RF Probes
What is an RF Probe? | DC vs RF probes | Types of RF Probe | RF Probe Calibration | S Parameter | Choosing an RF Probe
Radio frequency (RF) probes facilitate the accurate and reliable measurement of components that operate at high frequencies.
RF components (like amplifiers, mixers and data converters) operate at high frequencies, where the signal wavelength is comparable to the physical size of the circuit. This means that all the components start to behave like distributive circuit elements, rather than acting as isolated components or conductors. This is true even for the probes connecting the measurement equipment to the device, and at these frequencies, traditional probes begin to struggle.
Therefore, RF and microwave measurement require specific RF probes with controlled-impedance ground paths and geometries designed for specific frequency ranges.
What is an RF Probe?
An RF probe is an impedance-matched electrical interface that connects two components of a system for RF or microwave testing. For example, this could connect a device under test to a vector network analyzer (VNA) or oscillator.
Unlike standard probes, RF probes help maintain a controlled electrical impedance from testing system to device. This is fundamental for any lab carrying out high frequency RF or microwave measurement.
RF probes usually have multiple tip pathways including at least one signal path and one impedance-controlled ground return path at the point of contact. This allows the probes to exhibit high impedance and good high frequency behaviour. They are designed to minimize electromagnetic interference and be highly calibrated so that they have minimal distortions and energy loss.
Two important specifications matter when selecting an RF probe:
- Pitch - the spacing between isolated tips
- Frequency range - RF probes are rated to a maximum frequency and this should exceed the highest frequency of interest in the measurement.
DC vs RF probes
At DC and low frequencies, a simple metal needle probe works well. The wire simply acts as a conductor and small differences in geometry make little difference to a measurement.
However, as stated earlier, this changes at radio and microwave frequencies. As the frequency increases, the signal's wavelength becomes comparable to the physical size of the probe and the circuit, so the probe starts to behave like a circuit element rather than a passive connector. It can store energy, radiate energy, and reflect part of the signal back towards the source.
A standard DC needle probe has no defined return path for this signal, and its geometry introduces significant, uncontrolled inductance and capacitance. The introduces signal reflection, poor repeatability, and measurement error that gets worse as frequency increases.
RF probes are specifically designed to account for these effects. They have a defined ground pathway and controlled impedance guiding the signal along the tip properly at high frequencies.
Types of RF Probe
There are several types of RF probe, designed for different applications, measurement resolution and pad sizes.
- Ground-Signal (GS): The simplest 2 probe configuration
- Ground-Signal-Ground (GSG): A 3 probe configuration that is the most common. This balances simple design while maximizing isolation from external interference.
- Ground-Signal-Signal-Ground (GSSG): This offers the highest levels of signal isolation, suitable for advanced applications. This is also the only configuration that can process differential signals.
These tips can also be classified by their mechanical structure, which often informs their application. Some example probe geometries include:
-
Cantilever
Fine needle-like probes extending towards the device from a micromanipulator.
-
MEMS based
For microelectronic systems, people can design their own probes at much smaller scales on silicon wafers. This shrinks probe tip size substantially.
-
Co-axial and waveguide probes
These are the probes usually used in a lab, for connection to a manual probe station.
There are also active RF probes, that are named so because they work actively to counter act interference. The probes described so far in this article are passive probes.
Calibration
An RF probe alone does not guarantee an accurate measurement. Cables, connectors, and the probe itself all introduce their own loss and phase shift, and this needs to be removed before a measurement reflects only the device under test.
This is done by calibrating the VNA at the probe tips, using a set of known reference standards fabricated on an Impedance Standard Substrate (ISS). Common calibration methods include:
- SOLT (Short-Open-Load-Through) – uses four defined standards. This method is simple, but requires the well-characterized standards.
- TRL (Through-Reflect-Line) – uses transmission line standards. This method is often preferred at higher frequencies where ideal SOLT standards become harder to measure.
Calibration moves the measurement reference plane from the VNA ports to the probe tips to ensure that everything measured beyond that point belongs to the device, not the test setup.
RF Probe Measurement Example: S-Parameters
The S-parameters of a DUT is the best way to represent a networks RF characteristics. This will show you how much of a signal is reflected, transmitted or transferred between ports in a system.
| Notation | Name | Measurment | Meaning |
|---|---|---|---|
| Structure | Non-woven, rigid | Woven, flexible | Sintered or woven metal, highly compressible. |
| S11 | Input reflection coefficient | What percentage of the input signal at port 1 is reflected back to port 1. | How much of the input signal is reflected due to an impedance mismatch. |
| S12 | Reverse transmission coefficient | The fraction of the input signal at port 2 transmitted to port 1. | The efficiency of signal transmission from the output to the input. |
| S21 | Forward transmission coefficient | The portion of the input signal at port 1 transmitted to port 2. | The efficiency of signal transmission from the input to the output. |
| S22 | Output reflection coefficient | The percentage of the input signal at port 2 that is reflected to port 2. | The isolation of port 1 from signals entering port 2. |
From these, key properties such as impedance, gain, loss, and isolation can be derived, which is exactly what's needed to characterise the RF components (filters, amplifiers, mixers) described earlier.
Choosing an RF Probe
For effective probe selection, you must match the probe to both the device and the measurement:
- Choose a pitch or tip width that matches the device's pad spacing or diameter.
- Choose the right frequency rating that comfortably covers your frequency of interest.
- Consider the mechanical construction appropriate to your setup. Cantilever or coaxial probes are usually sufficient for single-device lab characterisation, while membrane or MEMS probes suit smaller or more sensitive devices.
- Check that a compatible calibration substrate is available for the probe's pitch and frequency range.
References
- Measurement of induced currents in radio frequency magnetic..., J.M. Jennings et al., AIP Advances (2020)
- On-wafer measurements of RF nanoelectronic devices, Wallis, T. and Kabos, P., Cambridge University Press (2017)