 Imagine spending loads of time and money on building a nuclear reactor only to intentionally destroy it. Well imagine no more because of course the US has been there and got the t-shirt. In 1964 and 1965 two nuclear reactors were prepared to be deliberately pushed beyond their limits to destruction. The Snapdragon series of tests involved testing a reactor for use in space travel and as such the experiments would be a vital stepping stone. The test in 1964 would involve submerging a reactor in water in order to simulate an aborted launch and subsequent ditching at sea. The test in 1965 would push the reactor out in the open simulating a crash in the desert. Today we're looking at the Snapdragon reactor tests and although it didn't injure anyone or cause massive destruction purposely pushing a unit beyond its limits is interesting all the same. And as such I'm going to rate it here number two on the plainly difficult disaster scale. NASA wanted to be the first to have a nuclear reactor in space and as part of its snapshot program. However our story starts in 1950s and the need for high power consumption spy satellites. In 1951 the US Atomic Energy Commission requested nuclear power plant studies to see if a reactor could be sent into space. By 1952 it was thought feasible although a series of test reactors would need to be built and destruction tested. Much like the HTRE project the AEC contracted two separate companies to work on two different concepts. Both sides of the project fell under the SNAP or systems for nuclear auxiliary power name. The first numbered in odd numbers starting with SNAP 1 employed a simple by nuclear standards RTG design using decay heat from a radioisotope to generate electricity. The concept was much the same as the RTGs that caused the issue in Lear using a thermobattery or thermocouple to transfer the heat to electrical power. This side of the project was contracted out to the Martin company and would continue on to SNAP 27 which made it all the way to the moon. Although interesting in concept it's the even numbered SNAP test units which are our focus today and they were much more complex in design as they were a fully fledged nuclear reactor. This part of the project was contracted to Atomics International Division of North American Aviation. The first reactors developed by AI was SNAP 2 and was developed to SNAP 8 and then finally 10A which was the first space ready type built by the company. The SNAP 10A consisted of a 15.6 inch long by 8.8 inch diameter tube holding 37 fuel elements. The reactor once in space was expected to produce a minimum of 500 watts of electricity for up to one year or even longer. Around the outside the core fixed beryllium blocks were employed to reflect neutrons back to the core. What's interesting about the design was that control wards were not employed. Instead additional beryllium reflectors were mounted to four drums which could be rotated towards and away from the reactor. With the reflector side facing the core allowing a chain reaction and reflector facing outward shutting down the reaction. Two of the drums were used to step up the power during startup. Upon launching into space a startup command was sent to the reactor. A pin was released by actuators and two control drums were immediately inserted and the fine control drums would then be stepped in. 50 seconds after the startup command the two fine control drums would take their first step and would continue to step in every 150 seconds until fully inserted. Approximately 7 hours after startup criticality was reached and would take around 72 hours for the fine control drums to reach the end of their steps. After which the reactor ran at a steady state of power without dynamic control. The reflectors were held in place by a retaining band and could buy an explosive bolt. If needed the bolt could be blown removing the reflectors from the core shutting it down. The reactor had 4752.5 grams of uranium-235 as uranium zirconium hydride fuel which were contained within the 37 fuel rods. Sodium potassium alloy was used as a coolant for the reactor design and was circulated through the core and the thermoelectric converters by a liquid metal pump. The converters would thus provide power for the satellites. Before NASA would allow the 10a for launch a number of destruction tests would be needed to be conducted to simulate the potential situations the reactor could be subjected to. For example being dropped into the sea or crashed into the desert. Before any radiological tests several unfueled 10a reactors and its components were stress tested in every imaginable way. This leads us on to the snaptran experience. The tran stood for transient. Snaptran 1 and 2 experiments were intended to investigate nuclear excursions occurring in an atmospheric environment during assembly or launch caused by unwanted movement of the reflectors. Second of which were nuclear excursions resulting from immersion of the reactor system in water or wet earth. This would become a snaptran 3 experiment. The test would be undertaken by Philips Petronium at the initial engineering test IET facility of the National Reactor Testing Station at Idaho National Laboratory. The first of the experiments unsurprisingly named snaptran 1 was a non-destructive test and used a modified 10a to subject the reactor to sudden changes in reactivity. The reactor assembly was mounted on a four-track test dolly that was constructed from two railway carriages. The initial non-destructive tests took place under a corrugated structure that was also mounted on the tracks. All operations were conducted by a nearby underground control room. For the test to be undertaken the reactor had its drum control modified splitting the four drums into two pairs of two. The first pair were the impulse drums and were intended to create sudden impulses of power by quickly turning the reflector side to the reactor core. The second pair were stepper drums and were intended to gradually increase power to test a much slower rise. The experiments made use of both drums and their respective form of movement to collect basic kinetics data for a wide range of transient conditions. The test began in May 1963 and according to a scheduling report from before the experiments would run until 1965. However snaptran 2 wouldn't be conducted until 1966. The first destructive test was the snaptran 3 and would simulate a launcher bought into the sea. There was a problem with this test as it wouldn't yield usable data by just dumping the reactor in the body of water and instead the designers thought up an ingenious way to immerse the unit. The reactor was permanently mounted in a fixed position and provision was made for water to be rapidly placed around the test unit before the reactor was allowed to reach criticality. Earlier non-nuclear tests showed that upon impact with the water it was likely that the beryllium reflectors and other external components would be separated from the unit. Because of this it was decided that only the reactor core would be destruction tested in water. The reactor vessel fully fuelled up would be mounted on a pedestal inside a 14 feet diameter open tank rigidly mounted to two railway flat carriages similar to the one used in the snaptran 1. The tank was surrounded by concrete one foot thick on the sides and two foot thick on the bottom. Due to not having the reflector drums installed a different form of power control was needed and this was provided by a poison sleeve actuated by a drive mechanism which could be fully withdrawn either slowly or rapidly as well as providing a scram facility. Intermentation was installed nearby as well as several miles away from the test site to gain information on the amount of radioactivity released into the environment. Before the destruction test a number of calibration tests were undertaken and this used a calorometer tank which can be filled with water after which the poison sleeve can be slowly raised until criticality is attained. During the calibration test the reactor operated at low power levels. For the destructive test the calorometer tank was removed and on the 1st of April 1964 operators at the NRTS prepared to purposely destroy a reactor. The sleeve removal mechanism was fired removing all poison from the reactor. The reactor experienced a sudden power increase to $3.60 above critical. Within milliseconds a Cherenkov glow masked the reactor and after 20 to 30 milliseconds the reactor experienced peak power. The power increase caused the reactor to blow apart. All 37 fuel elements and 6 beryllium reflectors were destroyed and a 40 foot high plume of water was blown out the top of the tank. The reactor was well and truly destroyed with debris scattered all around the tank. It was calculated that the reactor experienced a peak power transient of around 13,000 megawatts. All the parts and debris of the reactor were loaded onto a flatbed railway carriage and were shipped off to a hot shop for dissection and analysis. A radiation detector 20 feet away recorded 30 Runkins per hour at the time of the peak power excursion. Decaying to 0.055 120 minutes after the test and 24 hours later was at 2 milli Runkins per hour. Tracer mounts to the following isotopes were found, iodine 131, 132 and 133 and strontium 89 and 90. Right, that leads us on to Snapchat 2 and the beryllium reflected destruction test. This was to be conducted on 11th of January 1966. After the initial non-destructive test yielded satisfactory results the next step was to ramp up the stress on the 10A and undertake reflector criticality destruction tests. The purpose of Snapchat 2 was to find out total energy release, mode of shutdown and a degree efficient product released during the destruction of a 10A reactor in open air. The reflector drums returned for this test for control of the reactor. For the test a new similar core to the Snapchat 1 was used although some minor modifications were undertaken including a minor change to fuel weight and alterations to the way the drums were shimmed. Although the main change was the removal of the reactor's knack coolant this was done to allow high-speed cameras to observe the reactor core more easily. For this test the corrugated metal structure was wheeled away from the reactor to monitor the spread of radioactivity and the grid was set up to track any efficient product releases. Film badges, fallout plates, pocket dosimeters and nuclear accident dosimeters were used for monitoring. On January 11th 1966 after early morning preparations the reactor test pad and downwind areas were cleared of personnel. At 9.51am the test was initiated as all four brilliant sides were inserted facing the reactor core achieving peak reactivity reaching 74,000 megawatts but only for a few milliseconds before total destruction of the core. A visible smoke filled cloud arose immediately as the fuel burned and the reactor core blew itself apart around 4 seconds post peak power. 300 meters downwind from the reactor a rate of 2.7 runkins per hour was recorded. At 6 meters from the epicenter radiation was measured at 18 runkins per hour two hours post event eventually dropping off to 0.2 runkins an hour 48 hours later. Fuel pieces were found up to 100 meters away and debris from the beryllium reflector littered the test ground. In this before and after view of the reactor obviously not much was recoverable but in a test like this I assume that was considered a success. The many faces of strontium 90 and iodine 131 were found post test. Again like before all the reactor pieces were collected and sent to a hot shop for dissection and on the holder test went well. The data from all three tests were vital in being able to predict the environmental and radiological consequences of the reactor being crashed onto our planet. The snaptran 2 test seemed a little redundant however as the snapshot program had already launched the 10a reactor into space by April 1965. The actual reactor was successfully delivered to space marking the first to ever reach such a height. Although the experiment sadly failed 43 days later when a voltage regulator failed shutting down the reactor. The reactor was then left on a 1300km earth orbit after it shut down and will be up there for an expected duration of 4000 years. I hope you enjoyed the video. If you'd like to support the channel financially you can on Patreon from $1 per creation and that gets you access to votes and early access to future videos. I have YouTube membership as well from 99 pins per month and that gets you also early access to videos. Check me out on Twitter and also if you're on to wear my merch you can purchase it at my Teespring store. All that's left to say is thank you for watching.