Iec 60749 44 2016

Semiconduct or dev ices – Mechanical and cl mat ic test met hods – Part 44: Neutron beam ir adiat ed single ev ent efect SEE t est met hod for Partie 44: Mét hode d'essai des efet s d'un

Semiconduct or dev ices – Mechanical and cl mat ic test met hods –

Part 44: Neutron beam ir adiat ed single ev ent efect (SEE) t est met hod for

Partie 44: Mét hode d'essai des efet s d'un év énement isolé (SEE) ir adié par un

THIS PUBLICATION IS COPYRIGHT PROT CTED

Copyr ight © 2 16 IEC, Ge e a, Switzer la d

Al r i hts r es r ve Unle s oth r wis s e ifi d, n p r t of this p blc tio ma b r epr od c d or uti z d in a y f or m

or b a y me n ,ele tr onic or me h nic l in lu in p oto o yin a d microfim, with ut p r mis io in wr itin fr om

eith r IEC or IEC's memb r Natio al Commite in th c u tr y of th r eq e ter If y u h v a y q e tio s a o t IEC

c p r i ht or h v a e q iry a o t o tainin a ditio al r i hts to this p blc tio , ple s c nta t th a dr es b low or

y ur lo al IEC memb r Natio al Commite for f ur th r infor matio

Droits d r epro u tio r és r vé Sa f in ic tio c ntrair e, a c n p r tie d c te p blc tio n p ut êtr e r epr od ite

ni uti s e s u q elq e forme q e c s it et p r a c n pr oc d , éle tr oniq e o mé a iq e, y c mp s la p oto o ie

et le micr ofims, s n la c r d é r it d l EC o d Comité n tio al d l EC d p y d d ma d ur Si v u a e d s

q e tio s s r le c p r i ht d l EC o si v u d sire o te ir d s dr oits s p léme tair es s r c te p blc tio , uti s z

le c or do n e ci-a r ès o c nta te le Comité n tio al d l EC d v tr e p y d ré id n e

Th Inter natio al Ele tr ote h ic l Commis io (IEC) is th le din glo al or ga iz tio th t pr ep r es a d p bls e

Intern tio al Sta d r ds for al ele tr i al ele tr onic a d relate te h olo ie

A bout IEC publc tio s

Th te h ic l c nte t of IEC p blc tio s is k pt u d r c n ta t r eview b th IEC Ple s ma e s r e th t y u h v th

late t e itio , a c r r ig n a or a ame dme t mig t h v b e p bls e

IEC Catalo u - webstore.e ch/cat alo u

Th st an -alo e a plc at io for c on ult in t he e t ire

biblo ra hic al informat io o IEC Int ern tio al St an ard ,

Te h ic al Sp cific at io s, Tec hnic al R ep rt s a d ot her

d c ume t s A v aia le for PC, Ma OS, A ndroid Ta let s a d

iPa

IEC publc t io s s arch - w w w.e ch/s arch ub

Th a v anc ed s arc h e a le t o fin IEC p blc t io s b a

v riety of c riteria (efere c e n mb r, text, t ec hnic l

c mmit t ee,…) It als giv es informat io o projec t s, re lac ed

Ele tro edia - www.ele tro edia.org

Th world's le din o ln dic t io ary of elec t ro ic a d

elec t ric l t erms c ont ainin 2 0 0 t erms a d d finitio s in

En ls a d Fre c , wit h e uiv le t t erms in 15 a dit io al

la g a e A ls k nown a t he Int ern t io al Ele t rot ec hnic al

V oc ab lary (IEV ) o ln

IEC Glos ar y - st d.e ch/glos ary

6 0 0 elec t rote h ic l t ermin lo y e t rie in En ls a d

Fre c ex t rac t ed fom t he Terms a d Definit io s cla s of

IEC p blc tio s is u d sin e 2 0 Some e trie h v b e

c lec t ed fom e rler p blc at io s of IEC TC 3 , 7 , 8 a d

CIS R

IEC Cu t omer Servic Centre - webst ore.e ch/cs

If y ou wis t o giv u your fe d ac k o t his p blc t io or

n e furt her a sist an e,ple s c ont ac t t he Cu t omer Servic

Ce t re: c sc@ie c h

A pro os de l'IEC

L Commis io Ele trote h iq e Inter natio ale (IEC) e t la pr emièr e org nis tio mo diale q i éla ore et p ble d s

Nor me intern tio ale p ur to t c q i a tr ait à léle tr i ité, à léle tr oniq e et a x te h olo ie a p r enté s

A pro os de publc t io s IEC

L c nte u te h iq e d s p blc tio s IEC e t c n tamme t r ev Ve i e v u a s r er q e v u p s é e lé itio la

plu ré e te, u c r r ig n um o ame d me t p ut a oir été p blé

Cat alo u IEC - webstore.e ch/cat alo u

A pplc t io a t on me p ur c n ult er tou le re s ig eme t s

biblo ra hiq e s r le Norme intern t io ale ,

Sp cific t io s t ec niq e , R ap ort s t ec hniq e et a t re

d c ume t s d l EC Dis o ible p ur PC, Ma OS, t ablett es

A ndroid et iPa

R echerche de publc t io s IEC - w w w.e ch/se rch ub

L rec herc e a a c ée p rmet d t ro v r d s p blc at io s IEC

e ut ils nt difére t s crit ère (n méro d référe c , t ex te,

c mit é d’ét ud s,…) Ele d n e a s i d s informat io s s r le

projet s et le p blc t io s remplac ée o ret iré s

IEC J st Publshed - webstore.e ch/j st publshed

R est ez informé s r le n uv ele p blc at io s IEC Ju t

Pu ls e d t ai e le n u ele p blc tio s p ru s

Dis o ible e lg e et a s i u e fois p r mois p r emai

Ele tro edia - www.ele tro edia.org

L premier dic t io n ire e lg e d t erme élec tro iq e et

élec t riq e I c nt ie t 2 0 0 t erme et d finit io s e a glais

et e fa ç is, ain i q e le t erme é uiv ale t s d n 15

la g e a dit io n le Eg leme t a p lé V oc b laire

Ele t rot ec hniq e Int ern tio al (IEV ) e lg e

Glos aire IEC - st d.e ch/glos ary

6 0 0 e t ré s termin lo iq e éle t rot ec hniq e ,e a glais

et e fa ç is, ext raites d s article Terme et Définit io s d s

p blc t io s IEC p ru s d p is 2 0 Plu c rt ain s e t ré s

a t érie re ex t rait es d s p blc at io s d s CE 3 , 7 , 8 et

CIS R d l EC

Servic Clent s - webstore.e ch/cs

Si v u d sire n u d n er d s c mme t aire s r c t t e

p blc at io o si v u a e d s q e t io s c nt act ez-n u :

c c@ie c

Semiconduct or dev ices – Mechanical and cl mat ic t est met hods –

Part 44: Neutron beam ir adiated single ev ent efect (SEE) test method for

Partie 44: Mét hode d'ess i des efets d'un év énement isolé (SEE) ir adié par un

W arnin ! Mak e s re th t you o tain d this publc tion from a a thorize distribut or

A tt ention! Veui ez vou a s rer qu vou avez o te u c t te publc tion v ia u distribute r a ré

c olour

inside

CONTENTS

FOREWORD 4

1 Sco e 6

2 Normative ref eren es 6

3 Terms an def i ition 6

4 Test a p ratu 9

4.1 Me s rement eq ipment 9

4.2 Radiation source 10 4.3 Test sample 10 5 Proced re neutron ir adiated sof t er or test 10 5.1 Surf ace pre aration 10 5.2 Power s p ly voltage 10 5.3 Ambient temp rature 1

5.4 Core c cle time 11 5.5 Data p t ern 1

5.6 Numb r of me s rement samples 1

5.7 Calc lation f or time req ired in the b am 1

6 Evaluation 1

6.1 Me s rement an f ai ure rate estimation 1

6.2 Determination of MCU an MBU cros section 12 6.3 Determination of device FIT (event rate) from cros section 12 7 Summary 12 An ex A (informative) Ad itional information for the a pl ca le proc rement sp cif i ation 13 A.1 General 13 A.2 Des ription of the b am source 13 A.3 Des ription of the sample an test vehicle 13 A.3.1 Sample size 13 A.3.2 Vehicle des ription 13 A.4 Test des ription 14 A.5 Test res lts 14 An ex B (informative) White neutron test a p ratu 16 An ex C (informative) Fai ure rate calc lation 18 C.1 An in uen e of sof t er or f or actual semicon u tor devices 18 C.1.1 General 18 C.1.2 Duty deratin 18 C.1.3 Uti ty deratin 18 C.1.4 Critical y deratin 19 C.2 Fai ure rate calc lation in lu in deratin 19 Bibl ogra h 2

Fig re B.1 – Typical white neutron sp ctra with dif f erent s ield (p lyeth lene)

thic nes 16

Fig re B.2 – Typical neutron sp ctrum 17

Fig re B.3 – Comp rison of LANSCE (WNR) an TRIUMF neutron sp ctra with

ter estrial neutron sp ctrum 17

Fig re C.1 – Sc ematic image of d ty deratin 18

Fig re C.2 – Sc ematic image of memory ef f ective are f or uti ty deratin 19

INTERNATIONAL ELECTROTECHNICAL COMMISSION

Part 44: Neutron beam ir adiated single event ef f ect (SEE)

test method f or semiconductor devices

1 Th Intern tio al Ele trote h ic l Commis io (IEC) is a worldwid org niz tio for sta d rdiz tio c mprisin

al n tio al ele trote h ic l c mmite s (IEC Natio al Commite s) Th o je t of IEC is to promote

intern tio al c -o eratio o al q e tio s c n ernin sta d rdiz tio in th ele tric l a d ele tro ic f i ld To

this e d a d in a ditio to oth r a tivitie , IEC p bls e Intern tio al Sta d rd , Te h ic l Sp cif i atio s,

Te h ic l Re orts, Pu lcly Av ia le Sp cific tio s (PAS) a d Guid s (h re f ter refere to a “IEC

Pu lc tio (s)”) Th ir pre aratio is e tru te to te h ic l c mmite s; a y IEC Natio al Commite intere te

in th s bje t d alt with ma p rticip te in this pre aratory work Intern tio al g v rnme tal a d n

n-g v rnme tal org niz tio s laisin with th IEC als p rticip te in this pre aratio IEC c la orate clo ely

with th Intern tio al Org niz tio for Sta d rdiz tio (ISO) in a c rd n e with c n itio s d termin d b

a re me t b twe n th two org niz tio s

2) Th formal d cisio s or a re me ts of IEC o te h ic l maters e pre s, a n arly a p s ible, a intern tio al

c n e s s of o inio o th rele a t s bje ts sin e e c te h ic l c mmite h s re re e tatio f rom al

intere te IEC Natio al Commite s

3) IEC Pu lc tio s h v th form of re omme d tio s f or intern tio al u e a d are a c pte b IEC Natio al

Commite s in th t s n e Whie al re s n ble eforts are ma e to e s re th t th te h ic l c nte t of IEC

Pu lc tio s is a c rate, IEC c n ot b h ld re p n ible for th wa in whic th y are u e or f or a y

misinterpretatio b a y e d u er

4) In ord r to promote intern tio al u iformity, IEC Natio al Commite s u d rta e to a ply IEC Pu lc tio s

tra s are tly to th ma imum e te t p s ible in th ir n tio al a d re io al p blc tio s An div rg n e

b twe n a y IEC Pu lc tio a d th c r e p n in n tio al or re io al p blc tio s al b cle rly in ic te in

th later

5) IEC its lf d e n t pro id a y ate tatio of c nformity In e e d nt c rtific tio b die pro id c nformity

a s s me t s rvic s a d, in s me are s, a c s to IEC mark of c nformity IEC is n t re p n ible for a y

s rvic s c rie o t b in e e d nt c rtif i atio b die

6) Al u ers s o ld e s re th t th y h v th late t e itio of this p blc tio

7) No la i ty s al ata h to IEC or its dire tors, emplo e s, s rv nts or a e ts in lu in in ivid al e p rts a d

memb rs of its te h ic l c mmite s a d IEC Natio al Commite s for a y p rs n l injury, pro erty d ma e or

oth r d ma e of a y n ture wh ts e er, wh th r dire t or in ire t, or f or c sts (in lu in le al fe s) a d

e p n e arisin o t of th p blc tio , u e of, or rela c u o , this IEC Pu lc tio or a y oth r IEC

Pu lc tio s

8) Ate tio is drawn to th Normativ refere c s cite in this p blc tio Us of th refere c d p blc tio s is

in is e s ble for th c re t a plc tio of this p blc tio

9) Ate tio is drawn to th p s ibi ty th t s me of th eleme ts of this IEC Pu lc tio ma b th s bje t of

p te t rig ts IEC s al n t b h ld re p n ible for id ntifyin a y or al s c p te t rig ts

International Stan ard IEC 6 7 9-4 has b en pre ared by IEC tec nical commite 4 :

Semicon u tor devices

The text of this stan ard is b sed on the fol owin doc ments:

Ful inf ormation on the votin f or the a proval of this stan ard can b f ou d in the re ort on

votin in icated in the a ove ta le

This publcation has b en draf ted in ac ordan e with the ISO/IEC Directives, Part 2

A l st of al the p rts in the IEC 6 7 9 series, publs ed u der the general title Semico du tor

devic s – Mec a ic l a d climatic te t meth ds, can b f ou d on the IEC we site

The commit e has decided that the contents of this publcation wi remain u c an ed u ti

the sta i ty date in icated on the IEC we site un er "htp:/we store.iec.c " in the data

related to the sp cifi publ cation At this date, the publ cation wi b

• recon rmed,

• with rawn,

• re laced by a revised edition, or

IMPORTANT – The 'colour in ide' logo on the cov r pa e of this publ c tion indic te

understa din of its conte ts Us rs s ould therefore print this doc me t usin a

colour printer

SEMICONDUCTOR DEVICES –

Part 44: Neutron beam ir adiated single event ef f ect (SEE)

test method f or semiconductor devices

This p rt of IEC 6 7 9 esta l s es a proced re for me s rin the sin le event ef fects (SEEs)

on hig den ity integrated circ it semicon u tor devices in lu in data retention ca a i ty of

semicon u tor devices with memory when s bjected to at mospheric neutron radiat ion prod c ed

by c osmic ray s The sin le event ef f ects sen it iv it y is me s red w hie t he dev ic e is iradiat ed in a

neut ron b am of k now n flux This test method can be a pl ed to an typ of integrated circ it

NOT 1 Semic n u tor d vic s u d r hig v lta e stre s c n b s bje t to sin le e e t efe ts in lu in S B,

sin le e e t b rn ut a d S GR sin le e e t g te ru ture, f or this s bje t whic is n t c v re in this d c me t,

ple s refer to IEC 6 3 6-4 [2]

NOT 2 In a ditio to th hig e erg n utro s s me d vic s c n h v a s ft eror rate d e to low e erg (<1 eV)

th rmaln utro s For this s bje t whic is n t c v re in this d c me t, ple s refer to IEC 6 3 6-5 [3]

The f olowin doc ments, in whole or in p rt, are normatively referen ed in this doc ment an

are in isp n a le f or its a pl cation For dated ref eren es, only the edition cited a pl es For

u dated ref eren es, the latest edition of the referen ed doc ment (in lu in an

amen ments) a pl es

energ de osited in the si con of an electronic comp nent by a sin le ionisin p rticle

cau in more than one bit in the same word to b upset

Note 1 to e try: Th d f i itio of MBU h s b e u d te d e to th intro u tio of th d finitio of MC

3.4

multiple c l ups t

MCU

energ de osited in the si con of an electronic comp nent by a sin le ionisin p rticle

in u in es several bits in an integrated circ it (IC) to b upset at one time

3.5

sof t er or

er one u output sig al f rom a latc or memory cel that can b cor ected by p rf ormin one

or more normal f un tion of the device containin the latc or memory cel

Note 1 to e try: As c mmo ly u e , th term refers to a eror c u e b ra iatio or ele troma n tic p ls s a d

n t to a eror a s ciate with a p y ic l d f ect intro u e d rin th ma ufa turin pro e s

Note 2 to e try: Soft erors c n b g n rate fom S U, S FI MBU, MC , a d or S T Th term S R h s b e

a o te b th c mmercial in u try whie th more s e ific terms S U, S FI etc are ty ic ly u e b th

a io ic , s a e a d mi tary ele tro ic c mmu itie

Note 3 to e try: Th term “s ft eror” wa first intro u e (f or D AMs a d ICs) b Ma a d Wo d of Intel in th ir

Apri 19 8 p p r at th IRP a d th term “sin le e e t u s t wa intro u e b Gu n er, Wolc i a d Ala of

NRL in th ir 19 9 NSREC p p r (S U of D AMs b n utro s a d proto s)

3.6

single e e t ef f ect

SEE

resp n e of a comp nent cau ed by the imp ct of a sin le energetic p rticle

Note 1 to e try: Ex mple of e erg tic p rticle in lu e g la tic c smic ra s, s lar e erg tic p rticle , e erg tic

n utro s a d proto s

Note 2 to e try: Th ra g of e p n e c n in lu e b th n n-d stru tiv (or e ample u s t) a d d stru tiv (or

e ample latc -u or g te ru ture) p e ome a

3.7

single-e e t hard er or

SHE

sin le event in u ed hard er or

ir eversible c an e in o eration from a sin le radiation event that is typical y as ociated with

p rmanent damage to one or more of the device elements

Note 1 to e try: Ex mple in lu e p rma e tly stu k-bit in th d vic a d g te o id ru ture

f ai ure that can ot b reset other than by re o tin the s stem or by c cl n the p wer to the

relevant f un tional element

3.13

hard f ault

at the aircraft f un tion level, p rmanent fai ure of a comp nent within an LRU

Note 1 to e try: A h rd fa lt re ults in th remo al of th LR af fe te a d th re la eme t of th p rma e tly

d ma e c mp n nt b fore a s stem/s stem arc ite ture c n b re tore to ful f un tio alty Su h a fa lt c n

imp ct th v lu for th MT F of th LRU re aire

3.14

single e e t burnout

SEB

burnout of a p wered electronic comp nent or p rt there f as a res lt of the energ

a sorption trig ered by an in ivid al radiation event

Note 1 to e try: Ex mple of a c mple d vic in lu emicro ro e s rs

Note 2 to e try: This efe t h s s metime b e refere to a lo k p, in ic tin th t s metime th p rt c n b

p t into a “ o e ” state

3.16

single e e t gate rupture

SEGR

event in the gate of a p wered in ulated gate comp nent when the radiation c arge a sorb d

by the device is s f f icient to cau e gate rupture, whic is destru tive

3.17

single e e t latc up

SEL

event in a f our layer semicon u tor device when the radiation a sorb d by the device is

s f f icient to cau e a node within the p wered semicon u tor device to b held in a fi ed state

whatever input is a pl ed u ti the device is de-p wered

Note 1 to e try: Su h latc u c n b d stru tiv or n n-d stru tiv

3.18

single e e t tra sie t

SET

momentary voltage ex ursion (voltage spike) at a node in an integrated circ it cau ed by a

sin le energetic p rticle strike

Note 1 to e try: Th s e if i terms AS T a alo u sin le e e t tra sie t a d DS T digital sin le e e t tra sie t

c n b u e

spuriou digital signal or voltage, in u ed by the de osition of c arge by a sin le p rticle that

can pro agate throu h the circ it p th d rin one cloc c cle

3.21

multiple bit ups t

MBU

energ de osited in the si con of an electronic comp nent by a sin le ionisin p rticle

cau in upset of more than one bit in the same word

3.2

σ

<radiation terms f or p rticle interaction >

combination of a sen itive are an pro a i ty of an interaction de ositin the critical c arge

f or a SEE

The cros section (σ) is calc lated u in the fol owin formula:

σ = N / Ф

Where N, is the n mb r of er ors an Ф, the p rticle fl en e

Note 1 to e try: Th u its for cro s s ctio are cm

The eq ipment s al b ca a le of me s rin the f un tion of the integrated circ it devices,

an ca a le of me s rin the time taken for the c an e of stored data or other events by the

exp s re to energetic p rticles, s c as neutron , proton an alpha radiation to take place

(i.e the generation of a sof t er or) Alternatively, the test eq ipment (memory tester, etc.)

s al have the ca a i ty of cou tin the n mb r of sof t er ors in u it time The eq ipment

s al b ca able of identif yin when hard or f irm f aults oc ur; althou h these events are in

general les f req ent, their imp ct is hig er

NOT Th sta d rd IEC 6 7 9-3 c ntain a n n-a c lerate re l-time s ft eror te t.

4.2 Ra iation sourc

In order to p rform ac elerated ter estrial SER me s rements, a radiation source(s) is

req ired that is simi ar to the energ sp ctrum of ter estrial cosmic ray This can b

ac ompl s ed by a bro d sp ctrum b am or by u in multiple mono-energetic b ams The

radiation b am or b ams s ould cover the whole sp ctrum of atmospheric radiation takin

into ac ou t the energ dif feren es in the variou b ams Sp cial at ention is to b taken with

resp ct to the ef fects of s at ered radiation f rom the b am on the test setup Tec nical

p rson el o eratin the f aci ty are to b con ulted in terms of the relative f l x of the f orward

an b c ward s aterin distribution of the b am They s al also b con ulted on

ef f ectivenes of s ieldin materials f or the main b am an s at ered b am aten ation The

n mb r of b ard in f ront of the device u der test (DUT) an the distan e from the cou ter

s al b recorded to b u ed in aten ation calc lation The res lts of the testin s al b

d e to radiation ef fects on the DUT an not from interaction of radiation with other

comp nents in the test In p rtic lar, p wer s p l es can b v lnera le to radiation-in u ed

avalan he bre k own Sen itive electronic circ its in the tester an an device on the DUT

b ard (e.g buf f ers or registers) can also b af f ected These comp nents are to b moved as

f ar f rom the primary an s atered b am as p s ible or a pro riate s ieldin is to b u ed

Care is to b taken that the tester an p wer s p ly are not af fected by s at ered radiation

f rom the b am b fore con u tin tests in a new f aci ty or b f ore con u tin tests with a new

tester setup (in lu in modif ied s ieldin of the tester) To as ure this, the tester is to b

p sitioned an s ielded in exactly the same way as d rin actual tests ex e t for the DUT

that s al b p sitioned outside the b am or s ielded f rom the b am With the b am on an

the DUT s ielded or otherwise not exp sed to the b am, test the DUT Tester setup

verif i ation is s c es f ul if no fai ures are o served Unles otherwise sp cif ied, this tester

setup verifi ation test s al last as lon as a typical test Care s al b taken to prevent upsets

f rom stray sig als or noise in the ca les to the DUT A tester re dines c ec s al b

p rf ormed as p rt of the test seq en e to as ure electrical noise immu ity

4.3 Te t s mple

An typ of integrated circ its with memory can b tested The device p rameters

(ca acitan e of the memory cel in the DRAM, etc.) whic can af f ect the sof t er or rate s al

b wel u dersto d Modern complex devices in lu in a pl cation sp cif i integrated circ its

(ASIC) an f ield programma le gate ar ay (FPGA) can contain more than one typ of

memory These can have very dif ferent radiation upset sen itivities FPGA as an example wi

generaly contain con g ration memory, register (f lip/f lo ) memory an comp site SRAM

memory An ASIC for example general y contain register (f lip/flo ) memory an comp site

SRAM memory It is imp rtant that the distin tion is recog ised b twe n these elements an

e c of the SEE rates determined se arately f or e c typ of memory bit

5 Procedure neutron ir adiated sof t er or test

5.1 Surf ac preparation

The mould comp u d of the DUT do s not ne d to b etc ed of f b cau e the ran e of

neutron b am in the device is s f f iciently lon

5.2 Power s p ly volta e

Unles otherwise req ired, the p wer s p ly voltage s al b the nominal o eratin con ition

sp cif ied f or the device

In order to c aracterize cosmic ray sen itivity as a fun tion of Qcrit ( he minimum c arge

ne ded to upset a memory cel ), lower an hig er voltages are also p rmit ed

5.3 Ambie t temperature

Unles otherwise req ired in an sp cif i ation, the ambient temp rature s al b the nominal

o eratin con ition sp cified for the device

5.4 Core c cle time

The core c cle time is de en ent on the samples u der test (when req ired, the core c cle

time de en en e s al b me s red)

5.5 Data pat ern

This is de en ent on the samples u der test The stru ture of the data p t ern s al b

recorded (a c ec er b ard, al 0/1- e d/write p t ern, etc)

Record the imp ct of data p tern on the o served rates

5.6 Number of me s reme t s mple

Multiple samples s al b me s red to take into ac ou t me s rement variation If test

samples are mou ted alon the b am l ne with samples b hin the f irst sample, the b am

fl en e s al b calc lated at e c test location al owin f or b am aten ation at that sample

The maximum variation in f l x b twe n dif ferent p rts in the sample s al not ex e d 2 %

5.7 Calc lation f or time re uire in the be m

Inf ormation s al b provided on how to calc late, to a sp cif ied ac urac , the amou t of time

req ired in the b am, to o tain the req ired neutron fl en e This wi b b sed on b am typ

an b am fl x Some s ita le neutron b am test f aci ties are in lu ed in An ex B

6 Ev luation

6.1 Me s reme t a d f ai ure rate e timation

The set-up of DUTs an the me s rin s stem are the same as f or the other ac elerated tests

The ef fective SEU cros section σ

efover the sp cif i ran e of energ can b def i ed as

f ol ows:

),(

m ax

m inEEΦN

ma-) is the total f l en e in the energ ran e from E

in

to E

ma (neuton × cm

–2

)

On the con ition that the s a e of the white neutron sp ctrum is close enou h to the f ield

sp ctrum, Sof t er or rate (SER) or sin le event ef f ect rate (SEE rate) can b estimated by

),(

m ax

m inEESER

fie ld

e fffσ

ma) is the neutron fl x in the f ield (neuton.cm

-2

.s-

)

6.2 Determination of MCU a d MBU cros s ctions

These cros section are determined by analy in the er ors in the memory if two or more

er ors oc ur d rin the same re d c cle or in an adjacent re d c cle (se Note b low) then

these are cau ed by the same neuron event The MCU cros section is the n mb r of these

multiple events divided by the f l en e The n mb r of MBU events can b determined by

in p ction, these oc ur when more than 1 bit is upset in the same logical word The neutron

f l x an related event rate s al not b to hig ( he pro a i ty s ould b 10 or more times

smal er than the sin le event MCU pro a i ty) to avoid multiple upset cau ed by more than

one neutron in the same cloc c cle

NOT If th n utro e e t c u e S U in two s p rate word (ie MCU) after th f irst word h s b e re d in th

c cle,th n th two S U wi b id ntif i d in a ja e t re d c cle

6.3 Determination of d vic FIT (e e t rate) f rom cros s ction

Upset data expres ed as FIT/Mbit can b con erted to a cros section in u its of cm²/bit by

u in the con ersion f actor 7,7E-17 Mbit.cm

The f ol owin inf ormation s al b sp cif ied in the a pl ca le proc rement doc ment

(ad itional detai on these req irements is given in An ex A):

a) sample size;

b) vehicle des ription;

c) test des ription;

d) values of relevant test stres con ition f or example voltage, o eratin freq en y;

e) test res lts provided for e c SEE typ tested Cros section data on p r bit or p r device

f or al SEE me s red in lu in SEU, MBU, MCU, SEL, SEFI SET

The inf ormation provided in Clau es A.2 to A.5 may ad itional y b sp cif ied in the a pl ca le

proc rement doc ment (se Clau e 6)

A.2 Description of the beam source

The des ription of the b am source can in lu e the fol owin elements:

1) f aci ty an location, f aci ty contact information, des ription of the source generation,

p rticle typ (neutron, proton);

2) b am energ (mono-energetic) or energ sp ctrum des ription;

3) f ilters u ed, if an (e.g cadmium strip or b rated s ield f or thermal neutron );

4) variation in b am f l x or f l en e d rin testin , in lu in des ription of the monitor

tec niq e or method of estimation;

5) des ription of b am with resp ct to e c DUT, in lu in :

a) b am f l x den ity at DUT;

b) b am are an u iformity of b am acros DUT;

c) DUT orientation to in ident b am;

d) at en atin factors, in lu in a des ription of the typ s an thic nes es for materials

b twe n the tested si con c ip an the b am source if the thic nes is not u iform or if

the u iform thic nes at en ates the energ of the b am in ident on the si con

(in lu es at en ation f rom comp nent p c agin , he t sin s an thermal

en an ements, other tested devices, test f i tures an in lu es eff ects of materials

containin B-10)

A.3 Description of the sample and test vehicle

A.3.1 Sample size

The sample size (n mb r of devices) tested may also in lu e information on the circ its

(ar ay sizes, s an c ain size, etc.) tested on e c device

A.3.2 Ve icle de cription

The vehicle des ription may in lu e the fol owin elements:

1) circ it typ an s b-element (e.g SRAM, DRAM, f lip-f lo master, f lip-flo slave);

2) p c age des ription (e.g con ection to c ip, materials, an ge metries), in lu in an

modif i ation made f or SER testin (e.g non-stan ard he tsin );

3) s p l er p rt n mb r (an die revision, if a plca le);

4) o erational des ription of the circ it;

5) ECC des ription ( yp an coverage) or “ ested p r data s e t if the ECC is u k own”

A.4 Test description

The test des ription may in lu e the f ol owin elements:

1) voltage (external s p ly, u e of internal reg lated, b c bias if a plca le);

NOT 1 Re ortin a intern l re ulate v lta e le el is o tio al b t e c ura e wh re th p rta i ty of th

d ta to oth r d vic s is of intere t

2) test p t ern(s), in lu in logical data p t ern an , if k own, the ph sical data p t ern;

3) f l en e an test d ration;

4) core c cle time or freq en y with sp cial notation of c cle times dif ferent than prod ct

data s e t (f or d namic test or desig ation as “static”;

5) ref res rate, where a pl ca le;

6) temp rature d rin test (at minimum, ambient temp rature; if avai a le, ju ction

temp rature as wel Re ort the me n f or determinin the ju ction temp rature);

7) whic source an energ are u ed, if multiple sources an energies are u ed;

8) tester (commercial model an /or ph sical des ription);

9) pro lems or u u ual b havior of the devices d rin test;

10)f ai inf ormation:

i) cou t of e c er or typ ( ran ient sof t er ors, static sof t er ors, hard er ors);

NOT 2 Be a s te t d ratio s are of te relativ ly s ort, h rd eror o s rv tio s are ty ic ly

e c ptio al Wh re total d s eff ects driv th s erors, th y are te t artifa ts o ly a d th s

o s rv tio s are s e ialy id ntif i d a s c

i ) identif i ation of those soft er ors that are multiple-cel er ors;

i ) electrical sig ature of hard an f irm er ors / faults;

iv) f ai n logical ad res or ad res es;

NOT 3 Interpretatio of multi-c l erors is e h n e b a u d rsta din of th p y ic l relatio s ip of

fai n a dre s s

v) test con ition (voltage, ECC u age, data p tern, etc.) where multiple con ition are

a pl ed within the same test;

vi) f ai ure rate in test con ition Ide l y, the f ai ure rate s al b identif ied on b th a p r bit

(or other circ it element b sis as wel as a p r event b sis At minimum, the b sis for

an given f ai ure rate s al b cle rly identified

1 )p riodicity of test re douts;

12)the me s red SER; where avai a le, it in lu es the sin le-bit an multi-bit comp nents

an a des ription of how the multi-bit comp nent was determined;

13)f or devices inten ed for hars en ironments the electrical sig ature an sen itivity of hard

an f irm er ors / f aults are to b determined u der worst case con ition For example

SEL devices are general y at gre test sen itivity (rate) when s p ly voltage an

temp rature are at their hig est o erating l mits

A.5 Test results

The test res lts may in lu e the fol owin elements:

1) the me s red SEE rate an SEE cros section(s) at the DUT location if DUTs are not

u if orm in their exp s re to the b am (e.g stac ed DUTs, dif ferin distan es from the

b am);

2) where o served, fuly categorize by a) sin le-cel upset, b) multi-cel er ors, c) latc up, d)

ad res or comman er ors, e) upset of ed n an y latc es an provide a des ription of

how these elements were determined;

3) l ne rity whic s ows the f ai ure rate pro ortional ty to the fl x den ity;

4) multiple er ors – demon trate multiple er ors are f rom sin le energetic events

Annex B

(inf ormativ )

White neutron test apparatus

White neutron sp ctrum can b o tained by b mb rdin hig energ proton at a relatively

thic target o typical y, tu g ten or le d, whic simulates ter estrial neutron sp ctrum at the

grou d The b am l ne 4 P3 L at L s Alamos Neutron Scien e Center (LANSCE) is widely

k own an u ed as s c a neutron source As s own in Fig re B.1, the neutron sp ctrum in

the b am l ne is wel defi ed over 1 MeV by the combination of a f i sion c amb r an a TOF

(Time Of Flght s stem The neutron f l x is in the ran e of as hig as 10

6

neutron.cm

-2

.s-

is avai a le u in

a 4 0 MeV proton b am The neutron energ sp ctrum is s own in Fig re B.2, whic was

me s red by a TOF method u in l q id s inti ators A Comp rison of LANSCE (WNR) an

TRIUMF neutron sp ctra with ter estrial neutron sp ctrum is s own in Fig re B.3

The neutron energ sp ctrum of the atmospheric radiation ex e d 10 MeV an f aci ties

u ed for white neutron testin of devices s al have an up er energ ran e of gre ter than

10 MeV to en ure that the b am is re resentative Ad itional neutron b am f aci ties f or

white neutron testin are ANITA, at TSL, Up sala, Sweden maximum energ 18 MeV, NIF at

TRIUMF, Van ou er, Canada maximum energ 4 0 MeV an ISIS, Chipir at RAL, Harwel , UK

maximum energ 8 0 MeV

Figure B.1 – Typic l white ne tron spe tra with

dif f ere t s ield (poly thyle e) thic n s

Figure B.2 – Typic l ne tron spe trum

Ke

Gro n Sp ctrum [latitu e 4 º North]

Figure B.3 – Comparison of LANSCE (WNR) a d TRIUMF

ne tron spe tra with ter e trial ne tron spe trum

108

107

102

103

Annex C

(inf ormativ )

Fai ure rate calculation

C.1 An inf lue ce of sof t er or f or actual semic nductor de ices

C.1.1 Ge eral

Deratin with a con ideration of semicon u tor device an its a plcation u in con ition

may b a pled to extract actual sof t er or rate In An ex C, the deratin method is provided

C.1.2 Duty d rating

Normal sof t er or rate calc lation is done b sed on the p wer on time of the devices

Therefore, its er or rate can b decre sed by in erse pro ortion of the time an this

proced re is cal ed as “Duty deratin ”

For example, sof t er or rate of an a pl cation with 8 hours o eratin in a day b comes one

third (1/3)

In Fig re C.1 device A is an example of a case of non derated con ition with “Duty

deratin = 1” an device B is an example of a case of derated con ition “Duty deratin = 1/3”

Figure C.1 – Sc ematic ima e of duty derating

C.1.3 Uti ity derating

In general, al of memory are is not alway u ed The pro ortion of u ed data are an not

u ed data are in memory is cal ed as “uti ty deratin ” This deratin rate ne d to con ider

the variou situation Those are non-programed are , inef fective data are of programmed

data an inef f ective time rate of programmed data This calc lation may b complcated in the

a pl cation, so averagin is also avai a le

a) Al are u ed “Uti ty deratin = 1 b) Half are u ed “Uti ty deratin = 1/2

Figure C.2 – Sc ematic ima e of memory eff ectiv are for uti ty deratin

C.1.4 Critic l y derating

Critical y deratin is a pro a i ty of seriou nes to generate a critical pro lem in a pl cation

A 1 bit f ai ure in picture data is not detected by h man eye but a 1 bit er or in in u try or

automotive a pl cation mig t cau e a seriou pro lem in social activity Sof t er or in the f irst

case can b decre sed by digitizin its rate A s hematic image of memory ef fective are f or

uti ty deratin is s own in Fig re C.2

C.2 Fai ure rate c lculation including derating

In case these deratin f actors can b a pl ed, ef fective sof t er or rate is extracted as f ol ows

= 104

Uti ty deratin : 0,5 (5 % are u ed)

Critical deratin : 0,2 (seriou nes in the a plcation)

• Ef f ective upset rate of 3 MB with in a pl cation with deratin a ove (NYC eq ivalent

Ef f ective Upset rate (3 MB) in a pl cation = 1 × 10

Bibl ography

[1] IEC 6 7 9-3 , Semic nductor device – Mec a ic l a d climatic test meth ds –

P art 3 : Soft eror te t method for s mic nd uctor d evic s with memory

[2] IEC 6 3 6-4, P ro e s ma a eme t for a io ic – Atmo p eric radiatio efe ts –

P art 4: De ig of hig v lta e aircraft ele tro ic ma a in p tential sin le e e t

efe ts

[3] IEC 6 3 6-5, P ro e s ma a eme t for a io ic – Atmo p eric rad iation efe ts –

P art 5: As e sme t of thermal n utro flux es a d sin le efe ts in a io ic s stems

[4] Th rmal N eutro Sp ctra – J D Dirk, M E Nelson, J F Ziegler, A Thompson an

T.H Za el, Ter e trial Th rmal N eutro s, /IEEE Tran Nu l Sci/ vol 5 , no 6, p

2 6 -2 6 , Dec 2 0

[5] H ig Energ N eutro Sp ctra – J F Ziegler, Tere trial c smic ra inte sitie , IBM

J Res Develo , vol 4 , no.1, p 1 7-13 , Jan 19 8

[6] Bro d e erg ra g (thermal u to hig e erg ) – M S Gordon, P Gold agen, K P

Rodb l , T H Za el, H H K Tan , J M Clem, an P Bai ey, Mea ureme t of th

Flux a d En rg Sp ctrum of Co mic-Ra I ndu ed N eutro s o th Gro nd, IEEE

Tran Nu l Sci vol 51, no 6, p 3 2 -3 3 , Dec 2 0

[8] JESD8 -A M ea ureme t a d Re ortin of Alp a P article a d Tere trial Co mic Ra

-Induc d Soft Errors in Semic nd uctor De ic s

[9] JESD8 -1A Sy tem Soft Error Rate (SSER) Te t M eth d

[10] JESD8 -2A Te t M eth d For Alph Sourc Ac elerated Soft Error Rate

[1 ] JESD8 -3A Te t M eth d for Be m Ac elerated Soft Eror Rate

[12] T.C May an M.H Wo d , A N ew P hy ic l Me h nism for Soft Erors in Dy amic

Memorie , Proce din s of 16th An ual Rela i ty Ph sic Symp sium, 19 8, p 3 -4

[13] C.S Guen er, E.A Wolc i, an R.G Al as, Sin le e e t u s t of dyn mic RAM ’s b

n utro s a d proton , IEEE Tran Nu l Sci vol NS-2 , p 5 4 , 19 9

_ _ _ _ _ _