Imperfect Pulse Width?

[eddieenygma]

Hello to all.  I'm in need of some assistance in troubleshooting a Labview / data acquisition problem.  I have a USB-6218.  I want to use a counter to generate a pulse train (using Counter 0).  I then want to log the pulse train (I've tried using both digital inputs as well as analog inputs, but the example I am attaching uses analog inputs).  The block diagram for the pulse train generation is in Figure 1. The TDMS based logging block diagram is in Figure 2.  Frequency of generation is 1 Hz.  Duty cycle is 50%.  The resulting waveform is in Figure 3.  The "imperfect" pulses happen with some frequency, but it is unpredictable.  This occurs at other frequencies as well (10 Hz and 100 Hz) and also using a digital input instead of an analog input to log the pulse train data.  Any ideas as to why this is happening?  Thank you.

[Eric_NI]

Hi eddieenygma,

When you run the VI, how is the signal displayed in the screen? Is it showing the same data points? Have you tried the test panels in MAX do you get the same kind of signal? the log VI will save the data that is coming through the channel it is not supposed to add anything to it.

Eric NI

[deppSu]

Looks like there is no primitive show in TDMS logging.jpg.

[eddieenygma]

Eric:

Hi.  To keep things simple, I am not displaying the signal. I am only logging it using the TDMS format.  The image of the waveform comes from the TDMS file and plotted using StarOffice.  This was when I became very concerned, as the logged data was not what was expected.

-Jason

[eddieenygma]

Eric NI:

Hi.  To keep things simple, I am not displaying the signal. I am only logging it using the TDMS format.  The image of the waveform comes from the TDMS file and the plot is from OpenOffice.  This was when I became very concerned, as the logged data was not what was expected and not some graphical update issue.

-Jason

[eddieenygma]

Eric NI:

Hi.  To keep things simple, I am not displaying the signal. I am only logging it using the TDMS format.  The image of the waveform comes from the TDMS file and the plot is from OpenOffice.  This was when I became very concerned, as the logged data was not what was expected and not some graphical update issue.

-Jason

[Chris-NI]

Hi Jason,

We just spoke on the phone.  Since the USB-6218 is software timed, it may be getting sporadic measurements that are not representative of the actual counter/timer output.  Try putting a scope on the output to see if the odd pulse widths you are seeing are just an artifact of the measurement device.

Our X Series cards have hardware timing that can mitigate this timing issue.  The USB-6343 is a good card if you are using the 6218 and need hardware timing:

    http://sine.ni.com/nips/cds/view/p/lang/en/nid/209070

Have a great day!