Filter for noisy signal

[Peter Nachtwey]

Will it make a difference if I have more data points before and after the peak? I just did a drop test and exported the data. I'll do the analysis in Python and see how it comes out.

It depends how fast your sample rate is. Do the extra points look like they are part of the parabola.
So what Keith pointed out is interesting. The load cell is quite stiff and from that you can calculate the spring constant of the load cell and then how far the load cell will be depressed and from that the time to compress the load cell. See my post about the calculations.

From looking at your graphs, it looks like the impacts are quite short.

 

[drbitboy]

https://www.youtube.com/watch?v=ZwKV6co3Sjw


It's not exact because in the case of this thread there may be some heat loss (deformation).

 

[Peter Nachtwey]

https://www.youtube.com/watch?v=ZwKV6co3Sjw


It's not exact because in the case of this thread there may be some heat loss (deformation).

That is what I wrote above.
Yes, there is some but little heat loss. If the load cell can be calibrated accurately then the heat loss can be computed as the difference between mgh and the work done on the load cell. The heat loss will be small and you might as well include air resistance too if you are going to be that picky.


I think there is enough detail for a student project.