I am reading a 3120_0 FC22 Compression Load Cell (0-10 lbs) through 1018_2 PhidgetInterfaceKit 8/8/8. They work perfectly, however I am getting strange hysteresis behaviour.
These are two plots of forces I am reading by applying loads very gently:
In these plots, the vertical axis is in Newton and the horizontal axis is time in second. As you can see there are strange behaviour at the tip of the peak points in these two plots which I believe causing the hysteresis.
and these are the force-displacement plots showing the hysteresis loop:
In these plots, the vertical axis is force in Newton and horizontal axis is the angular position of a motor in degree that is applying loads to the load-cell through a this cable connected to a spring. As you can see in the plots the hysteresis effect is around 14 percent (0.2 N) which is a lot.
I have a motor with a pulley coupled on its shaft. When I actuate the motor, it rotates the pulley and pulls the cable that is connected to a spring from the other end. Therefore I produce tension in the cable. I have a mechanism based on what is used in tension meters (http://www.hans-schmidt.com/vendor/images/electronic_tension_meter_ZED_for_fibres_threads_and_fine_wires.jpg). Using this mechanism, I convert the tension in the cable to a normal loads for the load-cell.
When I actuate the motor, it pulls the cable and the load applied to the load-cell increases. Here is a plot showing the angular position of the motor shaft:
However, when it gets to the pick, the loads measured by load-cell keeps increasing where the motor rotation is slowing down as shown in the above plots for the forces.
I am wondering if anyone else has experienced such problem before or anyone can suggest a solution or explanation for this problem.