
Firehorn coupling systems design was performed as a semester project and its process is extensively detailed in its associated semester project report.
Following considerations were critical for the design process of the couplers:
- Nordend (2023) → heavy, complex (machining and operations), grip during assembly → avoid threaded coupling → modules assembly with screws for Firehorn and beyond

- Rocket structural philosophy same as Nordend: ears with CFRP rods and covering panels

- Compatibility must be possible with all current and future potential types of interfaces → flexibility
-- Reference design = rods/rods modules
-- Tank/rods → thermal compatibility (-183°C inside tank)
-- Tube (glued)/rods
-- Tube/tube (glued/glued), not in Firehorn
An outside screwing coupling system (called radax) was chosen to meet operations & interfaces flexibility needs. Considering all the iterations (see semester project report, Sect. 8), there are currently 3 radax design options: 30° and 45° options without short centering and a 30° option with short centering.
The main concern with such design is about the coaxiality that can be achieved with low angle conical interface. As there are 6 coupling locations, it may induce a curved rocket. Currently, the main solution to guarantee sufficient coaxiality would require the addition of a short centering feature with H7/js6 tolerances. However, as its centering diameter is around 170 [mm], costs increase (12 parts) may not be sustainable.
An overview of the design options is presented below.

A summary of their advantages/disadvantages and their specifications are shown in the table below.
| Option | Mass (male/female) [g] | [mm] | Height [mm] (ears excl.) | Advantages | Disadvantages |
|---|---|---|---|---|---|
| 30° standard | 1040 (390/650) | 170 | 34 | High integration diameter | Poorest coaxiality |
| 30° short centering | 1035 (425/610) | 170 | 34 | Greatest coaxiality, high integration diameter | Manufacturing (tolerances) cost |
| 45° standard | 990 (435/555) | 164 | 26 | Most balanced and efficient use of material between male/female parts more compact/lightest, enhanced self-centering capabilities compared to 30° standard | Screws less used in traction, very high preload to prevent slip (rotation) |
The 3 design options all share these same characteristics:
- Material: Al 6082-T6
- Ears: as their design is not completed yet by internal structure members, their dimensions were approximated to the worst-case. In addition, they do not reflect the number of screws and their configuration.
- Panel mounts:
– 4 M3x6 screws, undetermined type yet
– Li = 1D = 3 [mm]
– Lk = 1D = 3 [mm]- Coupling screws: 45° pattern between each (22.5° away from ears midplane)
– 8 M5x16 hex socket head cap screws with low head (10.9)
– Li = 1.5D = 7.5 [mm] (use of HELICOIL®)
– Lk = 8.5 [mm] with Lk,− = 5 [mm] (thickness of female part under screw head)
Screw plus tapping being the best way to secure bolted assemblies (especially compared to screw plus nut), the use of screw thread inserts (HELICOIL®) was considered. Yet, as one deals with aluminum parts, it is necessary to use these helical inserts to prevent damage to the aluminium tapping (seizing) since about 100 screwing/unscrewing cycles are planned. In addition, it better distributes the loads along the threads & reduces thread friction (which reduces the screw torsion).

A change in the integration diameter requirement during the project enabled the 45° option, which compared to 30° options provides the most balanced and efficient use of material for a radax system, beneficial for loads distribution. In addition, it is also the lightest option that may satisfy the mass requirement depending on final ears design and with theoretical enhanced self-centering capabilities. However, due to its higher angle, the screws are less used in traction and would require a higher preload to prevent slip.