How do i measure a pulse width ramp?

[XTrevor]

Hello together,

i´m looking for a solution to measure a pulse with ramps with NI 6602 and DAQmx .NET

- i want to know if the ramp is linear or changes in given limits

- Duty cycle starts with 0% and changes over certain time (2000 msec) to 100%.

- Duty cycle starts with 100% and changes over certain time (2000 msec) to 0%.

- Problem is to stop the measurement: at duty cycle 100% and 0% there is no more edge to count (permanent signal) and the buffer runs over in buffered semi-period measurement

Is there a possibiltiy to measure pwm over a certain time (not samples) and do the check in software?

Can anyone give me a solution or a concept to solve this problem?

Thank you in advance

XTrevor

[PeterAdelhardt]

Hi XTrevor,

this is quite the tricky problem.

The only option I see would be the following:

- feed the PWM signal to two counter inputs

- use a third counter for generating a high enough frequency to

- do two buffered edge count reads (one for the falling edges, one for the rising), start them synchronously

- now you will get a pulse start (rising edge) and a pulse stop (falling edge) information in counter 1 and 2, respectively

- a low-to high transition occured when the counter 1 is incremented but the counter 2 is not (hi-lo: vice versa)

- if the period is constant you could just evaluate equal-size blocks of the buffers

- the pw value is of course the time difference from unequal counter values to equal values to unequal again

e.g. 

1-1-1-1-1-2-2-2-2-2-2

1-1-1-1-1-1-1-1-1-1-2

^         ^         ^

Again, your sampling rate would have to be high enough to guarantee a decent resolution for each cycle. As you can tell, this is a more complicated approach than using the semi-period measurement but it should work for you.

Best regards,

Peter