I seem to have problems with the measurement on the 3406B. I am using PicoScope 126.96.36.199. The setup is watching a single channel A only, 500 us/div, 5 volt range, channel A on the compensation signal from the scope, so it is about 1 kHz. Trigger at 1 V in middle of display. I then enabled two measurements, A High Pulse Width and A Cycle Time. The measurements vary significantly based on how many samples I take. For example, some measurements I get by varying the number of samples:
Request 50 kS ==> 508.1 us and 1000 us (ave High Pulse Width and Cycle Time respectively)
Request 100 kS ==> 470 us and 900 us
Request 500 kS ==> 240 us and 500 us
Request 1 MS ==> 139 us and 270 us
Request 2 MS ==> 61 us and 133 us
Request 3 MS ==> --- and --- (measurement reading seem to stop)
Any request for more samples that this does not get a reading. Also, as the number of requested samples goes up, the readings start to bounce around more and more. For example, when requesting 1 MS, the cycle time "Value" can be seen anywhere in the range of 190 us to 375 us when you do single traces.
In all cases, the 2 volt square wave from the compensation output is displayed OK, with 5 cycles on the display.