Counter Frequency Twice the Known Input Value

[gregorio66]

Hi all,

I have a 50% CMOS counter signal coming from an edy current signal conditioning box.  The frequency of this signal changes directly proportial to the speed of a shaft with points I am picking up on.  Right now, I am feeding this signal to a PXIE-6358 module.  I get two different results in the following scenarios:

1) When I connect the positive lead to PFI9 (counter input) and the negative lead to the DI ground directly, the result is twice the frequency relative to what it should be.  I am using a DaqMX node for measurement.  The square wave when observed in this case goes from -2 to 5V.

2) When I connect the signal as stated in case 1, but "T" the signal to an Analog Input, the frequency is exactly what it should be.  The square wave when observed in this case goes from 0 to 5V.

Does anyone have any suggestions as to why this is occuring?

Thanks,

Greg

[Mr.F]

Hi Greg,

What did you use to observe the signal in both cases? In the 2^nd case the analog input make sense but in the first what did you use. I’m not sure about the different level from your sensor, but by adding the “T” to you input signal you could be changing the characteristic of the transmission line of your pulse train enough to filter out some noise and read the correct number of pulses and therefore the correct frequency. If your signal is noisy than for every actual rising edge, noise in your signal can register as two or more edges. You can try enabling digital filtering on the input lines to your counter task to removed unwanted noise. You can find an example of how to do this in Read Dig Chan-Digital Filter.VI in the example finder. You can use a channel property node like in the example to enable digital filtering on your task and specify a minimum pulse width. Let me know if that helps.

[gregorio66]

Matt,

This was actually figured out yesterday, but I forgot to update the forum.  I used an oscilloscope and saw overshoot on the rising edge.  I cleaned this up by attaching an RC ciruit at the digital input.  We are currently working through some other issues, but I believe them to be source (sensor) based.

Thanks for your help,

Greg