This section provides basic measurement techniques that make a big difference in debugging. First, it is valuable to use oscilloscope probe labels, if possible. Label each waveform in the picture, so that when it is captured it is clear what is being measured. If there are particular test points used by the oscilloscope probe, put that in parentheses. Second, use the graph marks for offsets. As Figure 2-1 shows, C1 (yellow) is 5 V/div and is offset by +5 V, C2 is 50 V/div and is offset by –150 V. Because of these results, it is apparent that C2 never goes above 100 V. If C2 was offset by –67.5 V, it would be difficult to determine the peak-to-peak voltage at a glance. It is obvious that C2 is the drain-to-source voltage of Q10 because of the oscilloscope probe label.

Another important tip is to use oscilloscope internal measurement functions when applicable. For example, most oscilloscopes can measure the Vpkpk of a particular channel. Oscilloscopes can make both vertical (y-axis) and horizontal (x-axis, which is usually time related) measurements. When setting the volts/div, try to use the largest V/div that will show the full waveform. If there are multiple signals in the same measurement, a smaller V/div can be used but use caution not to make it too small. For the time/div, there are two situations:

1. Figure 2-1 shows a successful start-up sequence. First VCC rises (blue), and then Vout, Iout and Vds begin to switch. This picture shows a successful start-up and a few time stamps of successful normal operation. Ensure scope shots show the entire sequence, but the individual parts can still be seen.
2. Figure 2-2 shows a switching waveform. The best practice is to show at least one full switching period. Optimal pictures show at least one but not more than three switching periods. The less switching periods shown the more detail can be seen in a single period.

Oscilloscopes also offer filtering of the probes. Figure 2-2 shows that both C1 and C2 have a BWL icon. This indicates that bandwidth limiting is turned on. Many scopes offer filtering for 20 MHz, 200 MHz, and no filtering at all. These filers act as low-pass filters at these set frequencies. Noise from around the lab can be introduced into the scope, so sometimes it is useful to filter that noise inside the oscilloscope itself (if checking if a MOSFET is switching, checking a start-up or shutdown, and so forth). However, when checking a maximum voltage on a MOSFET, or investigating a voltage spike, remove bandwidth limiting. Noise or real, it is critical to find the maximum voltage, so remove bandwidth limiting. If you are unsure, capture both. It is a simple setting change.

Another important measurement technique is called tip and barrel. This is best shown in a picture, please see below for reference. For oscilloscopes, tip and barrel commonly have a ground clip that attaches to the probe with a short wire. This short wire can pick up noise and add parasitic inductance, which can alter the waveform. So, when measuring critical voltage signals, whether that be a maximum voltage on a FET, output ripple, switching voltages on a gate, the current sense voltage for the feedback loop, use the tip and barrel method. The tip and barrel method is to remove the ground clip, and replace it with a smaller ground connection wire. The ground connection wire will be a small attachment of wire that is curly on one side and then a small straight portion with one corner. The curly portion is what holds the attachment to the probe, and the wire with the corner is the new ground connection. Many scope probes come with a cover that has a hook, which can be used to hook onto test points or wires. If this is removed, it exposes the probe tip and a small exposed ring of metal. This is the ground connection and the curly side connects to the ground of the probe. This method creates the smallest ground loop between the probe signal and ground, and therefore reduces noise.