After the Stratos II launch attempt had run its course and everyone had recovered from the two weeks in Spain, the discussion was started on how to continue the project. Should we build a new rocket and skip the 50 km mark altogether or improve the system and try again? As the majority of the team decided they were still up to launching Stratos II, it was soon decided that the objective for the coming year was going to be simple, make sure it works the next time.
For the capsule and recovery team, the major redesigns were based on operational requirements. This resulted in a redesign of the two load carrying bulkheads, for the flight model produced by Nieuwstraten Metaalbewerking BV, that form the basic interface with the payloads from Nijmegen and Delft Dynamics and with the Stratos II power systems and onboard computer.
Nieuwstraten BV made us very happy with two capsule bulkheads.
The second step was a new round of testing. Tests that were initially omitted during the busy period before the launch attempt in 2014. With another year and with, after modifications, a flight ready system we decided to look for opportunities to put the system through dynamic load testing as part of a proper flight qualification program.
The three main goals of the test were to verify continuous operation of the electronics, to ensure structural rigidity of the capsule when subjected to the dynamic conditions during flight, and to allow calibration of the accelerometers used for the Kalmann filter.
Together with TNO Delft we came up with a proper test proposal which we discussed with them in the weeks before the test. After a good discussion with TNO Test Engineer Cees van Everdinck and Dr. Stefan Vossen they gave us the green light for the test for the 26th and 27th of May.
Test readiness
To connect to the shaker table provided by TNO Delft, an interface was build quite similar to the test fixture used for the Stratos II propulsion tests. Welded steel plates are clamped to the shaker table and provide a common interface for tests in all directions. A test readiness review was held with all involved parties and in the week before the test, long hours were made to get the system flight ready.
On Friday evening everything was done and ready for the test the next week; Parachutes were packed, Radboud University Nijmegen had delivered and integrated their payload and the Stratos II flight computer was blinking happily.
The capsule in the test set-up at TNO
The DAYS at TNO
We arrived at TNO early in the morning with 5 people. It was decided earlier that due to the risk of damaging the flight system a small test would be performed first with the old (and obsolete) engine control unit. This allowed us to test at even higher load levels ramping up to way beyond the qualification levels for Stratos. The test was (fortunately) uneventful. Apparently our soldering and assemblies are good enough to withstand even the harshest environment (10 g rms).
On we went to the actual flight hardware: The test setup was quickly mounted to the table and in collaboration with the TNO Test Engineer the accelerometers were placed in and around the capsule.
Ready for testing.
In general the test follows a standard pattern. The shaker table performs a self test with the components to create a good estimate for the required input power. A low power sine sweep is made (at 0.5 g rms) to take a fingerprint of the system, indentify dangers such as loose connections and to use during later analysis. The actual test is a random vibration that is ramped up (typically in steps of 3 db), up to the required power and duration. Afterwards another sine test is performed and spectra are confirmed.
The test spectrum that we in the end decided upon was the same spectrum used in the past decades for payload qualification of the single stage Orion sounding rocket (see the 1999 NASA sounding rocket handbook). The vibration data from earlier propulsion tests were of insufficient quality to give a useful spectrum as it ran into the Nyquist frequency at 250 hz. Additionally, they provided only thrust oscillations and did not take into account any aerodynamic oscillations and flutter.
Thrust direction
The trust direction proved to be the most difficult direction to test in, as the test fixture itself proved to have a rather enthusiastic mode around 1800 Hz reaching 40 g rms. The machine was able to compensate mostly for the attenuation but it was not a clean run. It was however decided that, due the already high attenuation at the higher power levels and the excitation being primarily generated in the fixture, and not the capsule, to call the test a pass. In the end it went up and beyond the required power level. Everything was completed at 16:00 and the team decided to call it a day and return the day after for the lateral directions.
Isometric view of the CAD drawing of the test rig containing the capsule.
Lateral motion
The lateral directions were initially thought to be the most difficult modes due to constraints imposed by the test bench and the large pendulum motion that comes with the forcing function being introduced well outside the centre of mass of the capsule. To prevent any damage, an extra semi-rigid support was installed at the other end of the nosecone. This was enough to prevent RUD (rapid unintended disassembly) yet flexible enough to allow for excitation of the entire structure.
The tests looked more violent, since the tip swayed 3 cm peak to peak. Luckily the nosecone provided by Futura Composites is an extremely durable product and the test proved to be less of a problem than initially anticipated. After all, it was completed well ahead of schedule.
Tomas, one of our capsule team members, showing his happiness after a productive day of testing.
Conclusion
The Stratos Capsule, flight computer and Recovery system proved to be designed well beyond the required strength for the actual flight coming October. This is great news since it means that yet one more unknown has been tested, which gives us all the more confidence in the design!
We would like to specifically thank TNO in Delft and our TNO Test engineer for the change to use their facilities: We had a blast!
About Nieuwstraten Metaalbewerking BV
Nieuwstraten Metaalbewerking BV processes all metals and plastics for both single piece and series work using the most modern conventional and CNC lathing and milling machines. Furthermore, they process all electricity conducting materials using electrical discharge machining.
Nieuwstraten Metaalbewerking BV. http://www.nieuwstratenmetaalbewerking.nl/