Dear Rocketfans

We need your thoughts and ideas

A while ago, we had some tests to investigate the best landing configuration of the space capsule, to minimize the forces that our astronout will be exposed to. To do this we 3D printed a model of the space capsule, with two different end caps; a blunt “Apollo style” end cap and a cone-shaped end cap.

the scale model with the blunt end cap

In the model we placed a sensor that could measure the accelerations at a frequency of 5000Hz.


The test setup was relatively simple. We cut off the top of a pallet tank, filled it with water, and during the test we placed a set of cameras above and below the water.

Sarunas prepares a camera on the bottom

The drop itself was done by hanging the model at fixed height, at an angle created by a set of strings that forced the specific angles we wanted to test, from vertical to horizontal at 10 degree intervals.

Getting ready to drop the capsule


And now to the exciting part. Prior to the test, we expected to see reduced forces with the cone-shaped bottom. We also expected to see reduced forces with the blunt end cap if the space capsule landed at an angle.

Some of the landings gave some pretty impressive effects

But how big the effects were, and what their mutual relation would be, was what we looked forward to investigate.

Underwater video – at half speed

The results

This test wasn’t meant as scaled test, in order to predict the actual forces that the astronaut will experience, but rather a comparative test to get closer to understand what combination of end cap and landing angle would give the least accelerations to the astronaut. So please don’t put too much emphasis on the size of the values, but rather on the interrelationship between them.

Below you can see the results from the many drops we did. each dot represents numerous drops, to get a good data set. The graph shows the sum of accelerations of the three axis. the orange line represents the measurements with the blunt end cap, and the blue represents the cone-shaped end cap. For the cone-shaped end cap we were only interested in the vertical drops, but for the sake of further comparison, we did some more drops with the smaller angles.

The angles of the space capsule are represented on the X-axis, where vertical is 0 degrees and a horizontal landing ( equivalent to the astronaut landing on the back) is 90 degrees.

Remember that the astronaut is sitting almost vertically in the capsule, so a vertical landing would mean an almost vertical spine at impact.

Sum af accelerations, Vertical is 0 degrees, horizontal is 90 degrees

Some observations:

  • With the blunt end cap we get the lowest max. acceleration at a 40 degrees angle
  • With the cone shaped end cap we get the lowest max. acceleration at a vertical landing
  • With the blunt end cap we get the most consistent landing accelerations at a 61 degree landing angle
  • With the cone shaped end cap we get the most consistent landing accelerations at the vertical landing
  • Overall, we get the lowest maximum acceleration with the cone shaped end cap
  • With the blunt end cap we have a relatively large sweet spot around a 45 degree landing angle
  • The cone shaped end cap seems to be sensitive to the landing angle
  • The blunt end cap is generally giving more consistent landings

A few thoughts

With the blunt end cap the landing intensity seems to be more sensitive to the angle around vertical and horizontal landings. By having an angled landing we achieve greater certainty for not overloading the astronauts spine at splashdown. A landing with a crosswind can be compared to changing the angle of landing. By using the blunt end cap we get a larger sweet spot area, which can be interpreted to give us more consistent landing results regardless of waves, wind conditions and oscillations under the parachute at time of impact.

A 40 degree landing, seen from multiple angles

The above mentioned thoughts were based on the sum of accelerations. If we split up the values into the XYZ components, Where the axis parallel to the spine must be considered to be essential, we see more nuances to the data set.

Accelerations in all axis, at landing from vertical to horizontal (0 to 90 degrees). The X-axis is parallel to the spine, Y-axis is parallel with the shoulders, and Z-axis goes through the stomach.

We see the same basic pattern for the X-axis with the blunt end cap, with a minimum at a 40 degree angle. With the cone shape it becomes more evident that it is mainly the vertical landing that’s interesting. But we also see that it’s marginally better to land with the blunt end cap at 40 degrees than a vertical landing with the cone.

What now?

There’s no doubt that it looks interesting with the cone shape and a vertical landing, and we’re also planning to do another test with a pointier cone, to see if that gives us new insights.

As data has come out in this test, I would personally prefer to go for a setup where we outfit the space capsule with an Apollo-shaped end cap and land it at an angle between 40 & 50 degrees. As I see it, that would give the astronaut the least stress to the spine, and we will get the most freedom with regards to consistent landings under various weather conditions.

Splashdown at different angles


There are also other opinions, and I’m definitely not sure if we can pull out more knowledge from this data set. So it would be fantastic to hear your interpretations of the graphs, and if you have other suggestions to how we can proceed.

We therefore kindly ask you to throw questions, comments and thoughts after us in the comments field below, in order for us to make the best possible decision.

Ad astra,


Make a donation

Help us continue building rockets with a one time donation

Categories: Blog

Published by Mads Stenfatt on


Carlos A Rodriguez · 17th March 2020 at 12:01 pm

A quick end cap recommendation is a non-revolved shape at an angle. Messes with aero I know but may be worth looking into. Like look at a penguins beak but shorter. You also might already be planning this but you need to apply drift into the equation. Adding lateral motion will tend to effect the impact and spread of forces at varying angles like skipping a stone. Even at low velocities this can help or hinder impacts. Watch a couple videos on the Apollo capsule water pact tests as they had a hard time themselves figuring it all out. Their issue was more mass to shape related to keep it from tumbling but you guys have a slightly different concern. I know you said scale isn’t really accurately done but just make sure your mass you use is I think Raleigh scaled. If I remember correctly smaller objects need slightly more mass to replicate the same forces properly on water at a smaller scale…or that might be backwards. Double check that LoL. One of my old naval architecture professors did work on water impacts of objects like what your doing. I can see if they put out a paper on in. Her name was Leigh McCue-Weil if I remember correctly. Good luck and keep up the great work!

Mads Stenfatt · 17th March 2020 at 9:27 pm

Hi Carlos
We have played around with the Raleigh number and scaling quite a bit, and have come up with some pretty crazy numbers for what mass to use, which is why we in the end decided to just go for the comparative test in the first run.
It would be fantastic if you could find some literature for us to use, so please let us know if you find it.

Stephan · 18th March 2020 at 9:23 am

Blunt endcap between 30 and 50° looks good too me. I would be more concerned that the capsule dosen’t flip over as it did in those tests! On that note: maybe use two accelerometer (on in the top, one at the bottom of the capsule) to make sure the head isn’t rip off by centrifugal forces.
Also, can you add error/spread markers to your graphs?
How did you “add” the forces of the different axis into a single figure? (Absolute values, square and root, geometricly, …?)

Comments are closed.