1、Designation: F 1262M 95 (Reapproved 2002)METRICStandard Guide forTransient Radiation Upset Threshold Testing of DigitalIntegrated Circuits Metric1This standard is issued under the fixed designation F 1262M; the number immediately following the designation indicates the year oforiginal adoption or, i
2、n the case of revision, the year of last revision. A number in parentheses indicates the year of last reapproval. Asuperscript epsilon (e) indicates an editorial change since the last revision or reapproval.1. Scope1.1 This guide is to assist experimenters in measuring thetransient radiation upset t
3、hreshold of silicon digital integratedcircuits exposed to pulses of ionizing radiation greater than 103Gy (Si)/s.1.2 This standard does not purport to address all of thesafety concerns, if any, associated with its use. It is theresponsibility of the user of this standard to establish appro-priate sa
4、fety and health practices and determine the applica-bility of regulatory limitations prior to use.2. Referenced Documents2.1 ASTM Standards:E 666 Practice for Calculating Absorbed Dose from Gammaor X Radiation2E 668 Practice for the Application of Thermoluminescence-Dosimetry (TLD) Systems for Deter
5、mining Absorbed Dosein Radiation-Hardness Testing of Electronic Devices2F 867M Guide for Ionizing Radiation Effects (Total Dose)Testing of Semiconductor Devices Metric32.2 Military Standards:4Method 1019 in MIL-STD-883. Steady-State Total DoseIrradiation ProcedureMethod 1021 in MIL-STD-883. Dose Rat
6、e Threshold forUpset of Digital Microcircuits.3. Terminology3.1 Definitions:3.1.1 combinational logicA digital logic system with theproperty that its output state at a given time is solely deter-mined by the logic signals at its inputs at the same time (exceptfor small time delays caused by the prop
7、agation delay ofinternal logic elements).3.1.1.1 DiscussionCombinational circuits contain no in-ternal storage elements. Hence, the output signals are not afunction of any signals that occurred at past times. Examples ofcombinational circuits include gates, adders, multiplexers anddecoders.3.1.2 com
8、plex circuit response mechanismsFor mediumscale integration (MSI) and higher devices it is useful to definethree different categories of devices in terms of their internaldesign and radiation response mechanisms.3.1.3 over-stressed deviceA device that has conductedmore than the manufacturers specifi
9、ed maximum current, ordissipated more than the manufacturers specified maximumpower.3.1.3.1 DiscussionIn this case the DUT is considered tobe overstressed even if it still meets all of the manufacturersspecifications. Because of the overstress, the device should beevaluated before using it in any hi
10、gh reliability application.3.1.4 sequential logicA digital logic system with theproperty that its output state at a given time depends on thesequence and time relationship of logic signals that werepreviously applied to its inputs.3.1.4.1 DiscussionExamples of sequential logic circuitsinclude flip-f
11、lops, shift registers, counters, and arithmetic logicunits.3.1.5 state vectorA state vector completely specifies thelogic condition of all elements within a logic circuit.3.1.5.1 DiscussionFor combinational circuits, the statevector includes the logic signals that are applied to all inputs:for seque
12、ntial circuits, the state vector must also include thesequence and time relationship of all input signals. In thisguide the output states will also be considered part of the statevector definition. For example, an elementary 4-input NANDgate has 16 possible state vectors, 15 of which result in thesa
13、me output condition (“1” state). A 4-bit counter has 16possible output conditions, but many more state vectors be-cause of its dependence on the dynamic relationship of variousinput signals.1This guide is under the jurisdiction of ASTM Committee F01 on Electronicsand is the direct responsibility of
14、Subcommittee F01.11 on Quality and HardnessAssurance.Current edition approved Dec. 10, 2002. Published May 2003. Originallyapproved in 1995. Last previous edition approved in 1995 as F 1262M 95.2Annual Book of ASTM Standards, Vol 12.02.3Discontinued. Replaced by F 1893. See 1997 Annual Book of ASTM
15、Standards,Vol 10.04.4Available from Standardization Documents Order Desk, Bldg. 4, Section D,700 Robbins Ave., Philadelphia, PA 19111-5094, Attn: NPODS.1Copyright ASTM International, 100 Barr Harbor Drive, PO Box C700, West Conshohocken, PA 19428-2959, United States.3.1.6 upset responseThe electrica
16、l response of a circuitwhen it is exposed to a pulse of transient ionizing radiation.3.1.6.1 DiscussionTwo types of upset response can occur:(1) transient output error, for which the instantaneous outputvoltage of an operating digital circuit is greater than a predeterminedvalue (for a low output co
17、ndition) or less than a predetermined value(for a high output condition), and the circuit spontaneously recovers toits pre-irradiation condition after the radiation pulse subsides. Thepredetermined values mentioned above are agreed to by all partiesparticipating in the test and should be included in
18、 the test plan.(2) stored logic state error, for which there is a change in the stateof one or more internal logic elements that does not recover spontane-ously after the radiation pulse. Because the radiation changes the statevector, the circuit spontaneously recovers to a different logic state. Th
19、isdoes not imply the change will always be immediately observable on acircuit output. However, the circuit can be restored to its original statevector by re-initializing it afterwards.3.1.6.2 DiscussionAlthough the term upset response isusually used to describe output voltage responses, somedevices,
20、 such as open collector gates, are better characterizedby measuring the output current. Upset response also includesthe transient currents that are induced in the power supply leads(sometimes very large) as well as the response of the deviceinputs, although in most applications the input response is
21、 notsignificant.4. Summary of Guide4.1 For transient radiation upset threshold tests, the transientoutput voltage and the condition of internal storage elements,or both, is measured at a succession of radiation levels todetermine the radiation level for which transient voltage orfunctional test erro
22、rs first occur. An oscilloscope, digitalstorage oscilloscope, transient digitizer or similar instrument isused to measure the output transient voltage. Functional testsare made immediately after irradiation to detect internalchanges in state induced by the radiation. The device is initiallybiased an
23、d set up in a predetermined condition. The testconditions are determined from topological analyses or bytesting the device in all possible logic state combinations.4.2 A number of factors are not defined in this guide andmust be agreed upon beforehand by the parties to the test.These factors are des
24、cribed in the test plan. As a minimum thetest plan must specify the following:(1) Pulse width, energy spectrum, and type of radiationsource,(2) Voltage and electrical loading conditions on each pin ofthe device during testing,(3) Resolution and accuracy required for the upset responsethreshold of in
25、dividual devices, along with the method used tovary the radiation level,(4) Failure criterion for transient voltage upset, outputcurrent, and power supply current as applicable,(5) Measuring and reporting Ipp, transient output voltageand transient output current levels,(6) Functional test to be made
26、 after irradiation,(7) Power supply and operating frequency requirements,(8) State vectors used for testing,(9) Radiation levels to use for transient response measure-ments,(10) Recommended radiation level at which to begin the testsequence, and(11) Procedure to adjust the dose rate during testing.(
27、12) Device temperature during test.4.3 The state vectors in which the device is to be irradiatedare determined from the basic (see 8.2.1) and topologicalanalysis, (see 8.2.2) or both.5. Significance and Use5.1 Digital logic circuits are used in system applicationswhere they are exposed to pulses of
28、radiation. It is important toknow the minimum radiation level at which transient failurescan be induced, since this affects system operation.6. Interferences6.1 Accumulated Ionizing DoseMany devices may bepermanently damaged by the accumulated ionizing dose theyare exposed to during upset testing. T
29、his limits the number ofradiation pulses that can be applied during transient upsettesting. Accumulated ionizing dose sensitivity depends onfabrication techniques and device technology. Metal oxidesemiconductor (MOS) devices are especially sensitive toaccumulated ionizing dose damage. Newer bipolar
30、deviceswith oxide-isolated sidewalls may also be affected by lowlevels of accumulated ionizing dose. The maximum ionizingdose to which devices are exposed must not exceed 10 % (see8.4.5) of the typical ionizing dose failure level of the specificpart type.6.2 Dosimetry AccuracySince this guide ultima
31、tely deter-mines the dose rate at which upset occurs, dosimetry accuracyinherently limits the accuracy of the guide.6.3 LatchupSome types of integrated circuits may bedriven into a latchup condition by transient radiation. If latchupoccurs, the device will not function properly until power istempora
32、rily removed and reapplied. Permanent damage mayalso occur. Although latchup is an important transient responsemechanism, this procedure is not applicable to latchup testing.Functional testing after irradiation is required to detect internalchanges of state, and this will also detect latchup.6.4 Pac
33、kage ResponseAt dose rates above 108Gy (Si)/sthe response may be dominated by the package response ratherthan the response of the integrated circuit device being tested.For high speed devices, this may include lead/bondwire effectswith upsets caused solely by the radiation pulses rise and fallrates
34、rather than dose rate. Package effects can be minimizedby adequately decoupling the power supply with appropriatehigh-speed capacitors.6.5 Steps Between Radiation LevelsThe size of the stepsbetween successive radiation levels limits the accuracy withwhich the dose rate upset threshold is determined.
35、 Costconsiderations and ionizing dose damage limit the number ofradiation levels that can be used to test a given device.6.6 Limited Number of State VectorsCost, testing time,and cumulative ionizing radiation usually make it necessary torestrict upset testing to a small number of state vectors. Thes
36、estate vectors must include the most sensitive conditions inorder to avoid misleading results. An analysis is required toselect the state vectors used for radiation testing to make sureF 1262M 95 (2002)2that circuit and geometrical factors that affect the upsetresponse are taken into account.7. Appa
37、ratus7.1 The equipment and information required for this guideincludes an electrical schematic of the test circuit, a logicdiagram of the device to be tested, a transient radiationsimulation source, dosimetry equipment, and electrical equip-ment for the measurement of the device response and func-ti
38、onal testing. If the alternate topological analysis approach isto be used, (see 8.2.2) then a photomicrograph or compositemask drawing of the device is also needed.7.2 Radiation Simulation and Dosimetry Apparatus:7.2.1 Transient Radiation SourceA pulsed high energyelectron or bremsstrahlung source t
39、hat can provide a dose ratein excess of the upset response threshold level of the devicebeing tested at the pulse width specified in the test plan isneeded. A linear accelerator (LINAC) with electron energies of10 to 25 MeV is preferred (see Note 1), although in someinstances a flash X ray with end
40、point energy above 2.0 MeVmay be utilized (see Note 2 and Note 3). It is usually muchmore difficult to synchronize a flash X-ray pulse with circuitoperation, which limits the applicability of a flash X ray.NOTE 1Linac radiation pulses are made from a train of discrete“micropulses” occurring at the l
41、inac radio frequency (RF). This highfrequency pulse structure could cause erroneous results for high frequencydevices under test such as gallium arsenide. This has not yet been directlyobserved.NOTE 2The absorption coefficient of photons in silicon and packag-ing materials is relatively flat at ener
42、gies above 2 MeV, and has a nearlyconstant ratio to the absorption coefficient of typical dosimetry systems.At lower energies absorption coefficients increase, which can introducelarge dosimetry errors if the end point energy in a bremsstrahlung sourceis below 2.0 MeV.NOTE 3Because of dose enhanceme
43、nt and attenuation, a transportcalculation is generally required to relate the dose at the region of interestin the DUT to the dosimetry used if a low energy flash X ray is used.7.2.2 Ionizing Dose Dosimetry SystemA dosimetry sys-tem such as a thermoluminescent dosimetry (TLD) system orcalorimeter t
44、hat can be used to measure the absorbed ionizingdose produced by a single pulse of the radiation source isneeded (see Practice E 668).7.2.3 Pulse Shape MonitorA device for monitoring theshape of the radiation pulse such as a PIN diode is required. Insome instances it may be possible to directly dete
45、rmine thepulse shape by measuring the total beam current of theaccelerator with a current transformer or secondary emissionmonitor (SEM).7.2.4 Active Dosimetry StandardAn active dosimeter thatallows the dose rate to be determined from electronic measure-ments is needed. This may be a PIN detector, a
46、 Faraday cup, ora combination of a calorimeter and current transformer.7.3 Electronic Test Equipment:7.3.1 Radiation Test FixtureA test fixture that allows thedevice to be placed in the radiation beam with convenientconnection to external equipment (pulse generators, powersupplies, line drivers etc.
47、) is required for testing.7.3.2 Line DriversLine drivers that provide high imped-ance to the device under test and can drive the low impedanceof terminated output cables with adequate signal fidelity arerequired. The line drivers must be designed so that their ownresponse to transient ionizing radia
48、tion is much smaller thanthat of the circuit being measured (see Note 4).NOTE 4Although line drivers are normally not placed in the directradiation beam, there is always some stray radiation that may affect theline driver. Furthermore, replacement currents in the wiring that connectsthe line driver
49、to the circuit under test may also introduce a spuriousresponse.7.3.3 General Purpose Test EquipmentPower supplies,pulse generators, cables and termination resistors that arerequired to bias the device and establish its internal operatingconditions are needed.7.3.4 Transient Response Measuring DeviceAn oscillo-scope, transient digitizer or similar device shall be used tomeasure the transient response of the device under test. Thebandwidth and sensitivity of this equipment must be compat-ible with the pulse width and measurement criteria in the test
