The SPENVIS SEE Tool

SPENVIS User Workshop

Brussels 2013

The SPENVIS SEE Tool

E. De Donder

Belgian Institute for Space Aeronomy

(BIRA-IASB)



Brussels 2013

Outline

• Introduction

• How to access the SPENVIS SEE tool?

• Input and output

• SEE tool in the Next Generation SPENVIS



Brussels 2013

IntroductionSingle Event Effect (SEE): perturbation of the behavior of electronic (optoelectronic) devices,

circuits and/or systems produced by a single ionizing particle

Parameters to model SEE response of a device:

• Linear Energy Transfer (LET in MeV·cm2/mg) : rate of energy deposit per unit path length ≈ ρ-1·dE(x)/dx

• Cross section (σ in cm2/bit) : probability for SEE = #events/fluence

• Sensitive Volume (SV in µm3) : charge collection region

• Critical Charge (Qc in pC) : min. required charge for SEE



Brussels 2013

Access to SEE package


①①①①

②②②②

③③③③

Environment Definition

LET spec and Upset calculation

I

N

P

U

T

S/C trajectory definition


Brussels 2013

Input 1: Planet/orbit selection

• planet selection: Earth, Mars or Jupiter

• 2 ways to specify the trajectory:

1. Orbit generator

2. Upload trajectory file in SPENVIS format

�http://www.spenvis.oma.be/help/models/sapre_upl.html


Available if advanced level

Available if advanced level


Brussels 2013

Input 2: Environment specification

(see ECSS-E-ST-10-04C for required models)

• Trapped protons in radiation belts

• Solar particles (long, short term; Z=1-92)

• GCR particles (Z=1-92)

� Inclusion of magnetic shielding



Brussels 2013


Select particle source(s)

SEE tool

f(E’) : differential energy spectra at skin S/C

If advanced level

Shielding

�� differential energy spectra inside S/C

If advanced levelmax. 15

x


Brussels 2013

Input 3: Device characteristics

• Device Material: Si: CREME86 or SRIM2008

GaAs: SRIM2008 or Geant4.9

� diff. LET spectrum : f(S) = f(E)[dS/dE]

� int. LET spectrum with Emin = 0.1 MeV/n (before 1 MeV/n)

� sum over all species: composite integral spectrum

�to be used for SEU rate calculation


1.00E-07

1.00E-06

1.00E-05

1.00E-04

1.00E-03

1.00E-02

1.00E-01

1.00E+00

1 10 100

Dif

f. f

lux

(/m

2/s

/sr/

(Me

V/n

))

E(MeV/n)

Shielded (1g/cm2) particle spectra

H '

'He'

'C '

'N '

'O '

'Fe'

1.00E-161.00E-141.00E-121.00E-101.00E-081.00E-061.00E-041.00E-021.00E+001.00E+021.00E+04

1.00E+00 1.00E+02 1.00E+04 1.00E+06

LET

flu

x

LET(MeVcm2/g)

Composite Integral and Differential shielded LET spectra

Int. LET

Diff. LET


Brussels 2013


• SV parameters:

– Library (selection from literature)

– User defined:

• Shape of SV

• Cross section data

(given in report file)


Brussels 2013

Shape SV: (only for direct ionisation rates)

Y

– RPP (X x Y x Z): Z

Input: X, Y and Z or XY and Z X

�from manufacturer or from cs data i.e. X=Y=√σsat

�Z = 2µm (historical) or 1 µm (wc)

– Arbitrary: µµµµm units!!

Input: Top area (A), Total area (S), Volume (V) and

Differential pathlength distribution D(p)


(e.g. GEMAT with geantino’s)

Box, cylinder, L-shape


Brussels 2013


Cross section data:

for direct ionisation:

� Qc (pC) or LETth (MeV·cm2/mg) � Qc = γ·Z·LETth

� Weibull: W, LETth, S, σlim (�XY=√σlim)

� Cross section table: σ(cm2/bit) vs. LET (MeV·cm2/mg)

for proton induced:

� Bendel: A, B

� Weibull: Wp, Eth,p, Sp, σlim,p

� Cross section table: σ(cm2/bit) vs. E (MeV)

� PROFIT: W, LETth, S, σlim, T (=90°= wc) (only heavy ion data available)


Brussels 2013


0.00E+00

5.00E-03

1.00E-02

1.50E-02

2.00E-02

0.00 5.00 10.00 15.00 20.00 25.00 30.00 35.00 40.00

σσ σσ(c

m2/b

it)

LET (MeVcm2/mg)

Weibull fit to experimental data

LETth


Brussels 2013


+

(Option to ignore )


Brussels 2013


Rate calculation: upset if Edep>Ec (or Q>Qc or LET>LETth)

• Proton induced ionization:

Rate ̴ ∫ cross section data x diff. particle fluence

• Direct ionisation:

Rate ̴ ∫ SV data x pathlength distribution x (LET or ion) flux

algorithm: Default

� CREME (Constant LET) : Edep = ρ·LET·p with LET constant in SV

(�ok for long-range ions)

� Variable LET : Edep = ρ·∫LET(E)dp

� Slowing and Stopping : Edep = E – R-1(R(E)-p) � ion flux instead of LET flux

(precise formula’s: http://www.spenvis.oma.be/help/background/creme/creme.html#SEU)


Brussels 2013

Output


SEU rate along the ISS orbit

Shielded LET flux

Shielded p+ spectrum


Brussels 2013


worldmap Time plot

Mission total/Segment averaged SEU rate /bit or (bit/(s or day))

or total mission number (Long-term)


Brussels 2013


Petersen 2011 (126 comparisons of predicted and observed SEU

rates from 23 satellites):

“the methods of upset calculation are satisfactory”

But predictions may be wrong because of :

• Poor environment prediction e.g. dynamic nature of

radiation belt

• Part to part variation: part location, different lot, …

• Poor choice of device depth

• Insufficient test data

• Incorrect shielding distribution

• …


Brussels 2013

SEU tool in Next Generation SPENVIS


• New environment models e.g. AP9/AE9

• Link to Geant4 tools outputs:

• Realistic shielding distribution

• New materials and LET

• Inclusion of secondary particles

• Extension of device library:

• Inclusion of user defined devices � available in ≠ projects + sharing with

other users

• Link to ESCIES database

• Interface to framework for IC designers produced by the

“Evaluation of the Radiation Effects on Deep Sub Micron

CMOS Technologies” project

• Suggestions from YOU


Brussels 2013

Bibliography


• ECSS-E-HB-10-12A (space environment)

• ECSS-E-ST-10-04C (calculation of radiation and its effects and

margin policy handbook)

• Single Event Effects in Aerospace, E. Petersen (2011)


Brussels 2013

Back up slides



Brussels 2013


• CREME 86/96:

• Slowing down and stopping:

• Variable LET

( )( ) ( )

( )

min

0

0

min

min

min

E

dss

dssEs

sEs

s

=

⋅℘

⋅∆⋅℘

=∆

∫

∫

Documents

The SPENVIS SEE Tool