0
\$\begingroup\$

I need to measure the time period between points A and B of this signal.

Scope trace

The trace is 5 volts/div and 1 sec/div. There are 20 to 50 seconds of "noise" from point B ta A. The time between A and B varies from 3 seconds to 15 seconds.

I could use the circuit shown in EEstackexchange/ignore-pulses-less-than-a-given-width-in-a-pulse-train to eliminate the very short pulses from B to A, invert that signal and use another discriminator to eliminate the 1 second pulses from A to B, but that's a lot of circuitry.

Is there a simpler way to process this signal?

\$\endgroup\$
2
  • 1
    \$\begingroup\$ A state machine implemented on a micro controller is what springs to mind. Is the 'signature' of the A event 1 second high followed by 1 second low? Is B a high-going event followed by 1 second high? Are those 1 second marker pulses? What happens if B goes high during a marker? Once you've nailed down how to spot each event, you can build a state machine to parse what's happening. \$\endgroup\$
    – Neil_UK
    Commented Aug 7, 2017 at 8:40
  • \$\begingroup\$ Have you measured the form of this signal after filter it by a low-pass filter? \$\endgroup\$
    – pasaba por aqui
    Commented Sep 3, 2017 at 8:57

1 Answer 1

Reset to default
1
\$\begingroup\$

Thanks for your suggestion. I eventually got this to work on an RPi. I had to send the signal to two GPIO pins. The was set up to handle interrupts on the falling edge. The second was setup up the handle interrupts on the rising edge.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.