Tektronix Technical Forums are maintained by community involvement. Feel free to post questions or respond to questions by other members. Should you require a time-sensitive answer, please contact your local Tektronix support center here.

Calculate Period across two periods?

Questions in this forum area are community supported. Tektronix does not regularly monitor posts in this area.
Post Reply
stevenkanPRL
Posts: 7
Joined: December 21st, 2017, 3:59 pm
Country: United States

Calculate Period across two periods?

Post by stevenkanPRL » October 11th, 2019, 5:56 pm

We're rocking an old CSA8200, but it still works pretty well, I think!

I am playing around with a set of frequency multipliers from Mini-Circuits. These are unpowered devices that take an input signal, f, and output a signal with twice the frequency. Of course you lose a lot of power in the process, and the output waveform looks ugly as heck, but the 2f components are there:

Image

Image

So I'm starting with a 50 MHz TTL signal (0 - 2.2 V into 50 Ohms) that has reasonably good symmetry (Ch. 1, green trace). An identical copy of this TTL signal goes into the Mini-Circuits MK-2 multiplier. I'm inline-probing a bit of the MK-2 output and feeding it into Ch. 2, yellow trace, with the probe attenuation correctly set in the scope.

The rest of the MK-2 output is going into a comparator module with TTL outputs, and a settable input threshold of 0 V or +50 mV. The TTL output of that comparator is on Ch. 3, purple trace. Scope trigger is from another copy of the original 50 MHz source.

If I set the comparator input for 0-crossing, the output waveform alternates between fat pulses and skinny pulses, which is exactly what I'd expect, because 0 V is shown on the scope as the dashed horizontal cursor, and the yellow trace crosses that threshold narrowly for half the pulses (I'll call those the "even" pulses) and widely for the "odd" pulses.

But the "Prd" measurement on the scope shows a measurement of 12 ns with a standard deviation σ of only 3 ps. Given the bimodal distribution of the pulse widths, I'd expect the standard deviation to be on the order of a ns or more.

Which means I probably have the measurement set up incorrectly. It looks like it's only measuring the "even" period.

If I trigger instead on a copy of the 100 MHz signal, the Prd measurement now sees the variability of the periods, but it's not stable enough to take a measurement:

Image

Image

Furthermore, I can no longer see the true waveforms for Chs. 1 and 2, because they're being averaged.

Is there a way to set up the scope to trigger at 50 MHz, but take the Period measurement across two periods (including the variability thereof)?

p.s. in terms of the actual application, I would set up the comparator trigger point at +50 mV (the solid red horizontal cursor), and then the even/odd pulses are much closer in width, but I want to test out the measurement setup when they're visibly different, so I know that I'm taking the correct measurement. When I see visibly different widths/periods, but the scope is measuring σ = 3 ps, then I know I'm doing it wrong! Once I get the measurement set up correctly, then I'll tweak the comparator threshold to get the performance as good as possible.

Post Reply

Return to “Other or Discontinued Oscilloscopes”

Who is online

Users browsing this forum: No registered users and 1 guest