Aerospace Safety Results from a ROVER/NERVA Post-Operational Destruct Test (and memo response)

Page 1

Re: Your Paper Entitled "Aerospace Safety Results From This material was prepared for presentation and possible publication as a hand-cut at the meeting of the American Nuclear Society Winter Meeting, November 17, 1965 at Washingt This letter confirms our earlier notification of release. Th Classification and Invention Reports Divisions concur in thbot release of this material for the purpose stated. Because the material and the intended audience are primarily technica approval of the Public Relations Organization is not needed. It is required that all manuscripts and abstracts have the following statement at the bottom of Page One: This work is supported by the United States Atomic Energy Comnission. Distribution: (a) copy of paper (b) copy of transmittal memorandum

Page 2

Dticreerrioe Pres reerralc Sandia Corporation, Arauperque, %J.!, spec.ial eistrnuti.cn. These eata are a satizfactc, ry sour=tax' for the

Page 3

AEROSPACE SAFETY RESULTS FROM A ROM/IERVA Aerospace Nuclear Safety Division Sandia Corporation, Albuquerque, N.M. Prior to the use of nuclear material in space, ation of the safety criteria must be completed, The safety analysis deo.• pends on the method cf destruction proposed for the vehicle carryingthe For the ROVER/NERVA propulsion engine, a suitable nonnuclear:peep. Operational explosive destruct system has been developcd.by Aberdeen; Proving Ground and Picatinny Arsenal, but the character of the resulting: debris has net been fully defined, The characterization of'dibrieii source term for a computer analysis from which thereeulting dispersion,. fallout, and the Otimate safety can be established. To acquire this input information which includes three diMensionai: debris distribution, debris size and mass, and debris velocity, a full ecale mockup of the ROVER/NERVA nuclear propulsion engine uas Am.-*royed The destruction of the mockup ROVER/NERVA propulsion engine w33 requeztad by the joint tom. /NASA Space Nuclear Propulsion Office (SER0). de truetien be collected.

Page 4

The collection of explosive destruct data and analysis of thatelata was jointly done by the Army's Aberdeen Proving Ground and .Sandia Corpora-- tion. These two facilities independently instrumented and collected as much data as poesible to fulfill the safety re4piremenes. Figure 1 lists those data requirements considered to be fundaeentel, or that data which can be collected directly with on site instrumentation. Figure 2 lists those data rceuirements which are a direct result of fundamtntal data reduction and presented in usable format. Figure 3 lists those data requirements which require additional mathematical analysis to : remove the effects of gravity end the atmosphere on the debris pattern and velocity resulting from the explosive destruct test. The previous three Figures outline tbe data requirements and Figure 4 is a phetogreph of the mockup ROVERAMVA space propulsion engine destroyed with chemical eeplosives on June 22, 1965, at Aberdeen Proving Ground. space engine contained a simulated uranium core, control drums, reflector seements, core support plate, shield, and the outer; pressure veseel. That is, all components which were considered important to the veseel reaction to exploeive destruct were simulated. t:imulated entry holes we're placed in the vessel and four statically emplaced 105 mm special explosive charges were used for the destruct teat. .rte specs engine was surrounded by instrumentation as shoun schematically In Figure 5. This instrumentation is also shown photegrae!tically

Page 5

Duplicate instrumentation techniques . Were deed throughout to provide. a Mee confidence that the required data would be obtained. Description of Instrumentation The purpose of the air sampling program was to - collect atmospheric sample& in the vicinity of ground zero after the NE VA reactor had been destroyed with 111.17 pounds of high explosive, to determine the concentretion of graphite dust particlee in the resulting cloud and to ascertain the shape and size distribution of tha particles that remained airborne. The atmoseheric =pies were collected by Ream of midget impingers suspended from overhead cables. The cables were held about SO feat above and around greund zero. Three midget iopirgers were secured to each of 8 drop lince from the overhead cables and were 'located ea that samplers were fixed at 30 feet, 50 feet, and 70 feet above the ground. Each line of impingers was located 'Ce feet from ground zero, and the resulting arrey of 2e eemplers formed a cylinder 200 feet in diameter and 70 fete high* Tb4s array of samplers provided 360 ° eevereee and allowed a representative sample to be taken

Page 6

The blast pressures developed during the destruetton of the mockup space engine were recorded bn an Ampex CP 100 Magnetic Tape Recorder. Susquehanna ST-2 pressure trans' declare were connected through 1,000 feet of Microdot cable to an Endevco Model 2641M1 Charge Amplifier which provided' the input to the tape recorder.. Twelve transducers were mounted in the blast area, each 6 feet above ground level. Four transducers were located 20 feet from ground zero - one in a jet, one at '+ between jets 0 and two 5 on either side of a second jet. Four additional transducers were located at 30 feet' and also at 40 feet in the sate way as described above. To meacure the velocity of the particles resulting from the destruction of the test vessel, coneuctins, glzsa roes were employed. These rods were constructed of 3/16" x JiY gins tubing painted uith a conductive paint which pildcd a path of electrical continuity hatw,a.sn each end. Meacurements revealed that each 8•inch rod possessed a reeistence of approximately 2 ohms. Electrical connections l : ere made to the rods through th., use of fuse clips att3chod to each end of the rod.

Page 7

The glass rods were positioned above the simulated. nozzle, be/ow the sim►iated dome, adjacent to the pressure vessel wall and at 10 feet, 20 feet, 30 feat, 40 feet, 50 feet, 60 feet, and 70 feet along the west jet and also at 10 feet, 30 feet, and 50 feet along the south jet. •The glass rods were positioned to be broken Ly the. moving debris, either the metal components or the core material, which is deployed in the 5 major jet areas. 15 Rotating Single Dine The single disc was designed (a) to measure the time cf arrival of particles from which the average. velocity could be calculated, and (b) to catch and preserve the particles that reached it. The unit toil eists of a flat dies of 159 lbs/ft 3 polyetyrene foam 3" thick and /5" in diameter. The dice is driven at a constant rotational velocity of 1725 rpm by a 3/4, hp 220 volt AC single phase electric motor. The retating dice and motor are be”scd in a fabricated aluminum tha front fare of e ,hich contains a 4-3/4" x i" atetionary slit which faces the explosion center and allows a debris •ample to coter the polietyrene foem disc. A dual nark

Page 8

source, one from a modified transistorized.autemobi/e ignition system and one from s high energy capacitive. discharge system were mounted to provide a mark at detonation (zero time) on heat sensitive recorder tape attached to the periphery of the disco As the orplosien debris roaches the instrument, a line of particles are permitted to'enter•thraugh: the stationary slit and embed into the rotating polystyrene foam disco With the known disc angular velocity and the angle the particlos• (debris) mate with respect to the zero time mark, the arrival time by measurement; therefore, the average velocity (Va) is determined by the eq:lation: Va D/T6 Figure 7 is a schematic representation= of this engle diec particle collector and ve/eniCy devlae,. ; Thu rotating drum was designed to perform the Its principle and operation are the same except that the particles enter a l" r 18" slot and strike the dr= normel to the drum surface and axis. Further, the

Page 9

The drum is of polystyrene foam 18" long by 21" in diameter. Total weight of the drum assembly is appresimately 210 pounds. The drum and drive are housed in an aluminum skin and stringer structure stressed for 20 psi overpressure. Figure 8 is a schematic representation of the rotating drum particle collector and velocity device. 3. Rotating Twin Disc The twin disc was designed to provide: (a) average velocity, which is calculated by measuring time of arrival of particles, (b) final velocity, which is calculated by determining the time interval between particle penetration of the front disc and easing the average velocity and the final velocity found in (a) end (b), ard (d) particles for lator The twin disc consists of two spoked iliac arc4 8 inches apart on a comon driver shaft. raper in glued over each disc. The rear dies is hacked with fncm which capeurca anti preserves the Dual spark 3OUTCS3 are locAtcd inside elle housing along the periphery of the disc tkeembIy and are

Page 10

arranged to mar% the rotating assembly when energized, assembly, motor, and tine mark systems are housed in en aluminum skin and stringer structure stressed to withstand 20 poi overpressure. Figure 9 is a schematic representation of cuci rotating twin disc particle Collector and irelotitY As the explosion debris reaches the instrument, a line of particles is permitted to enter through stationary slit on the front cover. The particles pass through the paper on the front and rear discs and then cubed in the foam. The particle time oZ arrival can be calculated by knowing speed and by measuring the particle angular displace= =ant with respect to the zero time mark. The ayerage. velocity can be calculated by using the time of arrival and the distance the particles traveled from The final velocity is calculated by using the measured displacement of the particle penetration imprint on thu front disc paper with respect to the imrrint un the rear disc paper, the distance traveled 1:,Itucen the two disc papers (8 inches), and the

Page 11

The initial velocity can be computed using the calculated average velocity and final velocity in the Vi = initial velocity Each particle can be recovered from the forms and its size and shape correlated with its three velocities. E. Fixed Foam Particle Cellentors Tho polystyrene foam p/qptic placed in front of the rotating foam collectors hadpa threefold purpose: (a) collect (b) 0^t energy absorption media to protect the rotating loan collectors from large missiles!, and (c) show the relative o...7.antity of de3ris in tha sets and in between jetsa Figure 4 shows the fixed foam particle collectors arranged behind the mochup space propulsion engine,. The data estainci Erom this test were good and the test objectives sere aczamplirhed. Figure 10 east::: some of the deettliction inflicted on the inctrunentatioa; boweverp data tr,sre obtained and enough of the instruImad to give a good clata sample.

Page 12

Figures 11 and 12 show the resulting debris pattern when using four • explosive dharges. Figure 11 is a side view showing the two horizontal jets emanating from ground zero and a vertical jet above ground zero. Figure 12 is a photograph taken from a helicopter and chows the four strong jets originating at 450 between explosive charges. The sample of airborne core materieiobtained frotg. the midget iMpingers was quite small but representative of the material contained within that portion of the debris cloud. Because of the jetting action created by the four explosive :. charges and the variable wind at the time of firing, the percentage cf core material can only be estimated. Indications were that approximately 2 or 3 percent was contained the ibne ch The rare material that was collected debris i bean .5 and 3.5 micrcns, about 19% is between 3.5 and 10 microns, and the remainder in between IO and 20 microna. Further, the cezplcs collected contained no r.zeaPurabl- Guentity of uranium. 1%,:o techniques were Ear uranium analyses - Gamma counting and fluorometric. Both anatyzes showed no uranium present. There iL4 a vary slight variaticn of prenaurc_k when m:;asurad in a ;et or between jets but this is 8ms11. ftessurcs vare 23 psi at 20 feet t 8 psi at 30 fact,c.nd

Page 13

4,2 psi at 40 feet. These pressures are about 60% of that pressure expected ft-cm an uncontained explosion of the The velocities measured by the glass rods, rotating foaM velocity devices, and from the photographic film are The glass rod velocity data and the rotating foam • vela:.:ity data are acquired before the ewers is able to ace through the combustion cloud, However, the camera is able to provide data for a longer time after detonation, The metal debris from the explosive destruct of a mclo4p space propulstjn engine is generally distributed radiclly folvtke pressure vessel ruin, the simulatesi re€loctor segmcnts, and the s:;=lated central alrualn The notate it projected epuazd me.a0, er ef the reactor is distributed wrong the five jots. as.ch pertlele site ran recovered'' from the fiy:ar: foam, as

Page 14

well as the p!.0t of uranium found within each particle size range. "Observe the concentration of uranium at 75to' 200 microns. This is the appronimate size of the uranium particles which are imbedded in the graphite matrix.. Figure 16 is a plot of the accumulated weight percent of each particle size range, as well as the accumulated weight percent of the uranium recovered from each particle data obtained with respect to debris velocity, ai;;:e distribution,. acd spacial distribution agrees very closely with the data obtained by the Thanc Le := are acceptable aa the source tern for the computer enalysis - of the MVERINEVA Program Safety. The techniques used for data collection performed satisfactorily'and the data obtained by both Aberdeen Proving Ground and Sandia Corporation