Nasa_Pyros (1049401), страница 11
Текст из файла (страница 11)
A NASA Langley Research Center project,the Halogen Occultation Experiment (HALOE), used residual Viking pin pullers. The Jet PropulsionLaboratory (JPL) planned to go to the manufacturer of these pin pullers to produce another lot, conducta delta- qualification for their unique requirements, and fly it on the Magellan spacecraft. However,early in their evaluation effort, a unit only stroked half the required distance. Following a JPL analysisand resolution, another unit failed to stroke at all.
This design was then abandoned by JPL and anotherpreviously qualified pin puller was selected and used. Meanwhile, NASA Langley Research Center hadmade a commitment to use this device and elected to conduct a failure investigation.As shown in figure 42, the energy sources were the Viking Standard Initiators (VSIs), which are virtually identical to the NASA Standard Initiator (NSI). Firing either or both units would accomplish thefunction.
The outputs of the VSIs each pressurized a blind port, that has a 0.100-inch diameter orifice tovent the gas behind the piston. An 80-pound strength shear pin prevented premature motion. An energyabsorbing cup was crushed by the excess energy achieved by the piston/pin at the end of the stroke.The Viking development effort relied on monitoring the peak pressure produced in the pin puller toobtain an understanding of functional margin. A transducer was installed in the port opposite to the VSIthat was fired. It was found through off-loading of the pyrotechnic charge in the VSI that the pin pullerwould function with only half the normal peak pressure. Accordingly, the project assumed a functionalmargin of 2, or twice the capability that was necessary.
Furthermore, Viking never experienced a failure; more than 150 consecutive, successful go/no-go tests, including a rigorous environmental qualification program and a number of subsystem functional demonstrations were accomplished by the time thespacecraft flew. How could a “fully-qualified” device with such a pedigree fail to function 20 yearslater?60Figure 42. Cross sectional view of Viking pin puller.The Langley failure investigation revealed that peak pressure, as the only test parameter, meant virtuallynothing.
It was also found that the o-ring seals were inadequate: (1) the chemical chromate coating(Alodine) was wiped from the interior of the piston bore and adhered to the o-ring, preventing a seal,and (2) the molybdenum disulfide/graphite dry lubricant was wiped from the pin and piled up on theupstream pressure side of the pin o-rings and prevented a seal. The net effect was to decrease the pressures achieved in the working volume, and ultimately, to reduce the combustion efficiency and toquench the combustion of the cartridge mix. The bottom of the VSI port occasionally deformed to gripthe piston.The resolution was to change the pin puller’s housing material and the dry lubricant on the pin.
A steelbody was flown, but hard-anodized aluminum performed as well under additional testing. The dry lubricant was an electrolitically deposited nickel/Teflon coating. The energy required to function the pinpuller was obtained by dropping a small mass onto the pin; drop height, multiplied by the drop weight,produced a value of inch-pounds. The energy absorbing capability of the cup was calibrated by increasing the drop height.
Thus, after each firing, disassembling the pin puller and measuring the cup crushprovided an energy delivery value for the cartridge. These data are summarized in figure 43. A value of25 inch-pounds was determined to stroke the piston/pin and deform the energy absorbing cup to preventrebound. In a sample of only 5 pin puller functional tests, conducted using actual spacecraft structure,the average value of energy delivered by the cartridge was 165 with a standard deviation (sigma) of 22inch-pounds.
Assuming a normal probabilistic distribution, statistical tables indicated that the probability of failure for both pin pullers on the mission was equal to or less than 0.4%. That is, the probabilityfor success of both pin pullers was equal to or greater than 99.6%.61Figure 43. Statistical presentation of functional margin for redesigned HALOE pin puller.17.2 - Failure Investigation of Lockheed Super*Zip Separation Joint• Flew for 20 years as stage and shroud separation• Failure occurred in ground test at cold temperature• Contributors to failure• Structural material changed from fracture sensitive to fracture resistant• Did not adequately control thickness of structural material• Cold temperature had no effect on performance• Functional margin:• Flight explosive load 27% greater than needed• Capability of fracture 71% greater thickness than needed• Confinement margin was demonstrated by• Determining mechanism that caused tube rupture• Determined explosive load that induced tube rupture• Compared to flight load62References 2 and 39The Lockheed Super*Zip separation joint was developed and qualified in the late 1960’s for payloadshrouds (opening the shroud longitudinally and across the nose) and for payload release (circumferential, cylindrical severance, as shown in figure 44).
In 1984, during a cold-temperature ground-test demonstration of the Shuttle/Centaur system, the joint failed to separate around the entire circumference.This is one of the worst possible failure modes, in that a partial separation would prevent payloadrelease, requiring astronaut extravehicular activity to dump the payload. The Shuttle cannot land withthe partially released payload on-board. Landing loads might cause the remainder of the joint to fail,dropping the payload into the cargo bay, and destroying the Orbiter.The configurations of this separation joint are shown in figure 45. The principle of operation is theexplosive expansion of a flattened tube, which induces a tensile load in the material in the two sideplates between the notched areas and the tube to achieve separation.
The explosive cord, on detonating,transfers its energy through the rubber extrusion and into the tube. All products of the explosion arecontained by the tube. The material that was initially selected to be severed was fracture-sensitive aluminum, 7075- T6. The first question is, “Why are the three joints different?” One joint (Galileo) has oneexplosive cord and two (Centaur and IUS) have two cords. Firing either one or both cords achieves separation, but firing both cords can cause tube rupture. The first joint has a reduction in thickness at thebolt lines in the side plates, and the other two do not. The first joint has a material thickness at thenotches of 0.025 inch, while the other two have a thickness of 0.042 inch. Different fasteners are used.The flanges interfacing the two halves of the system to be separated are different to accommodate structural designs selected.
The Galileo design was selected for its lighter weight.63Figure 44. Shuttle/Centaur deployment system, using the Lockheed Super*Zip separation ring.Figure 45. Radial cross sectional views of three types of Super*Zip separation joints, and the programsto which they were applied.64Parameters within this system, a portion of which are shown in figure 46, were evaluated, and theireffects on functionality were quantified.
It was learned that the fracture mechanism was the following:1) the detonation of the explosive cord caused zones at the notches (ligaments) to be “bruised” orpreweakened by damaging grain boundaries, 2) the expansion of the tube forced the doublers to bend,hinging inboard of the line of fasteners, and an explosive impulse on the major axis of the tube created atensile load in the doubler to 3) structurally fail the ligament. The key point is the bending of the doubler, which is determined by the cube of the plate thickness. A decision was made by the Shuttle/Centaur Project Office to anneal the previously qualified 7075-T6 aluminum to a 7075-T73 condition toavoid concerns about corrosion-resistant properties of the T6 material.
A short panel was made up withthe T73 material, test-fired successfully and declared acceptable. Unfortunately, the functional marginof the system had been reduced to nearly zero. That is, an examination of the doublers in the failed testrevealed that thicknesses to 0.085 inch fractured successfully, and thicknesses above 0.086 experiencedseparation failures! Thus, while fracture properties of the material was the most important variable, aclose second was doubler thickness. The evaluation parameter used throughout the experimental effortto judge and compare performance was the doubler thickness.Figure 46. Identification of a portion of the parameters evaluated in the Super*Zip separaion joint.Using this doubler thickness as a performance parameter, a tapered doubler plate (figure 37) wasmachined to permit the evaluation of a particular variable within a length of 8 inches.
That is, as thedoubler thickness increased, it became stiffer to resist fracture. The doubler thickness was varied from0.065 to 0.123 inch to prevent total fracture within the limits of the variables evaluated, but allowedmaximum severance to be measured in each firing. Figure 47 shows the results of several variables,comparing doubler thickness for successful fracture to explosive load. The 7075-T6 doubler materialeasily produced the highest performance in the dual-cord flight configuration. The top curve indicatesthat a single, on-center cord is more efficient than the flight configuration at explosive loads to justunder 11 grains per foot.
![](https://s.studizba.com/z.php?f=/templates/si/i/noavatar.png&w=100&h=100&t=1"){kind=avatar}
