measuring varying frequency with 9401

[VPEA]

I need to measure a varying frequency with a 9401 in 9178 chassis.  I am using an Express VI.  I have the input connected to the 9401 terminal and can read counts just fine expect that the counts per second vary.  I compared this to a scope and see that the pulse width varies and thus my frequency varies.  When I attempt to measure frequency, I get errors that no data was measured.  I also see the note in the Express VI to connect the input to PFI0.  This I do not understand.  Can someone help me?

[James-B]

Hi VPEA,

Are you using the DAQ Assistant Express VI?  Also, the 9401 does not have access to the counters in the 9178 so it is not possible to configure a pure frequency measurement in the hardware.

What kind of frequencies are you looking to measure?

[VPEA]

I am using the DAQ Assist Exp VI.  For grins, and to learn, I connected my signal wires to the PFI0 BNC and got an RPM signal.  This answered my question if I really need to do this but it does not answer my question of why?  At this point that does not matter because ultimately I am not after RPM or frequency.  I am really after a steady pulse rate which I also learned is not possible.  The system I am using has an inherent varying flow rate which results in a varying pulse rate.

So here is where I am at.  I have two of my flow measurement systems in line measuring the same flow.  They are indicating about 0.3% differece after accounting for thermal expansion of the fluid.  The flow rate is on the order of 0.42 mL.  I am then converting this to gallons for the user to better understand but that makes the number really small.  I am suspecting that, because the flow rate is so small and because I am adding each second of mL flow, I am gettng an accumilation error with such small numbers even though I am using double-precision.  My next attempt is to not convert to gallons until all the calculations are complete.  This will allow larger values in the interium calculations.

Your thoughts?

[James-B]

Double precision floating point values have a precision of 53 bits or 16 decimal places, so that would only be the problem if you're adding values 15 orders of magnitude apart.  On that idea, though, what is the scale of the measurement?  If it's small (like 1-10mL) this is a reasonable measurement, but if it's large (on the order of 1L or so) the flow meter might not be precise enough for this measurement.

Let me know how that test you mentioned works out for you and we'll go from there.

[VPEA]

The meter outputs 90 pulses per CC (mL) - very sensitive.  I am moving liguid on the order up to 68 Liters per hour.

Turns out I had air in the lines.  Once I diligently purged all the air, the measurement became more stable and the two instruments began to correlate to within 0.1%.

Thank you for posting your thoughts.