When using current probes, a direct connection to the circuit under test is not required. This isolation is probably the single most important characteristics of current probes, as it allows one to measure the RF currents of a circuit with a minimal effect on the circuit. A current probe plays an essential role in EMC engineering by allowing engineers to:

•Measure RF current on cable structures, as demonstrated in case 1

•Predict radiated far-field performance of cable structures link

•Distinguish common mode and differential mode current on cable structures in case 3

•Troubleshoot ESD issues in case 4 

•Troubleshoot structural resonances that could lead to both emission (case 2) and immunity issues (case 5).

Theory

The theory and applications of RF current monitoring probes are detailed in this article. Essentially, a current probe is a type of current transformer. It differs from Hall-effect current probes and Rogowski coils, which work differently. As the name suggests, an RF current probe is not effective at low frequencies down to DC (unlike a Hall-effect current sensing probe). However, the great benefit of an RF current probe is its ability to accurately measure current up to the 1 GHz range.

The performance of a current probe depends on two main factors:

•The coupling mechanism between the probe and the cable (mutual inductance). It is always preferred to ensure that the wires under test are in the centre of the aperture, this is usually achieved by using some foams to fill in the gap in the current probe.

•The electric field shielding capability - since a current probe is designed to work with magnetic field coupling, we want minimal electric field coupling. The current probe housing has a gap in order to avoid magnetic short circuit, but at the same time, it is a weak point, because electric field can penetrate and couple into the inductor of the current probe.

The manufacturing of an RF current probe is a complex procedure, and as a result, a commercial-grade current probe is often expensive (over 2,000GBP). However, thanks to some manufacturers' efforts to reduce the cost, high-quality clamp-on current probes are now available for less than 1,000GBP.

Current probes are characterised by the parameter of "transfer impedance", the ratio of output voltage to current flowing through the current probe. Since it is the ratio of a voltage to a current, transfer impedance has units of ohms. This is calculated by Z_T=V/R, where Z_T is the transfer impedance of a current probe. In the log-log scale, this becomes Z_T(dB)=V(dBμV)-I(dBμA).

The following simulation shows the transfer impedance of a current probe TBCP2-500 compared with the manufacturing datasheet. As one can see, the output of the current probe for a constant current source rises with frequency until it reaches the corner frequency, after which it remains constant with increasing frequency. Before the corner frequency is the "voltage region", where the self inductance of the current probe coil does not affect the probe output by much (as at low frequency, the impedance of the self inductance is rather low). After the corner frequency point, the self inductance L, together with the terminating resistor R (the 50 ohm impedance of a spectrum analyser/oscilloscope) forms a low pass L/R filter, making the flat bit of the transfer impedance, this is the "current region", where most current probes are used.

Practical Applications

Case 1

In the first case study, an RF current probe is clamped onto the mains cable of an active clamp flyback converter that uses GaN devices. The RF current probe is configured to measure the common mode current in this set-up, with a distance of about 10 cm between the current probe and the device under test. Depending on the product, it may be necessary to move the current probe along the cable to capture the maximum RF current on the cable (due to standing waves).

The RF current probe is connected to a spectrum analyser, with a frequency range between 10 and 300 MHz. This set-up measures the RF current in the radiated emission frequency range, giving an indication of the far-field performance of the unit. Emissions peak above 40 dBμV in certain frequency ranges, enough to potentially cause a class B (residential) emissions failure.

Note that the reading on the spectrum analyser is always in dBμV. With knowledge of the probe's transfer impedance (in this case, 20 dBΩ), the RF current level in dBμA can be obtained by subtracting dBΩ from dBμV. Therefore, an RF current level above 20 dBμA on a cable generally raises "red flags" in general.

Case 2

In this instance, we investigated the issue further. An RF current probe is connected to the 50Ω input of an oscilloscope. Channel 2 of the oscilloscope (also configured with a 50Ω impedance) is connected to a near-field magnetic field loop. In this set-up, the oscilloscope is triggered by Channel 1 as the RF current probe on the mains cable picks up the common mode current. The near-field current probe hovers above the PCB to pick up the "ringing" during the converter switching stage. The ringing of the inductor shows a similar profile to the RF current measured on the mains cable, which helps to identify the noise source on a PCB. In this case, the converter has conducted emission failures between 10 and 30 MHz. By using this method, we can quickly locate the noise source and come up with solutions to fix the issue.

Case 3

An RF current probe can be configured to measure either common-mode or differential-mode noise depending on the wiring configuration, see the photo above. Measuring common mode current is a straightforward process, as both wires must pass through the same aperture of the probe, effectively cancelling out the differential mode current in this case. Measuring differential mode current, however, is slightly more complex, as one needs to reverse one of the pairs of wires passing through the probe to cancel out the common mode current in this case. Since the current probe measures current, we needed to convert the results into voltage. This conversion is done in the control software EMCView, where two settings are crucial.

The first setting is the LISN/Att Correction, where I selected the transfer impedance file of the current probe I used. Manufacturers of current probes should always provide you with the transfer impedance file for the probe you purchase. This file converts the voltage reading of a spectrum analyser to a current reading (dBμV to dBμA, where dBμA = dBμV - dBΩ, and dBΩ is the transfer impedance).

The second setting involves adding a 28 dB attenuation compensation. The reason for this is as follows:

Both LISNs used for the current probe set-up are terminated with 50Ω resistors. This is important. When I measured I_DM using a current probe, I measured 2×I_DM due to the wiring configuration of the current probe on the cable. Therefore, I needed to divide the value by 2, or subtract 6dB from the output. The differential voltage is measured on a 50Ω in the LISN voltage set-up, which means I needed to add 34dBΩ (50Ω) to obtain the DM voltage. That results in a 28dB compensation file in EMCView.

Similarly, when I measured I_CM using a current probe, I also measured 2×I_CM according to the picture below. So, I again needed to subtract 6dB from the output and then add 34dBΩ (50Ω) to obtain the CM voltage reading. This again requires a 28dB compensation file in EMCView.

You can check this video demo here.

Case 4

Unlike conventional ESD testing set-ups that strictly adhere to standard protocols, this method provides a rapid check. It allows us to quickly determine whether the ESD simulator’s current waveform aligns with expectations. Typically, heavily used ESD simulators may exhibit waveform distortion, characterized by inadequate rise peaks or irregularities in the low-frequency “hump.”

Test Procedure:

•Position the current probe close to the edge of the test ground plane, with only half of the probe on the plane and the other half in free space. This minimizes common-mode ESD current travelling along the coaxial shield to the scope.

•Employ ferrite cores on the coaxial cable at the scope to suppress common-mode ESD currents.

•Use a short, well-shielded coaxial cable (in this case, RG-316) for accurate results. The better the shielding effectiveness of the coaxial cable, the better.

•The ESD simulator’s grounding lead is connected to the horizontal ground plane on the wood table instead of earthing to the floor Reference Ground Plane.

I have opted for the Tekbox TBCP2-750 probe. It was important for me to ascertain whether the use of the TBCP2-750 would yield comparable results. The Tekbox TBCP2-750 can operate within a 1 GHz frequency range and boasts a transfer impedance of 20 dBΩ. In fact, the 3 dB bandwidth is typically 850 MHz, as the transfer impedance curve is shown in Figure 2. While a transfer impedance of 20 dBΩ renders the probe more sensitive, allowing for higher voltage readings for the same current magnitude, it introduces a challenge when connecting to an oscilloscope with a 50Ω input impedance. This is due to the fact that most oscilloscopes with a 50Ω input impedance impose an upper limit on the maximum input voltage, typically around 5V RMS. Consequently, when dealing with a 3 A peak current, the voltage measured by a TBCP2-750 probe can surge to approximately 30 V (given that 20dBΩ is equivalent to 10Ω (=1V/A in the flat region in the transfer impedance curve)). This voltage level substantially exceeds the oscilloscope’s upper limit.

To address this, an external 20dB attenuator is therefore used. This modification results in an effective transfer impedance of 0dBΩ (1Ω). Consequently, voltage measurements on the oscilloscope accurately represent the current (1V = 1A×1Ω). In order to achieve a good performance, use a coaxial cable and attenuator specified to 6 GHz.

Case 5

Current probes can be used to troubleshoot structure resonance issues as demonstrated in here. The picture is self-explanatory. 

Case 6

In certain situations, a current probe can also be utilized for noise injection since the coupling is mutual. However, it's important to emphasize that this is not the preferred option. Specialized probes, such as the BCI probe introduced in the conducted immunity session, are specifically designed for this purpose. When using pick-up probes for noise injection, exercise caution to avoid overstressing them.