An experimental program using small-scale model representing variations of the MX missile trench basing concept (shallow buried tunnel) was conducted in two parts: (1) a study of the effects of trench expansion and trench-to-surface venting on the attenuation of an airblast in the trench and (2) testing of reinforced concrete spur structures (underground shelters) at the 600-psi location to determine door loads and to investigate spur structural response. The trench models were tested at 1/26-scale and the spur models were tested at 1/23-scale. In both parts of the program, loads were generated with an explosively driven shock tube. The trench models were 60-feet-long, 6-in.-diameter fiber-reinforced concrete ribbed tunnels. The spur models measured 6 inches high and 54 inches long. The walls were nominally 1-in.-thick and were reinforced with 1/16-in.-diameter steel wires. They were loaded with a 700-psi incident shock. Pressure gages measured internal pressure, LVDTs measured wall motion, and high-speed cameras photographed roof and soil motion. Expansion and venting of the trench reduced the blast pressure at the spur location by about a factor of 3 compared to measurements in a rigid tube. Likewise, expansion are venting of the spur greatly reduced reflected pressure.

The purpose of this research was to assess the applicability of geometric scaling at very small scale to study the response of buried reinforced concrete vertical shelter structures (silos) subjected to airblast loading. The approach was to build and test two 1/30-scale models and compare the response with those from corresponding 1/6-scale tests. One of the structures tested was designed to respond elastically, and the other was designed to response inelastically. The 1/30-scale and 1/6-scale models were built with as much geometric and material similitude as practical. Special fabrication techniques were developed for the 1/30-scale models. The walls of the 1/30-scale structures were 0.4 to 0.8 in. thick. The steel reinforcing wires were 0.020 to 0.045 in. diameter; the wires were deformed and heat-treated to match the properties of the 1/6-scale reinforcement. At both scales, sand was used for the back-filled soil. The design airblast was simulated with a high explosive simulation technique (HEST).

For the elastic structure, concrete surface strains measured in the 1/30-scale test and reinforcing steel strains measured in the 1/6-scale test showed that the direct loading wave, the reflections from the base and the closure, the base and closure flexure, interface friction, and soil resistance to punchdown were all reproduced accurately at 1/30-scale. For the inelastic structure, the responses agreed well up to the time when the loading wave arrived at about the middepth of the models, where the 1/6-scale structure failed prematurely because of a locally weak section of concrete. Concrete surface strains measured in the 1/30-scale test and reinforcing steel strains measured in the 1/6-scale test showed excellent agreement above this location. The 1/30-scale strains throughout the structure were also in excellent agreement with the predictions of independent numerical analyses.

The objective of this work was to determine critical airblast loads for a shallow buried reinforced concrete box-type structure. The damage level associated with a critical load was assumed to be total collapse of the roof of the structure. The approach was to build several 1/32-scale models of a generic structure design and to determine the critical loads in static and dynamic tests. In all the tests, the surrounding soil was dry sand and the ratio of roof span to soil cover was 2. In the static tests, the surface load was applied by pressurizing a fluid-filled neoprene bladder. The dynamic tests were conducted in a shock tube using sheet explosive to produce the airblast. The static failure mode was a shear failure at the roof supports; a discrete central hinge did not form. In the dynamic tests, the roof responded in the three-hinge mode. In static and dynamic tests of a modified structural design that had essentially no shear reinforcement, membrane action in the roof became significant.

A 1/32-scale airblast test of a shallow-buried reinforced concrete structure was conducted, and the results and costs were compared directly with those of a previous 1/4-scale test of similar design. The comparisons showed that the fidelity of the response of the smaller-scale structure was quite good, and that the cost of the smaller-scale test was substantially less than that of the larger scale test.

Twenty-five 1/11-scale turbine missile impact tests were performed to determine the threshold velocity that would produce perforation of reinforced concrete panels representing a variety of designs of current reactor containment structures. The missiles represented 1/11-scale fragments of a turbine hub that disintegrates during an overspeed accident. They weighed from 2.5 to 6 lb and were propelled at speeds from 130 to 700 ft/s by an explosively driven launcher. Damage to the panels ranged from penetration into about half the panel thickness to complete perforation of the panels. A few of the tests replicated full-scale tests; the resulting damage also replicated the full-scale damage. High-speed movies in the 1/11-scale tests provided a record of the penetration and perforation histories. These data were used to evaluate current design formulas being used to assess turbine risk.