Sir,
I have only recently started working on the NI USB-6008 for my university project on temperature control of a plastic extrusion machine using adaptive fuzzy inference-based gain tuning PID controller. The controller implemented on the PID and Fuzzy Toolkit of LabVIEW works well on simulation. Now, I am testing it on actual hardware.
As an initial test, I am trying to control the temperature of water boiling in an electric kettle.
I am using a J-type thermocouple to measure the temperature of water. The results are attached herewith, with the amplitude representing the conditioned voltage in volts. The calibration is 10 mV for every degree Celsius temperature. As evident from the figures, the result I get is noisy, most probably due to vibration of water in the kettle while boiling it. I have attached the vi, in which I have only used an express VI acquiring the voltage and logging it into a spreadsheet file. I am not sure what kind of signal conditioning to do to get rid of this noise: filtering or moving average or anything else.
I tried filtering using a LPF, but it gives an error saying 0<low_freq<high_freq<(sampling freq/2). As far as I know, I kept the low freq to 100, high freq is default to 400 and sampling freq is 10KHz, I believe (USB 6008). What mistake am I making if filtering is required.
Otherwise, what other signal conditioning should I perform? Would it be advisable to perform the signal conditioning before data acquisition in hardware blocks or using LabVIEW itself after data acquisition?
