《DD-IEC-PAS-62396-5-2007.pdf》由会员分享，可在线阅读，更多相关《DD-IEC-PAS-62396-5-2007.pdf（18页珍藏版）》请在三一文库上搜索。
1、DRAFT FOR DEVELOPMENT DD IEC/PAS 62396-5:2007 Process management for avionics Atmospheric radiation effects Part 5: Guidelines for assessing thermal neutron fluxes and effects in avionics systems ICS 31.020; 49.060 ? Licensed Copy: London South Bank University, London South Bank University, Thu Dec
2、20 02:50:17 GMT+00:00 2007, Uncontrolled Copy, (c) BSI DD IEC/PAS 62396-5:2007 This Draft for Development was published under the authority of the Standards Policy and Strategy Committee on 30 November 2007 BSI 2007 ISBN 978 0 580 56737 7 National foreword This Draft for Development is the UK implem
3、entation of IEC/PAS 62396-5:2007. This publication is not to be regarded as a British Standard. It is being issued in the Draft for Development series of publications and is of a provisional nature. It should be applied on this provisional basis, so that information and experience of its practical a
4、pplication can be obtained. A PAS is a Technical Specification not fulfilling the requirements for a standard, but made available to the public and established in an organization operating under a given procedure. A review of this Draft for Development will be carried out not later than 3 years afte
5、r its publication. Notification of the start of the review period, with a request for the submission of comments from users of this Draft for Development, will be made in an announcement in the appropriate issue of Update Standards. According to the replies received, the responsible BSI Committee wi
6、ll judge whether the validity of the PAS should be extended for a further three years or what other action should be taken and pass their comments on to the relevant international committee. Observations which it is felt should receive attention before the official call for comments will be welcomed
7、. These should be sent to the Secretary of the responsible BSI Technical Committee at British Standards House, 389 Chiswick High Road, London W4 4AL. The UK participation in its preparation was entrusted to Technical Committee GEL/107, Process management for avionics. A list of organizations represe
8、nted on this committee can be obtained on request to its secretary. This publication does not purport to include all the necessary provisions of a contract. Users are responsible for its correct application. Amendments issued since publication Amd. No. DateComments Licensed Copy: London South Bank U
9、niversity, London South Bank University, Thu Dec 20 02:50:17 GMT+00:00 2007, Uncontrolled Copy, (c) BSI IEC/PAS 62396-5 Edition 1.0 2007-09 PUBLICLY AVAILABLE SPECIFICATION PRE-STANDARD Process management for avionics Atmospheric radiation effects Part 5: Guidelines for assessing thermal neutron flu
10、xes and effects in avionics systems DD IEC/PAS 62396-5:2007 Licensed Copy: London South Bank University, London South Bank University, Thu Dec 20 02:50:17 GMT+00:00 2007, Uncontrolled Copy, (c) BSI CONTENTS 1 General 3 2 Thermal neutron flux inside an airliner4 2.1 Definition of thermal neutron .4 2
11、.2 Overview.4 2.3 Background on aircraft measurements.5 2.4 Calculational approach6 2.5 Processing of in-flight neutron flux data.6 3 Thermal neutron SEU cross sections9 3.1 Overview of the issue9 3.2 Mechanism involved10 3.3 Thermal neutron SEU cross sections and Ratio-211 4 Recommendation for devi
12、ces in avionics at present time 13 Bibliography14 Figure 1 Atmospheric neutron spectra measured in four aircraft.7 Table 1 Tabulation of the various atmospheric neutron measurements used 6 Table 2 Comparison of thermal and high energy neutron fluxes and their ratios.8 Table 3 SRAM SEU cross sections
13、 induced by thermal and high energy neutrons12 DD IEC/PAS 62396-5:2007 2 Licensed Copy: London South Bank University, London South Bank University, Thu Dec 20 02:50:17 GMT+00:00 2007, Uncontrolled Copy, (c) BSI PROCESS MANAGEMENT FOR AVIONICS ATMOSPHERIC RADIATION EFFECTS Part 5: Guidelines for asse
14、ssing thermal neutron fluxes and effects in avionics systems 1 General The purpose of this PAS is to provide a more precise definition of the threat that thermal neutrons pose to avionics as a second mechanism for inducing single event upset (SEU) in microelectronics. There are two main points that
15、will be addressed in this PAS: 1) a detailed evaluation of the existing literature on measurements of the thermal flux inside of airliners and 2) an enhanced compilation of the thermal neutron SEU cross section in currently available SRAM devices (more than 20 different devices). The net result of t
16、he reviews of these two different sets of data will be two ratios that we consider to be very important for leading to the ultimate objective of how large a threat is the SEU rate from thermal neutrons compared to the SEU threat from the high energy neutrons (E 10 MeV). The threat from the high ener
17、gy neutrons has been dealt with extensively in the literature and has been addressed by two standards (21 in avionics and 1 in microelectronics on the ground). The two ratios that this PAS considers to be so important are: 1) the ratio of the thermal neutron flux inside an airliner relative to the f
18、lux of high energy ( 10 MeV) neutrons inside the airliner and 2) the ratio of the SEU cross section due to thermal neutrons relative to that due to high energy neutrons. These ratios are considered to be important because with them, once we know what the SEU rates are from the high energy neutrons f
19、or an avionics box, a topic which has been dealt with extensively, such as 1, then the additional SEU rate due to thermal neutrons can be obtained with these ratios. Thus, given the SEU rate from high energy neutrons, multiplying this by the two ratios gives the SEU rate from the thermal neutrons. T
20、he total SEU rate will be the combination of the SEU rates from both the high energy and thermal neutrons. The process for calculating the SEU rate from the thermal neutrons is shown in the following set of equations, (1) to (5). SEU Rate (Hi E, Upset/devh) Hi (neutron flux = 6000 n/cm2hr) (Hi E, SE
21、U X-Sctn. cm2/dev) (1) SEU Rate (thermal neutron, Upset/devh) therm Hi (neutron flux)(therm SEU X-Sctn.) SEU Rate (Hi E) (neutron flux)(Hi E SEU X-Sctn.) = (2) Ratio-1 thermal Hi (neutron flux) (neutron flux) = (3) Ratio-2 (therm SEU Cross Section) (Hi E SEU Cross Section) (4) 1 Numbers in square br
22、ackets refer to the bibliography. DD IEC/PAS 62396-5:2007 3 Licensed Copy: London South Bank University, London South Bank University, Thu Dec 20 02:50:17 GMT+00:00 2007, Uncontrolled Copy, (c) BSI SEU Rate (thermal neutron, Upset/devh) SEU Rate (Hi E neutron Upset/devh) Ratio-1 Ratio-2 (5) The obje
23、ctive of this PAS is to provide values of Ratio-1, the ratio of the thermal to high energy neutron flux within an airplane, and of Ratio-2, the ratio of the SEU cross section due to thermal neutrons relative to that due to high energy neutrons. We believe that Ratio-1 should be relatively similar in
24、 various types of commercial airliners, but it could vary significantly in other types of aircraft, such as military fighters. However, in the larger type of military aircraft, such as AWACS (Advanced Warning and Command System, E-3, which is based on either a Boeing 707-320-B or 767) and JSTARS (Jo
25、int Surveillance Target Attack Radar System, E-8C, which is based on Boeing 707-300 airframe), the ratio should be very similar to that in airliners. With regard to the ratio of the thermal neutron SEU cross sections, until recently, not very many such SEU cross sections were reported in the literat
26、ure. There were a few, and these were cited in 1, but they were relatively few. Due to the data that has recently become available, the number of devices in which the thermal neutron SEU cross section has been measured has increased significantly. This additional data allows us to have good confiden
27、ce on the values that have been measured and the resulting average value of the ratio. 2 Thermal neutron flux inside an airliner 2.1 Definition of thermal neutron Thermal neutrons have been given this name because while most neutrons start out with much higher energies, after a sufficient number of
28、collisions with the surrounding medium, the neutron velocity is reduced such that is has approximately the same average kinetic energy as the molecules of the surrounding medium. This energy depends on the temperature of the medium, so it is called thermal energy. The thermal neutrons are therefore
29、in thermal equilibrium with the molecules (or atoms) of the medium in which they are present. In a medium that has only a small probability of absorbing, rather than scattering, neutrons, the kinetic energies of the thermal neutrons is distributed statistically according to the Maxwell-Boltzmann law
30、. Therefore, based on this Maxwell-Boltzmann distribution, the neutron kinetic energy that corresponds to the most probable velocity is kT, where T is the absolute temperature of the medium and k a constant. For a temperature of 20 C, room temperature, this is 0,025 eV. This is based on a highly ide
31、alized model of elastic collisions between two kinds of particles, nuclei and neutrons, within a gaseous medium, and so there are departures from it in the real world. Therefore, even though a neutron energy of 0,025 eV is officially taken to be the true definition of thermal neutrons, for purposes
32、of this PAS, we will consider neutrons with energies 10 MeV) have had their energy reduced by about seven orders of magnitude. For practical purposes, we consider neutrons with E 10 MeV) neutron flux. A more recent paper by a group at EADS 10 that used a simpler detector system, again Bonner spheres
33、, but specifically designed to be used in an airliner was examined. Unfortunately, the high energy neutron fluxes from this paper are considered to be far too low to be realistic. Thus, we do not believe that the data collected by this detector system and DD IEC/PAS 62396-5:2007 5 Licensed Copy: Lon
34、don South Bank University, London South Bank University, Thu Dec 20 02:50:17 GMT+00:00 2007, Uncontrolled Copy, (c) BSI contained in 10 can be considered to be accurate enough and consistent enough to be used for our purposes of obtaining a reliable and representative value for Ratio-1. 2.4 Calculat
35、ional approach There is one paper in the literature 11 that represents a very significant step forward. It is based on applying an elaborate calculational method to a geometry consisting of a large airliner (a 747) and the atmosphere around it. The gross take-off weight of a large 747 is close to 1
36、million pounds (450 000 kg) and the overall internal volume is approximately 30 000 cubic feet (850 cubic metre) (based on the cargo capacity of cargo versions of the 747). The actual size is therefore enormous (length of aircraft is 250 ft (76 m) and wingspan of 225 ft (69 m) compared to most struc
37、tures or vehicles that are modelled for purposes of radiation transport calculations. Out of necessity, the calculation had to simplify the true geometry by orders of magnitude in order to be able to develop the model and carry out the calculations in a relatively short time. As a result, the full a
38、ircraft is described as being comprised of approximately 30 smaller volumes, into which the different proportions of the full 1 million pounds are distributed, using gross approximations for the various materials (fuel, baggage, aluminium structure, interior, etc.). Thus, it is unclear how accurate
39、the results of these calculations are, especially for the thermal neutrons. For the high energy neutrons, it is clear that for most locations the neutron flux should be very similar inside the airplane as it is outside the airplane, and that is true in the results of 11, so this serves as a consiste
40、ncy check. However, for the thermal neutrons, there are no consistency checks. The thermal neutrons are much higher everywhere inside the aircraft compared to outside within the atmosphere, so we have no idea of how accurate a result 11 represents. It may be correct, but it also may be that especial
41、ly for locations where the electronics are located, a much smaller model, greatly reduced in overall size but much more detailed in terms of the internal structures and the mass distribution that is used, would be needed to calculate the thermal neutron flux accurately. Therefore, we will use the re
42、sults from 11, but we will also compare them to the measurements from 6 and 7, to obtain Ratio-1. The results from 11 will represent the upper bound and the results from the in-flight measurements will represent a lower bound. 2.5 Processing of in-flight neutron flux data For the comparison of in-fl
43、ight measurements data is taken from four groups, 6, 7, 8 and 10, and in addition the calculations from two other groups, 11 and Armstrong 12 are used. First the measured spectra from the four aircraft measurements are shown in Figure 1, along with the calculated spectrum from 12. A tabulation of th
44、e main features concerning where the measurements were taken and which aircraft were used is given in Table 1. Table 1 Tabulation of the various atmospheric neutron measurements used Researcher Organization Detector Aircraft Year Altitude, Ft Ref. Goldhagen EML Bonner sphere ER-2 1997 40 000 (12,2 k
45、m) 8 Hubert EADS 7-detect spectrometer A300 2004 34 800 (10,6 km) 10 Hewitt NASA-Ames Bonner sphere C-141 1974 40 600 (12,4 km) 6 Nakamura Tohoku U. Bonner sphere DC-8 1985 37 000 (11,3 km) 7 Armstrong ORNL Calculation Atmosphere 1973 39 000 (11,9 km) 12 DD IEC/PAS 62396-5:2007 6 Licensed Copy: Lond
46、on South Bank University, London South Bank University, Thu Dec 20 02:50:17 GMT+00:00 2007, Uncontrolled Copy, (c) BSI Figure 1 Atmospheric neutron spectra measured in four aircraft All of the spectra have relatively similar shapes over 11 orders of magnitude, however two of the spectra seem to be l
47、ower than the other three, and these are the in-flight measurements by Nakamura over Japan and by the EADS group over the Atlantic. The differential neutron flux spectrum by Nakamura is lower than the others across the entire spectrum. The reason for this is that the measurements were made in an air
48、plane over Japan. The simplified Boeing model of the neutron flux as a function of latitude and longitude is not adequate to deal with this situation. Taking San Jose, CA as the approximate location for the ER-2 flights, the latitude for San Jose is approximately 37 which is similar to that for Nago
49、ya, Japan, the approximate location for Nakamuras measurements. The earths magnetic field varies with longitude as well as latitude. Although the variation is small in most locations, for other sites it can be large with the result that two locations very similar latitudes can have significantly different vertical rigidity cutoffs. In the case of these two cities, the rigidity cutoff ov