Codeofchina.com is in charge of this English translation. In case of any doubt about the English translation, the Chinese original shall be considered authoritative.
This part is drafted in accordance with the rules given in the GB/T 1.1-2009.
This standard is identical with International Standard IEC 62396-2:2012 Process Management for Avionics — Atmospheric Radiation Effects — Part 2: Guidelines for Single Event Effects Testing for Avionics Systems.
The Chinese documents consistent and corresponding with the normative international documents in this standard are as follows:
— GB/T 34956-2017 Atmospheric Radiation Effects — Accommodation of Atmospheric Radiation Effects Via Single Event Effects within Avionics Electronic Equipment (IEC 62396-1:2016, IDT)
For the purposes of this standard, the following editorial changes have also been made with respect to the IEC 62396-2:2012:
— the name of this standard is changed to Atmospheric Radiation Effects — Guidelines for Single Event Effects Testing for Avionics Systems.
This standard was proposed by Aviation Industry Corporation of China, Ltd.
This standard is under the jurisdiction of SAC/TC 427 (National Technical Committee 427 on Process Management for Avionics of Standardization Administration of China).
?
Introduction
This industry-wide standard provides additional guidance to avionics systems designers, electronic equipment component manufacturers and their customers to determine the susceptibility of microelectronic devices to single event effects. It expands on the information and guidance provided in GB/T 34956-2017.
This standard is provided on the use of existing single event effects (SEE) data, sources of data and the types of accelerated radiation sources used. Where SEE data is not available considerations for testing are introduced including suitable radiation sources for providing avionics SEE data. The conversion of data obtained from differing radiation sources into avionics SEE rates is detailed.
Atmospheric Radiation Effects — Guidelines for Single Event Effects Testing for Avionics Systems
1 Scope
This standard aims to provide guidance related to the testing of microelectronic devices for purposes of measuring their susceptibility to single event effects (SEE) induced by atmospheric neutrons. Since the testing can be performed in a number of different ways, using different kinds of radiation sources, it also shows how the test data can be used to estimate the SEE rate of devices and boards due to atmospheric neutrons at aircraft altitudes.
Although developed for the avionics industry, this standard may be applied by other industrial sectors.
2 Normative References
The following referenced documents are indispensable for the application of this document. For dated references, only the edition cited applies. For undated references, the latest edition of the referenced document (including any amendments) applies.
IEC 62396-1:2012 Process Management for Avionics — Atmospheric Radiation Effects — Part 1: Accommodation of Atmospheric Radiation Effects Via Single Event Effects within Avionics Electronic Equipment
IEC 62396-3 Process Management for Avionics — Atmospheric Radiation Effects — Part 3: Optimising System Design to Accommodate the Single Event Effects (SEE) of Atmospheric Radiation
IEC 62396-4 Process Management for Avionics — Atmospheric Radiation Effects — Part 4: Guidelines for Designing with High Voltage Aircraft Electronics and Potential Single Event Effects
IEC 62396-5 Process Management for Avionics — Atmospheric Radiation Effects — Part 5: Guidelines for Assessing Tthermal Neutron Fluxes and Effects in Avionics Systems
3 Terms and Definitions
For the purpose of this document, the terms and definitions given in IEC 62396-1:2012 apply.
?
4 Abbreviations
For the purpose of this document, the following abbreviations apply.
ANITA: Atmospheric-like Neutrons from thIck TArget (TSL, Sweden)
BL1A, BL1B, BL2C: Beam line designations at the TRIUMF facility (Canada)
BPSG: Borophosphosilicate glass
CMOS Complementary metal oxide semiconductor
COTS: Commercial off-the-shelf
D-D: Deuterium-deuterium
DRAM: Dynamic random access memory
D-T: Deuterium-tritium
DUT: Device under test
E: Energy
EEPROM: Electrically erasable programmable read only memory
EPROM: Electrically programmable read only memory
ESA: European Space Agency
eV: Electron volt
FIT: Failures in time (failures in 109 hours)
FPGA: Field programmable gate array
GeV: Giga electron volt
GNEIS: Gatchina Neutron Spectrometer (Russia)
GSFC: Goddard Space Flight Center
GV: Giga volt (rigidity unit)
IBM: International Business Machines
IC: Integrated circuit
ICE: Irradiation of Chips and Electronics
IEEE Trans. Nucl. Sci.: IEEE Transactions on Nuclear Science
IUCF: Indiana University Cyclotron Facility (USA)
JEDEC: JEDEC Solid State Technology Association
JESD: JEDEC standard
JPL: Jet Propulsion Laboratory
LANSCE: Los Alamos Neutron Science Center (USA)
LET: Linear energy transfer
LETth: Linear energy transfer threshold
MBU: Multiple bit upset (in the same word)
MCU: Multiple Cell Upset
MeV: Mega electron volt
NASA: National Aeronautical and Space Agency
PIF: Proton Irradiation Facility (TRIUMF, Canada)
PNPI: Petersburg Nuclear Physics Institute (Russia)
PSG: Phosphosilicate glass
QMN: Quasi-monoenergetic neutrons
RADECS: Radiations, effets sur les composants et systèmes.
RAM: Random access memory
RCNP: Research Center of Nuclear Physics (Osaka, Japan)
RVC: Result of voting (IEC)
SBU: Single Bit Upset
SDRAM: Synchronous dynamic random access memory
SEB: Single event burn-out
SEE: Single event effect
SEFI: Single event functional interrupt
SEGR: Single event gate rupture
SEL: Single event latchup
SEP: Solar energetic particles
SER: Soft error rate
SET: Single event transient
SEU: Single event upset
SHE: Single event induced hard error
SRAM: Static random access memory
SW: Software
TID: Total ionizing dose
TNF: TRIUMF neutron facility (TRIUMF, Canada)
TRIUMF: Tri-University Meson Facility (Canada)
TSL: Theodor Svedberg Laboratory (Sweden)
WNR: Weapons Nuclear Research (Los Alamos USA)
?
5 Obtaining SEE Data
5.1 Types of SEE data
The type of SEE data available can be viewed from many different perspectives. As indicated, the SEE testing can be performed using a variety of radiation sources, all of which can induce single event effects in ICs. In addition, many tests are performed on individual devices, but some tests expose an entire single board computer to radiation fields that can induce SEE. However, a key discriminator is deciding on whether existing SEE data that may be used is available, or whether there really is no existing data and therefore a SEE test on the device or board of interest has to be carried out.
5.2 Use of existing SEE data
5.2.1 General
The simplest solution is to find previous SEE data on a specific IC device. Data may be available on SEE caused by heavy ions, protons, high-energy neutrons, or thermal neutrons. Heavy-ion data is normally only applicable to space applications, where direct ionization by the primary cosmic ray flux is of concern. However, heavy ion data can be useful for screening purposes, as described in 5.2.2. Proton data is usually also gathered for space applications, where primary cosmic rays and trapped particles are of concern. However, high- energy protons provide a good proxy for neutrons in SEE measurements, as they undergo very similar nuclear interactions with device materials. Therefore, both existing neutron data and existing proton data may be applicable to the evaluation of SEE rates in a device of interest, as described in section 5.2.3. Low-energy (“thermal”) neutrons can also cause SEE in some devices but such data is only available on a very small number of devices (see section 5.2.4) and it involves neutron interactions with 10B rather than silicon.
5.2.2 Heavy ion data
An important resource that can be utilized to eliminate devices is the results from heavy ion SEE testing carried out to support space programs (~80% of the devices tested for space applications are tested only with heavy ions). This heavy ion SEE data can be used to calculate SEE data from high energy neutrons and protons by utilizing a number of different calculation methods, but this requires the active involvement of a radiation effects expert in the process. Heavy ion testing is characterized by the LET (linear energy transfer) of the ions to which the ICs are exposed. The LET is the energy that can be deposited per unit path length, divided by the density (units of MeV·cm2/mg). With neutron SEE, secondary particles or recoils created by the neutron interactions act as heavy ions, and the highest possible LET of neutron-induced recoils in silicon is ~15 MeV·cm2/mg [1, 2]. Thus, any device tested with heavy ions that has a LET threshold > 15 MeV·cm2/mg will be immune from neutron-induced SEE. In a recent paper summarizing SEE testing at NASA-GSFC[3], 21 ICs of various types were tested with only heavy ions and eight of them (~40%) had LET thresholds > 15 MeV·cm2/mg for diverse SEE effects.
However, for the rare commercial SRAMs that are susceptible to SEL from heavy ions [4], this susceptibility can be increased due to the presence of small amounts of high Z materials within the IC, e.g., tungsten plugs, because higher Z recoils are created which can cause SEE reactions due to their higher values of LET. The high Z materials also lead to higher proton and neutron SEL cross-sections due to the neutron/proton reactions producing these recoils with higher LET and energy. Therefore heavy ion SEL cross-sections need to be examined carefully for applicability to proton-neutron SEL susceptibility caused by embedded high Z materials in the SRAMs. A suggested conservative value of LET threshold above which a device can be considered immune from SEL induced by neutrons is 40 MeV·cm2/mg[4]. However, this caution does not apply to the primary rationale given above for eliminating some devices from consideration for neutron SEE sensitivity based on heavy ion SEE testing, since only some devices incorporate these higher Z materials and the limitation applies to SEL.
Heavy ion SEE data should not be used for application to the atmospheric neutron environment for calculation of neutron cross-section, except by scientists and engineers who have extensive experience in using this kind of data. Unless otherwise stated explicitly, when SEE data is discussed in the remainder of this international standard, it refers only to single event testing using a neutron or proton source, not to the results from testing with heavy ions.
5.2.3 Neutron and proton data
If SEE data on a device of interest is found from SEE tests using high energy neutrons or protons, it will still require expertise regarding how the data is to be utilized in order to calculate a SEE rate at aircraft altitudes. It mainly focuses on the following two points:
a) Neutron SEE sensitivity characterization unit: Data obtained by IC vendors for their standard application to ground level systems are often expressed in totally different units, FIT units, which is taken to apply at ground level but not apply at aircraft altitudes.
b) Process differences: IC devices are constantly changing. In some cases, devices which had been tested, become obsolete and are replaced by new devices which have not been tested. The fact that a device is made by the same IC vendor and is of the same type as the one it replaced does not mean that the SEE data measured in the first device applies directly to the newer device. In some cases, small changes in the IC design or manufacturing process can have a large effect in altering the SEE response, but in other cases, the effect on the SEE response may be minimal.
?
5.2.4 Thermal neutron data
There is little data on thermal neutron cross-section. However a number of the spallation neutron sources including TRIUMF, TSL and ISIS contain a substantial percentage of thermal neutrons within the high energy beam. Using thermal neutron filters or time of flight it is possible at such sources to determine thermal neutron cross-section. In addition there are a number of dedicated thermal neutron sources and these are listed in IEC 62396-1:2012.
A continuing problem with the existing SEE data is that there is no single database that contains all of the neutron or proton SEE data. Instead, portions of this kind of SEE data can be found published in many diverse sources. The SEE data in the larger databases is mainly on much older devices, dating from the 1990s and even 1980s, and is primarily from heavy ion tests that were performed for space applications and not from testing with protons and neutrons.
Foreword II
Introduction III
1 Scope
2 Normative References
3 Terms and Definitions
4 Abbreviations
5 Obtaining SEE Data
6 Availability of existing SEE data for avionics applications
7 Considerations for SEE Testing
8 Converting Test Results to Avionics SEE Rates
Annex A (Informative) Sources of SEE Data Published Before
Bibliography
