Displaying a hysteresis graph

[MagpieSwift]

I am trying to replicated what i created before using a load cell and linear transducer to create a hysteresis graph, simply force against distance displayed as per attachment.

I used before NI9219 for both load cell and linear transducer, this was ok but now i have a NI9237 which is better for the load cell, but now i seem to have a sampling issue that i can not seem to work out, it seems to be running too fast, actual sample rate says 1612.90.

This gives a jumping range for the load cell see attached NI9237, not quite so easy to read as previous graph.

If anyone can give me some help on this i would really appreciate it,

Thanks in advance

[Bob_Schor]

It seems to me that you have two issues (well, three -- you need to learn about sub-VIs to encapsulate your code so that it doesn't take multiple monitors to see it all):

1. Why does what seems to be innocuous DAQmx Read Analog Input report that it is running 1.6 times faster than the specified Sampling Rate?
2. Why does your output look so strange?

I don't have any good ideas for Problem #1 -- I'm unfamiliar with that device, and don't have its manual to consult, but if it has a clock and your sample rate specification is "reasonable" for the device, I would expect it to report a "true" sampling rate at, or possibly slower than, the specified rate, not faster ...

As for Problem #2, I have two recommendations.  First, you are collecting your data as an Array of Waveforms, the ideal (in my opinion) to get these data.  Do not convert them to Dynamic Data Wires -- they obscure and confuse the issue.  You can use Index Array (drag the bottom edge down to get two output terminals -- you don't need to wire the inputs in this case, the outputs will be the first two channels).  The second is to write a little test routine that generates "simulated data" that you can plot, and plot that.

Looking at your plot, what I see is a slowly-changing voltage and then some "noise spikes".  I'm assuming that the load cell and transducer are actually moving fairly slowly -- is it possible there are loose wires or poor connections that create those strange millisecond spikes?

Bob Schor

[CFER_STS]

The NI 9237 is a delta-sigma ADC on which acquisition is timed by the module.  See the datasheet for more information.  1.613 kS/s is the slowest acquisition rate supported by this module.  That said, for the best possible signal, you want to acquire data at a rate supported by the module and then use a block averaging filter to reduce noise in the signal you require.