I G Force Calculation Optimisation with camera, equations & accelerometer (drop testing)

[ao01]

TL;DR Summary

G Force Calculation Optimisation with camera&equations&accelerometer

Hello, we do drop testing on devices and we have an accelerometer. Our accelerometer does not always give accurate G values, that is, as the altitude increases, the G force should increase, but there are moments when it decreases. We also place impact labels on the box we throw away, but they cannot always verify this situation.

I'm trying to get more accurate results on the computer with the help of a camera. Peak accelerations average 441 in all tests (427-528). I tried to find a better result by dividing these by the collision time, and indeed I got slightly better results, but I'm never sure.

Apart from this verification, I also want to reach this result by making calculations. I don't know how the G-Force value is calculated, it is calculated differently in many sources. Maybe I'm wrong because I examined the calculations according to the impact label table.

To give an example, we carry out the tests by raising the height by 10 centimeters:

385G in 30cm
40 542G
50 550G
60 552G
70 524??
80 425??

We reach the values and the impact label of the device turns red at 80 cm, but the G value is lower than other heights.

I also want to develop a simulation. With which application can I find an easier and faster result while developing this simulation? I'm waiting for your help.

 

[berkeman]

Welcome to PF.

Can you upload some pictures of your test setup and sample preparation? (Use the "Attach files" link at the lower left of the Edit window.) It looks like you have some extra friction or binding in the higher drop tests...

 

[A.T.]

I tried to find a better result by dividing these by the collision time,

You should rather intergrate the acceleration over time, and compare that to the expected velocity change.

To give an example, we carry out the tests by raising the height by 10 centimeters:

Did you drop it from each height several times? What is the spread of values you get for a fixed height?

What is the temporal resolution of your accelerometer and data recording? Are you sure it can even capture short time peaks properly?

 

[ao01]

Welcome to PF.

Can you upload some pictures of your test setup and sample preparation? (Use the "Attach files" link at the lower left of the Edit window.) It looks like you have some extra friction or binding in the higher drop tests...

I'm sorry, I'm on leave for a while, but the system is like this. I am using pcc program with determine the speed and acceleration. If it might be useful to you, I can share the velocity and acceleration txt documents, but I don't think you can understand them alone.

You should rather intergrate the acceleration over time, and compare that to the expected velocity change.


Did you drop it from each height several times? What is the spread of values you get for a fixed height?

What is the temporal resolution of your accelerometer and data recording? Are you sure it can even capture short time peaks properly?

We could only do it once because we did not have enough impact labels.

 

[A.T.]

We could only do it once because we did not have enough impact labels.

I'm asking about your accelerometer values. How consistent are they for the same drop height?