Oscilloscope Probes - Eight Tips to Improve Your Oscilloscope's Probing Capabilities

Probing techniques are critical to high-quality oscilloscope measurements, and the oscilloscope probe is often the first link in the oscilloscope measurement chain.

If the probe performance is insufficient, you will see distorted or misleading signals on your oscilloscope. Selecting the right probe for your application is the first step to making reliable measurements. How you use the probe can also affect your ability to make accurate measurements, which can prevent you from getting useful measurements. This application note introduces eight important tips to help you select the right probe for your application and improve your oscilloscope probing capabilities. The following probing tips will help you avoid the most common probing pitfalls.

Tip 1: Passive or active probe?

Tip 2: Use dual probes to check probe loading

Tip 3: Probe compensation before use

Tip 4: High-sensitivity, wide dynamic range current measurement

Tip 5: Use differential probes for safe floating measurements

Tip 6: Check common-mode rejection

Tip 7: Check probe coupling

Tip 8: Damping resonance

Tip 1 - Passive or active probe?

Passive high-impedance probes are a good choice for low- to medium-frequency (less than 600-MHz) measurements. These rugged, economical probes have a wide dynamic range (greater than 300 V) and high input impedance to match the input impedance of an oscilloscope. However, passive probes have higher capacitive loading and lower bandwidth than low-impedance (z0) passive or active probes. In summary, high-impedance passive probes are an excellent choice for general-purpose debugging and troubleshooting of most analog or digital circuits.

For high-frequency applications that require accurate measurements over a wide frequency range (greater than 600 MHz), active probes are the best choice. Active probes are more expensive than passive probes and have limited input voltage, but they enable you to observe fast signals more accurately because they have significantly reduced capacitive loading.

In Figure 1-1, we see  a screenshot of a signal with a 500 ps rise time measured with  a 600 MHz oscilloscope (Keysight DSO 9064A) . The left image shows this signal measured with a Keysight N2873A 500 MHz passive probe. The right image shows the same signal measured with a Keysight N2796A  2 GHz single-ended active probe. The yellow trace in both images is the signal before probing. The green trace is the signal after probing, which is the same as the input to the probe. The purple trace shows the measured signal, which can also be called the probe output.

Figure 1-1. Measuring a signal with a 600 ps rise time using a passive probe and an active probe.

Keysight N2873A 500-MHz passive probe with 15-cm alligator clip ground lead:

– After loading the signal, it now has a 740 ps rising edge

– The probe output includes resonance, and the rising edge is measured at 1.4 ns

A passive probe introduces its own resistance, inductance, and capacitance (green trace) into the signal being loaded. You might expect an oscilloscope probe not to affect the signal of the device under test (DUT). However, in this case, the passive probe does affect the DUT. The rise time of the probed signal changes from the expected 600 ps to 4 ns, partly due to the probe’s input impedance and also because the probe has only a 500-MHz bandwidth when measuring a 583-MHz signal (0.35/600 ps = 583 MHz).

The inductive and capacitive effects of a passive probe also cause overshoot and fluctuations in the probe output. Some designers are not bothered by this amount of measurement error, but for others, this amount of measurement error is unacceptable.

Keysight N2796A  2-GHz active probe with 1.8-cm ground lead:

– Signal is not affected by the probe and still has a 630 ps edge

– The probe output matches the signal and measures the edge to be 555 ps

We can see that the signal is virtually unaffected when an active probe, such as  the Keysight N2796A  2 GHz active probe, is connected to the DUT. The signal characteristics are almost identical after probing (green trace) and before probing (N2796A 2 GHz trace). In addition, the rise time of the signal is also unaffected by the probe and remains constant at 555 ps. Furthermore, the output of the active probe (green trace) also matches the probed signal (purple trace), with a measured rise time of the expected 600 ps. This is made possible by the 2 GHz bandwidth of the N2796A active probe , its excellent signal fidelity, and its low probe loading.

Tip 2 - Use Dual Probes to Check Probe Loading

Figure 2-1: Checking probe loading using dual probes

Figure 1-2. Comparison of high impedance passive and active probes

Before probing a circuit, connect one probe tip to a point on the circuit, then connect the second probe to the same point. Ideally, you should see no change in the signal. If the signal changes, the change is caused by probe loading.

In an ideal world, an oscilloscope has clean wires (with infinite input resistance, zero capacitance, and zero inductance) connected to the circuit being measured, and it produces an exact replica of the signal being measured. But in the real world, probes are part of the measurement and they load the circuit.

To check the loading effect of a probe, first connect the probe to the circuit under test or a known step signal and the other end to the input of the oscilloscope. Observe this trace on the oscilloscope display, then save and recall it on the display to keep the trace on the display for comparison. Later, connect another probe of the same type to the same probe point and observe how the original trace changes when probing with both probes.

You may need to adjust your probing or use a probe with less loading to get better probing. In this case, shortening the ground lead works well. In Figure 2-2, the circuit ground is connected using an 18 cm (7 in) long ground lead.

Figure 2-2. Probe loading caused by a long ground lead

Figure 2-3. Using a short ground lead to reduce probe loading

In Figure 2-3, the same signal ground is shown using a spring ground lead. The ringing on the probe signal (purple trace) disappears when the shorter ground lead is used.

Tip 3 - Probe compensation before use

Most probes are designed to match the input of a specific oscilloscope model. However, each oscilloscope is slightly different from one another, and even from one input channel to another on the same oscilloscope. So before you connect a probe to the input of an oscilloscope, always be sure to check the probe compensation first, as the probe may have been previously adjusted to match a different input. To address this, most passive probes have a built-in compensated RC divider network. Probe compensation is the process of adjusting the RC divider so that the probe maintains its attenuation rate over the rated bandwidth.

Figure 3-1. Using a small screwdriver to adjust the probe's variable capacitance.

If your oscilloscope can automatically compensate for the probe performance, this feature is very useful. Otherwise, use manual compensation to adjust the probe's variable capacitance. Most oscilloscopes provide a square wave reference signal on the front panel to compensate the probe. You connect the probe tip to the probe compensation terminal and then connect the probe to the oscilloscope's input. Observe the square wave reference signal and use a small screwdriver to adjust the probe appropriately so that the square wave on the oscilloscope screen looks like a square wave.

Figure 3-2. Compensation adjustment corresponding to a flat square wave

The top diagram in Figure 3-2 shows how to properly adjust the compensation capacitor in the probe termination box. In the diagram you can see that if the low frequency adjustment is not done properly, there may be overshoot or undershoot in the square wave. This will cause inaccuracies in the high frequency measurement. Therefore, it is very important to ensure that the compensation capacitor is adjusted correctly.

Tip 4 - Low Current Measurement Techniques

As today’s battery-powered devices and integrated circuits become more environmentally friendly and energy efficient, engineers need the ability to measure low-level currents with high sensitivity to ensure that current consumption is within acceptable limits. The primary applications that require accurate power consumption measurements are battery-powered applications such as wireless mobile devices and consumer electronics. To maximize battery life, engineers need to minimize the power consumption of their products over their lifetime. Power is defined as P = V x I. The primary method for reducing device power consumption is to reduce the average current consumption of the device while keeping the supply voltage constant.

The main challenge in measuring the current consumption of battery-powered mobile devices, such as cell phones or tablets, is the very wide dynamic range of the current signal. Mobile devices often need to switch back and forth between an active state, where the peak current is very high and drains very quickly, and an idle or standby current mode, where very little DC and AC current is consumed.

Figure 4-1 shows the current consumption measured when making a call on a GSM mobile phone. The power peak in the active state is as high as about 2 A, while the current consumption in idle mode is very small.

Figure 4-1. Current consumption measured when making a call using a GSM mobile phone.

Figure 4-2. The simplest way to measure current with an oscilloscope is to use a clamp-on current probe, such as the Keysight 1147B or N2893A.

The simplest way to measure current with an oscilloscope is to directly monitor the current going into the device under test using a clamp-on current probe such as the Keysight 1147B or N2893A.

However, this method is not suitable for measuring small currents that change rapidly from less than 1 mA to several amperes, because the dynamic range and sensitivity of the clamp current probe are very limited, only a few mA. Taking the measurement of mobile phone current consumption as an example, the current in the idle state is difficult to measure because it is masked by the probe noise.