Nasa_Pyros (Раздаточные материалы), страница 7

Theenergy delivered by these fragments is obtained by calculating 1/2 mv2, where m is the mass of eitherthe end or side of the cup. The impact patterns, created by these fragments, are obtained using theacrylic witness plates. Explosive transfer line performance has been uniform and reliable over the 30year history of their application.39Figure 35. Schematic diagram of test fixture to monitor fragment patterns and velocities from rigidexplosive transfer line end tips.The energy required to initiate an acceptor explosive is obtained by impacting acceptors with fragmentsof known velocities.

To determine the minimum threshold for initiation, donor tips were manufacturedwith smaller quantities and densities of explosive materials, as well as using different types of donorcup materials.7.3.4 - Recommended Tests for Explosive Severance/Fracture• For linear explosive severance/fracture, use tapered plate• Flexible linear shaped charge (FLSC)• Mild detonating fuse• Lockheed’s “Super*Zip” separation joint• Measure maximum fracture capability in each test• Test setup must simulate flight applications• Structure• Material properties• Explosive load• Explosive backup• Functional margin achieved by sizing explosive load, based on the ratio of either:40• the explosive flight load to the minimum explosive load required to sever structureor• The minimum structural thickness severed by the flight explosive load to the flight structuralthicknessReferences 30 and 39The use of tapered plates, as shown in figures 36 and 37 provides a way of determining the limit performance of each length of linear explosive for every test.

That is, the maximum thickness of the plate isset so that the linear explosive cannot fracture along its entire length. With this method, systemparameters, such as explosive load, sheath materials and separation distance between the explosive andthe plate to be severed can be optimized.

For this information to be applicable to the flight system, thetest setup must accurately simulate the flight configuration. The linear explosive test fixture in figure 36shows a method to measure the severance capability, as well as the energy delivered by a length of theexplosive (using the McDonnell Energy Sensor), and the velocity of explosive propagation down thelength of the explosive. Figure 37 shows the test setup for the Lockheed “Super*Zip” separation joint inwhich an explosively expanded tube fractures plates on both sides of the tube at the notched area (ligament).

For this investigation the thickness of the plate was tapered and the depth of the notches was heldconstant. The most influential parameter in achieving fracture was the thickness of the plate; the platemust bend at the bolt line to fail at the notched areas in tension.41Figure 36. Test fixture to evaluate the output of linear explosives.Figure 37. Lockheed Super*Zip separation joint tapered plate test configuration.427.3.5 - Recommended Tests for Demonstration of Structural Containment• Successful functioning shall not create a hazard• Venting of fluids/gases• Fragments/debris• Require fail-safe demonstrations• Lock-shut (fire, while preventing mechanisms from stroking)• Function with minimum containment structure• Increase explosive load• Reduce initial free volume in combustion chamber• Functional margin based on comparing• Flight structure to minimum structure that maintained integrity• Explosive load where rupture occurred to flight explosive load• Pressure where rupture occurred to flight pressureNote: It is not necessary to test devices to rupture failure if acceptable structural margin can bedemonstrated.References 20 and 2143Chapter 8 - FUNCTIONAL RELIABILITY• Device reliability achieved by• Understanding of device mechanism through development• Effects of variables• Functional limits• Energy delivery mechanisms• Effects of environments• Determining functional margins• Qualification testing/environmental survivability• Measure performance before and after environments• Compile further performance data to substantiate functional margin demonstration• Go/no-go testing of a small number of devices does not measure reliability• Reliability can be determined by statistical comparison of energy required to energy deliverableReferences 1, 4 and 40Functional reliability must be evaluated on each device.

Each device should be evaluated through threephases: development, determination of functional margins, and qualification. This should be followedby system integration and demonstration. The information collected throughout this effort can be analyzed to produce classical statistical reliability estimations. Every test firing should be designed to produce functional data that will contribute toward defining reliability.For each component, functional evaluations should be conducted to understand the device’s mechanisms and the effect of parameter variability. Tests should be conducted to functional limits. That is,tests should be conducted to determine where devices fail to function to evaluate parameters, such asstructural strength and minimal pyrotechnic energy input, and where structural failures (rupturing)occur.

The manner in which energy is delivered within and by the device should be determined. Themost influential environment affecting combustion is temperature. At cold temperatures, both ignitionsensitivity and burn rate decrease. This combination normally produces the lowest energy output. Athigh temperatures, the reverse occurs. Also, at high temperatures, structural strength can be reduced.The combination of higher combustion efficiency and lower structural strength can push devicesbeyond structural limits.Currently applied go/no-go tests do not provide information on functional performance, and contributevery little to reliability predictions. When a device successfully passes qualification testing, which typically requires 10 to 200 firings, the specification reliability is assumed to have been met. The reliability/confidence predictions currently in existence have been created by compiling the success of pastdevices that are remotely similar to a particular device under evaluation.

Unfortunately, this logic isflawed, since the primary assumption in large-sample statistical evaluation is that all devices in the sample are identical. Examples of devices that have achieved significant, large-sample statistical demonstration from actual firings are explosive transfer lines and the NASA Standard Initiator (NSI).A statistical reliability estimate can be obtained through the analysis of all the functional performancedata collected through development, qualification, integration and checkout of the device under evaluation. As shown in figure 38, comparing the distributions of energy required to perform a function to theenergy delivered will provide a statistical baseline.

Clearly, the greater the separation between these twodistributions the more assurance that the function will be accomplished. the 3.89 standard deviationrefers to a typical distribution where 99.9% of the data is included. However, if the two distributions44overlap, failures can occur. Statistical tables, as described in reference 40, predict reliability, based onsmall-sample distributions.Figure 38. Graphic representation of statistical design margin, comparing normal distributions ofenergy required to perform a function to energy supplied by a cartridge.• System reliability enhanced by redundancy• Use of parallel, independent paths to perform a single function• Prevent single-point failures• Degree of redundancy depends on system tradeoffs• depends on criticality of function• achieved by design approach• Use of redundancy to assure success introduces contradictions• Increases complexity of systems• Increases chances of inadvertent functioning• Does not alleviate need for quality assembly or functional margin demonstrations of components andinterfacesReference 1Redundancy is defined as accomplishing a function through two completely independent paths.

Inexample A, figure 39, two electrical firing systems each give a command to Single Bridgewire ApolloStandard Initiators (SBASI, the predecessor to the NSI), which in turn ignites the cartridge energysources to one of two devices, such as pin pullers, either of which releases an interface to accomplish adesired function.