velocities (<7 km/s). Because of this lack of capability, all breakup models use data from tests at impact velocities lower than the typical LEO collision velocities. In addition, few facilities can perform large-scale collision tests in a controlled environment. (A very large chamber capable of conducting both explosion- and collision-induced breakup experiments has, however, been constructed in Russia [Fortov, 1993].) As mentioned in Chapter 2, without a controlled environment, data on the distribution of small particles generated by a breakup are suspect and data on the breakup-induced velocities of any size particle become difficult to obtain.
Finding 1: High confidence in the validity of (1) assessments of the response of spacecraft components and shield configurations to debris impacts and (2) component and shield qualification and acceptance tests can presently be provided only by hypervelocity impact testing, but current hypervelocity impact facilities cannot simulate the full range of debris impactor sizes, compositions, shapes, and velocities. As a result, spacecraft protection systems currently are designed to resist the type of projectiles that can be launched by these facilities (most typically aluminum spheres). Because actual debris objects typically have more complex shapes that are very likely to do more damage than spheres at LEO collision velocities, current spacecraft surfaces and shield designs may not provide the desired level of protection.
Finding 2: Facilities in a number of nations are capable of carrying out hypervelocity impact tests for debris research but information about and access to these facilities is often difficult to obtain, there is no coordinated interfacility approach to either impact research or new facility development, and the results of experiments are not widely available. The general inaccessibility of facility capabilities and of the impact data generated at these facilities has resulted in considerable duplication of effort, slowing the development of good models of debris impact damage.
Finding 3: Analytical models can be used to design spacecraft shielding and to predict impact damage for regimes that hypervelocity testing cannot simulate. Numerical simulations can be useful tools for predicting damage to spacecraft and determining the characteristics of breakup debris. Unless both of these methods are validated by comparison to experimental data, however, significant variability in predicted results can occur. When used together, hypervelocity testing and computer model