Weekly Report #17: Analysis

An Introduction to Frequency Response Analysis

We would really like to tell you how our Integration Progress Review (IPR) went on Wednesday, talk about the changes recommended by the experts and how we plan to implement them. There is a small problem however: The review did not take place. Due to the current airline pilot strikes in Germany, the experts told us on Tuesday evening that their flight from Bremen had been cancelled, so the IPR will be delayed to September 30.

In order to make up for the delayed report this week, here is a cool picture: The housing for our new HDRM prototype has been 3D printed, and we were able to assemble and test the entire mechanism!

Bild1 As you can see by the helpful scale-providing hand in the picture above, the mechanism is still quite large by CubeSat standards, so we will work together with our SMA spring supplier Ingpuls on scaling it down further. The design itself however is now finalized, and the prototype works like a charm.

As we mentioned in an earlier report, we conducted a finite element method (FEM) analysis of our experiment after the CDR. This revealed a small problem with our experiment frame:

Bild2The picture above shows the results from a frequency response analysis. This type of simulation basically shows what happens if the experiment is vibrated, which is what will happen on the rocket. The simulation places an acceleration load of about four times earth gravity (4 g) on the experiment in radial direction (i.e. perpendicular to the rocket’s flight direction). This acceleration is not constant, however, but varies over time between +4 g and -4 g, forming a sine wave. The frequency of this sine wave is then incrementally increased, which provides the graph on the right in the above picture.

The graph shows displacement of a corner of the experiment frame over frequency. The way to read it is like this: If you apply a sine form radial accelerating force with a frequency x to the experiment, it will start to vibrate. After some time, these vibrations will reach constant amplitude. The amplitude at the frame corner we are interested in is the displacement value y in the graph above.

Sorry if this explanation is hard to understand, FEM is complicated. The one important thing to take away from it all is this: At a certain frequency, the frame is deformed by up to 2.5 mm, which might cause damage to the frame, preventing the experiment from working correctly. This is a Bad Thing™.

We introduced two design changes in order to fix this: One, we will mount a stiffening beam on the frame:

Bild3And two, we will redesign the frame back plate to also include some stiffening beams. We are confident that this should solve the vibration problem.

So that’s it from the structure team for this week. Hopefully by our next report we can finally tell you the IPR results and give you an update on the smaller mechanism!