CHOOSING A GOOD CABLE
With an appreciation of how poor cabling and connectors can impact VNA measurements, how do you ensure using the right cables? Start by selecting good cables, following these guidelines:
Insertion loss and phase stability: Good test-port cables should be stable with temperature and when flexed or bent. Of course, the cable should never be bent beyond its minimum recommended bend radius. A good cable will typically have an insertion loss less than 2 dB across its entire frequency range, and the phase stability will be on the order of ±1 to 2 degrees.
Reinforced junctions: Bending cables to calibrate or connect to the product being tested puts strain on the interface between the cable and the connector. The cable can be damaged if this interface is not sufficiently reinforced.
Connection repeatability: The measurement should be repeatable when connecting to a device multiple times; repeatability is dominated by the quality of the connector on the cable.
Shielding: Good shielding is essential, especially if the measurements are in a noisy RF environment where electromagnetic interference and electrical noise can affect the signals.
Rugged jacket: Cables used in hostile environments, where they may be bent beyond the minimum recommended bend radius or dragged through a conduit, for example, should have a rugged jacket.
Return loss or match stability: While the return loss of the test-port cable is correctable with a user calibration, the accuracy of the correction degrades as the return loss of the cable approaches 0 dB. A return loss of 15 dB or greater should be sufficient for most applications. The stability of the return loss, however, is of paramount importance. Even small changes cause large errors if the return loss magnitude or phase changes as the cable is bent between calibration and measurement.
VERIFYING CABLES AND ADAPTERS
You may not always be the person buying the cables and adapters. Before making VNA measurements with whatever is available, the better course of action is to run the following simple tests to evaluate the cable and adapter quality:
SOL cal: Prior to measurements, perform a short-open-load (SOL) calibration on one of the ports of the VNA. A single port VNA is sufficient. To achieve accurate results, choose a low IF bandwidth and set the port power level to at least 0 dBm. Trace smoothing may be needed to reduce the trace noise to see small changes in the cable. In each case, adjust the reference level to center the trace on the VNA’s display and adjust the scale so the trace is somewhere between 25 and 75 percent of full scale. Depending on the VNA, the marker search function can help locate the worst-case point for each measurement.
Visual inspection: Before running any test, visually check to see if the cable or the connectors on any adapters are damaged or contaminated by dust or other particles. Even the slightest damage to the center pin can cause measurement errors or, worse, damage other components while attempting to connect to them. One way to clean the connectors is with a lint-free cotton swab dampened with isopropyl alcohol, then dried by low-pressure compressed air to clear away any remaining contaminants. You can also use a connector gage with the appropriate gender and type to measure the pin depth, comparing it with the connector’s specifications.
