Modal test of a CFRP rocket fincan for FEM correlation purpose

posted in: Portfolio | 0

Authors: Potito Cordisco, Senior Project Manager; Mauro Terraneo, Chief Technical Officer.

Vicoter (www.vicoter.it) is a consulting company operating in the field of structural dynamics, mainly from an experimental perspective, within the aerospace sector. For the past four years, it has been collaborating with the student association Skyward Experimental Rocketry (www.skyward.eu), supporting its various rocket development projects aimed at competing in the 3000 m category of the European Rocketry Challenge (EuRoC) (www.euroc.pt).

In the activity presented here, Vicoter carried out a modal test to verify the stiffness of a component against preliminary estimates obtained through a finite element model. In this way, both the design and the manufacturing process were validated.

The investigated component is the Fin Can, i.e. the aft section of the rocket that integrates the fins and plays a key role in stability during flight. It is manufactured as a single piece, in this case made of a laminate composed of multiple CFRP plies with different orientations. The junction area between the fins and the cylindrical body, which have different thicknesses, is particularly critical and difficult to model, despite having a strong influence on the vibration modes of the fins.

The structure was tested in free-free condition in order not to introduce sources of uncertainty due to the not ideal stiffness of the clamping system. Low-stiffness elastic was employed to suspend it. It was glued to the structure in two points positioned on the body and suspended to a frame by a crane.

Free-free boundary condition for modal testing of fincan.

A global reference cylindrical coordinate system was defined and  forty-two locations were measured: 9 on each of the three fins and the remaining ones on the body of the fincan.

To obtain the modes of the structure, a MIMO test in roving modality was carried out. Five accelerometers PCB 333B32 were used as references, in order to avoid nodes of the modes and to guarantee against the possibility to miss some modes, while impact location has been modified step by step to obtain all the 42 needed FRFs (Frequency Response Functions). Roving technique was preferred to the classic fixed accelerometers one due to the foreseen shapes of the modes, with only movements orthogonal to the planes of the layers, and the need to reduce at minimum the mass loading effect of the accelerometers.

Considering the participation of the investigated mode and the location of its nodal points, sensors were glued to the structure in points with the sensibility axis in the tangential direction for the first three ones that are collocated on the fins and in the radial direction for the other two points that are positioned on the case.

Thanks to the use of multiple references even double poles (modes at the same natural frequency), typical of axial symmetrical structures as this one, can be identified.

Installation positions of the reference accelerometers.

FRFs were acquired with a sampling frequency of 4096 Hz to cover with ease all the frequency band of interest. Each point was hammered 5 times in order to reduce measuring noise thanks to the averages and improve S/N ratio. Coherence analyses of acquired signals showed values very close to one in all the interested frequency band. So the method was considered suitable for test purposes.

Identification in terms of natural frequency, shape, damping and modal mass was carried out with software LMS-Test.Lab, using the algorithm PolymaxTM.

To remove the effect on the modes even of the 4 grams of each of the used accelerometers, a so called ‘Modification-Prediction’ procedure was applied at the previously identified modes, improving the feasibility of the obtained results.

Example of fin involving mode.
Example of body involving mode.

 

Quality of the obtained modes was evaluated by AutoMAC matrix and validation complexity indices such as Modal Phase Collinearity (MPC), Mean Phase Deviation (MPD), Mode Overcomplexity Value (MOV) and Mass Sensitivity.

Comparison of the modal test results with the FEM results shows a very good agreement in terms of shapes whereas the differences in the natural frequencies indicate the need to update the characteristics of the material once laminated.

Example of test and FEM shapes comparison

Vicoter greets all the members of Skyward and wishes them ‘In bocca al lupo’ for EuRoC ’25, hoping for a repeat of last year’s success.