Experimental characterisation of a launcher’s fairing separation shock and its influence on RF antennas’ supporting structures

Fairing separation is a critical event during a rocket’s launch as it generates mechanical shocks that can cause partial or total onboard equipment failure. This study details the experimental results of the fairing separation shock of the launcher RFA ONE and its influence on the supporting structures of RF antennas. Accelerometers were used at multiple points of neighbouring structures to characterise the separation devices’ shock during fairing separation tests. Moreover, the shock propagation on different material structures, such as aluminium, carbon fibre, and Viton® elastomer sealant, was investigated. The results showed that the presence of Viton® sealant increased shock transmissibility. Further investigation was conducted to study the influence of applied torque on the separation locks on the shock levels. The study revealed that higher torques lead to increased magnitudes of shock acceleration. Finally, the paper provides recommendations for reducing the shock levels. The experimental results and recommendations presented in this paper provide valuable insights for launcher designers and manufacturers to ensure the safety and reliability of shock-sensitive components during flight.


Introduction
During missions, most launchers experience a series of separation events, such as stage and fairing separation and spacecraft deployment.In these events, separation devices are usually used, which generate shock loads that propagate throughout the launchers' structure [1].These dynamic and nonperiodic excitations have short duration, high frequency, and high amplitude, consequently characterised by their abruptness and severity [1,2].
Due to this nature, shock loads can cause partial or total failure of the vehicle's structures or sensitive onboard equipment.Hence, testing the systems to verify that they can sustain the flight's shock environment is of paramount importance [1].This verification can be done via system-level shock tests [3], such as the Ariane 5/SHOGUN method [1], or by simulating the induced shock.This can be achieved via a mechanical excitation [4,5,6,7,8,9,10], an explosive excitation [1,11] or an electrodynamic shaker [1,12].
This paper presents an experimental study on the mechanical shock caused by the fairing separation event of the RFA ONE, a small launcher designed for Low Earth Orbit (LEO) launches.The study aims to contribute to the understanding of the launcher's separation device and to improve the reliability of the complete system.The separation is carried out by activating separation locks located at the fairing's interfaces with the vehicle.Full-scale separation tests are performed using sensors to measure acceleration levels at different points of the assembly.
Later, the Shock Response Spectrum (SRS) curves of the acceleration data are calculated, and the separation locks' shock is characterised.Conclusions are made regarding the shock propagation throughout the structure's different materials, including aluminium, carbon fibre, and Viton® elastomer sealant.Similarly, the effect of the shock load on a Radio Frequency (RF) antenna's supporting structure is analysed.The influence of the applied torque upon the assembly of the locks on the resulting shock levels was also investigated.Lastly, suggestions are given on approaches to reduce the severity of the shock.

Test item
Figure 1 shows an overview of the main components being tested: the two fairing halves, named Y POS and Y NEG; the second stage top separation flange, which makes the connection between the fairing and the launcher's interstage structures; and finally, the triangles that will support the RF antennas.The fairing assembly has eleven separation locks, two on each fairing half's bottom region and seven along their vertical connection.These locks are metallic components composed of a system secured by torque and released by a pressurisation system.This actuation imposes a significant shock load upon the surrounding structures.

Test campaign summary
A total of thirty-one tests were performed.Firstly, eight tests are carried out to repeatedly measure and characterise a particular separation lock's shock levels.Following this, another nine tests are performed to study the shock's propagation to the outer centre point of the triangle structure.
Next, eight partial activation tests are done where in each two, a different torque is applied upon the assembly of the separation locks: 0.72T, T, 1.45T and 1.81T, where T is the default set torque used on the locks on all other tests.These tests aim to investigate the influence of torque on the magnitude of the separation shock, along with acquiring acceleration data on the bottom region of the triangle to compare shock levels above and below the flange.Finally, another six tests are performed to study the influence of the sealant material on shock propagation.
During the test campaign, pressure sensors are used to monitor the pressure in the separation system, along with four uniaxial PCB 350C24 shock accelerometers.The accelerometers are attached to different points of the structure with minimal amounts of Loctite 454 superglue.

Test results and discussion
For calculating the maximax SRS curves, the standard damping factor, Q, of 10, is chosen, corresponding to a damping ratio, ξ, of 5%.The graphs are normalised with the maximum calculated acceleration response.The graph shows that the separation shock of the lock presents low variations between tests, which indicates a high degree of repeatability.

Shock propagation
In Figure 3, the SRS curves for the measured values below the flange, directly under the previous position, are shown to compare shock levels after propagation through distinct types of materials.The positions of both accelerometers are displayed in Figure 4.
It can be observed that there is increased variability among the tests than when measuring closer to the separation lock (Figure 2).Additionally, the curves exhibit lower overall acceleration levels, particularly for lower frequencies.This might be due to some energy of the shock being partially dissipated by the four aluminium and two Carbon Fiber-Reinforced Polymer (CFRP) plates in between.It is possible that most of the energy is being propagated through the CFRP sections, given that composite materials generally have a higher damping loss factor than aluminium alloys [13].Nevertheless, the curves mostly have a consistent overall trend when compared to each other.Figure 5 presents the SRS curves for the data acquired for the radial direction in the triangle's outer centre point (Figure 6).Since test 14 presents a variation not detected in other tests, the accelerometer may not have been adequately attached to the triangle.Finally, a future test campaign will be performed on the RF antenna and the other surrounding components on the triangle to qualify the equipment for the maximum measured shock environment.
Below are the SRS curves for the tests with and without Viton® sealant on the triangle's diagonal edge inside and outside (Figure 9).The sealant was applied for tests 26, 27 and 28 and removed for tests 29, 30 and 31.As depicted in Figure 7, the Viton® sealant has no significant impact on the acceleration on the outside face of the triangle's diagonal edge.On the other hand, Figure 8 shows that the shock levels inside the triangle's diagonal edge are higher when the sealant is applied between the component and the fairing.This might be due to the compression of the sealant material, which would heighten the material's rigidity [13].Further, without the sealant, the contact between the triangle and the fairing is imperfect, varying from direct metal-metal contact to no contact between them.Introducing the sealant as a filler would occupy any present voids in the middle of the components, increasing the contact surface between the two and, when pressed hard, intensify shock transmission [14].For frequencies higher than 3 kHz, the data in Figure 11 displays great variation between the plotted curves and is also significantly distinct from the data shown in Figure 2 for the same accelerometer position.However, the acceleration measured near the separation lock, below the flange, did not display this behaviour.An explanation might be the fixing of the accelerometer, more precisely the glue, as upon disassembly, it appeared to have not dried as expected, leading to an abnormally easy removal.In the same graph, one finds that overall, increasing the torque on the separation lock tends to increase the shock levels.This effect is better observed when inspecting the SRS for the measurements directly below the flange, shown in Figure 12.Both graphs confirm the previous conclusion regarding the effect of increased torque on shock levels.Comparing both figures, there is a noticeable decrease in the magnitude of the acceleration from below to above the triangle's bottom flange.This occurs for frequencies higher than 200 Hz despite the visibly similar curve trends.This indicates that, for this frequency range, some shock energy is absorbed by the CFRP second stage top flange and the aluminium triangle bottom edge.

Conclusion
This paper presented a test campaign that studied the generation and propagation of the mechanical shock caused by the fairing separation of the launcher RFA ONE.The results show that increasing the applied torque on the separation lock causes an increase in the intensity of the shock, which suggests that the minimum viable amount of torque should be applied upon assembly.Additionally, the presence of Viton® sealant between the triangle and the fairing intensifies shock transmissibility to the latter.The analysis also reveals that significant amounts of shock energy are dissipated through the CFRP and aluminium structures, which indicates that relocating sensitive equipment further away from the separation lock could mitigate the severity of the shock.Another approach would be to protect said components with damping materials or devices, keeping in mind the effect of compression on the materials' shock absorption efficiency.To lessen the shock transmissibility on the fairing/triangle interface, a more appropriate sealant material or sealant dimensions could also be chosen.Future work will be focused on developing a computational model of the structures, performing a finite element analysis, and validating said model with the test's data.Further testing with the fully assembled third stage will also be carried out.

Figure 1 .
Figure 1.Overview of the assembly.

Figure 2
presents the SRS curves for acceleration measurements on top of the flange at the nearest position possible to the separation lock.The partial activation tests are performed to characterise the locks' shock as closely as possible and to verify the shock's variability.

Figure 2 .
Figure 2. SRS curves -near separation lock above the flange.

Figure 3 .
Figure 3. SRS curves -near separation lock below the flange.

Figure 6 .
Figure 6.Accelerometers positioned on the triangle's outer centre point.

Figures 11 ,
12, 13 and 14 present the shocks measured for varying torque values applied to the separation locks.The acceleration values were measured simultaneously for different points of the structure.

Figure 10 .
Figure 10.Accelerometers positioned on the triangle's bottom edge.

Figure 11 .
Figure 11.SRS curves for different applied torques -near separation lock above the flange.

Figure 12 .
Figure 12.SRS curves for different applied torques -near separation lock below the flange.

Figures 13 and 14
Figures 13 and 14 show the resulting SRS curves for the measurements taken on the triangle's bottom edge, above and below the flange, as portrayed in Figure10.

Figure 13 .
Figure 13.SRS curves for different applied torques -on the triangle's bottom edge above the flange.

Figure 14 .
Figure 14.SRS curves for different applied torques -on the triangle's bottom edge below the flange.