Acquire PWM input using NI PXI-6528?

[ram301090]

Hi

Objective - My objective is to view the waveform of incoming PWM signal.

Queries:

1) Is it possible to acquire PWM signal of 250Hz through NI-PXI 6528 and plot it in digital waveform graph?

 2) Does PXI 6528 digital I/O supports the above  functionality through DAQmx Digital input continuous sampling method?

If any further detail or clarification is required, please let me know

Thanks

Sriram

[Kevin_Price]

From my skimming of the manual and specs, it does not appear that the 6528 supports hardware-clocked DIO.  It does support "change detection" but that alone won't give you the timing information needed to measure or properly display duty cycle.  You'd get exactly one sample per state, no matter the frequency and no matter the duty cycle.  Your graph would always look like a perfect square wave.

-Kevin P

[Lysandros]

It must be possible, but it really depends on your VI-architecture. Do you need to display/process the duty cycle % in real-time, or do you want to do this offline as post-processing?

Anyhow, you need to acquire the signal much faster than 250 in order to get a fine resolution of the pulse form and thus calculate its "high" period.

[ram301090]

Thanks Kevin.

If PXI-6528 does not support this feature, I also have PXI-6025E (multifunction DAQ).

Will that resolve my problem?

[ram301090]

Thanks Lysandros89

I need to display PWM signals in graph. Processing of PWM needs to be worked out in real-time.

Will PXI-6258 resolve the issue or PXI-6025E?

[Kevin_Price]

The 6025E (as well as any other E-series board) also does not support hardware-clocked DIO.  However, it does have 2 counters and you could use one of them to do semi-period measurement.  From there, you'll have to calculate your freq & duty cycle info and you'd have to do some further processing to generate a graph that looks like the actual digital waveform signal.   But it's at least feasible.

Note: I believe I recall from long, long ago circa Y2K that the legacy driver didn't support a polarity setting for E-series boards.  When doing semi-period measurement, there could be ambiguity about whether the odd/even time intervals are high/low times or low/high times.  I do not know if that's been remedied under DAQmx or whether it's a hardware limitation that will still be present under DAQmx.   It's mainly a problem if the real duty cycle can be near 50%, otherwise you can figure it out the correct polarity during post-processing.

-Kevin P

[Lysandros]

I only use a PXI chassis in combination with an external expansion chassis NI-1078 which houses C-Series modules (works with the Scan Engine principle), so I am not familiar with the specific PXI cards. However such a card definitevely provide the hardware for PWM measurement.

FPGA is very flexible for generating and measuring PWM signals. Do you have an FPGA capability in your system? If not, try the DAQmx: you could use the signal's rising edge itsself as the trigger start condition (beginning of the pulse) and afterwards captue a finite samples package during the following 1/(250Hz)=4ms, which can then be further processed online...

At the next rising edge, the next package of finite samples is captured.

Maybe this helps:

https://forums.ni.com/t5/LabVIEW/measure-PWM-duty-cycle/td-p/3313566

https://forums.ni.com/t5/LabVIEW/Two-Threshold-Analog-to-Digital/td-p/2738762