Having noticed a discussion of issues with frequency sweep behavior in the 2042X thread, I thought it would be interesting to see if the same behavior exists in the 1062X.
Short answer -- the amplitude step that is present in the 2042X is not present in the 1062X. That is the good news.
The less good news is that the lowest couple of octaves of the 1062X sweep have frequency errors.
A test was set up as follows:
- A table was created in Excel to generate the ideal frequencies at 100 points of a log sweep from 1 kHz to 10 MHz. These points were indexed by time, assuming a 100 msec sweep time.
- The SDG1062X was set up to sweep accordingly. Its trigger out was used to trigger the scope.
- The reference AWG, a Keysight 33521B, was set up with the same parameters, and triggered by the SDG1062X in order to display a time-synchronized sweep for comparison.
- The scope's zoom, delay, and measurement functions were used to take snapshots of the AWG outputs at selected points corresponding to the Excel sample points.
- The results of swept frequency versus time were plotted against the ideal curve.
The first chart shows the Siglent response. Anomalies are seen in the first 10-12 msec of the sweep.
The second chart shows the reference (Keysight) response. This matches the expected curve pretty closely.
The third chart shows a comparison of both AWGs versus the ideal.
Discussion:
During the first 12 -15 msec of the 100 msec sweep the frequency remains at roughly the starting frequency. Thereafter, it snaps to the correct value and continues the sweep without further issue. The changeover can be seen in the scope image at about 12 msec into the sweep.
This behavior suggests an issue with the SDG1000X sweep firmware. The AWG is running 1.01.01.33R1B6.