Consider a broadband coaxial example where it is desired to measure over as wide a frequency range as possible. Four shorts are available with three of them being from a standard 3659 0.8 mm calibration kit (those three shorts are normally used to cover 80-145 GHz as part of a broadband SOLT/SSST combined calibration). The basic setup is shown in Figure: Setup Dialog for an Example Measurement, using a broadband ME7838D system based on the MS464XX..

The VNA system is a ME7838D (70 kHz- 145 GHz) based on the MS464XX and these broadband systems are covered in more detail in Broadband/mmWave Measurements (Option 7, Option 8x) of this guide. The DUT for this experiment will be a stepped impedance line terminated in a pseudo-broadband load (thus creating a scalloped appearance of the reflection coefficient). The basic 4-standard mSSS calibration will be compared against variants with super-weighting applied and with the match measurement applied in addition using the protocol discussed in the previous section.

The fourth short’s offset length is quite disparate from those of the other shorts so one might expect low frequency coverage being possible but potentially with some holes where near-singularities could occur.

The measurement results are shown in Figure: Measurements of the example DUT for the three permutations of the mSSS calibration. and none of the variants are completely surprising. The match-enabled measurement helps resolve a near-singularity around 24 GHz and shows reasonable results (agreeing with expectations) all the way down to 10 GHz. The super-weighting helps resolve a smaller issue near 60 GHz and helps with some of the deep return loss areas at higher frequency (where it helps to de-emphasize the long offset short a bit more). One should not read too much into the deep notch measurements as repeatability is a contributor but it does illustrate a scenario where an unusual combination of shorts can be aided by weight adjustment.