The first stage of making a good high frequency measurement involves a device front panel connector and the connections made to it. Many times, you use the most accessible cables and connectors to make connections to instruments, and usually this results in a reliable measurement. However, every adapter, connector, and cable that exists between the instrument and the DUT adds uncertainty to the measurement. At higher frequencies or frequencies closer to noise, those uncertainties have a larger impact on measurements. Reducing uncertainty begins with proper adapter and cable selection.
Be attentive when selecting cables and adapters. Verify the integrity of the equipment selected. Cable integrity can be degraded due to age, improper use, or contaminants. Contaminants from cables and adapters are easily transferred to the input connector of the test equipment. Figure 1 shows an example of a contaminated cable
Figure 1: Contaminated SMA Cable
All of the metal fibers shown in Figure 1 are sources of uncertainty in measurements. Although the fibers may not contaminate any connector they interface with, they are very abrasive and can generate more contaminants as well as getting ground into the mating surfaces of both connectors. Using a connector with contaminants at low frequencies may not cause a problem, but using a connector with contaminants at high frequencies prevents a good match. The quality of the measurement is dependent on the match throughout the measurement system, which is dependent on the characteristic impedance of the various components. The characteristic impedance of the cable is dependent on the ratio of the diameter of the center conductor to the diameter of the outer conductor, as illustrated in Figure 2.
Figure 2: Characteristic Impedance of a Connector
Based on the preceding equation in Figure 2, little metal fibers act to decrease the diameter of the outer conductor, and this causes discontinuities in high frequency measurements.
Figures 3 and 4 show an example of how discontinuities can negatively affect high frequency measurements.
Figure 3 illustrates a 50 ohm coaxial semi-rigid cable measured on a network analyzer. Figure 4 shows the frequency response of the cable up to about 6 GHz.
Figure 3: Cable with Multiple Bends and an Intentional Crimp
Figure 4: Frequency Response of Cable Due to Bends Only
The cable was intentionally crimped near the connector (circled in red). Figure 5 shows the response of the cable with the crimp compared to the original cable with no crimp. The big discontinuity and the degradation in performance are caused by that crimp, which effectively moves the outer conductor closer to the inner conductor and changes the characteristic impedance. The effect is not limited to just a small portion of the measurement, but its effects can be seen across the full span of the measurement.
Figure 5: Frequency Response of Cable Due to Bends and Crimp
When the connector interface is contaminated, degradation in performance happens to a lesser degree.
Here are a few points to consider regarding cables and connector care:
- Contaminants follow the connector
- Contaminants “infect” mating connectors
- Contaminants can cause damage to expensive equipment
- Contaminants can cause discontinuities at high frequencies
- Good connections are key to precision RF measurements
- Dirty or damaged connectors are a significant factor in accuracy and repeatability