Having settled on a configuration and propulsion system for the probe it was necessary to ensure that the probe's structure could withstand the stresses placed upon it. The most severe dynamic loads would be placed on the probe during the injection into its intercept trajectory by the STAR 30E boost motor, when accelerations could approach 25g. Given the author's lack of experience in structural design a simple but conservative approach was adopted. A general structure was adopted that seemed to meet the requirements of the probe and some basic calculations were carried out to determine how well it would stand up to the loads imposed.

General Structure. The probe's configuration of a short, wide cylinder is inherently good at resisting compressive axial loads. A more difficult challenge comes from supporting systems within the probe. 'Floor' areas are required to mount equipment upon and these will be subjected to large bending loads due to the 'weight' of the equipment under high acceleration.

After considering general design guidelines for spacecraft structures [Larson92] the structure of Fig 17.1 was chosen as a design baseline. The main structural element is a 'thrust cylinder' that takes the axial load imposed by the boost motor. Part of the way up inside the thrust cylinder is the equipment floor, a platform that supports bulk of the spacecraft subsystems and science payload. The side array forms a non-strucural 'skirt' around the central cylinder suspended from the top end via braces that also support the front solar array. The rear solar array is mounted to the bottom end of the thrust cylinder and array skirt.

The use of a side array skirt was chosen over enlarging the thrust cylinder to the desired spacecraft diameter for two reasons. Firstly the smaller the radius of a cylindrical shell the larger the axial load it can support without buckling. Secondly the use of a smaller-diameter thrust cylinder allows the equipment floor to be smaller and thus better support the bending load it is subjected to. As it has already been decided to have a concentric array and antenna separated at 0.7 of the spacecraft radius it was decided to use this as the diameter of the thrust cylinder, giving it a radius of 35 cm and a height of 50 cm.

The thrust cylinder must be strong enough to support the axial load placed on it without either failing in compression or by bucking. The former can be checked by ensuring that the product of the cylinder's cross-sectional area and its yield tensile strength F_ty exceeds the applied load. Confirmation of the latter is more complex as it involves calculating the buckling stress for the cylinder. For a cylinder of radius R and thickness t made from material of Young's modulus E the critical buckling stress sigma_cr is given by

(17-1)

where gamma is an empirically-derived shape parameter given by

(17-2)

Taking the thrust cylinder to be made from sheet aluminium with F_ty = 460 MPa and E = 71 GPa [Larson92] then Fig 17.2 shows the compressive and buckling loads that it can support for thicknesses from 1 to 5 mm.

It can be seen that up to a wall thickness of 5 mm the cylinder fails by buckling before it would fail in compression. However, even a wall thickness of only 1 mm gives a buckling strength of 100 kN, 150% above the maximum thrust imposed by the boost motor. This was thus selected as the wall thickness, giving a cylinder mass of 3.11 kg.

The other main structural component sized was the equipment floor. This was modelled using a very basic approximation of a uniformly-loaded beam 0.7 m long and 0.25 m wide. This beam represented a wide chord across the floor; by assuming it carried the full weight that the floor would be required to support a considerable margin would be built into the calculations. From the mass budget for the probe it was estimated that some 69 kg of equipment and payload would need to be supported by the floor under an acceleration of (including further margin) 30g. The deflection of a uniformly-loaded beam of length L supporting weight W is given by

(17-3)

where E is the Young's modulus of the beam material and I the beam's area moment of inertia. A Mathcad model was set up to calculate the thickness of beam required to support its own mass and that of the equipment on it so as to give a central deflection of 1 mm. This figure was arbitrarily chosen as being an acceptable level of distortion to avoid damage to subsystems and connections. Three different structural materials were evaluated: sheet aluminium, sheet titanium and graphite epoxy composite [Fortescue95]. For each material the thickness was calculated for a solid floor and for one made of a honeycomb comprising two 1 mm thickness sheets of that material spaced by the floor thickness. The results, together with the masses of the whole equipment floor if made from that thickness of the material in question, are shown at Table 17.1.

It is immediately evident that honeycomb structures are preferable to solid ones, although a matter for consideration would be the crush strength of such honeycombs. This calculation is very simplistic in that a realistic design would almost certainly use a thin solid floor supported by flanges from the inside of the thrust cylinder. For the purposes of mass modelling, an equipment floor mass of 10 kg was assumed; this is considerably more than for any of the honeycomb masses and reflects the large margin of error in this calculation.

Other Probe Areas. Other elements of the probe whose load-bearing properties were not directly calculated were the front and back faces and the outer side wall, all of which supported solar arrays. Both front and back faces were annular in shape, 1 m in external diameter and 0.7 m in internal diameter. The outer side wall was a cylinder 1 m in diameter and 0.5 m deep. If made of aluminium honeycomb specified as above the mass of each end face would be 2.25 kg and the side wall 8.8 kg. The braces for the front face (shown schematically in Fig 17.1) were not modelled in detail but it was assumed that they would have a mass no greater than the thrust cylinder, i.e 3.11 kg.

The total mass budget for the probe's structure is thus as shown in Table 17.2.

Summary. An extremely basic effort at structural design arrived at a probe design built around a central thrust cylinder taking the axial load imposed by the injection boost motor. An equipment floor within the thrust cylinder supports payload and spacecraft systems. Modelling indicates that a an aluminium thrust cylinder and aluminium or graphite epoxy honeycomb equipment floor of sufficient strength would have a total mass of approximately 13.1 kg, whilst other structural elements would bring the total structural mass up to 29.52 kg.