Denhac, we decided that it would be fun to keep it going with more talks on electronics! While everyone seemed to dig what I had to say, I think there was probably too much talking/drawing and too little showing. Remember, this is meant to be as fun as possible! To address this, I thought lecture #3 should focus on electronic measurement equipment. This will give me a chance to bring in a bunch of equipment, talk about each unit, and put the concepts we talked about on display with a few demo circuits. The best part is that with multiple devices, the attendees can spend some time playing around with the stuff, looking at their datasheets/prices, and asking me about them. Nice!
A big thank you to the folks who made suggestions on the last lecture. If anyone has ideas for additions, exclusions, and corrections, please leave a comment below and I'll incorporate them. I expect this to be another relaxed setting, with those attending always welcome to interrupt me with questions/comments/anecdotes. After all, it is on a SATURDAY! Saturday 4/28 at 1:30pm MDT at 975 E 58th Ave, Unit N Denver, CO 80216 to be specific - Hope you can make it!
A Voltmeter measures voltage (clever name, I know). It is one of the more simple measurement tools, which is why you can find them in $5 bins among other crappy tools at Harbor Freight. The digital-readout units work by adjusting the input signal (usually with op amps) to a level that an analog-to-digital converter can measure. The value is then displayed on the LCD readout. The Ideal Voltmeter has infinite input impedance (which means that no current can flow in to the meter). This prevents the act of measuring the voltage from loading down the source and changing the result by taking the measurement (Dead Cat? Live Cat? What???)
Voltmeters are great for getting very accurate measurements of DC signals or periodic AC signals. It is one of those tools that most engineers have a few of around the bench, ranging from crappy $5 hand-held units to nicer $1,000 units that are faster, more accurate, and have lower noise.
The most important deviations from the ideal volt meter is its accuracy and noise. Accuracy is spec'd in terms of 'offset' and 'gain' errors. This allows engineers to understand the error of any measured voltage by y = M*x + B where x is the voltage being measured, M is the gain error (in %), and B is the offset error (in Volts). Noise is also a concern, which represents the how much the measurement will appear to randomly change despite the measurand remaining perfectly constant.
There are some features to look for when buying a voltmeter in addition to accuracy and noise. First, if it is a battery powered unit, it should have an auto-off feature to prevent the waste of (usually) 9V batteries, which are excruciatingly expensive for the little energy they hold. Similarly, it should have an indicator to tell you when the battery is running low. When a unit has a low battery, the measurement error starts to increase as the battery voltage continues to fall below a certain point. And finally if you are going to be measuring AC signals that are not perfect sine waves, a 'True-RMS' meter will be sure these measurements are accurate, since cheap meters simply do an AC calculation assuming a perfect sine wave.
Ammeter (pronounced Ammeter because saying 'Ampmeter' or 'Amperemeter' is awkward)
An ammeter measures current, and is basically a volt meter with a small tweak. The only difference is that the signal is sent through a 'shunt resistor' to cause a voltage that can be measured. If the ammeter knows exactly what the shunt resistance is, a measured voltage can lead to a calculated current value thanks to Ohm's Law, V = I*R.
The ideal ammeter will simply act link a wire that can count the electrons that flow over a period of time. In reality, amp meters must add some resistance to the circuit being measured. But perhaps more surprising and problematic than the shunt resistance is that they cannot respond to fast changes in current levels. For instance, if you are trying to measure the current going into a circuit and power the unit on, it may jump from 0A to 3A very quickly. Because the amp meter cannot quickly change the required current range, the switching may create a strange start-up condition in the product that changes its behavior. It is best to manually use the highest current range in these cases.
Since an amp meter is a voltmeter that has the input signal adjusted, all of the purchasing considerations and types of signals that it performs best with are basically the same.
An ohmmeter is another re-use of voltmeter circuitry in order to measure resistance. In the ideal case, a resistance meter applies an exact current signal (maybe 1mA), and measures the resulting voltage. Again, Ohm's Law, V = I*R can take a known current signal and measured voltage to calculate the resistance.
Ohmmeters are great for measuring individual parts on the bench that are not plugged into boards. They can also be used to measure other things such as the polarity of a diode by checking to see which way current flows. They may not be good for measuring resistances of parts that are in a circuit since the applied current may be able to flow through other parts that are connected to the same node. Similarly, the resistance cannot be measured when the unit is on since there are other sources for current in the circuit that the meter does not know about.
Because the unit is only a DC volt meter with the added current source, the main purchase considerations are the accuracy of the current source, the accuracy of the volt meter, and the noise.
Because of the similarities of voltmeters, ammeters, and resistance meters, they are usually sold as one unit with all functions, called a 'DMM' or 'Digital Multi-Meter'.
A power supply is often used like a battery with a voltage adjustment knob so the user can dial in the desired voltage. They can provide lots of power in order to run an entire circuit without much worry about running out of juice. The most common unit will work by the control knob adjusting a potentiometer in a circuit that then controls a circuit that can output high power levels (called an output stage). Some units also have a current knob where the user can set the maximum current, which is handy for preventing part damage when something is connected wrong.
There are many different (and cheaper) ways to give a desired voltage on the bench – wall warts, batteries, home-brewed voltage sources, etc... However decent power supplies start at ~$100 and it is wonderful to have two variable +30V supplies and a fixed +5V supply always within arms reach. When looking to buy a unit, the most important factors are the voltage, current, and power output. Just because a unit is rated to output 30V AND 3A doesn't mean it can put out 3A AT 30V (90W), so watch out. Accuracy is not a huge concern because an accurate DMM can be used to determine how to accurately set the knob. Output noise, however, is a large concern because your circuit might simply pass any power supply noise right to the output.
A function generator can make a signal of your choice so you can create a sine wave, square wave, and triangle wave. Some newer units even let you define any shaped waveform on your computer to be downloaded to the function generator, which is called the 'Arbitrary' or 'ARB' function. These units are brilliant at giving you complete, easy control over a waveform that can be used to test a circuit. Want to see how your filter performs? Use a function generator as the input so you can vary the input frequency!
The latest entry-level units that can make ARB waveforms are designed are as glorified Digital-to-Analog converters. The microcontroller tells the A/D chip what waveform to make, and then the output circuitry sends the signal out at the desired amplitude. The biggest error that users find when using a function generator is that its output is not very powerful. It is easy to load it down when trying to drive something similar to an 8 ohm speaker. Even when driving a 500 ohm resistor, your programmed output amplitude might be loaded down, giving a lower-than-desired signal output.
When buying a unit one should be concerned with the type of waveforms that it can output, the range of output voltages, and noise. It is difficult to nail down an exact spec on output impedance, which tells you how easily it can be loaded down. To cope with this variation, designers have a '50-ohm' mode, which makes the assumption that your load is exactly 50 ohms. It costs about $200 to get into the function generator game, with the entry-level ARB models starting at around $400.
An oscilloscope is similar to a volt meter except that it draws a picture of what the voltage level is over time instead of displaying a voltage on a slow-changing 7-segment display. This is the piece of equipment that is always shown in the movies to make a bench look serious. The goal with an O-Scope is to be able to actually SEE what a signal is doing. What is the rise time of that signal? Are there blips? Any noise? What frequency is the noise at? Scopes are great for answering these questions since it is a live graph with time as the X-axis and voltage as the Y-axis.
Oscilloscopes are particularly useful because you can zoom out on a signal to look at the entire waveform, or zoom in to take a close look at the exact point where a sine wave crosses from negative voltage to positive voltage.
When buying a scope, the most important spec is the 'speed'. A 100MHz scope is advertized as being able to resolve signals up to 100MHz (although one would want at least 2x headroom, so a 100MHz scope is really only good for 50MHz signals). A scope is not an accurate way to measure absolute voltage so it is not a concern, but the noise is another big deal since you don't want to see a ton of noise that isn't really on your circuit. It is possible to get into a reasonable scope for about $200.
And there you have it. A bench with a DMM, power supply, function generator, and o-scope would be able to build and troubleshoot most electronic circuits. For a casual person looking to jump in with both feet, it would cost well under $1,000 to get totally set up – which is a pretty inexpensive buy-in for a new hobby! As I mentioned, you can come see all of these devices in action at Denhac on Saturday 4/28 at 1:30pm; Hope to see you there!
