Frequently Asked Questions (FAQs)

Summary

Sweep speed (also referred to as sweep time) may be interpreted and specified in a multitude of ways. In general sweep, speed refers to the time required for a vector network analyzer (VNA) to complete one sweep of the source and acquire data over the defined frequency range.

Question

What is the difference between "sweep speed", "cycle time" and "time per data point" as related to a vector network analyzer?

Answer

Sweep speed (also referred to as sweep time) may be interpreted and specified in a multitude of ways. In general sweep, speed refers to the time required for a vector network analyzer (VNA) to complete one sweep of the source and acquire data over the defined frequency range.

Many analyzers' technical specifications report a number in the form of "time per data point". To assume that the sweep time may be computed by multiplying this number by the number of points in a trace is typically in err.

Often, the sweep time per data point is the raw, uncorrected, best-case data acquisition time in the fastest single frequency band of the VNA. (Analyzers may also have a specification parameter for hardware sweep time.) Note that users may never achieve this "optimized" sweep time in their measurements because in reality what they obtain is the "cycle time."
This cycle time includes:

1. Sweep (hardware) set-up time
2. Band-switch times (when the source or receiver crosses frequency bands)
3. Data-acquisition time
4. Retrace time (for the source to move from the end of one sweep to the start of the next one)
5. Data-calculation and formatting time
6. Display update time

Refer to the figure below

Additionally, time to compute and apply error-correction may not be included. For example, if the analyzer is applying and measuring an n-port calibration (where n = 2,3,4, etc.) the analyzer requires additional sweeps (based on the number of corrected s-parameters).

For most applications the VNA's sweep speed under actual test conditions is not equivalent to "time-per-point" number published in the technical specifications. As noted previously, within most technical specifications, the value reported is a best-case number and is often derived under one (or all) of the following optimized conditions:

1. The instrument's widest IF bandwidth may be activated. (Impact - In some applications this may result in excessive trace noise and a minimization of the dynamic range)
2. The instrument's frequency span may be set such that no band crossings occur. (Impact - Many applications will incur at least one band crossing)
3. Number of point set to a large number in order to spread out the effects of overhead items such as sweep set-up time and obtain the smallest time value per point.

Actual sweep speed is closely tied to an instrument's set-up parameters, including the number of points, frequency range (including any band crossings), IF bandwidth setting (directly affects the degree of accuracy and amount of dynamic range), the calibration method, and many other factors. Therefore it is impossible to predict the measurement throughput performance for each unique analyzer configuration. Sweep speed for each unique configuration should be determined empirically.