• • • TEST & MEASUREMENT • • • Probe Positioning and
Figure 3: Step Response for different probe tip orientations. To make the measurements in Figures 3 and 4 a Keysight MX0031A 52GHZ probe with a MX0041A 52GHZ tip was connected to a UXR0702A 70 GHz oscilloscope. The tip was connected to a MX0030-60001 PV/Demo Fixture in a manner similar to that described in the performance verification section of the user guide for MX0031A. The source is a Keysight N2126A 70 GHz calibration module.
Figure 4: Frequency response for different probe tip positions.
the frequency of interest, more significant effects and distortions in the signal may occur. An engineer would likely want to understand the circuit’s behaviour without the probe and visualise voltage waveforms as they would appear with an ideal probe. To do so they can apply corrections by leveraging oscilloscope software or post-process the data using simulation tools. Essentially, understanding how the circuit behaves independently of the probe allows users to compensate for the effects of probe loading to obtain more accurate measurements and analysis results.
Manufacturer and user generated loading s-parameters can be entered into an oscilloscope correction software to show the response of a measurement as if the loading was not present. When probing a 50-ohm system, assumptions can be made, and limited corrections can be implemented in software. If a measurement is
made without loading a correction in real-time, a loading model can help subtract the effects of the probe from the resulting measurement in simulation. Overall, these are powerful tools that will help users to understand the circuit behaviour and how it may differ from a measurement.
Despite the corrections available, the load will still be present on the node being probed. To get the best possible test result, a probe with the highest impedance at the frequencies of interest should be selected. Example plots of probe input impedance versus frequency for both RC and RCRC probes are shown in figure 1. Impedance is shown on the y axis and frequency on the x axis. After some frequency, impedance decreases and more current will be drawn away from the circuit. When selecting a probe, this impedance needs to be considered so as not to negatively impact the circuit functionality.
Undesired Signal Coupling Another issue of concern is high-frequency electromagnetic (EM) signals coupling into the measurement system. Most of an oscilloscope probe is shielded to prevent unwanted EM signals from coupling into the signal path, but there is often a short section of the tip that is unshielded between the input resistors and the amplifier. Additionally, some probe tips may not be shielded on all sides. These gaps in shielding exist for practical usage and manufacturing reasons but they aren’t desirable. If the probe picks up high-frequency portions of the signal, not attenuated by the input impedance, it may show an undesired initial porch or step in the measurement of the signal transition. Besides altering the signal’s appearance and showing irregularities, this can also impact frequency responses leading to reduced measurement accuracy, reducing 10-90% rise time and decreasing digital eye openings. Otherwise, if the probe tip is capacitively loaded by a DUT, it may result in a lower measured bandwidth. Since these issues are not present in the DUT, but only in the measurements of it, troubleshooting may be tricky. To minimise these issues, fixture the probe tip to be perpendicular to the DUT. Some probe tips have fixtures or positioners to aid in this setup. If an angle is required, it is better to have the probe leaning away from the incident signal. Sometimes various loading models are provided to better match the orientation of the probe tip to the DUT in use. Figures 2, 3 and 4 show the effect of various positions of a 52GHz probe tip on the resulting bandwidth and step response. The difference in - 3dB bandwidth between the optimal tip position and having the tip with the signal trace side leaned 45 degrees toward the incident waveform is approximately 3 GHz and the step response suffers greatly. In the extreme case with the signal trace of the probe tip flat against the signal trace of the DUT on the incident side, there is also a significant suckout evident. Based on the geometry of the probe tip, how undesired energy couples into it, and the electrical design of the tip, the effects of position may be even more significant for other probing solutions.
Summary It is often critical to understand what is happening in a circuit, and this is where measurement equipment is necessary. It is also important to consider what impact the observational tools are having on the circuit and how this can be minimised. Engineers need to ensure they select the right probe, account for probe load, and optimally attach the probe to the DUT. This can save a lot of time and effort, so as not to chase measurement artefacts. To get the best understanding, consider how software may help. Device specific recommendations are often found in the user manual and should be followed. Making and understanding good measurements will ultimately help meet research and design goals faster.
electricalengineeringmagazine.co.uk
ELECTRICAL ENGINEERING • OCTOBER 2024 35
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50
