Naval Research Laboratory "d

Washlngton, DC 20375-6000 NRL Memorandum Report 6001 December 31, 1908

to

Cosmic Ray Effects on Microelectronics, Part IV

JAMES H. ADAMS, JR.

Gamma and Cosmic Ray Astrophysics Branch

Space Science Division

DTIC
St
ELECTE

FB 09 1987U

Approved for public release; distribution unlimited.

.200
. ,S,•,
• . , - • •,-,. •,.,.•,. ,....•.e ," , . , ., ,, ,,,, , • , , ,, -. . •, , " ",", . "• .• -
 SURITY CLASSIFICATION OF THIS PAGE ..... 4/"rL
4 ..
REPORT DOCUMENTATION PAGE
I&. REPORT SECURITY CLASSIFICATION lb RESTRICTIVE MARKINGS
UNCLASSIFIED
2a. SECURITY CLASSIFICATION AUTHORITY 3. DISTRIBUTION/AVAILABILITY OF REPORT

2b. DECLASSIFICATION /DOWNGRADING SCHEDULE Approved for public release; distribution

unlimited.
4. PERFORMING ORGANIZATION REPORT NUMBER(S) S. MONITORING ORGANIZATION REPORT NUMBEP(S)
NRL Memorandum Report 5901
Ea. NAME O PERFORMING ORGANIZATION 6b. OFFICE SYMBOL 7a. NAME OF MONITORING ORGANIZATION
Naval Research Laboratory (if applicable)

Sc. ADDRESS (City, State, and ZIPCode) 7b. ADDRESS (City, State, and ZIP Code)

Washington, DC 20375-5000
Ba.ORGANIZATION
NAME OF FUNDING/SPONSORING 8b. OFFICE SYMBOL 9. PROCUREMENT INSTRUMENT IDENTIFICATION NUMBER
(if applicable)
Office of Naval Technology
Sc. ADDRESS (City, State, and ZIP Code) 10. SOURCE OF FUNDING NUMBERS
PROGRAM PROJECT TASK WORK UNIT
Arlington, VA 22217 ELEMENT NO. NO. XF1ý142 NO. IACCESSION NO.
62712N I100
11. TITLE (Include Secu rty Classification)
Cosmic Ray Effects on Microelectronic, Part IV
12. PERSONAL AUTHOR(S)
Adams, James H., Jr.
13a. TYPE OF REPOKT 13b. TIME COVERED 14. DATE OF REPORT (Year, Month, Day) 1S. PAGE COUNT
Final FROM 10183 TO 10186 1986 December 31 114
16. SUPPLEMENTARY NOTATION

17. COSATI CODES 18. SUBJECT TERMS (Continue on reverse if necessary and Identify by block number)
FIELD GROUP SUB-GROUP Soft upsets) -Cosmic rays, Geomagnetic cutoff)-,/
Soft errors, Propagation, Single event upsets,,
Microelectronics)
19. ABSTRACT (Continue on reverse If necessary and identi(y by block number)

''Single Event Upset (SEU) vulnerability must be considered in the. design of all modern digital electronic
systems for use in space. This has resulted in a need to accurately estimate the SEU rate for each type of digi-
tal microelectronic component used in the electronic system of any spacecraft on any orbit.

This report presents a method of mating such estimates. This method, considers the effects of device
parameters, material shielding, orbital parameters, interplanetary weather conditions and solar activity.

' /'

20. DISTRIBUTION /AVAILABILITY OF ABSTRACT ' 21. ABSTRACT SECURITY CLASSIFICArION

I9 UNCLASSIFIEDIUNLIMITED 1" SAME AS RPT. DTIC USERS UNCLASSIFIED
22a. NAME OF RESPONSIBLE INDIVIDUAL 22b. TELEPHONE (Include Area Code) 22c. OFFICE SYMBOL
J•.,a,• H. Adams, Jr. (202) 767-2747 Code 4154
DO FORM 1473, 84 MAR 83 APR edition may be used until exhausted. SECURITY CLASSIFICATION OF THIS PAGE
All other editions are obsolete
*UL Gownv.mmtftbltt Ottier 1111111 0417

%--v %~
 CONTENTS0

I
1.0 IN T R O D U C T IO N .......................................................................................................................
2
2.0 REVISIONS AND EXTENSIONS OF PREVIOUS WORK ........................................................
2
2.1 Update of the near-earth particle environment ....................................................................
4
2.2 Extension of the geomagnetic cutoff transmission function ................................................
4
3.0 RADIATION TRANSPORT THROUGH THE WALLS OF THE SPACECRAFT ....................
.5
4.0 CALCULATION OF THE LET SPECTRUM .............................................................................
THE DIRECT IONIZATION OF
5.0 CALCULATION OF SEU RATES FROM
6
C H A RG ED PA RTIC LES .........................................................................................................

6.0 CALCULATION OF SEU'S RESULTING FROM NUCLEAR REACTIONS,

8
C A U SED BY PRO TO N S .........................................................................................................
8
7.0 DESCRIPTION OF THE CREME COMPUTER PROGRAMS ..................................................
9
7.1 How to run the CREM E program s ............... :.......................................................................
9
7.2 The m ain CREM E program s ...............................................................................................
10
7.3 A uxiliary C R EM E program s ................................................................................................
I.'.. <V
II
7.4 C R EM E subprogram s ..........................................................................................................
12
7.5 Options and parameters in the CREME programs ...............................................................
13
8.0 DISCUSSION OF RESULTS OBTAINED WITH THE CREME PROGRAMS .........................
19
9.0 R E FE R EN C ES ............................................................................................................................

APPENDIX 1: THE ANALYTIC MODEL FOR THE CHARGED

36
PARTICLE ENVIRON M ENT ..........................................................................................
46
APPENDIX 2: THE GEOMAGNETIC CUTOFF TRANSMISSION FUNCTION ............................
......... 49
APPENDIX 3: PARTICLE RANGE VS. ENERGY FOR VARIOUS IONS IN ALUMINUM
53
APPENDIX 4: PARTICLE STOPPING POWER VS. ENERGY IN ALUMINUM ...........................
57
APPENDIX 5: PARTICLE STOPPING POWER VS. ENERGY IN SILICON ..................................

APPENDIX 6: THE WORLDWIDE VERTICAL GEOMAGNETIC CUTOFFS

60
AT 20 KM A LTITU D E ....................................................................................................
69
APPENDIX 7: SOURCE CODES OF THE CREME PROGRAMS ...................................................
 COSMIC RAY EFFECTS ON MICROELECTRONICS
PART IV

1.0 INTRODUCTION

This report is intended to provide some of the tools needed to estimate

single event upset (SEU) rates. An SEU occurs when a single ionizing
radiation event produces a burst of hole-electron pairs in a digital
microelectronic circuit that is large enough to cause the circuit to change
state. In space, these unplanned changes of state usually result from the
direct ionization of single heavy ions originating outside the spacecraft.
SEU's may also be caused by the products of nuclear reactions initiated by
particles that originated outside the spacecraft.

SEU's are causing operational difficulties of various kinds on more than a

dozen spacecraft. Obviously, vulnerability to SEU's must be considered in the
design of future spacecraft. In order to assess the vulnerability of any
proposed design, engineers must have a reliable means of estimating the SEU
rates in the radiation environments that can be expected during the mission.
This requires experimental measurements and circuit modeling to establish the
parameters that determine the SEU sensitivity of each device in the circuit.
It also requires a reliable model of the near-earth particle radiation
environment.

The first report in this series (Adams et. al., 1981a) provided a model of
the near-earth particle environment at the orbit of the earth, but outside the
magnetosphere. The second report (Adams et. al., 1983) described a method for
computing the orbit-averaged geomagnetic cutoff transmission function that is
used to modulate the environment outside the magnetosphere to any spacecraft
orbit. The third report of the series (Tsao et. al., 1984) described the
radiation environment high in the earth's atmosphere.

This report first describes some improvements in the model of the

near-earth particle environment and extends the method of calculating the
geomagnetic cu off transmission function to elliptical orbits. Next, the
Smethod of transporting the radiation environment through the walls of the
spacecraft is presented. The report goes on to describe the transformation
from energy spectra to LET spectra and the computation of single event upset
rates caused by the direct ionization of particles originating outside the
spacecraft. Several examples of SEU rate calculations are given and the SEU
rates due to direct ionization are compared with those that result from the
products of nuclear reactions caused by particles that originate outside the
spacecraft. Computer programs are given that allow the user to compute the
SEU rate of any device inside any spacecraft on any orbit.

Manuscript approved November 6, 1986.

I
2.0 REVISIONS AND EXTENSIONS OF PREVIOUS WORK

Our earlier work on cosmic ray effects on microelectronics has been

revised and extended as described in the following sections.
2.1 Update of the near-earth particle environment
S.i.nce the publication of the first report in this series (Adams et. al.,
1981a), new data on the near-earth particle environment has been published.
Using these results and others, the model presented in the first report has
been improved and extended (see Appendix 1).
We used the data from the French-Danish experiment on HEAO-3 (Engelmann
et. al., 1983, and Juliusson et. al., 1983) to improve the fits to the
galactic cosmic ray data for the elements up to nickel.

Figures 1 and 2 give the new fits.to the helium and iron differential
energy spectra for galactic cosmic rays. The lower solid curves in each
figure are fits to the pure galactic cosmic ray spectra at solar maximum.
This is the mildest radiation environment in the 11-year solar cycle. As can
be seen in figure 2, no data have yet been published on the Fe spectrum at
solar maximum. The upper solid curves are fits to pure galactic cosmic ray
spectra at solar minimum. The galactic cosmic ray flux is highest at the
minimum of the solar activity cycle. The dashed curves are spectra that
include contributions from co-rotating interaction regions, the anomalous
component, and small solar flares. These spectra are chosen to present fluxes
that are so high at each energy that flur, levels exceeding these occur only
10% of the time. The new fits should predict the absolute cosmic ray flux
within a factor of two. This uncertainty arises primarily from our inability
to predict future levels of solar modulation. The new fits predict the
relative abundances of the elements below nickel to ±15%, on average, at the
energies where the particle fluxes are most intense.

The model was extended to include all the elements up to uranium. This
was done in the following way: The data from the HEAO-3 Heavy Nuclei
Experiment (Binns et. al., 1981; Binns et. al., 1982, and Binns et. al., 1983)
provided measurements of the elemental abundances for the elements heavier
than nickel. These measurements did not, however, resolve the individual
elements. It was necessary to determine the individual elemental abundances,
in most cases, by matching a cosmic ray propagation calculation (for an
example, see Tsao et. al., 1981) to the HEAO measurements. The cosmic ray
source composition was initally taken to be the general abundance of elements
(Cameron, 1980) and then the source composition and propagation conditions
were adjusted to reproduce the measured abundances at earth. The "arriving"
elemental abundances at earth from this propagation calculation were taken as
the correct elemental abundances for galactic cosmic rays.

The description of the contribution from the anomalous component (Adams

et. al., 1981) has been extended to allow- investigation of the effect it would
have on spacecraft, were it to be singly-ionized. This has been done by
independently modeling the anomalous component spectra, based on the data of
Webber et. al. (1979). Both the singly-ionized anomalous component and the
galactic cosmic ray component of each particle spectrum must be modulated to
the orbit-averaged geomagnetic cutoff (Adams at. al., 1983) for the spacecraft
according to their different magnetic rigidities at the same energies. The
two modulated components are then combined to form the orbit-averaged
differential energy spectrum incident on the skin of the spacecraft.

2
 Proton differential energy spectra for three solar flare modlels were
presented in Adams et. al. (1981a). The ordinary and worst case flare models
were for large flares, .e., those Pith one week integral proton fluxes above
10 MeV exceeding 2.5x10 protons/ca . These models were based on fits to the
log-normal distributions of fluxes above three energy thresholds. We have
revised these fits by extending the data base for integral proton fluxes above
10 MeV through April, 1984 using GOES satellite data (Coffey, 1984). This
allowed the amplitudes of the model spectra to be adjusted. See eqs. 33 and
34 of Appendix 1 for details.

The solar flare particle event descriptions in the original model (Adams
et. al., 1981a) used the proton spectra, scaled by the heavy ion to proton
ratios, to obtain the spectra of heavier ions. These ratios have been revised
on the basis of additional data on solar energetic particle composition
(Mewaldt, 1980; Zinner, 1980; Briggs, 1979; and Van Hollebeke, 1975). They
have also been extended to include the spectra for all the elements heavier
than nickel. The mean composition was extended using the data on the general
abundances of the elements from Cameron (1980). To get the worst-case
composition for these same elements, we used the charge-dependent enhancement
factor suggested by Dietrich and Simpson (1978), and assumed that it could be
extended to uranium. The only data on heavy ion enrichment beyond zinc comes
from tracks in meteoritic and lunar olivine samples (Goswami et. al., 1980).
These data show an even greater heavy ion enrichment than predicted by the
formula we used here. We will, however, continue to use the Dietrich and
Simpson formula until solar flare composition measurements, made with
detectors designed for this purpose, become available. In this report, we
have dropped the descriptions of the mean flux from various flares given in
Adams et. al. (1981), because only instantaneous flare fluxes are relevant to
the SEU problem.

We also introduced another flare spectral model (Adams and Gelman, 1984).
This model is intended to display all the worst features of all the recorded
flares in the last thirty years. Lingenfelter and Hudson (1980) investigated
the statistical data on solar flare particle fluxes. They concluded that
there is a maximum size for solar particle events and that events larger than
those thus far observed are very rare. We therefore will assume that our
composite worst-case flare model is the most extreme event that can occur. As
such, it presents the kind of envirorunent that spacecraft should be designed
to endure if they are intended to function without interruption.
The updated model is given in Appendix 1. We have compared this model
with the results of Chenette and Dietrich (1984). We find that its
predictions are in agreement with the results reported by these authors. This
updated model should be compared with Appendix 1 of Adams et. al. (1981a) to
see the detailed changes.

In recent years, evidence has been accumulating that solar energetic heavy
ions may not be fully ionized. This is certainly the case in the 0.5 < E <
2.5 MeV/u energy range, as has been shown by A. Luhn et al. (1984). Ai thti
higher energies of interest here solar energetic heavy ions may not be fully
ionized (as has been generally assumed up to now). Although heavy ions with
energy >10 MeV/u would be fully ionized by passing through sufficient matter,
the available data only place an upper limit their pathlength in matter
(Mewaldt and Stone, 1983).

3
 Fischer et al. (1984) report evidence that solar energetic heavy ions in
the energy range 5 < E < 20 MeV/u are not fully ionized. These authors report
upper limits on the charge to mass ratio of heavy ions as low as 0.1 ( this
ratio is -0.5 for fully ionized heavy ions). Breneman and Stone (1985) have
obtained indirect evidence that solar energetic heavy ions in the energy range
3.5 to 50 MeV/u have the same distribution of charge states as measured for
0.5 to 2.5 MeV/u ions by Luhn et al. These authors have shown that the
systematic abundance variations in SEP heavy ions compared to solar coronal
abundance can be understood if the change state distributions measured by Luhn
et al. are assumed for these higher energy heavy ions.

The SEP model described in this report assumes that the SEP heavy ions are
fully ionized. This assumption may be incorrect from the evidence discussed
above. If this is the case, the SEU rates due to SEP's will be systematically
underestimated for spacecraft in low earth orbit, because geomagnetic
shielding will not be as effective as the present model assumes.

Under the present circumstances, the charge state of SEP's is uncertain,

so it's not clear how the model for SEP's should be altered to account for the
SEP charge states. Therefore, we recommend continuing to use the present
model. A conservative calculation can always be made by neglecting the
protection afforded by the geomagnetic cutoff (assuming the geomagnetic cutoff
transmission function is 1.0 for all energies).
2.2 Extension of the geomagnetic cutoff transmission function

The technique for calculating the geomagnetic cutoff transmission function

described in Adams et. al. (1983) has been extended to elliptical orbits. See
Appendix 2 for the details.
3.0 RADIATION TRANSPORT THROUGH THE WALLS CF THE SPACECRAFT

The orbit-averaged differential energy spectrum of each element at the%

skin of the spacecraft can be found using the model spectra from Appendix 1,
and modulating them to the orbit of spacecraft, using the geomagnetic cutoff
transmission function calculated by the method described in Adams et. al.
(1983) and Appendix 2. To find the differential energy spectra at the
microelectronic components inside the spacecraft, a transport calculation must
be performed. This is best done by the method of Adams (1983). This method
takes into account the effects of energy loss and particle losses through
total inelastic collisions. No account is taken of the way in which the
projectile fragments from these collisions contribute to the differential
energy spectra of lighter ions. This omission results in a systematic
underestimate of the particle fluxes, especially for the elements from argon
through manganese. This underestimate is so small, it can be neglected for
shielding thicknesses typically found in spacecraft.

14 The differential gnergy spectrum, f(E), inside the spacecraft and behind a
thickness, t (in g/cm of aluminum or equivalent) of shielding is,
f(E) = f'(E')[S(E')/S(E)]exp(-at), and (1)

c ([5X10-2 6
1(A 1 / 3 + 271/3 _0.4)2]/27. (2)

I~
Where:
f'(E') is the differential energy spectrum at the skin of the spacecraft;
. R-1
E' is the energy at the skin of the spacecraft, i.e., E' R 1R(E) + -1
t], where R(E) is the residual range of an ion having an energy E and R
is the inverse function of R(E);

E is the energy inside the spacecraft;

S(E) is the stopping power of an ion having an energy E;

A is the atomic mass of the ion; and

n is Avogadro's number.

The range-energy and stopping power data come from Adams et. al. (1987). The
tables needed for aluminum are reproduced in Appendices 3 and 4. Equations
(1) and (2) give estimates of the differential energy spectra inside
spacecraft that are satisfactory for estimatig SEU rates provided the
shielding thickness does not exceed 50 g/cm . For greater shielding
thicknesses, the method of Adams (1983) seriously underestimates the SEU rate.
For further discussion of this point, see Adams (1983). Precise estimates can
be made by an exact transport calculation, following the methods of Tsao et.
al. (1984).

4.0 CALCULATION OF THE LET SPECTRUM

The 'next step is to calculate the linear energy transfer (LET) spectrum
from the differential energy spectra at the microelectronic components inside
the spacecraft. LET is the rate at which energy is deposited per unit
pathlength of an ionizing particle. For the present purpose, it is equivalent
to the rate of energy loss per unit pathlength for the same particle, i.e.,
dE/dx or stopping power. Figure 3 shows the stopping power, S, versus energy
(taken from Adams et. al., 1987) for several ions in silicon. Appendix 5
gives these same data in tabular form.
The transformation from a differential energy spectrum to a differential

LET spectrum is simply,

f(S) = f(E)KdE/dS]. (3)

By examining Figure 3, it can be seen that Eq. 3 has three and sometimes four
singularities, at the points where dS/dE becomes zero (and therefore dE/dS
becomes infinite). When the transformation is done on a computer, the ratio
of finite differentials is used to approximate dE/dS and if the boundaries of
the intervals in E and S are properly chosen, the ratio can be kept finite.

To see an example of how this is done, consider the three iron

differential energy spectra in Figure 4. The lowest is pure cosmic ray iron
at solar maximum, the middle curve is pure cosmic ray iron at solar minimum,
and the upper curve is the 90% worst case environment spectrum for iron. Each
of these spectra is behind 0.025 inches of aluminum shielding and outside the
magnetosphere, so there is no geomagnetic cutoff. Figure 5 shows the result
of the transformation of these spectra to differential LET spectra. Three
spikes can be seen in these spectra, corresponding to the three singularities
in dE/dS. These singularities make differential LET spectra awkward to handle
on a computer. We prefer, instead, to use integral LET spectra. Figure 6
shows these same three spectra in integral form.

The final step is to repeat the application of Eq. 3 to the differential

energy spectra for all the elements in cosmic rays (i.e., protons to uranium)
and sum the resulting integral LET spectra to form one composite integral LET
spectrum. This spectrum can then be used to estimate SEU rates that result
from the direct ionization of charged particles.

5.0 CALCULATION OF SEU RATES FROM THE DIRECT IONIZATION OF CHARGED PARTICLES

An SEU occurs when a sufficiently large burst of charge is collected on a

critical node in one of the digital microcircuits on a chip. The minimum
charge required to produce an SEU is called the critical charge. This burst
of charge can come from a segment of the ionized trail left by the passage of
an intensely ionizing particle. It is assumed that each critical node is
surrounded by a sensitive volume (idealized as a rectangular parallelepiped)
and that the charge deposited in this volume gets collected (Pickel and
Blandford, 1980). The dimensions of the sensitive volume are related to those
of the critical node. The sensitive volume is not, however, Just the
dimensions of this feature. Charge is also collected from the silicon
surrounding the node by diffusion. The efficiency of charge collection from
beyond the node falls off with distance. Pickel and Blandford (1980) discuss
how the critical charge and the dimensions of the sensitive volume are found.
In general, experimental measuraments of the operational SEU cross section for
the device, as well as design data supplied by the manufacturer, are required.
It is often necessary to interpret these data using detailed circuit modeling
computer programs before the device SEU parameters can be determined (see
Price et. al., 1981, Zoutandyk, 1983, and Zoutendyk et. al., 1984, for
example).

This simple model, described above, predicts that when the critical charge
has been collected, an SEU will occur. The amount of charge collected depends
linearly on the LET of the ionizing particle and the length of its path within
the sensitive volume. There is, however, another effect that extends the size
of the sensitive volume. The intense trail of ionization left by the charged
particle alters the electric field pattern in the neighborhood of the feature.
The field forms a funnel along the particlets track and this enhances the
efficiency with which charge is collected (Hsieh et. al., 1981). This funnel
effect can be partly accounted for in the simple model discussed above if the
dimensions of the sensitive volume are experimentally determined. Several
attempts have been made to model the funnel effect explicitly (see McLean and
Oldham, 1982; Messenger, 1982; and Oldham and McLean, 1983). The model of
Oldham and McLean (1983) appears to correctly describe the experimental data
qualitatively, and is in quanitative agreement for lightly ionizing particles.
This model predicts that the collected charge will vary with LET to the 4/3
power and that SEU sensitivity may depend on the charge collection time, that
in turn depends on LET. These are the practical differences between the model
of Mclean and Oldham (1983) and the Pickel and Blandford model. Because
detailed models that explicitly include the funnel effect have not yet
"matured", the simpler model of Pickel and Blandford will be used here.

6
 This method for estimating soft upset rates due to the direct ionization
by particles originating outside the spacecraft is given below in the form
presented by Adams (1983). The upset rate, Ne, in upsets/bit-sec is:

Lmax

Ne = 22.5,rA Qcrit D[p(L)]F(L)/L 2 dL, (4)

22.5Q crit/pmax

where,
A is the surface area of the sensitive volume in m

Q i"is the minimum charge required to produce an upset, in

pigooulombs,
r, 2
L 1.05xi05 MeV cm /g, the highest LET any stopping ion can
dmefver,
in g/cm2
Pmax is the largest diameter of the sensitive volume
L is LET in MeV cm2 /g,

F(L) is the integral LET spectrum in Particles/m22 ster-sec,

and
D~p(L)] is the differential pathlength distribution in the sensitive
volume of each memory cell in cm /g,

where,

p(L) = 22.5 Qcrit/L is the pathlength over which an ion of LET, L, will
produce a charge Qcrit The constant 22.5 is the conversion from picocoulombs
to MeV, assuming 3?. v per hole-electron pair.
Equation (4) contains the implicit assumption that the LET of each ion is
essentially constant over the dimensions of the critical volume. Of course,
this is not true for stopping ions very near the end of their range. Eq. (4)
will assume that the maximum LET of the stopping ion applies over its entire
pathlength in the sensitive volume. This assumption can result in calculated
energy depositions that exceed the residual energy of the ion. The problem is
especially acute for large sensitive volume dimensions, and threshold LET
values just below the maximum LET of an ion that is much more abundant than
all heavier ions. Fortunately, this circumstance arises rarely. Eq. (4) will
be accurate if the flux of stopping ions is small compared to fast ions having
the same LET. Care should be taken in the use of this formula when the
threshold LET is just below the edge of a "cliff" in the integral LET spectrum.

Eq. (4) assumes one continuously sensitive critical node per bit. In
general, there may be several critical nodes per bit, each with its own
sensitive volume dimensions and critical charge. In addition, these nodes may

~~X,
only be sensitive part of the time, making it necessary to calculate partial
upset rates for each node and then combine the results, weighted by the
fractional "live" time of each node. The reader is referred to Nichols et.
al. (1983) for a collection of test data and to Pickel and Blandford (1980),
Price et. al. (1981), Kolasinski et. al. (1983), Zoutendyk (1983), and
Zoutendyk et. al. (1984) for examples of how device SEU parameters can be
determined.

6.0 CALCULATION OF SEU'S RESULTING FROM NUCLEAR REACTIONS, CAUSED BY PROTONS

The intensely ionizing particles that cause an SEU can be the fragments of
a silicon nucleus struck by a particle originating outside the spacecraft.
The most common particle in space capable of causing such a nuclear reaction
is the proton. At present, the most practical method for calculating
proton-induced SEU's is to mersure the operational SEU cross section at one
proton energy and then use the method of Bendel and Petersen (1983) to find
the SEU rate in any proton envirorment. Bendeý and Petersen give the SEU
operational cross section in upsets per proton/cm per bit as,

a = 1x10 12 (24/A) 1 4 [1-exp(-O. 18YO' 5 )] 4 (5)

where,
Y = [(18/A) 0 ' 5 ](E-A), (6)

where E and A are in MeV.

Proton differential energy spectra can be obtained from Sawyer and Vette
(1976) for the trapped radiation, and Appendix 1 of this report for the
exomagnetospheric components. The exomagnetospheric components can then be
modulated to any orbit by the method discussed in Adams et. al. (1983) and
Appendix 2. The combined proton differential energy spectrum incident on the
skin of the spacecraft can then be propagated to the electronics inside using
a simplified version of Eq. 1, given below:
f(E) = f'(E')[S(E')/S(E)] (7)
The final step is to integrate the cross section over the proton differential
energy spectrum, i.e.,

R Z(IXI0"4 ) 41r f(E)c(E)dE (8)

0
where f(E) is in protons/(m2 ster.sec. MeV) and R is in upsets per bit-sec.

7.0 DESCRIPTION OF THE CREME COMPUTER PROGRAMS

The CREME programs are a group of FORTRAN routines that calculate

differential and integral energy and LET spectra of cosmic rays incident on
the electronics inside any spacecraft in any earth orbit and the single event
upset rates that result. Input parameters for running these programs describe

.
' -' ., ,- ,'. . •V % -... ,. AN %
-'Y!
the interplanetary and magnetospheric weather conditions, the spacecraft's
orbit, the shielding surrounding the electronics, and the characteristics of
the device under consideration. Input data files contain tabulations of
stopping powers and ranges of cosmic ray nuclei in aluminum and silicon, and
geomagnetic cutoffs. The program's output files containing energy and LET
spectra, and single event upset rates. Descriptions and instructions on how
to run the CREME programs are given below. The various options given by the
programs are described here following the descriptions of the programs
themselves. Listings of all of the CREME programs are given in Appendix 7 of
this report.

7.1 How to run the CREME programs

First, compile all the programs and subroutines, and link the programs
with all appropriate subroutines. (See Sections 7.2 and 7.3 to see which
subroutines must be linked with each program). For each run of the CREME
programs, follow the below instructions:
1. Decide all the options you want in your run. These deal with: your
spacecraft's orbit; which cosmic ray environment to consider; material
shielding surrounding your device; and specifics about the device. (See
Section 7..5 for details).

2. If you wish to consider geomagnetic cutoff and trapped proton effects

(see Section 7.5), then run the two auxiliary programs STASS and GEOMAG2 (see
Section 7.3).
3. To get differential or integral cosmic ray flux for any element versus
particle energy, run SPEC (see Sectilon 7.2).
4. For differential or integral cosmic ray flux versus particle LET, run
LET (see Section 7.2).

5. To find upset rates from nuclear reactions caused by protons, run

BENDEL (see Section 7.2).

6. To find upset rates caused by direct ionization, you must run two
programs. First run LET to get an integral LET spectrum; then run UPSET to
get the upset rate (see Section 7.2).
7.2 The main CREME programs
The main CREME programs are SPEC, LET, BENDEL, and UPSET. SPEC and LET
produce spectra; UPSET and BENDEL produce upset rates.
SPEC outputs the differential and/or integral energy spectra for a single
element. SPEC options are: name(s) of output file(s); number of data points
in output file(s); the atomic number of the ion whose spectra you want; the
year; whether to include geomagnetic cutoff and trapped protons; the
interplanetary weather index; and the thickness of material shielding. SPEC
must be linked with the subroutines INSIDE, CUT, CRF, DEDXAL, and RAL. if
geomagnetic cutoff and trapped protons are included, the intermediate data
files GTRANS. DAT and STASS.DAT (created by the auxiliary programs GEOMAG2 and
STASS) must be present. The output files of SPEC are in two columns: the

.-
 first column is differential2 flux in particles/(m2 .ster.sec)/( VeV/u) or
integral flux in particles/(m .ster.sec); the second column is energy in
MeV/u.
LET outputs differential and/or integral LET spectra from a range of
elements. Unlike SPEC, which simply tabulates the results of its subroutines,
LET must reparameterize the fluxes. Using stopping power functions, LET
transforms the energy spectra into LET spectra. The LET program has the same
options as SPEC. In addition, the range of atomic numbers of elements to be
included in the LET spectre is required. LET must be linked witn the
subroutines INSIDE, CUT, CRF, DEDXAL, RAL, and DEDXSI. As with SPEC, the
intermediate data files GTRANS.DAT and STASS.DAT are needed if cutoff and
trapped protons are to be included. The output files of LET have the same
format as t ose of SPEC, but w~th different units: differential flux iý
perticles/(m .ster.sec)/(MeV.c /g) or integral flux in particles/(m
.ster.sec); versus LET in MeV.cm /g.

BENDEL outputs the upset rate of a microelectronic device due to nuclear

reactions caused by protons. The program integrates the formula of Bendel and
Petersen (1983) over the differential energy spectrum of protons. BENDEL
options are: number of points used to calculate the proton energy spectrum;
the year; whether to include geomagnetic cutoff and trapped protons; the
interplanetary weather index; the thickness of material shielding; and either
Bendel's 2 parameter A or a measured upset cross-section, in upsets/(bit
proton/cm ), at some energy, in MeV. BENDEL uses the same subroutines and
data files as SPEC. It outputs upsets/(bit.sec) and upsets/(bit.day).

UPSET is a more complicated program that calculates the upset rate from
the direct ionization of particles originating outside the spacecraft. UPSET
assumes that each bit is stored in a microcircuit which has one continuously
sensitive node. This node is surrounded by a sensitive region which is
assumed to have the shape of a rectangular parallelepiped. To calculate the
upset rate using UPSET, one must first use LET to create an integral LET
spectrum. UPSET then integrates this spectrum over the differential
distribution of pathlengths through the sensitive region of the one sensitive
node in the microcircuit to obtain the SEU rate. UPSET's input parameters
are: the critical charge required to produce an SEU in the device; the linear
dimensions of the parallelipiped representing sensitive region of a
microcircuit;
spectrum. and the name of the file containing
UPSET must be linked with the subroutine DIFPLD.the input integral LET
UPSET outputs

upsets/(bit.sec) and upsets/(bit.day).

7.3 Auxiliary CREME programs

STASS provides input data to LET, SPEC, and BENDEL on the trapped proton
flux, averaged around the spacecrafts' orbit. It is a simple program that
transforms unformatted hand input into formatted output for use as input data
in CREME main programs. Data for STASS should come from the "Averaged Differ.
Flux, #/om**2/sec/keV" column of "Composite Orbit Spectrum" tables. These are
typically included in a National Space Science Data Center report such as
Stassinopoulos (1981) or Stassinopoulos (1982). STASS needs no subroutines or
Sinput data files. It outputs a two-column file: the first ýaolumn is energy
in MeV; the second column is differential flux in protons/(cm sec)/keV. This
file must be renamed to STASS.DAT before it can be used as input. (The

!0

%''? d N8 % *
subroutine CRF converts the units in this file to the units used in the main
programs). Note that the trapped proton contribution to SPEC, LET, or BENDEL
can be omitted by creating a STASS.DAT file with zero fluxes.

GEOMAG2 is a long, complicated program that computes the geomagnetic

cutoff transmission function corresponding to a given orbit. The program
traces the orbit in small time steps for two days of the orbit, calculating
the position and then the cutoff 200 times. These cutoffs are integrated into
a "transmission function", see Adams et. al. (1983) and Appendix 2 of this
report for details on this process. GEOMAG2 options are Just those parameters
of the orbit that are mentioned in Section 7.5. They should describe the
orbit corresponding to the STASS.DAT file being used. GEOMAG2 does not need
to be linked with any other program files; it uses the input file CUTOFF.DAT,
a table of cutoffs at 20 km altitude (see Appendix 6). GEOMAG2 outputs a
two-column data file: the first column is cutoff in GeV/ec; the second column
is the transmission function, which varies from 0 to 1. This file, called
GTRANS.DAT, is used as input, along with a corresponding STASS.DAT file, for
the main programs--if the "geomagnetic cutoff and trapped protons" option is
chosen.

7.4 CREME subprograms

INSIDE is a subroutine that accounts for the effect of material shielding

on particle fluxes and energies. It considers energy loss effects and nuclear
fragmentation, but does not keep track of the fragments. The shielding is
assumed to be aluminum. INSIDE is called by SPEC, LET, and BENDEL; it calls
CRF, CUT, RAL, and DEDXAL.

CUT applies the geomagnetic cutoff transmission function to individual

particle fluxes and adds the singly-ionized anomalous cosmic rays (if that
weather condition has been chosen). CUT is only called when the "geomagnetic
cutoff and trapped protons" option has been chosen. In this case, CUT is
called by INSIDE; it calls CRF and uses the GTRANS. DAT data file.

CRF uses formulas to calculate the differential energy spectrum of solar

flare particles and/or cosmic rays of a given element, energy, year, and
"weather condition". In addition, CRF has an entry point called PROTON that
extracts orbit-averaged trapped proton fluxes from STASS.DAT. CRF is called
by INSIDE or CUT; it calls no subroutines and uses the data file STASS.DAT (if
the "geomagnetic cutoff and trapped protons" option has been chosen).

DEDXSI, DEDXAL, and RAL are subroutines that interpolate from range and
stopping power tables (see Appendixes 3, 4, 5). DEDXSI is called by LET;
DEDXAL and RAL are called by INSIDE. The subroutines use the input data files
SILICN.DT1, ALUMNM.DT1, and ALUMNM.DT2, respectively.

DIFPLD computes the differential distribution of linear path lengths in a

rectangular parallelepiped, from the formula of Bendel (1984). DIFPLD is
called by UPSET; it calls no subroutines.

11
7.5 Options and parameters in the CREME programs

The auxiliary program GEOMAG2 requires an exact description of the

satellite's orbit, which the program traces. To describe the orbit, one must
input the altitude at apogee, the altitude at perigee, and the orbital
inclination. To describe the satellite's position at the start of the
calculation, one must specify the initial longitude of the ascending node, the
initial latitude of the ascending node, and the displacement of the perigee
from the ascending node. (This last parameter is not needed for a circular
orbit). In addition, GEOMAG2 asks whether there is a "magnetic storm" and
whether to take into account the fact that the earth casts a shadow on the
spacecraft in cosmic rays. See Adams, et. al. (1983), for details.

The main CREME programs (those that output energy spectra, LET spectra, or
upset rates) require further information: First, the cosmic ray flux
subroutines require the date, the range of nuclear charges considered, and an
"interplanetary weather index". The date (in years) tells where one should be
in the solar cycle. The weather index M picks which of several sources and
solar conditions to use in calculating the cosmic ray flux near the earth. A
weather index of M=1 gives our best approximation to the galactic cosmic ray
flux at the given date. This flux is included in all other "weather
conditions", except for 3. This value (M=3) gives worst-case galactic cosmic
ray fluxes that allow uncertainties in flux data and solar activity. These
fluxes are so severe that they have only a 10% chance of being exceeded by
actual fluxes at any moment. When M=2, the anomalous component, assumed fully
ionized, is added to galactic cosmic rays. When M=4, a singly-ionized
"anomalous component is assumed. The singly-ionized particles are affected
differently than fully-ionized particles by geomagnetic cutoff. When this
weather condition is chosen, the "geomagnetic cutoff and trapped protons"
option must be chosen as well.

Weather indexes from 5 through 12 add solar flare particles to the

galactic cosmic rays:

M-5: peak ordinary flare flux and mean composition;

M=6: peak ordinary flare flux and worst-case composition;
M:7: peak 10% worst-case flare flux and mean composition;
M:8: peak 10% worst-case flare flux and worst-case composition;
M:9: peak Aug. 4, 1972, flare flux and mean composition;
M=10: peak Aug. 4, 1972, flare flux and worst-case composition;
M=11: peak composite worst-case flare flux and mean composition;
M:12: peak composite worst-case flare flux and worst-case composition.

The four flare options--ordinary, 10% worst-case, Aug. 1972, and

composite-describe different flares. The "peak" flux is the maximum flux
resulting from the flare. "Mean" and "worst-case" composition describe flares
with normal and worst-case (high) ratios, respectively, of heavy ion fluxes to
proton fluxes. See Appendix 1 for details on all of these cases.

The thickness of shielding surrounding the device must be entered in the

runs of the main programs. Shielding thickness is entered as the thicknrss
equivalent in inches of aluminum, which the programs convert into g/cam
Shielding of other materials can be approximated by the equivalent thickness
of aluminum; i.e., corresponding to the same thickness in mass per unit area.

12
 Another option of the main programs is whether to account for the
geomagnetic cutoff and trapped protons. This decision should be made based on
the spacecraft's orbit: calculations for low orbits should include these
effects, while high orbits (near geosynchronous and beyond) need not. The
CREME programs do not compute trapped proton fluxes; these must be entered
using the auxiliary program STASS.

Finally, the programs that calculate upset rates need device parameters.
BENDEL, which computes the upset rate due to proton-induced fragmentation,
needs the parameter A. If A is u~known, a measured value of upset
cross-section, in upsets/(bit proton/cm ), and the proton energy, in MeV,
corresponding to this measurement may be given (see Bendel and Petersen,
1983). UPSET, which finds the total upset rate due to direct ionization,
requires the linear dimensions of the rectangular parallelipiped representing
the sensitive region associated with each bit and the critical charge required
to produce an upset.

8.0 DISCUSSION OF RESULTS OBTAINED WITH THE CREME PROGRAMS

The set of programs described in section 7.0 has been used to investigate
the effects of interplanetary weather conditions and geomagnetic and material
shielding on SEU rates. As Adams and Gelman (1984) discussed, the critical
charge of a device corresponds approximately to a threshold value of LET
required to produce an SEU. It follows then that the particle flux above this
LET threshold is approximately proportional to the SEU rate. With this in
mind, integral LET spectra can be compared directly to see their relative
effect on the SEU rate of any proposed device.

Adams (1982) discusses how the integral LET spectrum changes under various
conditions. Figure 1 of his paper shows the enormous dynamic range in the LET
spectra that may occur outside the earth's magnetosphere. The particle flux
behind 0.025 inches of aluminum shielding and above any LET threshold can
change up to 5 orders of magnitude depending on the interplanetary weather.
Fortunately, most of this dynamic range is due to infrequent solar flare
particle events. The largest of these occur only once a decade, and are at
their worst for only a few hours. These facts lead to the first important
conclusion: "It's much easier to build a spacecraft to work 99% of the time
than it is to build one that will work 100% of the time".

Figure 2 of Adams (1982) demonstrates the effect of material shielding on

the solar particle event resulting from the August 4, 1972 flare for the
elemental spectra of hydrogen through nickel. Figure 2 of Adams (1983) shows
the effects of material shielding on a Aug. 4-like flare particle event with a
90% worst case enrichment in the heavy element spectra. These results include
the contributions of the elemental spectra of all the elements found in
nature. Adams and Gelman (1984) present a composite worst-case flare particle
event model. In Figure 4 of their of their paper, they show the effect of
material shielding on this flare. In each of the above examples, it can be
seen that material shielding is quite effective in reducing the SEU rate in
these severe environments. The second conclusion: "Material shielding is
useful in moderating solar flare particle environments and reducing the
dynamic range of SEU rates, where no geomagnetic shielding is involved".

13
 It should be noted, however, that even large amounts of shielding cannot
reduce the SEU rate to the levels normally occurring in the interplanetary
medium. Comparing Figures 2 and 3 of Adams (1983), we can see that 4 c? (1.6
inches) of aluminum shielding will still leave the SEU rate from the Aug. 4 -
like flare 1000 times worse than the cosmic ray background environment. If we
compare Figure 3 of Adams (1983) with Figure 4 of Adams and Gelman (1984), it
is clear that even 2 inches of aluminum shielding will leave the composite
worst-case flare particle event 10,000 times more severe than the cosmic ray
background. Conclusion 1 is still valid for any practical shielding
thickness.

Adams (1983) also investigated the effect of material shielding on the

cosmic ray background. Figure 3 of Adams (1983) demonstrates that thick
shielding is required to obtain even modest reductions in the cosmic ray
induced SEU rate. The third conclusion: "Material shielding is ineffective
against galactic cosmic rays".

Besides material shielding, the earth's magnetic field provides protection

for spacecraft that are at low altitude and near the equator. Adams (1983)
shows how the orbit-averaged LET spectrum is affected by orbital altitude and
inclination. While SEU's resulting from direct ionization will be less for
lightly shielded devices at lower inclinations over the normal range of values
of the critical charge, the degree of reduction will increase, generally, with
the critical charge (see Fig. 4 of Adams, 1983). Secondly, at low
inclinations, and for ligh1ly shielded devices, there is a range of critical
charge, above 2x10 MeV cm /g, for which devices appear to be immune to SEU's.
It is unlikely that such devices would really be free from upsets because in
practical spacecraft electronic components will not be lightly shielded from
all sides. Figure 8 of Adams (198V demons-ates t~at heavy shielding will
restore the LET spectrum in the 2x10 to IxIO MeV cm /g range. Figure 11 of
the same paper demonstrates that the SEU rate for a SBP-9900 in fact goes
through a minimum and then increases at greater shielding depths. In such
cases, additional shielding can actually cause the SEU rate to increasel The
fourth conclusion: "One must always consider the combined effects of magnetic
and material shielding on the SEU rate".

The effectiveness of geomagnetic shielding depends on the charge to mass

ratio of the ions. If these heavy ions are completely stripped of their
electrons, this ratio will be approximately 0.5 and magnetic shielding will be
as effective as discussed above. It is certainly true that galactic cosmic
rays are fully stripped of their electrons. The chemical and isotopic
composition of arriving galactic cosmic rays can 2,only be explained if these
particles have passed through approximately 7 g/cm of interstellar gas before
reaching us (Adams et. al., 1981a). That is more than enough matter to bring
these ions into their equilibrium charge state, and their energy is so high
outside the heliosphere that their equilibrium charge state is fully ionized.

There is a component of the particle radiation outside the magnetosphere

that may be singly ionized. It is called the anomalous component. It is
discussed in detail by Adams et. al. (1981a). If this component is singly
ionized it will penetrate the earth's magnetic field with greater ease and
reach a spacecraft as lower energy, higher LET radiation. The existing data
shows some indirect evidence that the anomalous component is singly ionized.
Its charge state will not be known for certain until an experiment currently

14
in space, onboard the NASA's LDEF spacecraft (Adams et. al., 1981b) is
analyzed. Figure 11 of Adams (1983) shows that some devices like the SBP-9900
are quite sensitive to the charge state of the anomalous component when they
are in a sracecraft that's in a 28.5 inclination orbit. The SEU rate for the
SBP-9900 is a couple of orders of magnitude higher than normal for light
shielding, if the anomalous component is in fact singly ionized. In general,
the charge state of the anoalous component will influence the SEU rate for
low altitude orbits below 60 inclination. The fifth conclusion: "Until the
charge-state of the anomalous component has been determined, it's prudent to
assume that it's singly ionized."

Unlike cosmic rays, solar flare particles have not been through much
matter before reaching earth. Mewalt and Stone (1983) eport that solar flare
particles at 50 MeV/u have traversed less than 30 mg/cm of matter. In light
of this result one must consider the popsibility that *olar flare particles
are not fully ionized. Gloeckler et. al. (1981) measurfod the charge state of
various solar flare ions from He to Fe in the energy range from 0.3 to 2.4
MeV/u. The ionic charge states of Fe were found to be distributed from +5 to
+20, with a broad maximum around +13. At higher energies, Fischer et. al.
(1984) recently reported, many ions in the 5 to 20 MeV/u range from the solar
flare of Feb. 13, 1978, were not fully ionized. Using the geomagnetic cutoff,
these authors were able to determine the maximum ionic charge that the
observed ions could have possessed. They report upper limits on the ionic
charge as low as +3 for silicon and +2 for oxygen. Based on these results, we
cannot be certain of the ionic charge of solar flare ions at any. energy.
Further experiments will be needed to establish the charge state of solar
flare particles as a function of atomic number and energy. The sixth
conclusion: "The effectiveness of geomagnetic shielding for heavy ions from
solar flare particle events is uncertain. A conservative approach is to
ignore geomagnetic shielding for solar flare heavy ions."

Spacecraft in earth orbit are exposed to the trapped radiation which

contains protons and possibly heavier ions. The trapped proton population has
been modeled by Sawyer and Vette (1976). There may also be significant
numbers of heavier ions trapped in the inner belt. This subject has been
explored by Adams and Partridge (1982). These authors found that, based on
the scanty data available, the contribution to the SEU rate from upsets caused
by trapped heavy ions is likely to be important at least under some conditions
and that they could not rule out the possibility that trapped heavy ions may
cause SEU rates that are orders of magnitude larger than those predicted from
the trapped proton population. Our present knowledge of trapped heavy ions at
the energies needed to produce SEU's is so poor that it's not possible to set
a credible upper limit on them. Recently, Adams et al. have reported
detecting trapped He ions but no heavier trapped ions at space shuttle
altitudes. At the low altitudes investigated by these authors, the trapping
lifetime for heavy ions is quite short (Blake and Fresen, 1977). As a result,
the failure of Adams et al. to detect them does not imply their absence at
somewhat higher altitudes. Seventh Conclusion: "Estimates of the SEU rate
for orbits below 7000 and 200 m.mi. are likely to only be lower limits".

Within the magnetosphere, SEU's may result from the direct ionization of
heavy ions coming in from outside the magnetosphere, or from heavy ions
trapped in the earth's magnetic field. SEU's may also result from the
intensely ionizing fragments from nuclear reactions caused by protons. This

15
indirect mechanism may be important in the inner Van Allen belt due to the
intense proton fluxes trapped there. Unfortunately, too little is known about
trapped heavy ions to assess their relative contribution to SEU rates (perhaps
this question will be resolved by the Chemical Release and Radiation Effects
Satellite mission). The relative importance of the other radiation components
will be investigated below.

The SEU rates for three devices will be examined as a function of orbital
altitude for circular orbits at 600 inclination. Calculations were done at
orbital altitudes of 200, 400, 600, 800, 1200, 1667, 2593, 3889, 5186, 6389,
and 10371 km. The trapped proton environments were obtained from
Stassinopoulos(1981) and Stassinopoulos and Barth (1982). The parameters for
these devices are listed in Table 1. These parameters are intended only for
use in the example calculations presented here and may not be accurate enough
to obtain exact SEU rates.
Table 1: The Device Parameters Used to Calculate the SEU Rates in Figs. 7,
8, 9, 11, 12, and 13.

DEVICE SENSITIVE VOLUME CRITICAL CHARGE "A" PARAMETER REFERENCES

(in micrometers) (in picocoulombs)

HM6508RH 20xi 5x2.2 0.82 1

SBP9989 10xlOxl.8 0.36(30%) 25.2 2,3,4

10x10x1.8 0. 10(70%x15%) 5,1
1 xl xl. 8 0. 02 (70%x15%)

AMD2901B 80x80 x3 0.25 23 5,1,6,7,8

NOTE: In the SBP9989 entry, 30% and 70% refer to the fraction of critical
collectors of each type. The 15% refers to the fraction of time the collector
is sensitive. The last two entries for the SBP9989 reflect the reduced cross
sections with which these low values of critical charge are observed.

REFERENCES:

1. Petersen et. al. (1983).

2. Price et. al. (1981).
3. Pickel ( 981).
14.Pickel (1983).
5. Nichols et. al. (1983).
6. Zoutendyk (1983).
7. Zoutendyk et. al.(1984)
8. Zoutendyk (1984).

Figure 7 compares the SEU rates for a AMD-2901B from trapped protons and
heavy ions from outside the magnetosphere as a function of orbital altitude
for a 60 degree inclination orbit. The environment in the interplanetary
medium is assumed to be the 90% worst case environment, and the electronic
components are assumed to be shielded uniformly by 0.1 inches of aluminum.
Because of the large critical volume, this device presents a large target for

16
cosmic rays. At the same time, the relatively high critical charge limits the
sensitivity of the device to proton-induced upsets. The result is that
proton-induced upsets are dominant only in the heart of the inner belt,
between 1700 and 3600 kin.

In this calculation, it was found that Eq. (4) incorrectly took into
account contributions from trapped protons. This problem arose, as discussed
in Section 5, because of the large device dimensions and the value of the
critical charge. The corner to corner distance across the sensitive volune is
113 micrometers. Multiplying this length by the maximum LET of a stopping
proton (see Appendix 5), we get an energy deposition of 13.8 MeV or .62 pC,
which is greater than the critical charge. However, the energy of a proton
having a residual range of 113 micrometers is only 3.35 MeV or 0.15 pC,
therefore protons cannot actually cause SEU's in a AMD2901B by their own
direct ionization. The problem was solved by omitting the proton contribution
from the integral LET spectra used to 'calculate SEU rates with the UPSET
routine.

Figure 8 gives the sare results, as Figure 7 for the SBP-9989. Here both
sources of SEU's are lower, but this time the proton-induced SEU's are more
important. This is because the relatively small device dimensions limit the
comsic ray contribution, while the low critical charge means that
proton-induced SEU's depend more on the device dimensions. Figure 9 compares
the total SEU rates for these devices with the rate for the HM6508RH that is
not vulnerable to proton-induced SEU's. The HM6508RH is the least sensitive
of the three devices considered here. Since the HM6508RH is not sensitive to
proton-induced SEU's, its SEU rate declines smoothly with orbital altitude,
while the other devices show peak SEU rates at about 2600 km. It should be
emphasized here that Figure 9 should be regarded as a lower limit on the SEU
rate, especially below 7000 km since the contribution from trapped heavy ions
was not included.

So far this discussion has delt with orbit-averaged spectra. It is

important to note that the geomagnetic cutoff is far from constant around an
orbit. The geomagnetic cutoff will be lowest at the northern- and
southern-most extremes of the orbit, and highest at the points where the orbit
crosses the geomagnetic equator. This means that the SEU rate due to
particles coming from outside the magnetosphere will vary around the orbit,
more dramatically so at high orbital inclinations. SEU's from particles
trapped in the magnetosphere will also vary around the orbit. Stassinopoulos
(1981) and Stassinopoulos and Barth (1982) computed the fraction of mission
time that 2 satellites spend in regions that are free from trapped radiation (<1
proton/cm sec). Figure 10 shows a plot of one minus this fraction (i.e. the
fraction of time spent in the belts) and the fraction spent in the high
intensity regions of the belts (>1xI0 3 protons/ca see) as a function of
orbital altitude for 60 degree inclination orbits. From this plot, it's clear
that proton-induced SEU's occur in only a portion of the orbits, and a small
portion at low altitude orbits.

To explore this further, we compute the instantaneous SEU rate around one
orbit of a satellite in a 60 inclination orbit at 1667 km altitude for each
of the devices in Table 1. As above, the 90% worst case environment is
assumed to exist beyond the magnetosphere and the components are assumed to be
behind 0.1 inches of aluminum. The instantaneous trapped proton flux was

17
 :en from Stassinopoulos (1981). Figure 11 compares the SEU rates due to
mic rays and trapped protons for an AMD-2901B as a function of mission
Scime. We see that trapped protons appear to dominate only in the heart of the
radiation belt passes. Cosmic rays are at their worst near the magnetic poles
and their mildest at the equator, as expected. Figure 12 shows the same
comparison for the SBP9989. Here the peak SEU rates in the radiation belt
passes are more pronounced, though lower than in Figure 11. Figure 13
compares the total SEU rates versus mission time for the AMD2901B, the
SBP-9989 and the HM6508RH. Because the HM6508RH is not vulnerable to trapped
protons, its SEU rate is at a minimum when the SEU rates in the other devices
are at their peaks. The SEU rates of the AMD29O1B and the SBP9989 are much
closer in the radiation belts than near the magnetic poles because their
sensitivity to cosmic rays differs more than their sensitivity to
proton-induced upsets.

For all these devices, however, the IEU rates are so low that much less
than one upset is expected per orbit in each device. At the SEU rates
calculated for these devices, the SEU rate should be no more than one per
10,000 bits per orbit. Therefore variations in SEU rate around an orbit
cannot be easily observed in real time, unless very sensitive memory devices
are used to construct very large memories. Conclusion 8: "While the
instantaneous SEU rate can vary dramatically around an orbit, these variations
occur on time scales that are usually small compared to the mean time between
SEU's. Therefore the orbit-everaged SEU rate is a useful guide for designing
spacecraft digital electronics. Only when the orbit averaged SEU rate gets as
large as the inverse of the orbital period should the designer be concerned
with the instantaneous rate around the orbit."

18

'ýIVA.' I
9.0 REFERENCES

Adams, Jr., James H., "The Natural Radiation Environment Inside Spacecraft",
IEEE Trans. on Nucl. Soi., Vol. NS-29, 2095-2100, 1982.

Adams, Jr., James H., "The Variability of Single Event Upset Rates in the
Natural Environment, IEEE Trans. on Nuol. Sci., Vol. -S-30, 4475-4480, 1983.

Adams, Jr., J.H. and K. Partridge, "Do Trapped Heavy Ions Cause Soft Upsets on
Spacecraft?", NRL Memorandum Report 4846, Oct. 12, 1982. (AD-A119909)

Adams, Jr., James H., and Andrew Gelman, "The Effects of Solar Flares on
Single Event Upset Rates", IEEE Trans. on Nuclear Science Vol. NS-31, 1212-1216

Adams, Jr., James H., R. Silberberg, and C. H. Tsao, "Cosmic Ray Effects on
Microelectronics, Part I: The Near-Earth 'Particle Environment", NRL Memorandum
Report 4506, August 25, 1981a. (AD-Al03897)

Adams, Jr., James H., M. M. Shapiro, R. Silberberg, and C. H. Tsao, "Fast

Heavy Ions in the Heliosphere", Adv. in Space Res., Vol. 1, 169-172, 1981b.

Adams, Jr., James H., J. R. Letaw, and D. F. Smart, "Cosmic Ray Effects on
Microelectronics, Part II: The Geomagnetic Cutoff Effects", NRL Memorandum
Report 5099, May 26, 1983.(AD4128601)

Adams, Jr., James H., James Bellingham, and Peter E. Graney, "A Comprehensive
Table of Ion Stopping Powers and Ranges", NRL Memorandum Report (to be
published), 1987.

Bendel, W. L., "Length Distribution of Chords Through a Rectangular Volume,"

NRL Memorandum Report 5369, July 3, 1984. (AD-A143302)

Bend el, W. L., and E. L. Petersen, "Proton Upsets in Orbit", IEEE Trans. on
Nucl. Sci., Vol. NS-30, 4481-5, 1983.

Binns, W. R., R. K. Fickle, T. L. Garrard, M. H. Israel, J. Klarmann, E. C.

Stone, and C. J. Waddington, "Cosmic-ray Abundances of Elements with Atomic
numbers 26<Z<401, Astrophysical Journal, Vol. 247, L115-L118, 1981.

Binns, W. R., R. K. Fickle, T. L. Garrard, M. H. Israel, J. Klarmann, E. C.

Stone, and C. J. Waddington, "The Abundance of the Actinides in the Cosmic
Radiation as Measured on HEAO-3", Astrophysical Journal, Vol. 261, L117-L120,
1982.
Binns, W. R., R. K. Fickle, T. L. Garrard, M. H. Israel, J. Klarmann, K. E.
Krombel, E. C. Stone, and C. J. Waddington, "Cosmic-ray Abundances of Sn, Te,
Xe, and Ba Nuclei Measured on HEAO 3", Astrophysical Journal, Vol. 267,
L93-L96, 1983.

Blake, J. B. and L. M. Friesen, "A Technique to Determine the Charge State of

the Anomalous Low-Energy Cosmic Rays", Proc. of the 15th International Cosmic
Ray Conference Plovdic, 1977, Volume 2, p. 341

Breneman, H. H. and E. C. Stone, "Solar and Photospheric Abundance from Solar

Energetic Particle Measurements," Astrophysical Journal, Vol. 299, L57-L61,
1985.

19
Priggs, P. R., T.P. Armstrong and S. M. Krimigis, "Hydrogen Over Helium
Enhancement in Successive Solar Flare Particle Events From the Same Active
Region", Astrophysical Journal, Vol. 228, L83-L87, 1979.

Cameron, A. C. W., "Elementary and Nuclidic Abundances in the Solar System",

Harvard-Smithsonian Center for Astrophysics Preprint Series No. 1357 (1980).

Chenette, D. L., and W. F. Dietrich, "The Solar Flare Heavy Ion Environment
for Single Event Upsets", IEEE Trans on Nucl Sci, Vol. NS-31, 1217-1222, 1984

Coffey, Helen A., ed., Solar Geophysical Data, No. 476, Part II, Published by
the Environmental and Infornation Data Service, NOAA, Boulder, Colorado,
April, 1984.

Dietrich, W. F., and J. A. Simpson, "Preferential Enhancements of the Solar

F1 are Accelerated Nuclei Carbon to' Zinc from -20-300 MeV/Nucleon",
Astrophysical Journal, Vol. 225, L41-L45, 1978.

Engelmann, J. J., P. Goret, E. Juliusson, L. Koch-Miramond, P.Masse, A.

Soutoul, P. Byrnak, N. Lund, B. Peters, I. L. Rassmussen, M. Rotenberg, and N.
J. Westergaard, "Elemental Composition of Cosmic Rays from Be to Ni as
Measured by the French-Danish Instrument on HEAO-3", Proceedings of the 18th
Inter. Cosmic Ray Conf., Vol. 2, 17-20, Bangalore, India, 1983.

Fischer, S., M. Vand as, K. Kudel a, S. N. Kuznetsov, V. N. Lutsenko,

"Determination of the Effective Charge of Solar Cosmic Ray Nuclei Using the
Earth's Magnetic Field", Advances in Space Research, Vol. 4, 169-172, 1984.

Gloeckler, G., H. Weiss, D. Hovestadt, F. M. Ipavich, B. Klecker, L. A. Fisk,

M. Scholer, C. Y. Fan, and J. J. O'Gallagher, "Observations of the Ionization
States of Energetic Particles Accelerated in Solar Flares", Proc. of the 17th
Intl. Cosmic Ray Conf., Paris, Vol. 3, 136-9, 1981.

Goswami, J.N., D. Lal, and J. D. Macdougall, "Charge Composition and Energy

Spectra of the Ancient Solar Flare Heavy Nuclei", in "The Ancient Sun", Ed. by
R. 0. Pepin, J. A. Eddy and R.B.Merrill, 347-64, Pergamon Press, 1980.

Hsieh, C.M., P. C. Murley, and R. R. O'Brien, "A Field-funneling Effect on the

Collection of Alpha-particle Generated Carriers in Silicon Devices", IEEE
Electron Device Letters, Vol. EDL-2, 103, 1981.

Juliusson, E., J. J. Engelmann, J. Jorrand, L. Koch-Miramond, P. Masse, N.

Petrou, N. Lund, I. L. Rassmussen, and M. Rotenberg, "The Galactic Cosmic Ray
Energy Spectra as Measured by the French-Danish Instrument on HEAO-3",
Proceedings of the 18th Intl. Cosmic Ray Conf., Vol. 2, 21-24, Bangalore,
India, 1983.

Kolasinski, W. A., R. Koga, D. L. Chenette, "Heavy-Ion Induced Single Event

Upsets in a Bipolar Logic Device", IEEE Trans. on Nucl. Sci., Vol. NS-30,
4470-4, 1983.

Lingenfelter, R.E., and H. S. Hudson, "Solar Particle Fluxes and the Ancient
Sun", in "The Ancient Sun", Ed. by F. 0. Pepin, J. A. Eddy and R.B.Merrill,
69-79, Pergamon Press, 1980.

20
 Luhn, A., B. Klecker, D. Fovestadt, G. Gloeckler, F. M. Ipavich, M. Scholer,
C. Y. Fan and L. A. Fisk, "Ionic Charge States of N, Ne, Mg, Si, and S in
Solar Energetic Particle Events", Advances in Space Research, Vol. 4 161-164,
1984.

Messenger,.G. C., "Collection of Charge on Junction Nodes from Ion Tracks",

IEEE Trans. on Nucl. Sci., Vol. NS-29, 2024-31, 1982.
3
Mewaldt, R. A., and E. C. Stone, "A Search for Deuterium, Tritium, and He in
Large Solar Flares", Pull. of the Am. Phys. Soc., Vol. 28, 742, 1983.

Mewaldt, R. A., "Spacecraft Measurements of the Elemental and Isotopic

Composition of Solar Energetic Particles", in "The Ancient Sun", Ed. by R. 0.
Pepin, J. A. Eddy and R.B.Merrill, 81-101, Pergamon Press, 1980.

McLean, F. B., and T.R. Oldham, "Charige Funneling in N- end P-Type Si

Substrates", IEEE Trans. on Nucl. Sci., Vol. NS-29, 2018-23, 1982.

Nichols, D. K., W. E. Price, and C. J. Malone, "Single Event Upset (SEU) of

Semiconductor Devices - A Summary of JPL Test Data", IEEE Trans. on Nucl.
Sci., Vol. NS-30, 4520-5, 1983.

Oldham, T. R., and F. B. McLean, "Charge Collection Measurements for Heavy

Ions Incident on n- and p-Type Silicon", IEEE Trans. on Nucl. Sci., Vol.
NS-30, 4493-4500, 1983.

Petersen E. L., J. B. Langworthy, and S. E. Diehl, "Suggested Single Event

Upset Figure of Merit", IEEE Trans. on Nucl Sci., Vol. NS-30, 4533-9, 1983.

Pickel, James C., "Cosmic Ray Effects in Integrated Injection Logic Devices",
General Order No. 6301, Rockwell Intl. Science Center, Anaheim, CA. June, 1981.

Pickel, James C., private communication, 1983.

Piokel, James C., and James T. Blandford, Jr., "Cosmic-Ray-Induced Errors in

MOS Devices", IEEE Trans. on Nucl. Soi., Vol. NS-27, 1006-1015, 1980.

Price, W. E J. C. Pickel, T. Ellis, and F. B. Frazee, "Cosmic Ray Induced

Errors in I2L Microprocessors and Logic Devices", IEEE Trans. on Nucl. Sci.,
Vol. NS-28, 3946-3954, 1981.

Sawyer, D. M, and J. I. Vette, "AP-8 Trapped Proton Environment for Solar

Maximum and Solar Minimum", Report NSSDC/WDC-A-R&S 76-06, Greenbelt, Md., 1976.
Shea, M.E. and D.F. Smart, "tTables of Asymptotic Directions and Vertical
Cutoff Rigidities for a Fine Degree by Fifteen Degree World Grid as Calculated
Using the International Geomagnetic Reference Field for Epoch 1975.0",
AFCRL-TR-75-0185, Hanscom, AFB, Mass., 1975.(AD-A012509)
Stassinopoulos, E. G., "Orbital Radiation Study for Inclined Circular
Trajectories", NASA-X-601-81-28, Goddard Space Flight Center, Greenbelt, Md.,
1981. (AD-A141849)

21
I
-)*. *
Stassinopoulos, E. G., and J. M. Barth, "Non-Equatorial Terrestrial Low
Altitude Charged Particle Radiation Environment", NASA-X-601-82-9, Goddard
Space Flight Center, Greenbelt, Md.,1982.

Sterne,T.E.,"An Introduction to Celestial Mechanics," Interscience Publishers,

1960.

Tsao, C. H., R. Silberberg, M. M. Shapiro, and J. H. Adams, Jr., "Origin,

Transformations and Propagation of Cosmic Rays with Z>50", Proc. of the 17th
Intl. Cosmic Ray Conf., Vol. 9, 130-3, Paris, 1981.

Tsao, C. H., R. Silberberg, J. H. Adams, Jr., and J. R. Letaw, "Cosmic Ray

Effects on Microelectronics, Part III: Propagation of Cosmic rays in the
Atmosphere", NRL Memorandum Report 5402, 1984. (AD-A145026)

Van Hollebeke, M. A. I., "Relative Abundance of Proton to Helium Nuclei in

Solar Cosmic Ray Events", Proc. of the 14th Intl. Cosmic Ray Conf., Vol. 5,
1563-7, 1975.

Webber, W. R. , E. C. Stone, and R. E. Vogt, "The Elemental Composition of

Quiet Time Low Energy Cosmic Rays Measured on the Voyager Spacecraft", Proc.
of the 16th Intl. Cosmic Ray Conf., Vol. 5, 357-362, Kyoto, 1979.

Zinner, E, "On the Constancy of Solar Particle Fluxes from Track,

Thermoluminescence and Solar Wind Measurements", in "The Ancient Sun", Ed. by
R. 0. Pepin, J. A. Eddy and R.B. Merrill, 201-26, Pergamon Press, 1980.
Zoutendyk, J. A., "Modeling of Single-Event Upset in Pipolar Integrated
Circuits", IEEE Trans. on Nucl. Sci., Vol. NS-30, 4540-5, 1983.

Zoutendyk, J. A., private communication, 1984.

Zoutendyk, J. A., C. J. Malone, and L. S. Smith, "Experimental Determination

of Single-Event Upset (SEU) as a Function of Collected Charge in Bipolar
Integrated Circuits," IEEE Trans. on Nucl. Sci. Vol. NS-31, 1167-1174 Dec., 1984.

22

v ve
 1111 11 1r 0

Cc
40 o 7

> 4j

/ 0)
> =

c0 -

L t

I. CO 1i0.

0 0

op'
00

0 a4

- _ o
(WAeV4 *09 JG~s- 2w)/S3-101Ii8d IxLO

23
 \ -

IRON
\03

C -
> 0 4

E- -
GARCIA-MUNOZ ET AL. (1975)
PRICE ET AL. (1973)

co) U WEBBER ET AL. (1973b)

L 4 SCARLETT ET AL. (1978)
j 0 SIMON ET AL. (1979)
SJUIUSSON (1974)
SIO. LEZNIAK AND WEIBBER (1978)

ORTH ET AL. (1978)

FrEIER AND WADDINGTON (1968)
46 WEBIER AND ORMES (1967)
SEAMES AND FICHTEL (1967)
WEBBER IT AL. (1979)
10 ET AL. (1979) --
: COMSTOCK (1969)
S JULUSSON ET AL. (1983)

102 1024
10 5
1i0 106
I01

KINETIC ENERGY (MeV/ u)

Figure 2: The cosmic ray iron spectrum: The solid curves are for solar
maximum (lower) and solar minimum (upper). The dashed curve in the 90%
worst-case iron spectrum, which is implied by comparison with the cosmic ray
helium spectrum.

24

'. \'
 z z
0 z z W
~z Z wo
o a- Zg Dmc
<~~ Z>Ce.>4c

0
cu

- C

0Z

zW0
o. CLC
z
F-

0
094.
0
N 0. 4.)

1 0

B/ZWO AOVY NI UB3MOd O)NlddOLS

25
 *~~~~~~~~~ EWhaar44) -. - * -

f4;

Cx
914~

4.0 0

0 , 8 "
M 0

(44-

0
LLn

L- .0 -

fl~v'/AOVI DOS JOIS f/3 fN

26

~ ýxl A
 I4I4

20

0 E
40 U')

0 0

B/aw 019 AGN099 4/13io0

270
 0 4.

04)(

%0

0 0

CU4
u,

N/Glo
Ogg Jol

28'.
 0

0
0 C
0
0 0 e4 C
2 0 , 0

00 ".4(
- 0 WO ;,s
'U0 M
00 °

o
0(9 L 1

0 "4.0
00

,w~ .Ll r= 0
1.13.0 Q

4) 0~a

(00
-f4 4

0 0 0
AVG L1/S.LSd0

29-
 00
_0
CLL 0F4
O w __, 7

Ix 000 __

a) 2 z
00

01

(0 0

101

400

300
 0 0)4.
C0 0- I. L.
0 M.
CD 0

00

(0 45

4~)

ot
C'.+ 00

_ 0) 0, J00

)~'Q LI8/i3~df

310
 1.0

F TIME SPENT IN THE

0.8- RADIATION BELTS

TIME SPENT IN HIGH-

W INTENSITY REGIONS

z
Now0.6-
Clo

* 0
z
S0.4-

0.2-

0
0 4000 8000 12000
ALTITUDE (kin)
Figure 10: The fraction of mission time tliat a spacecraft spends in the
earth's radiation belts versus orbital altitude. The calculations are for
circular orbits at 60 0 inclination.

32

MO.. S. ~f.
 0o

0
a) 0E-

4-)

0O (04
(0
co CO~ c

0
,4 e -4
0)

C/) = 0
o 10

z 00
FLBSJ3d

03

Ac
 m 4.) 0

0 4

0=
0) 0.
a.)
Cl)U
C4ct

OD E c~ N

41
X Eo

0 40. )

4) 4) 4.

w~ 0 Cf

Qoc
00 0

00

00

00
N~~~
0 400
w ol 0)

34,

~ ~V-
 0

I"0..
O 10C 00 -

oO

00
_m 0

CD. 0 CrA

C)j

o•o

N 0 00 00
00)

o /)3:

CD - I-- 4'

35
APPENDIX 1: THE ANALYTIC MODEL FOR THE CHARGED PARTICLE ENVIRONMENT

Galactic cosmic rays consist of electrons and the nuclei of all the
elements in the periodic table; the first 28 elements are the most important
for cosmic ray effects on microelectronics. These particles are from outside
the solar system and their flux at low energies is anti-correlated with solar
activity (i.e. more cosmic rays at solar minimum). The differential energy
spectra in particles per square meter - steradIan - second - million electron
volts per atomic mass unit (i.e. particles/(m .ster.sec.(MeV/u))) are given
in the following paragraphs.
The spectra for protons (hydrogen nuclei), a-particle (helium nuclei), and
iron nuclei are given below for energies above 10 MeV/u:

F(E,t) = A(E) sin[W(t-t 0 )J'+ B(E), (0)

where
W 0.576 radian/year,

tO 1950.6 A.D. date,

t current date in years,

E particle energy in MeV/nucleon,

B(E) = 0.5 [fin (E) + fax (E)], (2)

A(E) = 0.5 Ifmin (E) - fax (E)]. (3)

fain and fax differ only by the choice of constants in the equation

f(E) = 10m(E/E 0 )ab (4)

where a = a0 (1 - exp[-X 1 (log 1 0 E)b)) (5)

m = C1 exp[-X 2 (log 1 0 E) 2 - C2 . (6)

The values of the constants a, EO, b X X, C, and C2 are given in Table 1

for each of the elements hy rogen 1),helum (He), and iron (Fe) for the
conditions of solar maximum and solar minimum.

TABLE 1. Constants Used in Eqs. 5 and 6 to Compute the Differential Energy

Spectra of H, He, and Fe at Solar Maximum and Solar Minimum

Element a0 E0 b XI X2 C1 C2

H-min -2.20 1.1775x00 55 2.685 0.117 0.80 6.52 4.00

H-max -2.20 1.1775x10 2.685 0.079 0.80 6.52 4.00
He-min -2.35 8.2700x0 4 2.070 0.241 0.83 4.75 5.10
He-max -2.35 8.2700x10 5 2.070 0.180 0.83 4.75 5.10
Fe-min -2.14 1.1750x10 5 2.640 0.140 0.65 6.63 7.69
Fe-max -2.14 1.1750x10 2.640 0.102 0.65 6.63 7.69

36

n1 K,.
 The differential energy spectra for carbon (C), oxygen (0), fluorine (F),
neon (Ne), sodium (Na), aluminum (Al), and phosphorus (P) are obtained by
multiplying the helium spectrum (from Eq. (1)) by the scaling factors listed
in Table 2.

Element Ratio Element Ratio

C 3. 04x 0-2 Na 1. 02x1 0-3

0 2. 84xl 0- 2 Al I. 07 x1 0-3
F 6.06xl 0-4 P 2. 34 x1 0-4
Ne 4 63 x1 0-3

The differential energy spectra for calcium (Ca), cobalt (Co), and nickel
(Ni) are obtained by mutliplying the i~on spectrum (from Eq. (1)) by the
scaling factors listed in Table 3.

TABLE 3. The Ratios of the Abundances of Various Nuclei to Iron

Element Ratio

Ca 2.1 x10'
3
Co 3.4x1l0
Ni 5.0x10"2

The spectra of the elements lithium (Li), beryllium (Be), and boron (B)
are obtained from the helium spectrum FHe , modified by the equation below.

0.021 FHe E < 3000 MeV/u

F* = (7)
0.729 E043 FH E > 3000 MeV/u

F* is the combined spectrum of (Li + Be + B). This value is multiplied by

the ratios in Table 4 to obtain the individual element spectra.

TABLE 4. The Relative Fractions of Li, Be, and B in the Combined Total
Ahtndance of Li + Be + B

Element Ratio

Li 0.330
Be 0.176
B 0.480
The spectrum of the element nitrogen (N) is obtained by modifying the
helium spectrum FHe as shown:

FN (8.7x10- 3 exp(-0.4(log 10 E - 3.152]

+ 7.6xi0 3 exp[-0.9(log1 0 E - 0.8)2)) FHe* (8)

37

,Z ýZ f
 The spectra of the elements magnesium (Mg), silicon (Si), and sulfur (S)
are obtained by modifying the helium spectrum FHe as shown:

F - H SFHe' E > 2200 (9)

(1 + 1.56x10- (E-2200))FHe. E > 2200

The individual spectra for these elements are obtained by multiplying F by

the ratios of Table 5.

TABLE 5. The Ratios of Mg, Si, and S to an Adjusted Helium Spectrum

Element Ratio

Mg 6.02xl0-3
Si 4.63xI04
S 9.30x10

The spectra for the elements chlorine (Cl), argon (Ar), potassium (K),
scandium (Sc), titanium (Ti), vanadium (V), chromium (Cr), and manganese (Mn)
are all obtained by modifying the iron spectrum FFe as shown below:

F* Q(E)FFe, where (10)

Q(E) = 16 (1 - exp(-0.075E0 .4 )] E"O' 33. (11)

The individual spectra for these elements are obtained by multiplying F* by

the ratios of Table 6,

TABLE 6. The Fractional Abundance of Each Element in the Sub-Iron Group

Element Ratio Element Ratio

Cl 0.070 Ti 0.147
Ar 0.130 V 0.070
K 0.090 Cr 0.1140
Sc 0.0142 Mn 0.100
The differential energy spectra for elements from copper to uranium are
obtained by multiplying the iron spectrum (from Eq. (1)) by the scaling
factors listed in Table 7.

TABLE 7. The Ratio of the Abundances of Various Nuclei to Iron

Element Ratio Element Ratio

-4-7
Cu 6.8x10.4 Pm 1.9x10 7
Z6 8.8xi0.5 Sm 8.7xi- 7
Ga 6.5 x10 Eu 1.5 x10 7
Ge 1.4x10_- Gd 7.0X10 7
As 9.9X10-6 Tb 1.7x10
Se 5.8x10.- Dy 7.0x10 7

Br 8.3x10 HO 2.60-7

38

47)
 Kr 2. 3x10-5 Er 3X1 0-7
Rb 1. 1X10 5 Tm 8. 9X10- 8
Sr 3.•x1 0-5 Yb 4.4xi 0-7
Y 6.8x1 0"6 Lu 6.4xI 0-8
Zr 1. 7x10"5 Hf 4. Ox 0-78
Nb 2.6xi0-6 Ta I.6xi 0-8
Mo 7. 1 x1 0" W 3. x1 0-7
To 1. 6xi -6 Re 1. 3x1 0-7
Ru 5. 3xl0 66Os 5.6010-7
Rh 1. 5xi 0_ Ir 3.7x10 7
Pd 4.5x00 6 Pt 7.2x10 7
Ag 1. 3xI0-6 Au 1. 3x10-7
Cd 3.6x10" 6 Hg 2.3xI007
In 1.4xI0- 6 T 1.8x1 0-7
Sn 7.5x10- 6 Pb 1.7x10-6
Sb 9.9x1 0-7 Bi 9.OX1 0-8
Te 5.7x1 0 6 Po 0
I 1.5x1 O At 0
Xe 3.5x0 0- Rn 0
Cs 5.8xi 0 Fr 0
Ba 6. Ox.O_ Ra 0
La 5.3xi0 7 Ac 0
Ce 1.6x1 0-6 Th 9. Oxl 0-8
Pr 3. Ox10-7 Pa 0
Nd 1.1x10-6 U 5. 4x10-8

The recipe given above is correct for quiet periods in the interplanetary
medium when only the galactic cosmic rays are present. These conditions are
often disturbed, especially at low energies, by small solar flares,
co-rotating events, etc. To allow for typical disturbed conditions, we
recommend that a worst-case spectrum be used. With 90 per cent confidence,
the instantaneous particle flux should never be more intense than described by
this case at any energy.
To construct the worst-case spectrum for protons, compute the "H-mmn"
spectrum (using Eq. 4) and then compute FH-worst as shown below:
FH worst : 1189 7 e-E/ 9 "6 6 + 1. 64 IF Hmi. (12)

This applies for E < 100 MeV/u.

In like manner, the solar minimum case heliun and iron spectra (also from
Eq. (4)) are multiplied by:
28. 4e"E1 1 3 . 84 + 1.64 (13)

for E < 100 MeV/u.

The worst-case spectra of H, He, and Fe for any element for E > 100 MeV/u
are approximated by a multiple of the solar minimum spectra:

Fworst =1. 64 Fmin. (14)

39
 The resulting spectra are used as described above to obtain the other
elemental spectra, i.e. in the same way as FHe and FFe were used.

In addition to galactic cosmic rays, some particles are believed to be

accelerated in the interplanetary medium. The most important of these are
called the anomalous component. Its greatest contribution is to the helium
spectrum. We recommend that, for the period 1982-1990, the cosmic ray helium
spectrum be modified as follows:

1. Determine the maximum values of the helium spectra from Eq. (4) using
the He-max and He-min constants from Table 1.

2. Modify Eq. (4) so that each maximum value applies to all energies
below the energy at which the maximum occurs; i.e., for solar minimum:

0.33, E < 200 MeV/u

f*-
He-min f He-min (from Eq. (15)
(4)). E > 200 MeV/u

3. Make the same kind of modification, f* He-max for solar maximum:

0.079, E < 300 MeV/u

f He-max, =(1 fiHe-max f* (from Eq. (4)). E > 300 MeV/u 6)

4. Combine the resulting spectra as before using Eqs. (1-3).

NOTE: This applies only to He; use the regular He spectra of Eqs. (1-6)
to get the spectra of the other elements.

The anomalous component contributes to the spectra of oxygen and nitrogen,

as well as helium, at low energies. For the years 1982-1990, the following
contributions may be added to the galactic cosmic ray oxygen and nitrogen
spectra.

For oxygen, use:

f(E) = 6x100 expr-(ln(E)-1.79) 2 /0.70] particles/(m2 .ster.sec.MeV/u). (17)

This spectrum crosses the galactic spectrum near 30 MeV/u. The two should be
matched at that point with Eq. (17) replacing the galactic spectrum at lower
energies. Similarly, for nitrogen:

f(E) : 1.54x10 2 exp[ (2n(E)-1.79) 2 /0.70)

in particles/(m .ster.sec.MeV/u). (18)

Again, this crosses the galactic cosmic ray spectrum near 30 MeV/u and should
replace it below this energy.

The spectra of the remaining elements are unaffected or affected at too

low an energy to matter.

40

"I". ~
i•,~ ~ ~I , ' . • " " . '.• • " ',."• ", " • .", ", t . .,". , •- • :•
 There is a possibility that the anomalous component is singly ionized. If
this is so, it will have an extraordinary ability to penetrate the earth's
magnetosphere. In this case, the differential energy spectra shown below are
assumed for the helium, carbon, nitrogen, oxygen, neon, magnesium, silicon,
argon, and iron spectra of the anomalous component. There probably are
anomalous components in the spectra of the nuclei heavier than iron, but there
are no dot,:- -n them at this time.

For singly-ionized helium,

.4, E < 195

1.54XF 4 E-2 E > 195

For singly-ionized carbon,

4.00xi0 3 exp[-(lnE - 1.79)2 /0.7], E < 10
F = (20)
0.27 E"* E > 10

For singly-ionized nitrogen,

1.54xi0"2 exp[-(lnE - 1.79)2 /0.7], E < 20

F: - >20 (21)
0.773 E . E > 20

For singly-ionized oxygen,

6.00x10-2 exp[-(lnE - 1.79)2 /0.7], E < 30

F - (22)
1.32 E"* E > 30

For singly-ionized neon,

8.00x0- 3 exp[-(lnE - 1.79)2 /0.7), E < 20

F - (23)
0.40 E E > 20

For singly-ionized magnesium,

8.MOxM0 exp[-(lnE - 2.30)2 /0.7], E < 20

F 0.16 -
E=2 E > 20 (24)

For singly-ionized silicon,

1.00x10- 3 exp[-(lnE -2.20)2 /0.4], E < 10

F = (25)
0.10 E"2 E > 10

For singly-ionized argon,

5.40x10"4 exp[-(lnE - 1.79)2 /0.7], E < 20

41

SY
 F E>(26)
0.028 E"-. E > 20

For singly-ionized iron,

6.000x0 exp[-(InE - 2.48)2 /2.0], E < 30
F -3 (27)
0.35 E E > 30
Since these anaoalous component particles are assumed to be
singly-ionized, they will have a higher magnetic rigidity than galactic cosmic
rays of the same energy. The magnetic rigidity of galactic cosmic rays is

R = (A/Z) (E2 + 1862.324 E) 1 / 2 /1000, (28)

in GeV/ec. The rigidity of singly ionized nuclei is

R = A (E 2 + 1862.324 E) 1 /2 /1000. (29)

To edd the singly-ionized anomalous component, it is necessary to modulate
both the galactic cosmic ray spectra and the anomalous component spectra given
above using the geomagnetic cutoff transmission function. The method for
calculating this function is discussed in Adams et. al. (1983) and in Appendix
2 of this report. The resulting modulated spectra are then added together.

Solar flare particle events are sporadic occurrences lasting 1-5 days.
When these events occur they can be the dominant cause of SEU's. For
statistical treatment, they are broken down into two classes: ordinary (OR)
and anomalously large (AL). The probability of having more than a number n of
events in a time t is given by:
n
P(n,t,N,T) = 1 - z (i+N)I(t/T)i /[ilNl(1+t/T)1+i+N ], (30)
i=O

where T and t are in years, and N is the number of flares that have occurred
in T years.

For ordinary events, Eq. (30) becomes:

aFOR = P(n,t,24,7) for 1981-1983

and OR(31 )
SPOR = P(n,t,6,8) for 1984-1988,

where there is a probability P. of having more than n ordinary events in t

years. Similarly, for anomalously large events:
PAL =P(n,t,1,7). (32)

If there is an unacceptable risk of an AL event then it will be the worst-case

flare for the mission.

42
 The peak proton flux differential energy spectrum for ordinary events is,
typically
F2x1 4 -F./27. 5 -E/4 2
SOR 2.45x10 (e 2 +. 173e ) protons/m2 .ster.sec.MeV. (33)
and no worse than
5 (e"E/24.5 + 63.6eE/ 4 ) protons/m2 .ster.seo.MeV, (34)
FWOR a 2.06x10

with a confidence of about 90 per cent.

Using the August 1972 flare as a model Al event, the peak proton flux
differential energy spectrum is

9.3x10 9 (dP/dE) exp(-P/0.10), E < 150 MeV

fAL 1.76x105 (dP/dE) P-9 E > 150 MeV (35)
2
in protons/2 .ster.sec.MeV, where

P = [(E/1000) 2 + 1.86 x 10-3 E 11/2. (36)

To model the worst flare that is ever likely to occur, we use the
composite of the two worst flares ever recorded: the August, 1972, flare and
the February, 1956, flare. The composite worst-case flare proton spectrum is
taken to be the peak of the 1972 spectrum, as given by Eq. (35), and the 1956
peak spectrum, given below.

f 1956 = 1.116x10 8 (E" 1 . 2 4 8 )(0.248 + 2.5x10 5 xl.7xEPOW) EXPOW

+ L4.7x10 1 9 (E"5 . 3 )(I4.3x(I-EXPOW) - 6.32xi0" 1 5 x4.7xEPOWxEXPOW), (37)

where 1.7
EPOW• =
and EXPOW exp(-2.5xlO"5 EPOW).

The composition of flare particles also varies greatly from flare to

flare. Table 8 gives the composition relative to hydrogen of the elements
through nickel. Both mean and (90 percent confidence level) worst-cases are
given. Multiply the abundance ratio from Table 8 by the appropriate flare
proton spectrum to get the flare spectrum of any element in the table.

TABLE 8. Mean and Worst-Case Flare Compositions

Mean Case Worst Case Mean Case Worst Case

2. 3X1 0 7 00 -6
H 1 1 P
He .Oxo- 3.3x10-2 S 8.0x106 5010 5
Li 0 0 Cl 1.7x10- 7 8.0x10"-
Be 0 Ar 3.3x10-6 1.800O'
B 0 0 K 1.3x10-_ 6.Ox10-7
C 16x- 4.0 x10-4 Ca 3.2x1i06 2.0x10 5
N 3.8x10"- 1.1xl- S 0 0
0 3.2x10 1.x100 3 Ti 1.0x10 7 5.0x10 7

43

It*
F 0 0 4 V0-
Ne 5.1X1C0- 1.9X10" Cr 5.7x10 7
4.0xi10-6
Na 3.2x10 1.3x10 5 Mn 4.2xl0 5 2.3x10=
Mg 6.4x10 5 2.5x1 0_ Fe 4.1x10- 4.0xl 0
Al 3°5x10 - 1.4x10•5 Co 1.0x10- 7 5.5x10 _7
Si 5.8x1O 1:.9x10O Ni 2.2x1 0 2.0x1 0

The mean case compositions for the elements from copper to uranium are
taken from Cameron (1980). The ratios of these abundances to hydrogen are
given in Table 9.

TABLE 9. Mean Flare Compositions

Cu 2.Oxl 0-88 Pm 0
Zn 6.Ox1O- Sm -1.0X10
Ga 2.0x10• Eu 4.x-1
As *.0X01 0 Gd 2.OxlO 1
A 3.OxO-10 Tb 3.-ox1 1
Se 3ox1 0-9 Dy 2.Oxl o-1
Br Oxl0 10 Ho
H. 4.0x10-1 2
Kr 2.0x10 9 Er 1.0x10 1 1
Rb 3.OxlO 0 Tm 2.Ox10- 12
Sr 1.0x10 9 Yb 9.0x10 _
Y 2.oxlo.-10 Lu 2.OxlO - 2
Zr 5. Ox10 10 Hf 8.OxlO-12
13
Nb 4.OxlO -1 -10 Ta 9.0x10--11
Mo 2°.Ox 1 0 W 1.0x10. 1
Tc 0 -11 Re 2.0x10---11
Ru 9. 0 x1U1 1
Os 3.0 x10. 1 1
Rh 2.0xl0-1 1 Ir 3.OxlO
Pd 6"0xlu-11 Pt 6.0x101 111
Ag 2.0 x10U1 1 Au 1.0x10 -O1
Cd 7.Oxl1_ 12 g 1.0oxlo-12
In 9.0xlu10 .- 1O--10 Tl 9.0x10-12
Sn 2.Oxlu-11 Pb 1.0x10 12
Sb 1
1.4x1o0 Bi 6.0x0O
Te 3.•Oxl .1 Po 0
I 6.-Oxl U10 At 0
Xe 2.70xl1 1 Rn 0
Cs 201 -10 Fr 0
Ba 2.ux10 1 1 Ra 0
La 2 . x1 _OH Ac 0 12
Ce 5.0x10. 12 Th 2.Ox1O
Pr 8.ox10o1• Pa 0
Nd 4. Ox1O- U 1.2xl -12

The worst-case compositions of the elements from copper to uranium are

obtained by multiplying the abundance ratios of Table 9 by:

(CW (0)/Ct (0))0.48 exp(Z 0 .7 8 /6.89), (38)

where C (0) and Ct (0) are the worst-case and mean abundance coefficients for
oxygen Yn Table 8.

44
 There are several good mathematical models for the trapped proton
environment. We recommend "AP-8 Trapped Proton Environment for Solar Maximum
and Solar Minimum," Donald M. Sawyer and James I. Vette, Report No.
NSSCS/WDC-A-.R and S 76-06, Dec., 1976, NASA-Goddard, Greenbelt, Md.

The environment of trapped a particles and heavier nuclei is very poorly

known. The reader is referred to Adams and Partridge (1982) for a discussion
of the data base on trapped ions heavier than protons.

The modulation of cosmic ray spectra by the earth's magnetic field is

discussed in Adams et. al. (1983) in Appendix 2 of this report. Briefly, the
geomagnetic cutoff is a value of magnetic rigidity below which cosmic rays
will not reach a specified point in the magnetosphere from a specified
direction. The magnetic rigidity P in GeV/ec may be computed from a
particle's energy by:

P = (A/Z)[(E/1000)2 + 1.86x10- 3 I /2 (39)

where A and Z are the atomic mass and charge of the nucleus in question.

The cutoff at any point for particles arriving from the zenith is most
simply computed with:
P0 = 15.96/L 2 . 00 5 GeV/ec, (40)

I where L isMcllwain's L parameter (i.e. the radial distance, in earth radii,

from the center of the earth to the point in the geomagnetic equitorial plane
it is crossed by the magnetic field line that also passes through the
where of
point observation).

Detailed calculations of the cutoff are available from Shea and' Smart
(1975). Transmission functions for satellite orbits may be computed using the
techniques described in Adams et. al. (1983) and Appendix 2.

The transmission functions are useful in modulating the cosmic ray

spectra. Some thought must be given to their use on solar flare spectra
because: (1) the flare particle intensity changes on a time scale comparable
to or shorter than an orbital period; (2) there is no certain proof that solar
flare particles are fully ionized; (3) the geomagnetic cutoff is suppressed to
some extent during a flare. We recommend that the geomagnetic cutoff during a
flare, PF , be computed from the "quiet time" cutoff P0 using:

AP/P 0o 0.54 exp(-P 0 /2.9) (41)

and
PF PC - AP, (42)

where PF' 0 and AP are in GeV/ec.

45

NI N
APPENDIX 2: THE GEOMAGNETIC CUTOFF TRANSMISSION FUNCTION
The geomagnetic cutoff transmission function is calculated using the
method described in Adams et, al. (1983). Here we will describe how this
method has been extended to elliptical orbits.

Section 4.0 of Adams et. al. (1983) describes the procedure for finding
the geographic position of satellites in circular orbits. For spacecraft in
elliptical orbits, the offset dipole coordinates of the position must be
calculated in three steps. First, the position of the satellite along its
orbit must be calculated versus time. Second, the geographic latitude and
longitude of the sub-satellite point must be found. Third, the geographic
coordinates must be transformed into the offset and the tilted dipole
coordinates required by the Stormer approximation.

Numerous celestial mechanic texts describe elliptical orbits of satellites

constrained by Newton's laws. For convenience, this description will follow
the notation of Sterne (1960). Elliptical positions are described by an angle
a called the "true anomaly" and the distance r from the satellite to the
earth's center. These in turn are formulated in terms of the "mean anomaly"
M, the "eccentric anomaly" E, the orbit's period P, eccentricity e, and
semi-major axis a.

The period is:

3 /GM )1/2
P 21r (a e

where G is the gravitational constant and M is the earth's mass. The mean
anomaly is given by: e

M = (2r/P) t,

where t is the time since perihelion. The eccentric anomaly is then found by
iteratively solving the equation:

M = E - e sin E.

The values r and a are calculated from E:

r = a (1 - e cos E),

tan (W/2) ((1+e)/(1-e)) 1 /2 tan (E/2).

This last equation is solved uniquely with the added condition that 0/2 and
E/2 be in the same quadrant. For purposes of calculating geographic
coordinates, the displacement of the perigee from the ascending node is added
to this to obtain a final value of a.

The spacecraft's geographic latitude a and longitude * are calculated from

the above r and A, and the parameters of the orbit: the orbital inclination
60; the initial longitude, * , of the ascending node, and the angular
velocity, w, of the earth. The latitude is simply:

e =/ 2 - cos" (sine 0 sine).

46
The longitude is:
where 0 - /2 - wt+ tan" S +tan"I S2

sin(1 /2 (i/2-e 0 ) Icos[1 /2 (1r/2-0)3

S "------'---''
- ' -'---''--- -
sin[ 1/2 (r/2+e 0 )I sin( 1/2 ( /'2-1)]

and
cos[1 /2 (w/ 2 -eo )Icos[1 /2 (f/ 2 -0)]
S2 - ---------
2
cosC1 /2(w/ +eO )Isin[1 /2(w/2-S)J

The final step--transforming to offset-tilted dipole coordinates--is done

using the method of Smart and Shea (1977).

Appendix B of Adams at. al. (1983) contained some typographical errors.

This appendix is presented in its correct form as Appendix 6 of this report.

I\

47
 APPENDIX. 3
PARTICLE RANGE VS. ENERGY FOR VARIOUS IONS IN ALUMINUM

Preceding Page Blank

49
-Y
-.'
--
-.
- ~ - - -% -0 m o v.c.-~'- 0 u. - 0 N U -

U-C-=, Mc
U -N0 a 0 U'.~m 0 a, C-0r
V 0 0 S4: a''0 C- -r" U.0 0
o UN.C- - N *M%'. U'.
'.0 C --. N N N N f". z'.tt'.**U'.0 L--0,- - ýN

U, U'.Lngt r --r = zn 0 - - mMV' MMMMMM InmMM0

V' (" N N NNN

C\0%J 0M w0 ON'
U'. CV. LAO
e, M0 Q 10U'NC-- C- %Q
- C-
N -VCV'.0CV'.
N '.0 -
* ý
0% . . . .
#N0*0 V.U.C0 '00-'. 0ý 0ý C. 0% 0ý Ný '.

-w f- w - w waw-
-r- 0 4 CCV'.0%00U'.V'm aU'.o0%.
NU,0CyCV'. CU.N
%0 Ut.- U. 0 M' N0 00 M Q\ o
t".r- - C\- 0% -4 *ý 0 0A =w 0; 0;,0 9

t- *80Ci'.0N
0C-W *IW w vwwm 1l NW 1,*
9L.
za

1.4) 00

0% LJ C CaC? aJ iCal 71?(T 1?

.4 N
C) NNO0%N.-1
~0j WW14VANWS
00 0C q-
NU -%N 48N#4 0%C-
,- , W oww

M- CCal. ýlv ý 1: 1

CiiCVi Na
Cii -CJa--r LA
ii l\ 1-ii00i i CVi W\ CiMMU100
C\A l LIn

- N; -w 0N 00rm-r Am
Ctr a'00 ' en
='. 0% 0 M MV'M M M M %m M
-0 U' 0 C'-U'. Al'

oa U.0
C- '.0U .- 0-- 0%
W U'. Z%- z
00 C- cmo % N0c C- 0 C-M.-
* OU'. o MM
C\
0 0 C-N*00*
% 0
C-. 0*000:C . - C-*-0N,~'00VCY'.

ý~~~~N(.
ri 0% - N N N Nx

C.- tn N
00 Ci\
'. s 0 %aUt'.0 ChN * CV.- 0 0 t 7 t--
NOOM
NC ,0 C0-0%
.- * Nw%-0U-0--**r Ll 0 l
0% 0%0

cy~~ L 0000

0 N0 M0" 0 C-O N10 00 U'.' 90 ONU0 U'0000coo

C3d aC aC oooooaoooo
;d ;C
oooo CDoa 0 0~- N CV Mx i %0r-CO

50 ,o
 LI 00 a ON N 0 Cy0 u- 4 I c- t- MYCO U'0 U* 0L- N0 0 0 m (ON M
-0 t'0 4,
0-

0s0w4T-0. 00.-NCe~ s

NN N N ~jj-
N~ - - -- .- - -0000 -- - N N-

0 t ON U)- t---0 %a
000 % r W,0o - 0%-- '\aL\ - N n NNwNmN

0+ 0+

- d W
C-%0,WUW mNW 0%N
%*0 mw w -44C
W wt'iwN 19 IWO8 9WN .~C~
NN

Cit-
N N - C-
-r- - 0 0 m -\ .-. NNNNI

m4jM~ar %0 N0 M'U UML-

-- %N 4'0 -- ONN m(1' - M w N
00 ~ j<4 U4

-- '0M 4710
40-' Nm~ a~
0%Ct UC- -~ 0 0%' N t- ~.- j 0 00 Ln
M% U. %'. -N N
004 0 r 1 CO 4 0-' \ T LA. 4 rn t'
N''C\- U'l N
0 -- U'
- '

co00 -N('.U.C -
NmU.C - - - ---- --- ~xU.C
- -

Z7N~ m C% w C %g Qwý w 8ý U'=.0-

9,64 ~ 'NO t" %-NrN - m N
a%0 . m r m 0 ao - 0% A o w~ A r0 o m

.- - 0 0 0 0 - - - N N- N% Ný N p. Y. C' Y' .. ( C l'

.41.- ma U'.N CO - --
n.%Q -o -NCV' c-
W%0~ -, Nm N 4 %, '.C - - - Nv N CV'r44

Ln
ON 0 0\ - w
'. = 0 - %v ON (Y) - 0% '7 \0N I0
0mo Ncn4 ooC- '0ý
n
NNC- 00% ( LO 00%0 40%0-Cl'- OC'' O 0Qm
~c

W,~~ ~~ ~~ ~ w N N'.' qwW ., NW CV' t'-COmm

en%0
t- 00\ 51rcr-
U% \ w -tLo t r'QC CjC-
 APPENDIX 4
PARTICLE STOPPING POWER VS. ENERGY IN ALUMINUM

53
Preceding Page Blank
 C-4r ~ ~ ~ ~'0
~ q Lm srf M AL10 .' S%0%a%
-0% 47 0%4 0%

N- VN Z! M- CA (v (%I - cm (\l N
NNN (44 AA -w VN C.- c- 0A ad
LA a%0
LA Lo
zLc \a,

4r U0NM N N CV0
N - C- 5 m
4 N F-q %5 0r
mA t- cm
0%o0 4 C-OW04
L-- %
O~- gr m %C\ c* M%' =-
0(\s00 ( t.-
N - 0 N
(ý .. .. .% .. L ýý ý% t.ýý ýC
oV A%'j ~4 -- 4 44- 44--44N4M44444a
0z444n0 N L "I0 N 4

-D 0 '0 % N
RNN00 MC-t0%.A0o.m t-C-ON*UL- 0 0CN
000
(C- C-4 NO 0% ON ( j C'0N Mcvn MA'
'0 M- M- M- re'0
(V MMM O M

C%%0 0 0 0 (7%0 N NNNNN

NON-0ý Trt-NN n t-N MN.
N

C6:x
W . I
0 ma
Qr
%CL
~0ZZ'
0% C- m t-
0
0N w
N
LA
V%0
--
4r
~ 'o'J0
a,' ~ C-
N O
-

UU

0''M

0 1-L4 LAL
ct%-A0A0 0w w -W 0v%'0A V 8 t-,WIN A~0
9-%%o
0 % a --r'.NL
OSW (Y -0 ' -0 Mom
4 %%%

- ~~ ~ ~ - ~N NM N '0M M
0C - -C .Nv 'N' N. NALN NA NA ~ N. N N N'

Nv N

M w -w ko- 0%0n '0V 0% N-40% %0,~-~N0L

-V M- 0 00 t a -% LA 0 N N0%
C4 14 1ýO. .......... . . . . . .4 4. . . . . . . .

N NN
M mmw At wf ~ ~ ~ _r Cv (1 (r 'V Nr N N N NN N N N %Q
N N M

NtLA No
m% - N N N N11 LN
N0 a - 0 0 a- 0 0Ca C- N0
0% N 0% 0-L

0 o LA D 00
00000000 0N 0 LA 00 l" - I0 00000 0 0 00
0 ~~ ~ ~ ~ L0 0000 C- 0o 0~C9 4~:;
0000000~aC~c o

54

'Ur'k
 co0 t-No j o
acoL * -r( mmm A00 14
0N LA - (

'o ~( N0 N N -r '- - co 0% t- 00L AL AL t

UUUL c AL A

N .0 O CO
'0 ' 0 00 L- U% M ;; a (v U srv N 04 tN '- m
*V 0
Nw N - (A q cu-0
4 03'
Cl tLd N; LA UU
N Nr L¶
N u; NNNN LA N; N N

'o0c o . W rmt

CV
N Z
c..G~o

QA z

I -' AU*~mNNN.. -0 -C-00'L AL AL AL AL AL

C6Iww2 W w

cm0NC-' LUULLN r m CA( \ -- - - - - - - --

(flcn
N N NN NR NKNw N N Nw N NN WN N
.

.ý p% p0 pý. . . 0: W! t, . . . .pý p.1 p

cr'0-, LAL rA f cm
N Nm Nv y '- -- co t-PO
c- ' LA
o0 LA
JA LA
oA oAo' 8

~
2 LA
(V Cf NLA~
%Q%Q0%t.
Mv~0'Lt-A
O0'-( 0%0N
N-LAAt(1 ý"I-U
Q LA-
tO g%0U 0
(I &~A
%
L aUl
N
t- tJ N
enN( 0%LA Ua 00' NrMMN
CVr ul '-- N N -
-C-C' - 0 '0''' -% -0 -

N N
MONOt-ONO U'90 00000 00000oo000

WE ''-'0 -0 AUU~f - - - N-
-N -(n -n -C

~ 000 000 00 000 00 000 00 0 0 00

 APPENDIX 5
ENERGY IN SILICON
PARTICLE STOPPING POWER VS.

57 Preceding Page Blank

57
 CfL000zr~
LA 0. O0U,-"L
ov0 '

~ ~ ~~ ~N ~ ~ ~Y ~ ~AL~~c
1- 0'1100 !0 %W!qU4 AY
%. 1A m
-- m 4 S3

-r 0 Ar *r = xv
*aaaa a at-i
gri Aa ZarrTW az ar iz -V a- r

w
1.1. 'JW KWWMa~1 wIWW

NY Nv N
C4- N N N N M Ar W
MA; W 0 0A LA LIN 0 %0 %0 LA LA UN L

'00itN o-'0 400W

=' Rr'0o0MLLYL 'W0Sri 0wo

UNNLNUNO-N
*r ~ .I2t .nQx toP ** **
'00
\W W i '*L Lew) W0
m% K g m W
N LA00 %. --- NNNN~ Le~c~tLc~r~m~ .
0 ~aaa t aaawaaaaa

w~ 0; 't:,.a000e ''

%0
C u Wts ~q1W Af-am A, ý fs9 is0i AW l R (Y0 LA
fA-W.1N-a('flW ý

m; 1-'QV\0 OU'D0- W-0% - q- 4 C4

NN C4 NN N N; N4 NY 4 N4 N

"n 0

N11-1- Ln ar CIDCYO
N o%01-1 WA0% 0% QO% N -aAn

-. . . . . . . . . . . . . . .

A I.00

.~ N
t-0-00L- *'--000a-LaOaa 00% Lc-A .
o Daa
LAO 0 LAL0'-AO.NYIL 00 -OL-0aea0% 0a. 01-LA

JW M W a2W W W NO*2W(*2WE-Zm.2WWWLZ (0

0 - -- N N e~Lmm a -v uL LLALA LA a xr *-t ar mY

a mP
(A N N 4Nu

NNN N NNCL LLLL N . MMV N N N N NM N4 NM N NJ N N NY N

CY a

%
LnLA -NsO*%I 1: LA N- aa W--*% a0~
%*r O m'% 00NN 00

moot= N 00N'aL tn#MN-0%G

A LAo 00 LAOLACLA0I(-Mc tn-'

P.~P ma o u n "- 0 . . 0. . 0 0 PO- c

LA LA ,-- 80-0 -. - a.- - 0% 00 aMMNNN.
8LA

Z--0w M0
-
0 0 0
~M1 O LA0 LA
%'
N 0
-'
0 0 0
%L
a LA0
' -ALL-1
YCM.-p 10%
L00
ýA &ý-Q
L
E

58

P. 4 .
 ~l ii o
irM".1MOMM~ro i ii ii i ll

s 1I mm g w w g** I W t * W W N W a W A W N N g * w W

m7 40 0 04
11: um lp c
o tW( A4 ýu;t tuu ýL - nt

iS I 1 11 . 1*1 I 9 1 5' 11 1 1 * 11111

LAWN Cs-O00A
WW*N Wiq W 'Wig *KS

C.)gl LAW 'rAA 4 44 4 N4N mc

Air AAAJV7nnn(mnnm (M M MMnMnn( n m

i-Sol IMK

0- N* I 9ý7 I I T I 11 I * *
7 1 1 17 1 * 1* 1

mt-.U W 0' fA~-t

IQ ~
N le
so L

OCOOC OCO 0 000 0 0 0O0

C*;Ca
04!-4 * ;Zl IIm tviCY 9 474 aolC£IA %ii%0l W %W
U-% Il% L
U-% A 0 l ClN WW

Ln. go 40 - k-Lt-Q 000

Lt-.-2r 00 rn 0t00 0 0O
tgO 0 m
."nj (U(Ufr
COW!N4ý *t-0(N A0W 200 C'0
,LA' 0 000

11 9iTiI 1 3 T91 ýi7

ON -no 4: 'P
APPENDIX 6: THE WORLDWIDE VERTICAL GEOMAGNETIC CUTOFF AT 20KM ALTITUDE

LATITUDE LONGITUDE RIGIDITY LATITUDE LONGITUDE RIGIDITY

(GeV/ea) (GeV/eo)

80.00 0.00 0.02 70.00 0.00 0.26

80.00 15.00 0.04 70.00 15.00 0.34
80.00 30.00 0.06 70.00 30.00 0.41
80.00 45.00 0.09 70.00 45.00 0.47
80.00 60.00 0.09 70.00 60.00 0.49
80.00 75.00 0.10 70.00 75.00 0.51
80.00 90.00 0.10 70.00 90.00 0.52
80.00 105.00 0.11 70.00 105.00 0.55
80.00 120.00 0.11 70.00 120.00 0.59
80.00 135.00 0.11 70.00 135.00 0.60
80.00 150.00 0.08 70.00 150.00 0.62
80.00 165.00 0.06 70.00 165.00 0,56
80.00 180.00 0.04 70.00 180.00 0.47
80.00 195.00 0.00 70.00 195.00 0.36
80.00 210.00 0.00 70.00 210.00 0.23
80.00 225.00 0.00 70.00 225.00 0.13
80.00 240.00 0.00 70.00 240.00 0.06
80.00 255.00 0.00 70.00 255..00 0.00
80.00 270.00 0.00 70.00 270.00 0.00
80.00 285.00 0.00 70.00 285.00 0.00
80.00 300.00 0.00 70.00 300.00 0.00
80.00 315.00 0.00 70.00 315.00 0.05
80.00 330.00 0.00 70.00 330.00 0.11
80.00 345.00 0.00 70.00 345.00 0.18
75.00 0.00 0.10 65.00 0.00 0.58
75.00 15.00 0.14 65.00 15.00 0.72
75.00 30.00 0.18 65.00 30.00 0.80
75.00 45.00 0.20 65.00 45.00 0.89
75.00 60.00 0.23 65.00 60.00 0.93
75.00 75.00 0.25 65.00 75.00 0.97
75.00 90.00 0.25 65.00 90.00 1.01
75.00 105.00 0.26 65.00 105.00 1.03
75.00 120.00 0.27 65.00 120.00 1.12
75.00 135.00 0.28 65.00 135.00 1.19
75.00 150.00 0.26 65,00 150.00 1.20
75.00 165.00 0.24 65.00 165.00 1.13
75.00 180.00 0.20 65.00 180.00 0.95
75.00 195.00 0.14 65.00 195.00 0.74
75.00 210.00 0.09 65.00 210.00 0.53
75.00 225.00 0.03 65.00 225.00 0.32
75.00 240.00 0.00 65.00 240.00 0,17
75.00 255.00 0.00 65.00 255.00 0.09
75.00 270.00 0.00 65.00 270.00 0.05
75.00 285.00 0.00 65.00 285.00 0.014
75.00 300.00 0.00 65.00 300.00 0.08
75.00 315.00 0.00 65.00 315.00 0.16
75.00 330.00 O.02 65.00 330.00 0.28
75.00 345.00 0.07 65.00 345.00 0.42

60
LATITUDE LONGITUDE RIGIDITY LATITUDE LONGITUDE RIGIDITY
(GeV/ea) (GeV/ec)

60,00 0.00 1,14 50.00 0.00 3.21

60.00 15.00 1.34 50.00 15.00 3.54
60.00 30.00 1,46 50.00 30.00 3.81
60.00 45.00 1.57 50.00 45.00 3.97
60.'00 60.00 1.61 50.00 60.00 4.14
60.00 75.00 1.67 50.00 75.00 4.27
60.00 90.00 1.73 50.00 90.00 4.36
60.00 105.00 1.82 50.00 105.00 4.37
60.00 120.00 1.95 50.00 120.00 4.68
60.00 135.00 2.05 50.00 135.00 4.93
60.00 150.00 2.05 50.00 150.00 4.92
60.00 165.00 1.99 50.00 165.00 4.67
60,00 180.00 1.75 50.00 180.00 4.27
60.00 195.00 1.40 50.00 195.00 3.38
60.00 210.00 1.00 50.00 210.00 2.81
60.00 225.00 0,65 50.00 225.00 2.03
60.00 240.00 0.40 50.00 240.00 1.41
60.00 255.00 0.22 50.00 255.00 0.95
60.00 270.00 0.16 50.00 270.00 0.73
60.00 285.00 0.14 50.00 285.00 0.69
60.00 300.00 0.21 50.00 300.00 0.89
60.00 315.00 0.38 50.00 315.00 1.34
60.00 330.00 0.59 50.00 330.00 1.98
60.00 345.00 0.86 50.00 345.00 2.65
55.00 0.00 1,94 45.00 0.00 4.77
55.00 15A00 2.28 45.00 15.00 5.12
55.00 30.00 2.47 45.00 30.00 5,36
55.00 45.00 2.61 45.00 45.00 5.51
55.00 60.00 2.68 45.00 60.00 5.73
55.00 75.00 2.78 45.00 75.00 5.90
55.00 90.00 2.85 45.00 90.00 6.11
55.00 105.00 2.92 45.00 105.00 6.29
55.00 120.00 3.12 45.00 120.00 6.57
55.00 135.00 3.31 45.00 135.00 6.86
55.00 150.00 3.35 45.00 150.00 6.86
55.00 165.00 3.15 45.00 165.00 6.33
55.00 180.00 2.88 45.00 180.00 5.59
55.00 195.00 2.22 45.00 195.00 4.85
55.00 210.00 1.75 45.00 210.00 4.08
55.00 225.00 1.23 45.00 225.00 3.16
55.00 240.00 0.78 45.00 240.00 2.37
55.00 255.00 0.50 45.00 255.00 1.74
55.00 270.00 0.36 45.00 270.00 1.32
55.00 285.00 0.36 45.00 285.00 1.22
55.00 300.00 0.46 45.00 300.00 1.49
55.00 315.00 0.75 45.00 315.00 2.21
55.00 330.00 1.13 45.00 330.00 3.16
55.00 345.00 1.59 45.00 345.00 4.20

61
LATITUDE LONGITUDE RIGIDITY LATITUDE LONGITUDE RIGIDITY
(GeV/eo) (GeV/eo)

40.00 0.00 6.75 30.00 0.00 11.30

40.00 15.00 7.27 30.00 15.00 11.71
40.00 30.00 7.48 30.00 30.00 12.13
40.00 45.00 7.70 30.00 45.00 12.67
40.00 60.00 8.19 30.00 60.00 13.34
40.00 75.00 8.73 30.00 75.00 14.07
40.00 90.00 9.14 30.00 90.00 14.37
40.00 105.00 9.29 30.00 105.00 14.40
40.00 120.00 9.49 " 30.00 120.00 14.26
40.00 135.00 9.89 30.00 135.00 13.95
40.00 150.00 9.74 30.00 150.00 13.44
40.00 165.00 8.95 30.00 165.00 12.72
40.00 180.00 7.86 30,00 180.00 11.65
40.00 195.00 6.46 30.00 195.00 10.48
40.00 210.00 5.41 30,00 210.00 9.63
40.00 225.00 4.55 30.00 225.00 8.78
40.00 240.00 3.61 30.00 240.00 7.00
40.00 255.00 2.76 30.00 255.00 5.60
40.00 270.00 2.07 30.00 270.00 4.44
40.00 285.00 1.93 30.00 285.00 4.07
40.00 300.00 2.42 30.00 300.00 4.87
40.00 315.00 3.41 30.00 315.00 6.98
40.00 330.00 4.82 30.00 330.00 9.67
40.00 345.00 5.92 30.00 345.00 10.60
35.00 0.00 9.54 25.00 0.00 13.10
35.00 15.00 9.89 25.00 15.00 13.64
35.00 30.00 10.10 25.00 30.00 14.10
35.00 45.00 10,53 25.00 45.00 14.53
35,00 60.00 11.15 25.00 60.00 15.06
35.00 75.00 11.44 25.00 75.00 15.58
35.00 90.00 11.52 25.00 90.00 15.85
35,00 105.00 11.71 25.00 105.00 15.79
35,00 120.00 11,93 25.00 120.00 15.49
35.00 135.00 12.04 25.00 135.00 15.03
35.00 150.00 11.55 25.00 150.00 14.44
35.00 165,00 10.60 25.00 165.00 13.76
35,00 180.00 9.49 25.00 180.00 13.07
35,00 195.00 8.97 25.00 195.00 12.43
35.00 210.00 7.65 25,00 210.00 11.81
35,00 225.00 6.12 25.00 225.00 10.98
35.00 240.00 5.21 25.00 240.00 9.74
35.00 255.00 4.25 25.00 255.00 7.89
35.00 270.00 3.19 25.00 270.00 6.08
35.00 285.00 2.89 25.00 285.00 5.44
35.00 300.00 3.58 25.00 300.00 6.56
35.00 315.00 4.98 25.00 315.00 9.65
35.00 330.00 6.80 25.00 330.00 11.47
35.00 345.00 8.70 25.00 345.00 12.46

62
 LATITUDE LONGITUDE RIGIDITY LATITUDE LONGITUDE RIGIDITY
(GeV/ec) (GeV/ec)

20.00 0.00 14.11 10.00 0.00 14.73

20.00 15.00 14.62 10.00 15.00 15.26
20.00 30.00 15.09 10.00 30.00 15.80
20.00 45.00 15.57 10.00 45.00 16.34
20.00 60.00 16.12 10.00 60.00 16.94
20.00 75.00 16.63 10.00 75.00 17.44
20.00 90.00 16.87 10.00 90.00 17.67
20.00 105.00 16.75 10.00 105.00 17.56
20.00 120.00 16.37 10.00 120.00 17.18
20.00 135.00 15.83 10.00 135.00 16.65
20.00 150.00 15.20 10.00 150.00 16.10
20.00 165.00 14.55 10.00 165.00 15.61
20.00 180.00 13.93 10.00 180.00 15.16
20.00 195.00 13.40 10.00 195.00 14.75
20.00 210.00 12.91 10.00 210.00 14.39
20.00 225.00 12.33 10.00 225.00 14.00
20.00 240.00 11.47 10.00 240.00 13.44
20.00 255.00 9.69 10.00 255.00 12.55
20.00 270.00 7.82 10.00 270.00 11.50
20.00 285.00 6.84 10.00 285.00 11.07
20.00 * 300.00 8.65 10.00 300.00 12.16
20.00 315.00 11.45 10.00 315.00 13.09
20.00 330.00 12.76 10.00 330.00 13.76
20.00 345.00 13.54 10.00 345.00 14.25
15.00 0.00 14.61 5.00 0.00 14.50
15.00 15.00 15.14 5.00 15.00 14.99
15.00 30.00 15.65 5.00 30.00 15.52
15.00 45.00 16.17 .5.00 45.00 16.10
15.00 60.00 16.74 5.00 60.00 16.71
15.00 75.00 17.25 5.00 75.00 17.22
15.00 90.00 17.47 5.00 90.00 17.47
15.00 105.00 17.34 5.00 105.00 17.42
15.00 120.00 16.93 5.00 120.00 17.11
15.00 135.00 16.37 5.00 135.00 16.66
15.00 150.00 15.76 5.00 150.00 16.21
15.00 165.00 15.17 5.00 165.00 15.83
15.00 180.00 14.63 5.00 180.00 15.49
15.00 195.00 14.16 5.00 195.00 15.15
15.00 210.00 13.76 5.00 210.00 14.82
15.00 225.00 13.30 5.00 225.00 14.47
15.00 240.00 12.62 5.00 240.00 14.03
15.00 255.00 11.27 5.00 255.00 13.41
15.00 270.00 9.49 5.00 270.00 12.67
15.00 285.00 8.46 5.00 285.00 12.40
15.00 300.00 10.71 5.00 300.00 12.78
15.00 315.00 12.52 5.00 315.00 13.36
15.00 330.00 13.43 5.00 330.00 13.80
15.00 345.00 14.07 5.00 345.00 14.11

63

.- .
LATITUDE LONGITUDE RIGIDITY LATITUDE LONGITUDE RIGIDITY
(GeV/ec) (GeV/eo).

0.00 0.00 13.94 -10.00 0.00 12.11

0.00 15.00 14.37 -10.00 15.00 12.31
0.00 30.00 14.87 -10.00 30.00 12.71
0.00 45.00 15,46 -10.00 45.00 13.29
0.00 60.00 16.10 -10.00 60.00 13.89
0.00 75.00 16.62 -10.00 75.00 14.35
0.00 90.00 16.90 -10.00 90.00 114.68
0.00 105.00 16.94 -10.00 105.00 14.90
0.00 120.00 16.73 . -10.00 120.00 14.91
0.00 135.00 16.38 -10.00 135.00 14.80
0.00 150.00 16.05 -10.00 150.00 14.74
0.00 165.00 15.81 -10.00 165.00 14.84
0.00 180.00 15.59 -10.00 180.00 14.94
0.00 195.00 15.32 -10.00 195.00 14.90
0.00 210.00 15.03 -10.00 210.00 114.76
0.00 225.00 14.71 -10.00 225.00 14.55
0.00 240.00 14.33 -10.00 240.00 14.29
0.00 255.00 13.06 -10.00 255.00 13.94
0.00 270.00 13.32 -10.00 270.00 13.51
0.00 285.00 12.95 -10.00 285.00 13.10
0.00 300.00 13.05 -10.00 300.00 12.92
0.00 315.00 13.38 -10.00 315.00 12.82
0.00 330.00 13.58 -10.00 330.00 12.55
0.00 345.00 13.69 -10.00 345.00 12.20
-5.00 0.00 13.13 -15.00 0.00 10.75
-5.00 15.00 13.45 -15.00 15.00 10.91
-5.00 30.00 13.91 -15.00 30.00 11.23
-5.00 45.00 14.50 -15.00 45.00 11.69
-5.00 60.00 15.14 -15.00 60.00 12.18
-5.00 75.00 15.65 -15.00 75.00 12.70
-5.00 90.00 15.97 -15.00 90.00 13.00
-5.00 105.00 16.10 -15.00 105.00 13.25
-5.00 120.00 16.00 -15.00 120.00 13.25
-5.00 135.00 15.77 -15.00 135.00 13.32
-5.00 150.00 15.58 -15.00 150.00 13.48
-5.00 165.00 15.50 -15.00 165.00 13.78
-5.00 180.00 15.42 -15.00 180.00 14.09
-5.00 195.00 15.25 -15.00 195.00 14.24
-5.00 210.00 15.02 -15.00 210.00 14.24
-5.00 225.00 14.74 -15.00 225.00 14.15
-5.00 240.00 14.41 -15.00 240.00 13.97
-5.00 255.00 14.01 -15.00 255.00 13.69
-5.00 270.00 13.53 -15.00 270.00 13.30
-5.00 285.00 13.14 -15.00 285.00 12.89
-5.00 300.00 13.08 -15.00 300.00 12.59
-5.00 315.00 13.19 -15.00 315.00 12.31
-5,00 330.00 13.15 -15.00 330.00 11.76
-5.00 345.00 13.04 -15.00 345.00 11.15

64
LATITUDE LONGITUDE RIGIDITY LATITUDE LONGITUDE RIGIDITY
(GeV/ec) (GeV/ea)

-20.00 0.00 9.21 -30.00 0.00 6.36

-20.00 15.00 9.06 -30.00 15.00 6.02
-20.00 30.00 9.29 -30.00 30.00 5.86
-20.00 45.00 9.70 -30.00 45.00 5.79
-20.00 60.00 10.21 -30.00 60.00 5.42
-20.00 75.00 10.25 -30.00 75.00 5.23
-20.00 90.00 10.41 -30.00 90.00 5.10
-20.00 105.00 10.65 -30.00 105.00 5.16
-20.00 120.00 10.84 -30.00 120.00 5.19
-20.00 135.00 10.62 -30.00 135.00 5.39
-20.00 150.00 10.74 -30.00 150.00 5.89
-20.00 165.00 11.93 -30.00 165.00 6.58
-20.00 180.00 12.78 -30.00 180.00 7.99
-20.00 195.00 13.19 -30.00 195.00 9.45
-20.00 210.00 13.44 -30.00 210.00 9.43
-20.00 225.00 13.51 -30.00 225.00 10.64
-20.00 240.00 13.46 -30.00 240.00 11.85
-20.00 255.00 13.28 -30.00 255.00 12.00
-20.00 270.00 12.94 -30.00 270.00 11.88
-20.00 285.00 12.53 -30.00 285.00 11.43
-20.00 300.00 12.14 -30.00 300.00 10.75
-20.00 315.00 11.65 -30.00 315.00 9.83
-20.00 330.00 '10.73 -30.00 330.00 8.39
-20.00 345.00 9.78 -30.00 345.00 7.09
-25.00 0.00 7.55 -35.00 0.00 5.24
-25.00 15.00 7.50 -35.00 15.00 4.59
-25.00 30.00 7.72 -35.00 30,00 4.45
-25.00 45.00 7.88 -35.00 45.00 4.31
-25.00 60.00 7.86 -35.00 60.00 4.07
-25.00 75.00 7.37 -35.00 75.00 3.72
-25.00 90.00 7.08 -35.00 90.00 3.35
-25.00 105.00 7.24 -35.00 105.00 3.32
-25.00 120.00 7.43 -35.00 120.00 3.37
-25.00 135.00 7.71 -35.00 135.00 3.65
-25.00 150.00 8.47 -35.00 150.00 4.10
-25.00 165.00 9.55 -35.00 165.00 4.90
-25.00 180.00 10.22 -35.00 180.00 5.65
-25.00 195.00 11.22 -35.00 195.00 6.54
-25.00 210.00 12.01 -35.00 210.00 7.88
-25.00 225.00 12.63 -35.00 225.00 9.11
-25.00 240.00 12.76 -35.00 240.00 9.58
-25.00 255.00 12.72 -35.00 255.00 11.12
-25.00 270.00 12.46 -35.00 270.00 11.16
-25.00 285.00 12.04 -35.00 285.00 10.67
-25.00 300.00 11.53 -35.00 300.00 9.90
-25.00 315.00 10,74 -35.00 315.00 8.72
-25.00 330.00 9.63 -35.00 330.00 7.16
-25.00 345.00 8.36 -35.00 345.00 6.18

65
LATITUDE LONGITUDE RIGIDITY LATITUDE LONGITUDE RIGIDITY
(GeV/ec) (GeV/ec)

-40.00 0.00 4.29 -50.00 0.00 2.87

-40.00 15.00 3.74 -50.00 15.00 2.37
-40.00 30.00 3.40 -50.00 30.00 1.95
-40.00. 45.00 3.21 -50.00 45.00 1.60
-40.00 60.00 2.83 -50,00 60.00 1.27
-40.00 75.00 2.43 -50.00 75.00 0.94
-40.00 90.00 2.08 -50.00 90.00 0.66
-40.00 105.00 2.00 -50.00 105.00 0.554
-40.00 120.00 2.01 . -50.00 120.00 0.53
4-40.00 135.00 2.21 -50.00 135.00 0.60
-40.00 150.00 2.65 -50.00 150.00 0.884
-40.00 165.00 3.214 -50.00 165.00 1.15
-40.00 180.00 4.11 -50.00 180.00 1.63
-40.00 195.00 14.76 -50.00 195.00 2.24
-40.00 210.00 5.56 -50.00 210.00 2.94
-40.00 225.00 6.65 -50.00 225.00 3.76
-40.00 240.00 8.20 -50.00 240.00 4.48
-40.00 255.00 9.75 -50.00 255.00 5.57
-40.00 270.00 10.18 -50.00 270.00 7.02
-40.00 235.00 9.77 -50.00 285.00 7.57
-40.00 300.00 8.98 -50.00 300.00 6.98
-40.00 315.00 7.61 -50.00 315.00 5.68
-40.00 330.00 6.42 -50.00 330.00 4.51
-40.00 345.00 5.31 -50.00 345.00 3.50
-45.00 0.00 3.46 -55.00 0.00 2.23
-45.00 15.00 2.96 -55.00 15.00 1.79
-45.00 30.00 2.53 -55.00 30.00 1.42
-45.00 45.00 2.30 -55.00 45.00 1.12
-45.00 60.00 1.93 -55.00 60.00 0.87
-45.00 75.00 1.53 -55.00 75.00 0.53
-45.00 90.00 1.28 -55.00 90.00 0.34
-45.00 105.00 1.12 -55.00 105.00 0.23
-45.00 120.00 1.11 -55.00 120400 0.21
-45.00 135.00 1.25 -55.00 135.00 0.26
-45.00 150.00 1.51 -55.00 150.00 0.38
-45.00 165.00 2.04 -55.00 165.00 0.59
-45.00 180.00 2.72 -55.00 180.00 0.90
-45.00 195.00 3.33 -55.00 195.00 1.38
-45.00 210.00 4.24 -55.00 210.00 1.88
-45.00 225.00 4.93 -55.00 225.00 2.64
-45.00 240.00 5.91 -55.00 240.00 3.38
-45.00 255.00 7.83 -55.00 255.00 4.20
-45.00 270.00 9.00 -55.00 270.00 4.96
-45.00 285.00 8.76 -55.00 285.00 5.19
-45.00 300.00 7.84 -55.00 300.00 5.02
-45.00 315•00 6.91 -55.00 315.00 4.45
-45.00 330.00 5.63 -55.00 330.00 3.67
-45.00 345.00 4.32 -55.00 345.00 2.93

66
 LATITUDE LONGITUDE RIGIDITY LATITUDE LONGITUDE RIGIDITY
(GeV/ec) (-eVlec)

-60.00 0.00 1.78 -70.00 0.00 0.89

-60.00 15.00 1.32 -70.00 15.00 0.64
-60.00 30.00 1.03 -70.00 30.00 0.47
-60.00 45.00 0.75 -70.00 45.00 0.31
-60.00 60.00 0.49 -70.00 60.00 0.18
-60.00 75.00 0.30 -70.00 15.00 0.08
-60.00 90.00 0.15 -70.00 9@.tou 0.00
-60.00 105.00 0.08 -70.00 IOr.i•U, 0.00
-60.00 120.00 0.06 -70.J0 . f, 0.00
-60.00 135.00 0.08 -70 ,) 135,00 0.00
-60.00 150.00 0.14 -70.00 1 00 0.00
-60.00 165.00 0.27 -70.0G0 165.00 0 J3
-60.00 180.00 0.48 -70.00 161. 00 C,10
-60.00 195.00 0.79 -70.00 "111.00 0.22
-60.00 210.00 1.18 -70. 'n u210.On 0.41
-60.00 225.00 1.62 -70.% Z-5
2`3 o.
-60.00 240.00 2.23 -70.OU 2 ZJ..O0 0-96
-60.00 255.00 3.00 -70.00 25--,. Of I. 20
-60.00 270.00 3.77 -70.00 270.00 1o2.0
-60.00 285.00 3.95 -70.00 285.00 1.75
-60.00 300.00 3.97 -70.00 300.00 1.80
-60.00 315.00 3.52 & -70.00 315.00 1.67
-60.00 330.00 2.88 -70.00 330.00 1.39
-60.00 345.00 2.27 -70.00 345.00 1.14
-65.00 0.00 1.30 -75.00 0.00 0.59
-65.00 15.00 0.98 -75.00 15.00 0.43
-65.00 30.00 0.72 -75.00 30.00 0.30
-65.00 45.00 0.50 -75.00 45.00 0.19
-65.00 60.00 0.30 -75.00 60.00 0.10
-65.00 75.00 0.15 -75.00 75.00 0.04
-65.00 90.00 0.06 -75.00 90.00 0.00
-65.00 105.00 0.00 -75.00 105.00 0.00
-65.00 120.00 0.00 -75.00 120.00 0.00
-65.00 135.00 0.00 -75.00 135.00 0.00
-65.00 150.00 0.03 -75.00 150.00 0.00
-65.00 165.00 0.11 -75.00 165.00 0.00
-65.00 180.00 0.23 -75.00 180.00 0.05
-65.00 195.00 0.43 -75.00 195.00 0.12
-65.00 210.00 0.71 -75.00 210.00 0.23
-65.00 225.00 1.06 -75.00 225.00 0.36
-65.00 240.00 1.51 -75.00 240.00 0.54
-65.00 255.00 1.98 -75.00 255.00 0.72
-65.00 270.00 2.53 -75.00 270.00 0.91
-65.00 285.00 2.71 -75.00 285.00 1.02
-65.00 300.00 2.72 -75.00 300.00 1.03
-65.00 315.00 2.50 -75.00 315.00 1.05
-65.00 330.00 2.10 -75.00 330.00 0.88
-65.00 345.00 1.61 -75.00 345.00 0.76

67

ij
LATITUDE LONGITUDE RIGIDITY
(Ge V/ec)

-80.00 0.00 0.37

-80.00 15.00 0.28
-80.00 30.00 0.19
-80.00 '45.00 0.13
-80.00 60.00 0.07
-80.00 75.00 - 0.03
-80.00 90.00 0.00
-80.00 105.00 0.00
-80.00 120.00 0.00
-80.00 135.00 0.00
-80.00 150.00 0.00
-80.00 165.00 0.00
-80.00 180.00 0.03
-80.00 195.00 0.08
-80.00 210.00 0.13
-80.00 225.00 0.21
-80.00 240.00 0.30
-80.00 255.00 0.40
-80.00 270.00 0.48
-80.00 285.00 0.55
-80.00 300.00 0.56
-80.00 315.00 0.54
-80.00 330.00 0.49
-80.00' 345.00 0.42

68
 APPENDIX 7: SOURCE CODES OF THE CREME PROGRAMS

PROGRAM STASS
C
C THIS PROGRAM PREPARES ORBIT-A•ERAGED TRAPPED PROTON DATA
C FILES FOR THE LET PROGRAM.
C THE TRAPPED PROTON DATA CAN BE OBTAINED
C FROM REPORTS PUBLISHED BY THE NATIONAL SPACE SCIENCE DATA
C CENTER AT GODDARD SPACE FLIGHT CENTER. THE
C PROGRAM IS SET UP TO TAKE THE TABULAR DATA UNDER
C 'AVERAGED DIFFER. FLUX' IN THE 'COMPOSITE ORBIT
C SPECTRUM' TABLE OF ONE (OF THESE REPORTS. CARE SHOULD BE
C TAKEN TO BE SURE CUNTRIBUTIONS FROM SOLAR FLARE PROTONS
C HAVE NOT BEEN INCLUDED IN THIS TABLE.
C THE DATA SHOULD BF ENTERED IN PAIRS OF POINTS (I.E. THE
C PROTON ENERGY FOLLOWED BY THE CORRESPONDING FLUX).
C THE ENERGY SHOULD BE IN MEV AND THE FLUX SHOULD BE IN
C #/CM**2/SEC/KEV. DATA BELOW 10 MEV NEED NOT BE ENTERED.
C ENTER A PAIR OF ZERO VALUES TO CLOSE THE FILE AFTER ALL
C THE DATA HAS BEEN ENTERED. THIS FILE MAY BE EDITED
C AFTERWARDS TO CORRECT ANY ERRORS IN THE DATA YOU ENTERED.
C
CHARACTER* 2 ALPHA
WRITE (6, 100)
100 FORMAT(' ENTER THE NAME OF THE FILE TO CONTAIN THE PROTON DATA
1: ',$)
READ (5,200 )ALPHA
C
C NOTE: THIS FILE MUST BE RENAMED TO STASS.DAT BEFORE IT
C CAN BE USED IN THE LET PROGRAM.
C
200 FORMAT(Al2)
OPEN (UNIT-1-,STATUS 'NEWI ,FILE:ALPHA)
WRITE (6,300)
300 FORMAT(' ENTER THE DATA POINT PAIRS, STARTING WITH'/
2 ' 10 MEV AND PROCEEDING MONOTONICALLY TO THE HIGHEST'/
4 ' ENERGY. THE PAIRS ARE TO BE TAKEN FROM THE COMPOSITE'!
5 ' ORBIT SPECTRUM. THE PAIRS ARE ENERGY LEVEL IN MEV, FOLLOWED'/
6 ' BY THE AVERAGED DIFFER. FLUX IN #/CM**2/SEC/KEV.'/
7 ' ENTER A PAIR OF ZERO VALUES WHEN YOU ARE DONE.')
1:1
6 CONTINUE
WRITE(6,500)I
500 FORMAT(' ENTER PAIR ',12,': ',$)
ACCEPT *,E,F
IF (E.LE.O.) GOTO 7
WRITE (1, 600 )E, F
600 FORMAT0(X,F6.1,1X,E10.3)
I :1 +1
GOTO 6
7 CONTINUE
END

69
 PROGRAM GEOMAG
C THIS PROGRAM CALCULATES THE 2 DAY ORBIT AVERAGE OF
C THE GEOMAGNETIC TRANSMISSION FUNCTION FOR CIRCULAR ORBITS.
C WHEN YOU RUN THIS PROGRAM, ALL INPUT
C DATA WILL BE REQUESTED IN PROMPTS. OUTPUT DATA TABULATED IN
C INTERVALS OF .1 GV ARE STORED IN THE OUTPUT FILE GTRANS. DAT.
C (EXCEPTION: THE HIGHEST RIGIDITY VALUE WITH A ZERO
C TRANSMISSION FUNCTION (I.E., A PARTICLE WITH THAT RIGITITY
C CANNOT REACH THE SPACECRAFT AT ANY POINT ON THE SPACECRAFT'S
C ORBIT) MAY BE INTERPOLATED TO GIVE A VALUE NOT DIVISIBLE BY
C . 1 GV. THE RIGIDITIES WILL BE IN INCREASING ORDER IN ANY CASE.)
C AN INPUT FILE CALLED "CUTOFF.DAT" CONTAINS THE TABULATION
C OF WORLDWIDE VERTICAL GEOMAGNETIC CUTOFFS AT 20KM ALTITUDE,
C TAKEN FROM M. A. SHEA AND D. F. SMART, REPORT NO.
C AFCRL-TR-75-0185, HANSCOM AFB, HASS., 1975.
C AZ IS THE AZMUTH ANGLE OF THE PARTICLE WRT THE SPACECRAFT.
C ZE IS THE ZENITH ANGLE OF THE PARTICLE WRT THE SPACECRAFT.
C THESE ANGLES ARE BOTH IN DEGREES AND ARE DATAED IN AS ZERO.
C THEY CAN BE RESET TO ANY OTHER ARRIVAL DIRECTION.
C
DIMENSION MAT(200),CUTOFF(33,24),T(201),CF(201)
DATA (MAT (J),J=1,200)/2000/
DATA AZ/O./,ZE/O./,AZG/O./,ZEG/O./
OPEN (UNIT:=5, READONLY, SHARED, STATUS= 'OLD' ,FILE= 'CUTOFF. DAT')
OPEN(UNIT=16,FILE='GTRANS.DAT' ,STATUS='NEW')
C
C ASK FOR INPUT.
C IS THE SHADOW OF THE EARTH ON THE SPACECRAFT
C TO BE ACCOUNTED FOR IN THIS CALCULATION?
C
WRITE (6,411)
ACCEPT *, ISHADOW
C
C IS THE MAGNETOSPHERE QUIET OR STORMY?
C
WRITE (6,412)
ACCEPT *, ISTORM
C
CALL ORBIT(1,PERIODZLON,ZLAT,IRADIUS)
WRITE (6,1) PERIOD
1 FORMAT(C THE ORBITAL PERIOD IS ',E12.5,' SECONDS.')
C
C READ IN THE TABLE OF WORLD WIDE VERTICAL
C GEOMAGNETIC CUTOFFS AT 20KM ALTITUDE.
C THESE DATA ARE TABULATED EVERY 5 DEGREES IN LATITUDE
C AND EVERY 15 DEGREES IN LONGITUDE.
C THEY ARE TAKEN FROM M. A. SHEA AND D. F. SMART, REPORT NO.
C AFCRL-TR-75-0185, HANSCOM AFB, MASS, 1975.
C
DO 20 1=1,33
II =34-I

70
 DO 20 J=1,24
READ(15,405) CUTOFF(II,J)
C THIS ACCOUNTS FOR GEOMAGNEC.C ý,UTOFF SUPRESSION DURING
C LARGE MAGNETIC STORMS (FOLL9WTING ADAMS ET AL., 1981).
C
IF(ISTORM.EQ.1) CUTOFFCII, C,,vi(IIJ)*(1. -. 54#*XP(
1 -CUTOFF(II,J)/2.9))
20 CONTINUE
C
C AVERAGE THE CUTOFF AROUND THE 80 DEGREE LATITUDE LINES,
C NORTH AND SOUTH, TO APPROXIMATE THE CUTOFFS AT 90 DEGREES.
C
CN=O.
CS-O.
DO 30 J=1,24
CN =CN+C UTOFF (33,J)
30 CS--CS+CUTOFF(1,J)
CN :CN/24.
CS:CS/24.
C
C COMPUTE THE TOTAL NUMBER OF STEPS IN TWO DAYS IF WE MAKE
C 200 STEPS PER ORBIT.
C
JMAX=INT(2.*200.*86400./PERIOD + 1.5)
C
C COMPUTE THE STEP SIZE IN SECONDS.
C
STEP=PERIOD/200.
C
C COMPUTE THE VERTICAL CUTOFF AT THE SPACECRAFT
C POSITION FOR EVERY TIME STEP.
C
DO 50 J:1,JMAX
TIME --FLOAT (J -1) *STEP
CALL ORBIT (2, TIME, ZLON, ZLAT, ALT)
C
C COMPUTE THE TABULAR POSITION OF THE VERTICAL CUTOFF.
C
ZIzZLAT/5. +17.
ZJ-ZLON/15.+1.
ILOzINT (ZI)
IUP :ILO+I
JLO=INT(ZJ)
JUP :JLO+I
IF (JUP.EQ. 25) JUP=I
C
C INTERPOLATE THE VERTICAL CUTOFF TO THE EXACT LOCATION
C OF THE SPACECRAFT USING STORMER THEORY.
C
IF(ABS(ZLAT).GE.80.) GO TO 100
DI =ZI -FLOAT (ILO)
DJ=ZJ-FLOAT(JLO)

71
 XORB: (6371. 2.ALT )/6371 .2
RORD= 6391.2/6371.2
ZLONLO=(JLO-.1 )*15.
ZLONUP=ZLONLO.1 5.
ZLATLO=IL0*5. -85.
ZLATUP--ZLATLOS.5
SC $TORM ER (ZLAT,ZLON XORB AZ ,ZE )
SCLL=STORMER(ZLATLOZLONLORGRD,AZG,ZEG)
SCUL::STORMER (ZLATUP, ZLONLO, RORD, AZG, ZEG)
SCLU:STORMER(ZLATLOZLONUP,RGRD,AZGZEG)
SCUOUSTORMER (ZLATUP, ZLONUP, RGRDtAZG, ZEG)
Y1 :SC*CUTOFF(ILO, JLO)/SCLL
Y2= SC*CUTOFF(IUP,JLO)/SCUL
Y3= SC*CUTOFF(ILO,JUP)/SCLU
Y4=: SC*CUTOFF(IUPJUP)/SCUU

GO TO 200
C
C FOR ABS(LATITUDE).GT.80 USE THE CUTrOFFS AT THE POLE INSTEAD OF
C THE CUTOFFS AT FOUR NEARBY LOCATIONS.
C
100 CONTINUE
IF(ZLAT.LE.-8OJ)GO TO 110
SC=STORMER (90., 0.,XORP,AZ,ZE)
SCX=STORMER (90. ,O0.
,RGRD, AZG, ZEG)
C :SCCN/ SCX
GO TO 200
110 CONTINUE
SC=STORMER(-90. O.,XORB,AZtZE)
SCX:STORMER (-90., 0.,fGRD,AZG,ZEG)
C=SC*CS/SCX
200 CONTINUE
C
C HISTOGRAM THE CUTrOFFS.
C
IDEX=INT (C*10. )+1
MAT (IDEX )=MAT (IDEX ).1
C TO PRINT OUT THE LONGITUDE, LATITUDEt AND CUTOFF AT EACH
C POINT ARROUND THE SPACECRAFTfS ORBIT, JUST ENABLE THE FOLLOWING
C STATEMENT BY REMOVING THE COMMENT C.
C WRITE (60 435)ZLON,ZLAT, C
50 CONTINUE
CMAT :0.
C
C SAVE THE THRESHOLD.
C
JSAV =0
DO 300 J=1,200
C
C CONVERT THE HISTOGRAM TO TRANSMISSION.

72
"• CMAT-FLOAT (MAT (J) )/FLOAT
(JMAX)+.CMAT
IF((JSAV.EQ.O).AND.(CMAT.GT.O)) JSAV•J+I
CF(J+1 )rFLOAT(J )/l O.
300 T(,+l )=CMAT
CF(1 )=0.O
T(I )vO.0
C INTERPOLATE THE HIGHEST ZERO TRANSMISSION VALUE TO BE
C JUST AT THE CUTOFF THRESHOLD.
C
DT=T (JSAV+I )-T (JSAV)
IF(DT.EQ.O.) GO TO 301
CFT--CF(JSAV)-T (JSAV)*(CF(JSAV+1)-CF(JSAV) )/DT
CF(JSAV-1 )=AMAX1 (CFT,CF(JSAV-1))
301 CONTINUE
C
C THIS IS A CORRECTION FOR THE EARTH'S SHADOW ON THE SPACECRAFT
C ACCORDING TO SIMPLE GEOMETRICAL OPTICS.
C
IF(ISHADOW.NE.1) GO TO 390
DO 380 J=1,200
T (J)=T (J)*(I. -0.5'(I,. -((6371.2+ALT )**2. -(6371.2)*'2. )**.5/
1 (6371.2+ALT)))
380 CONTINUE
390 DO 400 J-1,200
400 WRITE (16, 410 )CF(J ), T(J)
405 FORMAT(17X,F6.2)
410 FORMAT(X, F6.3, 5X, F8.6)
411 FORMATO(X,'DO YOU WANT TO INCLUDE THE EFFECT OF THE SHADOW' ,/,
1 ' OF THE EARTH? (1 FOR YES; 0 FOR NO): ',$)
412 FORMAT(IX,'ENTER THE MAGNETIC WEATHER CONTITION: 1 FOR STORMY;
1 0 FOR QUIET: ',$)
435 FORMAT(' LONGITUDE ',F6.2,', LATITUDE ',F5.2,' CUTOFF -,
1 F6.3,'.')
END

73
 FUNCTION STORMER (GCLATDGCLONDRGCAZ,ZE)
C WE DID NOT WRITE THIS SUBROUTINE. WE HAVE MADE NO CHANGES IN
C IT IN 1984. C

COMMON/KARL/RED, EDLAT, AZM, ZEM ,GAMMA

C THIS FUNCTION TRANSFORMS A GEOGRAPHIC LOCATION AND ARRIVAL

"C DIRECTION INTO OFFSET DIPOLE COORDINATES, THEN COMPUTES THE
C STORMER CUTOFF IN GV AND RETURNS THE RESULT. THE OFFSET DIPOLE
C COORDINATES ARE AVAILABLE IN THE COMMON BLOCK /KARL/.

C GCLATD IS GEOCENTRIC LATITUDE IN DEGREES

C GCLONG IS GEOCENTRIC LONGITUDE IN DEGREES
C RGC IS
RADIAL DISTANCE FROM4GEOCENTER IN EARTH RADII
C AZ IS
GEOGRAPHIC AZIMUTH
C ZE IS
GEOGRAPHIC ZENITH
C RED RADIAL DISTANCE FROM OFFSET DIPOLE POSITION IN
IS
EARTH RADII
C EDLAT IS THE GEOMAGNETIC LATITUDE IN OFFSET DIPOLE COORDINATES
C AZM IS GEOMAGNETIC AZIMUTH IN OFFSET DIPOLE COORDINATES
C ZEM IS GEOMAGNETIC ZENITH IN OFFSET DIPOLE COORDINATES
C GAMMA IS GAMMA ANGLE MEASURED FROM MAGNETIC EAST
C
DATA JDATA, NOPT/2 0/,PI, RAD, PI02, XEDFGC, YEDFGC, ZEDFGC, CP, SP
1 ,ST,CPCTCPST,SPCTSPST,XGMEDYGMED,ZGMED/16*-8000./
DATA SMALL/i o OE-35/
DPEC(U):-SIGN(I./SNGL(DSQRT(1DO÷DBLE((U/ZEDRTL)*2 ))),ZRTL*ZEDRTL)
IF(JDATA.EQ.77) GO TO 10
PI - ACOS(-I.O)
RAD 180.O/PI
PI02 = PI/2.0
TWOPI u PI*2.0
DATA ERAD, THETAD, PHID, R1KM ,TH IDEG, PHIDEG/6371.2,
1 11.4354,-290.2392,450.2586,72.8278,148.7753/
NOPT z 0
SQRT3 z SQRT (3.0)
C ENTER GEOMAGNETIC DATA, IGRF 1975
C SEE JGR, 81, 5163, 1976
C DT IS NUMBER OF YEARS SINCE 1975
DT - 5.0
G01 z -30186.0 + 25.6*DT
G02 -1898.0 - 24.9*DT
Gl1 -2036.0 + 10.O'DT
G12 2997.0 + 0.7*DT
Hl1 5735.0 - 10.2*DT
H12 z -21241.0 - 3.0ODT
G22 z 1551.0 + 4.3'DT
H22 = -37.0 - 18.9*DT
IF(NOF¶.EQ.1)PRINT 1000, GO1, G02, Gil, G12, G22, H11, H12, H22
C COMPUTE POSITION OF OFFSET DIPOLE

74
. . .. I.S.
 HO --SQRT ( G01 *r01..G11'G11.sHll 'H11
HOSQ HO'HO
ELO 2.0 1 001N002+(Gll1 012.4H11IHl2)*SQRT3
ELi -0114G02+(G01*Gl2+Gll*G22+sHll"H22)*SQRT3
EL2 *-Hll*GO2+I(GO1*Hl24ill*G224GI11H22)*SQRT3
C E :(ELOIGO1+EL1NGll.EL2IH11)*4.O*HOSQ
E =(ELO0~01+EL1'Gl1+EL2*Hll)/(4.0'HOSQ)
IF(NOPT.EQ.1) PRINT 1011, ELO, EL1, EL2t El, HO
1011 FORMAT(1H , 8E15.5)
C XEDFGC aERAD*(EL1-.G11'E)/(3.0O'HOSQ)
XEDFGC z (EL1-G11*E)/(3.O'HOSQ)
C YEDFGC aERAD*(EL2-HhllE)/(3.O'HOSQ)
YEDFGC = (EL2-H11'*E)/C3.O*HOSQ)
C ZEDFGC =ERAD*(ELO-GO1'E)/(3.O'HOSQ)
ZEDFGC = (ELO-.GO1*E )/(3 Q*HOSQ)
REDFGC =SQRT (XEDFGC*XEDFGC.YEDFGC*YEIFGC.ZEDFGC*ZEDFGC)
IF(NOPT.EQ.1)PRINT 3001, XEDFGC, YEDFGCq ZEDFGC, REDFGC
3001 FORMAT O1H , 4F1O.14 3X, fXEDFGC, YEDFGCq ZEDFGC, REVFGC')
1000 FORMAT O1H ,10F13.5)
1010 FORMAT(1H0, 815.5/1H 98F15.5)
THETA =THETAD/RAD
PHI :PHID/RAD
CP COS (PHI)
SP SIN(PHI)
ST SIN (THETA)
CT COS (THETA)
CPCT =CP*CT
CPST a CP*ST
SPCT =SP*CT
SPST =SP*ST
RlER = R1KM/ ERAD
TH1RAD zTHiDEG/RAD
PHlRAD =PHlDEG/RAD
IF(NOPT.EQ. 1)PRINT 1000, R1KM,THlDEG, PH IDEG, RIER, THiRAD, PH iRAD
XGMED =X)EDFGC*CPCT -YEDFGC 1 SPCT -ZEDFGC*ST
YGMED = XEDFGC*SP +YEDFGC*CP
ZGI4ED aXEDFGC*CPST -YEDFGC*SPST 4.ZEDFGC*CT
IF(NOPT.EQ.1)PRINT 3002, XGMED, YGMED, ZGMED
3002 FORMAT(1H , 3F10.Z4, 13X9 'XGMED, YGMED, ZG?4ED')
IF(NOPT.EQ.1)PRINT 1010, CF. SF, CT, ST, CPCT, CPST, SPCT, SFST
JDATA =77
10 CONTINUE
C ITERATE TO FIND COORDINATES OF OFFSET NORTH DIPOLE AT ANY
C LATITUDE
C FIRST GUESS FIND OFFSET NORTH DIPOLE AT DISTANCE RGC
ZDEDNP zRGC
100 XODNF XGMED*CPCT + YGMED'SP +~ZDEDNP'CPS'T
YODNP -XGMED'S PCT + YGMED*CP - ZDEDNF'S PST
ZODNP -XGMED*SP +. ZDEDNP*CT
DODNP aSQRTCXODNP*XODNP + YODNP#YODNP + ZODNF'ZODNP)
DIFLA =DODNP - RGC
IF(ABS(DIFLA) - 1.OE-5) 120, 120, 110

15

jwml
 110 ZDEDNP z ZDEDNP - DIFLA
4001 FORMAT O1H , 5XtfODC 0, 0, 9 F7.5, GC X, Y, Z OF 13F8. 5,
1 1 DODNP ='F9.59' DIF OF 'F9.6, I AT LOND LATIF1O.4, F8.4)
GO TO 100
120 CONTINUE
PHINOF a ATAN2(YODNPXODNP)*RAD
IF(PHINOF. LT.O0.0 PHINOF zPHINOF + 360.0
TNOF m -ACOS(ZODNP/DODNP)*RAD +. 90.0
IF(NOPT.EQ. 1)PRINT 4001lZDEDNP, XODNP, YODNP,ZODNP, DODNP, DIFLA,
1 PHINOFTNOF
SGCLATD vSIN (GCLATD/RAD)
CGCLATD =COS(GCLATD/RAD)
SGCLOND zSIN (GCLOND/RAD)
CGCLOND -- COB (GCLOND/RAD)
C GET GEOCENTRIC X Y Z COORDINATES
XGC =RGC*CGCLATD*CGCLOND
YGC zRGC*CGCLATD*SGCLOND
ZGC zRGC*SKgLATD
GCT =(90.0 - GCLATD)/RAD
SGCT SIN(GCT)
CGCT COS(GCT)
C FIND X Y Z IN LOCAL COORDINATES OF X=O, Y=-O, Z
C THE LOCAL COORDINATE Z AXIS PASSES THRU P
C THE LOCAL COORDINATE X,Z PLANE CONTAINS P
GCROTL ATAN2(YGC,XGC)
IF(NOPT.EQ.1)PRINT 2001, XGC, YGC, ZGC, GCROT
2001 FORMAT(1H , WF10. 4, 3X, 'XGC, YGC, ZGC, GCROT')
SGCROT =SIN(GCROT)
CGCROT =COB (GCROT)
XRL zXGC*CGCROT*CGCT +1YGC*SGCROT'CGCT - ZGC*SGCT
YRL m-XGC'SGCROT + YGC'CGCROT
ZRL =XGC*CGCROT*SGCT + YGC*SGCROT'SGCT +ZGC*CGCT
2002 FORMAT(1H , 3F10.41, 13X, 'XRL, YRL, ZEL')
IF(NOPT.EQ.1)PRINT 2002, XRL, YRL, ZRL
C DETERM4INE LOCATION OF OFFSET DIPOLE CENTER IN THESE SAME
C ROTATED LOCAL COORDINATES
*1XEDRL =XEDFGC*CGCROT*CGCT +4 YED)FGC"SGCROT*CGCT - ZEDFGC*SGCT
YEDRL = -XEDFGC*SGCROT + YEDFGC*CGCROT
ZEDRL =XEDFGC*CGCROT'SGCT + YEDFGC*SGCROT'SGCT + ZEDFGC*CGCT
IF(NOPT.EQ.1)PRINT 3003, XEDRL, YEDRL, ZEDRL
3003 FORMAT (0H, 3F10.4, 13X, IXEDRLM YEDRLI, ZEDRLI)
XEDRTL =XEDRL
YEDRTL =YEDRL
ZEDRTL =ZEDRL - ZRL
IF(NOPT.EQ.1)PRINT 2303, XEDRTL, YEDRTL, ZEDRTL
2303 FORMAT OHf t 3F10.4I, 13Xt, 'XEDRTL, YEDRTK, ZEDRTLI)
C TRANSLATE TO LOCAL COORDINATE SYSTEM WITH ORIGIN AT SURFACE
XRTL zXRL
YRTL =YRL
ZRTL :--ZRL
XEDP =XRTL + XEDRL
YEDP xYRTL + YEDRL

76
 ZEDP a ZRTL + ZEDRL
RED c SQRT (XEDP*XEDP+YEDP*YEDP+ZEDP'ZEDP)
2302 FORMAT (OHI 3F10.4, 13X, 'XRTL, YRTL, ZRTL')
IF(NOPT.EQ.1)PRINT 2302, XRTL, YRTL, ZRTL
C. EARTHS SURFACE AT A SPECIFIED ALTITUDE
C POSITION OF OFFSET NORTH DIPOLE IN LOCAL COORDINATE SYSTEM
XODNPR: XODNP*CGCROT*CGCT + YODNP*SGCROT*CGCT - ZODNP*SGCT
YODNPR= -XODNP*SGCROT + YODNPOCGCROT
ZODNPRz XODNP*CGCROT*SGCT + YODNP'SGCROT*SGCT + ZODNP*CGCT
XODNPT = XODNPR
YODNPT = YODNPR
ZODNPT = ZODNPR - ZRL
IF(NOPT.EQ.1)PRINT 1103, XODNPT, YODNPT, ZODNPT
1103 FORMAT(1H ,10X,IXODNPT :'F1O.L4,2X,'YODNPT :'F10.4,2X,'ZODNPT ='
1 F10.4,3X, 'OFFSET N DIPOLE IN LQCAL COORDINATESt)
ROTM a ATAN2(YODNPT, XODNPT) + PI
C FIND ANGLE FROM GEOGRAPHIC NORTH
C NEGATIVE - ROTATION FROM GEOGRAPHIC NP CLOCKWISE
C POSITIVE - ROTATION FROM GEOGRAPHIC NP CCW
SROTM a SIN(ROTM)
CROTM = COS(ROTM)
RCTMD = ROTM*RAD
2327 FORMAT (OH , F15.5, 3X, ' ROTM IN DEGREES MEASURED CCW SO -X
1 WILL POINT TOWARD OFFSET NORTH DIPOLE AXIS')
IF(NOPT. EQ. 1)PRINT 2327, ROTMD
C FIND COMPONENTS OF UNIT VECTOR AT ARBITARY AZIMUTH AND ZENITH
PLAZ z -AZ/RAD + PI
TLZE a ZE/RAD
SPLAZ = SIN(PLAZ)
CPLAZ = COS(PLAZ)
STLZE = SIN(TLZE)
CTLZE = COS(TLZE)
XLD z STLZE*CPLAZ
YLD c STLZE'SPLAZ
ZLD z CTLZE
IF(NOPT.EQ.1)PRINT 2005, XLD, YLD, ZLD, AZ, ZE
2005 FORMAT (0H, 5F10.4, 3X, 'UNIT VECTOR COMPOENTS AT AZ & ZE')
C
C FIND COMPONENTS OF UNIT VECTOR IN DIPOLE RADIAL COORDINATES
C
C ROTATE AROUND Y AXIS SO -Z AXIS PASSES THROUGH XED, 0, ZED
C NEW VECTOR IS VA = ZRTL + ZEDRTK + XEDRTL
C ANGLE BETWEEN VECTOR FROM POINT LOCAL ORIGIN TO GEOCENTER
C AND VECTOR FROM POINT LOCAL ORIGIN TO XED, 0, ZED

C JIM LANCMORTHY 'S FIX

"CA=DPEC (XEDRTL)
C CA z ZRTL*ZEDRTL/(ABS(ZRTL)*SQRT(ZEDRTL#ZEDRTL + XEDRTL*XEDRTL))
A = ACOS(CA)
IF(XEDRTL.GT.O.O) A -A
SA a SIN(A)

77

*
Al
 ADEG aA*RA D
IF(NOPT.EQ. 1)PRINT 1000, CA, At SA, ADEG
XLP = XLD*CA + ZLD *SA
YLP = YLD
ZLP =-XLD*SA + ZLD*CA
IF(NOPT.EQ.1)PRINT 5001, XLP, YLP, ZLP
5001 FORMAT(1H , 3F10.~4, 13X,'XLP, YLP, ZLP ')
C ROTATE AROUND X PRIME AXIS SO -Z PASSES THROUGH XEDv YED, ZED
CB=DPEC(YEDRTL)
C CBE ZRTL*ZEDRTL/(ABS(ZRTL)*SQRT(ZEDRTL*ZEDRTL + YEDRTL'YEDRTL))
B z ACOS(CB)
IF(YEDRTL.GT.O0.O) B =-B
SB =SIN(B)
BDEG --B*RAD
IF(NOPT.EQ.1)PRINT 1000, CE, B, 6B, BDEG
XLPP xXLP
YLPP =YLP*CB + ZLP*SB
ZLPP =-YLP*SB + ZLP*CE
IF(NOPT.EQ.1)PRINT 5002, XLPP, YLPP, ZLPP
5002 FORMAT(1H , 3F10.4, 13X,'XLPP, YLPP, ZLPP ')
C ROTATE AROUND ZPP AXIS SO -X AXIS PASSES THROUGH NORTH
C OFFSET DIPOLE AXIS
ZLDM =ZLPP
X.LD?4 = XLPP*CROTh + YLPP'SROTh
XLDM a XLPP*CROfl4 -YLPP*SR0TM
YLDM =-XLPP*SROTh + YLPP*CROTM
YLLM - XLPP*SROfl4 + YLPP*CROTh
IF(NOPT.EQ.1)PRINT 1101, XLD, YLD, ZLD, XLDM, YLDM, ZLDM
1101 FORMAT OHM , 'UNIT VECTOR IN LOCAL COORDINATES 1, 3F8.5,5X,
1 'UNIT VECTOR IN LOCAL MAGNETIC COORDINATES', 3F10.5)
C FIND AZIJMITH ANGLE OF UNIT VECTOR IN LOCAL DIPOLAR RADIAL COOR
IF((ABS(YLDM).GT.SMALL).OR.(ABS(XLDM)IGT.SMALL)) GO TO 1102
PAZM=O. 0
Go TO 1104I
1102 PAZM ATAN2(YLDM,XLDM)
1104~ AZM (PI - PAZM)*RAD
IF(AZM.GT.360.O) AZM =AZM - 360.0
ZEM ACOS(ZLDM)*RAD
C FIND GAMMA ANGLE
GAMMA = ACOS(YLDM)*RAD
C TRANSFORM TO OFFSET DIPOLE COORDINATES
XED1 :XGC-XEDFGC
YED1 :YGC-YEDFGC
ZEDI =ZGC -ZEDFGC
C FIND THE Z COORDINATE IN OFFSET DIPOLE COORDINATES
ZED2 mXED 1*C PST-YED 1 *SPST+ZED 1 CT
C FIND THE GEOMAGNETIC LATITUDE
EDLAT=RAD*(PI02-ACOS(ZED2/RED))
COSLDA -COS (EDLAT/RAD)
STORMER:60.*COSLDA'*4./(RED*RED*(1.+SQRT(1 .-COSLDA**3.*YLDM))*2)
RETURN
END

78*
 SUBROUTINE ORBIT (NTIME,ZLON,ZLAT,RADIUS)'
o aaaa*a*a*a*aaa*aaiaaaaaaa*aaa*aa*aa****aaa*aa*a***.*a**** **a***
C THIS SUBROUTINE ACCEPTS INPUT CONCERNING SATELLITE
C ORBITS AND CALCULATES THEIR GEOGRAPHICAL LOCATION.
C NaO: ACCEPT DATA ON ORBIT (FAST MODE).
C N=1: ACCEPT DATA ON ORBIT (COMPLETE MODE).
C N=2: CALCULATE ORBIT AS A FUNCTION OF TIME.
C CLEARLY DATA MUST BE INPUT BEFORE COMPUTATIONS.
C ON DATA INPUT, TIME RETURNS THE ORBITAL PERIOD.
C DURING ORBIT CALCULATIONS, TIME IS AN INPUT VARIABLE.
C
"DATA PIALTA,ALTPREE,Al,A2,A3,XI,W1,RMAJ,W2,FACT
1,THO,PHO,PSIXIO /17'-8000./
C
C THESE VARIABLES HAVE BEEN DATAED IN HERE TO FORCE THE COMPUTER
C TO CREATE STORAGE LOCATIONS FOR THEM BETWEEN CALLS OF THIS
C SUBROUTINE. THE DATAED-IN VALUES ARE NEVER USED BY THE PROGRAM.
C
PI1:355./I 13.
IF (N.EQ.2) GO TO 1000
C
C ACCEPT ORBITAL DATA.
C aaaJaJJaaaaaaaaaaaaaaaaaaaaaJaaaaaJJaaaaaaaaaaaJaJaJaaaaaaJaaaa
C ALTA:ORBITAL ALTITUDE AT APOGEE (KILOMETERS).
C ALTP=ORBITAL ALTITUDE AT PERIGEE (KILOMETERS).
C RE:RADIUS OF EARTH (KILOMETERS).
C E=ORBIT ECCENTRICITY,
C Ai=ORBITAL INCLINATION (DEGREES).
C A2iINITIAL LONGITUDE OF ASCENDING NODE (DEGREES).
C A3:INITIAL DISPLACEMENT FROM ASCENDING NODE (DEGREES).
C XIfDISPLACEMENT OF PERIGEE FROM ASCENDING NODE (DEGREES).
C
C WHAT IS THE ALTITUEDE AT APOGEE?
C
WRITE(6,420)
ACCEPT , ALTA
C
C WHAT IS THE ALTITUDE AT PERIGEE?
C
WRITE(6,400)
ACCEPT *, ALTP
RE =6371.2
E=(ALTA-ALTP)/(ALTA÷ALTP+2.*RE)
IF (E.LT..O0001) E=O.
C
C WHAT IS THE ORBITAL INCLINATION?
C
WRITE(6,405)
ACCEPT , Al
A2--O.
A3=O.

79
 XI -0.
IF (N.EQ.O) GO TO 100

C WHAT IS THE INITIAL LONGITUDE OF THE ASCENDING NODE?

C
WRITE(6,410)
ACCEPT *, A2
C
C WHAT IS THE INITIAL DISPLACEMENT FROM THE ASCENDING NODE?
C
WRITE (6, 415)
ACCEPT *, A3
IF (E.EQ.O.) GOTO 100
C
C WHAT IS THE DISPLACEMENT OF THE PERIGEE FROM THE ASCENDING NODE?
C
WRITE (6,425)
ACCEPT *, XI
100 CONTINUE
C
C WI=ANGULAR VELOCITY OF EARTH (RADIANS/SEC).
C RNAJ=SEMI-MAJOR AXIS (KILOMETERS).
C W2=MEAN ORBITAL ANGULAR VELOCITY (RADIANS/SECOND).
C TIME=ORBITAL PERIOD (SECONDS).
C FACT=A USEFUL FACTOR.
C
W1=7.27E-5
RMAJ=(ALTP+RE)/(1.-E)
W21I. 24E-3*(RE/RMAJ )*1I.5
TIME :2. *PI/W2
FACT=SQRT((I .+E )/(I .- E))
C
C DEFINE MORE USEFUL ANGLES.
c
THO =PI *A1 /180.
PHO=PI*(A2-90.) /180.
PSI=PI*A3/1 80.
XIO, PI *Xl /180.
C FORMAT STATEMENTS.
C
400 FORMAT(IX,'ENTER ALTITUDE AT PERIGEE (KILOMETERS): '$)
405 FORMAT(IX,'ENTER ORBITAL INCLINATION (DEGREES): '$)
410 FORMAT(IX,'ENTER INITIAL LONGITUDE OF ASCENDING NODE',
1 ' (DEGREES): '$)
415 FORMAT(1X,'ENTER INITIAL DISPLACEMENT FROM ASCENDING',
I ' NODE (DEGREES): '$)
420 FORMAT(IX,'ENTER ALTITUDE AT APOGEE (KILOMETERS): '$)
425 FORMAT(OX,'ENTER DISPLACEMENT OF PERIGEE FROM',
1 ' ASCENDING NODE (DEGREES): '$)
RETURN
1000 CONTINUE

80
I
C
C COM4PUTE SATELLITE POSITION.
C
C QM:MEAN ANOMALY.
C
IF (E.NE.0.) GOTO 1005
CM -W2'TIME+PSI
GOTO 1010
1009 CONTINUE
YSS:(PSI-XIO)/2.
QEO=2 .'ATAN2 (SIN (YSS) ,FACT*COS (YSS))
Qt4O-QEO-E'SINCQEO)
CM -W2 'TIME -QMO
1010 CONTINUE
C
C QE=ECCENTRIC ANOMALY.
C
QE -QM
DEL: 1.
1014 CONTINUE
QTEMP :QE
QE=QM.E'SIN(QE)
DEL :QE-.QTE?4P
IF (ABS(DEL).GT..00O1) GOTOb1014
C
C QT:TRUE ANOMALY.
C
IF (E.NE.O.) GOTO 1019
QT zQE
GOTO 1020
1019 CONTINUE
QECYC=INT(QE/2./PI)
QERED :QE-2.'P1 'QECYC
AA1=F'ACT'SIN(QERED/2.)
AA2 =COS (QE RED /2.)
QTREDx2.'ATAN2(AA1,AA2)
IF (QTRED.LT.O.) QTRED:QTRED,2.'PI
QT-QTRED+2. 'P1 QECYC+XIO
C
C NOTE* ANOMALY COMPUTATIONS ARE DONE FROM PERIGEE
C WHILE ORBIT COMPUTATIONS BELOWJ ARE DONE FROM THE
C ASCENDING NODE. THE FACTOR OF XIO CORRECTS THIS.
C
1020 CONTINUE
C
C ZLAT=LATITUDE.
C
R1=SIN(THO )'SIN(QT)
THP zACO0S(H 1
ZLAT:-90. -1 80.'*THP/PI
C
C ZLON=LONGITUDE.

81
C
RP--. 5*(PI /2.+THO)
RM:.5 (PI /2.-THQ)
RF: 5 (PI /2. -QT)
IF (SIN(RF).NE.0.) GOTCr 1029
PH P:PI+PHO-W I *TIME
GOTO 1030
1029 CONTINUE
SQlzSIN(RH) t COS (RF)/SXN(RP)/SIN(RF)
S2--COS RM)*COS (RF)/cos (RP)/SrN (RF)
SLJH=ATAN (Si )+ATAN(32)
PH? :PHO-W1 'TIMF+SUM
1030 CONTINUE
IF (PHP.GE.0.) GOTO 1034~
PUP =PHP.2. *PI
GOTO 1030
103J4 CONTINUE
IF (PHP.LT.2. PI ) GOTO 1035
PH? =PtiP-P. *PI
GOTO 1034s
1035 CONTINUE
ZLON =1 80. 'PHP/PI
C
C RADIUS=ALTITUDE (KILOMETERS).
C
RADIUS =RMAJ'(l .-E*COS (C:Z ))-RE
RETURN
END

82
 PROGRAM LET
C THIS PROGRAM AuoMPUTES INTEGRAL AND DIFFERENTIAL LET SPECTRA.
C THE PARTICLE SPECTRA IN THE INTERPLANETARY MEDIUM ARE IN
C THE CRF SUBROUTINE.
C THE GEOMAGNETIC CUTOFF IS APPLIED BY THE CUT SUBROUTINE.
C PROPAGATION THROUGH SHIELDING IS HANDLED IN THE INSIDE ROUTINE.
C THE DEDXSI ROUTINE RETURNS LET IN SILICON.
C THE CONTROL PARAMETERS ARE:
C Y a YEAR OF THE PARTICLE ENVIRONMENT (1975. 1-44 WAS SOLAR MIN.
C AND 1980.598 WAS SOLAR MAX.);
C IT =-ORBITAL INDEX (0 z HIGH ORBIT: NC .U'TOf, &',14O TRAPP.ý-D
C PROTONS; 1 = LOW ORBIT: CUTOFF AND TRALPED P.ROTONS);
C MQ = INTERPLANETARY MEDIUM WEATHER INDEX (SEE NOTES IN CRF
C ROUTINE);
C THK SHIELDING THICKNESS .(ALUMINUM EQUIVALENT) IN G/CM**2;
C IDEM NO. OF LOG STEFS IN ENERGI;
C IELM LIGHTEST ELEMENT INCLUDED;
C JELM HEAVIEST ELEMENT INCLUDED (JELM MUST NOT EXCEED 92).
C THE OUTPUTS ARE THE DIFFERENTIAL AND INTEGRAL LET SPECTRA.
M
C THE UNITS ARE PARTICLES/((M*"2)*STER*SEC'MEV*CM *2/G) AND
C PARTICLES/((M**2)*STER*SEC) RESPECTIVELY
C VERSUS MEV*CMe*2/G.
C
C DXMAX AND DXMIN THE MAXIMUM AND MINIMUM LET VALUES CONSIDERED.
a'lhN IS THE MINIMUM ENERGY CONSIDERED.
C ABOVE EMAX, LET IS CONSTANT.
C
REAL N(IO1),INSIDE
CHARACTER *12 ALPHA,BETA, GAMMA
DIMENSION DX(101 ),F(10OO),DEDX(iO00)
DATA DXMAX/1.IE+5/,DXMIN/1.6/,aEMIN/.1/,EMAX/20000. /
DATA F/1000*0.0/,GAMMA/' '/,NDIF/O/,NINT/O/
C ASK FOR ALL THE CONTROL DATA NEEDED FOR A RUN.
C
WRITE (6,101)
10 FORMAT(' ENTER THE NAME OF THE DIFFERENTIAL LET SPECTRUM'/
1 (ENTER NOTHING IF YOU DO NOT WANT ONE): ',$)
ACCEPT 102, ALPHA
102 FORMAT(A12)
IF(ALPHA.EQ.GAMMA) GO TO 104
OPEN (UNIT=40,FILE=ALPHA)
WRITE (6,103)
103 FORMAT(' ENTER THE NUMBER OF POINTS IN THE TABULATION OF'/
I ' THE DIFFERENTIAL LET SPECTRUM (MAXIMUM" 1000): ',$)
ACCEPT *,NDIF
I10 WRITE (6,105)
V 105 FORMAT(' ENTER THE NAME OF THE INTEGRAL LET SPECTRUM'/
1 ' (ENTER NOTHING IF YOU DO NOT WANT ONE): ',$)
ACCEr7 102, BETA
IF(BETA.EQ.GAMMA) GO TO 107
OPEN (UNIT =45,FILE =BETA)

83
 WRITE (6,106)
106 FORMAT(' ENTER THE NUMBER OF POINTS IN THE TABULATION OF'/
1 THE INTEGRAL LET SPECTRUM (MAXIMUM: 1000): ',$)
ACCEPT *,NINT
107 NPT =MAX0 (NDIF, NINT)
WRITE (6,108)
108 FORMAT(' ENTER THE ATOMIC NUMBER OF THE LIGHTEST ELEMENT'/
1 ' TO BE INCLUDED IN THE LET SPECTRA: ',$)
ACCEPT *,IELM
WRITE (6,109)
109 FORMAT(' ENTER THE ATOMIC NUMBER (<93) OF THE HEAVIEST ELEMENT'!
1 ' TO BE INCLUDED IN THE LET SPECTRA: ',$)
ACCEPT *,JELM
IF (JELM.GT.92) JELM=92
WRITE (6,110)
110 FORMAT(' ENTER THE YEAR FOR WHICH YOU WANT THE SPECTRUM'/
1 (1975.144 FOR SOLAR MIN.; 1980.598 FOR SOLAR MAX.): ',$)
ACCEPT *,Y
WRITE (6,111)
111 FORMAT(' ENTER THE ORBITAL INDEX: 0 FOR NO CUTOFF AND NO ',
1 `TRAPPED PROTONS;'/" 1 FOR CUTOFF AND TRAPPED PROTONS',
2' (NOTE: STASS.DAT AND ITRANS.DAT'/
3' MUST BE SUP7LrED WHEN 1 IS ENTERED HERE): t,$)
ACC,"2, -ý,IT
WRITE (6,112)
112 FORMAT(' ENTEI ThE IINTERPLkNETi.X :4EATHER INDEX: 'uS)
ACCEPT *, MQ
WRITE (6,113)
113 FORMAT(' ENTER THE nHIT-DING THICKNESS IN INCHES OF'/
I ' ALUMINUM (OF EQUIVA'EaI): ',$)
ACCEPT *.THK
C
C CONVERT ?. THICK ESS TO G/CM**2 (ALUMINUM EQUIVALENT).
C
THK:THK *6. 1528
WRITE (6,11') IELM,JELM
116 FORMAT(//' THE LET SPECTRA WILL INCLUDE THE CONTRIBUTIONS'/
1 ' OF THE ELEMENTS ',I2,' THROUGH ',12,'.')
IF(NDIF.GT.0) WRITE (6,114) ALPHA,NPT
IF(NINT.GT.O) WRITE (6,115) BETA,NPT
114 FORMAT(' THE FILE ',A12,' WILL CONTAIN THE DIFFERENTIAL LET'/
1 ' SPECTRUM, TABULATED IN ',I4,' DATA POINTS.')
115 FORMAT(' THE FILE ',A12,' WILL CONTAIN THE INTEGRAL LET'/
1 ' SPECTRUM, TABULATED IN ',I4,' DATA POINTS.')
WRITE (6,117) Y
117 FORMAT(' THESE LET SPECTRA WILL BE FOR THE YEAR ',F8.3,'1')
IF(IT.EQ.O) WRITE (6,118)
118 FORMAT(' THESE LET SPECTRA WILL BE UNSHIELDED BY THE'/
1 ' EARTH''S MAGNETIC FIELD, AND THE CONTRIBUTION FROM TRAPPED'/
2 ' PROTONS WILL NOT BE INCLUDED.')
IF(IT.EQ.1) WRITE (6,119)
119 FORMAT(' THrSE LET SPECTRA WILL INCLUDE THE EFFECTS OF'/

84
 1 ' GEOMAGNETIC SHIELDING ACCORDING TO THE GEOMAGNETIC CUTOFF'/
2 TRANSMITTANCE FUNCTION TABULATED IN GTRANS.DAT. THE'/
3 ' CONTRIBUTION FROM TRAPPED PROTONS WILL BE INCLUDED'/
4 ACCORDING TO THE ORBIT-AVERAGED TRAPPED PROTON SPECTRA'/
5 ' TABULATED IN STASS.DAT.')
WRITE (6,120) MQ
120 FORMAT(' THE INTERPLANETARY WEATHER INDEX IS ',12,'.')
WRITE (6,121) THK
121 FORMAT(' THE SHIELDING THICKNESS IS ',F7.3,' GM/CM**2
1 ALUMINUM (OR EQUIVALENT).'//)
DLL-ALOG(DXM IN)
DDm (ALOG (DXMAX)-DLL) /NPT
DLX =DLL
IDEM=I00
JDEM-IDEM÷I
C
C SET UP LOG ENERGY STEPS.
c
ELNX-ALOG (EMIN)
DELN- (ALOG (EMAX)-ELNX)/IDEM
C
C DO ONE ELEMENT AT A TIME.
C
DO 41 IZ-IELM,JELM
C
C SET LOWER BIN BOUNDARIES.
C
L=I
E=EMIN
K:1
ELN =ELNX
ZFLOAT(IZ)
C
C GET LET.
C
DX(1 )=DEDXSI(Z,E)
C
C GET INTERNAL FLUX.
C
FL=INSIDE (ZE,Y,MQ, IT, THK)
C
C RETURN TO HERE AFTER FLUX IN EACH BIN HAS BEEN CALCULATED.

10 L=L+1
C
C RETURN TO HERE IF D(LET)/DE IS SINGULAR TO INCREASE THE BIN
C WIDTH.
C
11 K=K+I
C
C WHEN K EXCEEDS JDEM YOU ARE THROUGH.
C

85

N ~A~A ~ ~
L~X~gWgW A~&L%
J't .t'~ '~j~~ '.A l' 4,jro'Ch-R''
 IF(K.GT.JDEM) GO TO 20
C
C STEP TO UPPER ENERGY BIN BOUNDARY.
C
ELNzELN+DELN
C
C COMPUTE UPPER ENERGY BIN BOUNDARY.
C
EU=EXP (ELh)
C
C COMPUTE UP'ER LET BIN BOUNDARY.
C
DU=DEDXS I (Z, EU)
C
C TEST FOR D(LET)/DE SINGULARII'Y.
C
T=ABS (DX (L-I )-DU )/DU
IF(T.LT.1.OE-6) GO TO 11
C
C COMPUTE FLUX IN THE BIN.
C
FU=INSIDE(Z,EU,Y,MQ, IT,THK)
N(L-1 )=0.5*(FU+FL)*(EU-E)
C
C THE UPPER BIN BOUNDARY BECOMES THE LOWER BOUNDARY OF NEXT BIN.
C
DX(L)=DU
E=EU
FL=FU
C
C CLOSE THE LOOP.
C
GO TO 10
C
C DO THE HIGH ENERGY TAIL IN THE LAST BIN, ASSUMING AN E**(--2.5)
C POWER LAW FOR HIGH ENERGY COSMIC RAYS.
C
20 LL=L-2
N(LL+1)=0.6666*FL*E
C
C PRESERVE THE LET BIN BOUNDARIES.
C
DLU =DLL
DL=DXMIN
C
RE-BIN EACH ELEMENT IN A COMMON SET OF LET' BINS.

DO 41 KK=1,NPT
C
C SET UP THE COMMON LET BIN BOUNDARIES.
C
DLU =DLU+DD

86

rt~~Kw % mNIW.%WI-,I A I A?.N ?..I kr. j n14f%

 DU mEXP (DLU)
C
C SPIN THROUGH THE ELEMENT BIN SET ADDING FLUX TO THE
C COMMON BIN.
C
DO 40 I--,LL
DXI -DX (I)
DX2=DX(I+I)
c
C SKIP THE LOOP FOR ELEMENT BINS NOT OVERLAPPING THE COMMON BIN.
* C
IF ((DX1.GE.DU).AND.(DX2.GE.DU)) GO TO 40
IF ((DX1.LE.DL).AND.(DX2.LE.DL)) GO TO 40
IF(DX1.LT.DX2) GO TO 35
IF(DX1.GE.DU) GO TO 32
IF(DX2.LT.DL) GO TO 31

F(KK ) =F(KK )÷+N (I)

GO TO 40
31 F(KK )-F(KK )+N (I )* (DX1-DL)/(DX1-DX2 )
GO TO 4O
32 IF(DX2.GE.DL) GO TO 33
F(KK)-F(KK)+N (I)*(DU-DL) /(DX1-DX2)
GO TO 40
33 F(KK)=F(KK )+N (I)*(DU-DX2)/(DXI-DX2)
GOTO 4O
35 IF(DX2.GE.DU) GO TO 37
IF(DX1.LT.DL) GO TO 36
F(KK)=F(KK)÷N (I)
GO TO 40
36 F(KK)=F(KK)+N (I )*(DX2-DL)/(DX2-DX1)
GO TO 40
37 IF(DX1.GE.DL) GO TO 38
F(KK ) =F (KK)+N (I) (DU-DL) / (DX2-DX1)
GO TO 40
38 F(KK)=F(KK)÷N (I )*(DU-DXI )/(DX2-DX1)
40 CONTINUE
C
C DO THE INTEGRAL BIN FOR THE HIGH ENERGY TAIL.
C
IF((DX2.GE.DL).AND.(DX2.LE.DU)) F(KK)=F(KK)÷N(LL+I)
C
C FINISH STEPPING THROUGH THE COMMON LET BINS
C THEN GO BACK AND DO THE NEXT ELEMENT.
C
41 DL=DU
C
C COMPUTE THE DIFFERENTIAL LET SPECTRA.
C
DL=DXMIN
DO 50 Kz-,NPT
DLX--DLX+DD
DULEXP (DLX)

87
 G F(K)/(D(J-DL)
DEDX(K) aO.5*(DLJ+iDL)
IF(NDIF.GT.0) WRITE(401600) G, DEDX(K)
600 FORMAT(2X,E12.5t2X,E12.5)
50 DL:DU
summO. 0
C
C INTEGRATE THE LET SPECTRA.
C
DO 60 IXNPT, 1,-i
SUM=SUM+F(I)
DL:DEDX (I)
IF(NINT.GT.0) WRITE(45,600)SUt4,DL
60 CONTINUE
END

88
 FUNCTION DEDXSI(ZPE)
C THIS SUBROUTINE RETURNS THE STOPPING POWER IN SILICON IN
C MEV*CMW*2/G.
C Z IS THE ION ATOMIC NUMBER.
C E IS THE ION ENERGY IN MEV/AMU.
C ZN ARE THE ATOMIC NUMBERS FOR WHICH THE STOPPING POWERS HAVE
C BEEN TABULATED.
C KK ASSOCIATES EACH ELEMENT WITH ONE WHOSE STOPPING POWER HAS
"C BEEN TABULATED.
C
DIMENSION EN(77),RN(9,77),ZN(9),KK(92)
DATA IST/O/,ZN/l.,2.,6.,8.,18.,26. 36.,54.,92/
DATA IZS,J/O,3/,KK/1,2*2,3*3,7*4, 8*5,9*6
1 ,14'7,28*8,20*9/
C
C THE STOPPING POWER TABLES ARE IN THE SILICN.DT1 FILE.
C
C
C READ THE STOPPING POWER TABLES UPON THE FIRST ENTRY.
C
IF(IST.NE.O) GO TO 5
IST--
OPEN(UNIT=51, READONLY, SHARED,STATUS-'OLD' ,FILE-'SILICN.DT1')
DO 3 I=1,77
READ(51,2) EN(I),(RN(J,I),J=1,9)
2 FORMAT (6X,F9.3,9 (3X,E9.3))
3 CONTINUE
CLOSE (UNIT-51)
5 CONTINUE
IZ=Z+0.5
C
C COMPUTE THE APPROXIMATE STARTING INDEX FOR TABLE LOOKUP.
C
IJ= (ALOG(E)+4. 065 )*4. 85567-1 .0
IJ=MAXO(IJ,I)
IJ=MINO(IJ, 77)
C
C SEARCH FOR THE ENERGY TABULAR POINT JUST ABOVE E.
C
10 CONTINUE
DO 11 I=IJ,77
IF(EN(I).LT.E) GO TO 11
J =I
GO TO 13
11 CONTINUE
J =77
C
C CALCULATE THE INTERPOLATION FACTOR.
C
13 CONTINUE
F=(E-EN(J-1 ))/(EN(J)-EN(J-1))

89
 F =AMIN I(F, 1.0)
C
C FIND THE NEAREST TABULATED ELEMENT.
C
14 CONTINUE
K=KK(IZ)
C
C INTERPOLATE THE STOPPING POWER FOR THE TABULATED ELEMENT.
C
DEDXSIzF*(RN(K,,J)-RN(K,J-1 ))+RN(K,J-1)
ZNE=ZN (K)
C
C IF THE ION IS THE ONE TABULATED, YOU ARE DONE.
C
IF (Z.EQ.ZNE)RETURN
ZE -Z
EC=I.+2.*Z
C
C IF THE ENERGY IS HIGH ENOUGH, THE ION IS FULLY STRIPPED.
C
IF(E.GT.EC) GO TO 21
C
C COMPUTE THE EQUILIBRIUM CHARGE STATE.
C
B--SQRT (I.-(E*. 001073927+I )** (-2.) )
ZE =Z* (1.-EXP (-130.*B*(Z**( -. 6666) )) )
ZNE=ZNE*(I.-EXP(-130.*B*(ZNE**(-.6666))))
C
C INTERPOLATE TO THE ION IN QUESTION.
C
21 CONTINUE
DEDXSI=(ZE*ZE)/(ZNE*ZNE )*DEDXSI
RETURN
END
 FUNCTION INSIDE (Z,E,YMIT, TH)
C
C TRANSPORTS THE PARTICLE SPECTRA THROUGH THE SPACECRAFT WALL
C CONSIDERING ENERGY LOSS EFFECTS AND NUCLEAR FRAGMENTATION,
C BUT DOES NOT KEEP TRACK OF THE FRAGMENTS.
C Z IS THE ATOMIC NUMBER OF THE ION.
C E IS THE ENERGY IN MEV/AMU.
C Y IS THE YEAR OF THE ENVIRONMENT.
C M IS THE INTERPLANETARY MEDIUM WEATHER INDEX (SEE CRF.FOR).
C IT IS THE ORBITAL INDEX: 1 FOR LOW ORBITS, 0 FOR HIGH ORBITS.
C TH IS THE THICKNESS OF ALUMINUM SHIELDING IN G/CM**2.
C THIS ROUTINE RETURNS THE INTERNAL FLUX IN PARTICLES/((M**2)*
C STER*SEC*MEV/U).
C
C AN ARE THE AVERAGE ION MASSES EOR EACH ELEMENT (TAKEN FROM
C THE HANDBOOK OF CHEMISTRY AND PHYSICS, 59TH EDITION).
C
REAL INSIDE
COMMON/MASS/AN (92)
DATA (AN(I),I=1,92)/1.,4.,6.9,
1 9.,10.8, 12.,1I4.,16.,19.,2O.2,23.,24.3,27.,28.,
2 31.,32.,35.5,39.9,39.,40.,45.,47.9,50.9,
3 52. ,5L4.9,55.8,58.9,58.7,63.5,65.4,69.7,72.6,
4 74.9,79.,79.9,83.8,85.5,87.6,88.9,91.2,92.9,95.9,97.,101.,
5 102.9,1O6.4,107.9, 112.4,114.8,118.7,121.8,127.6,126.9,1 3 1. 3 ,
6 132.9,137.3,138.9,140. 1, 140.9,144.2, 145.,150.4,,152.,157.3,
7 158.9,162.5,164.9,167.3,168.9,173.,175.,178.5,180.9,183.9,
8 186.2,190.2, 192.2,195.1,197. ,200.6,204.J4,207.2,209. ,209.,
9 210.,222.,223.,226.,227.,232.,231.,238./
IZ-Z+.5
A=AN (IZ)
C
C COMPUTE THE TOTAL INELASTIC CROSS SECTION USING C.H.
C TSAO'S VERSION OF THE BRANDT-PETERS FORMULA.
C USE A SIMPLER FORMULA FOR PROTONS.
C
IF (Z.NE.1.) COEF=((6.02E+23)*(5.OE-26)
1 *(A"*(l./3.)+AN(13)**(1./3.)-0.4)*"2)/AN(13)
7F (Z.EQ.1.) COEF-O.

C FIND THE RESIDUAL RANGE OF THE ION INSIDE THE SPACECRAFT.

C
RuRAL(Z,A,E)
C
C FiND THE RESIDUAL RANGE OF THE SAME ION OUTSIDE THE SPACECRAFT.
C
RfR÷TH
C
C FIND THE CORRESPONDING ENERGY OUTSIDE.
C

91
 ES-EAL(Z, A,R)
C
C FIND THE STOPPING POWER INSIDE AND OUTSIDE.
C
SS=DEDXAL(Z, ES)
S=DEDXAL( Z,E)
C
C GET THE FLUX OUTSIDE AND COMPUTE IT INSIDE.
C
C IT=O FOR AN ORBIT SO HIGH THAT CUTOFF IS INSIGNIFICANT
C AND TRAPPED PROTONS ARE INSIGNFICANT.
C
C ITz1 FOR LOWER ORBITS: INCLUDE CUTOFF AND TRAPPED PROTONS.
C
IF (IT.EQ.0) INSIDE=CRF(Z,ES, YwM)*(SS/S)*EXP(-COEF*TH)
IF (IT.GT.O) INSIDE=CUT(Z,ES,Y,M)*(SS/S)*EXP(-COEF*TH)
RETURN
END

92
 FUNCTION DEDXAL(Z,E)
C
C THIS SUBROUTINE RETURNS THE STOPPING POWER IN ALUMINUM IN
C MEV*CM**2/G.
C Z IS THE ION ATOMIC NUMBER.
C E IS THE ION ENERGY IN MEV/AMU.
C
C ZN ARE THE ATOMIC NUMBERS FOR WHICH THE STOPPING POWER HAS BEEN
C TABULATED.
C KK ASSOCIATES EACH ELEMENT WITH ONE WHOSE STOPPING POWER
C HAS BEEN TABULATED.
C
DIMENSION EN(66),RN(9,66),ZN(9),KK(92)
DATA IST/O/,ZN/l.,2.,6.,8.,18..,26.,36. ,54.,92/
DATA IZS,J/O,3/,KK/1,2*2,3*3,7*4U 8U5,9'6
,14*7,28*8,20O9/
C
C THE STOPPING POWER TABLES ARE ON THE ALUMNM.DT1 FILE.
C
C
C READ THE STOPPING POWER TABLES UPON THE FIRST ENTRY.
C
IF(IST.NE.O) GO TO 5
IST--1
OPEN(UNIT=50, READONLY,SHARED,STATUS='OLD' ,FILE-'ALUMNM.JDT1')
DO 3 I=1,66
READ(50,2) EN(I),(RN(J,I),J=1,9)
2 FORMAT (6X,F9. 3,9 (3X,E9.3))
3 CONTINUE
5 CONTINUE
IZ-Z+0.5
C
C COMPUTE THE APPROXIMATE STARTING INDEX FOR TABLE LOOKUP.
C
IJ=(ALOG(E)÷4. 605 )*4. 85567-1.0
IJ =MAXO(IJ, 1 )
IJ=MINO (IJ, 66)
C
C SEARCH FOR THE ENERGY TABULAR POINT JUST ABOVE E.
C
10 CONTINUE
DO 11 I-IJ,66
IF(EN(I).LT.E) GO TO 11
J -I
GO TO 13
11 CONTINUE
J =66
C
C, CALCULATE THE INTERPOLATION FACTOR.
C
13 CONTINUE

93
 F= (E-EN(J-1 ) /(EN(J )-EN (J-1))
F:AMINI(F, 1.0)1
C
C FIND THE NEAREST TABULATED ELEMENT.
C
14 CONTINUE
K:KK (IZ)
C
C INTERPOLATE THE STOPPING POWER FOR THE TABULATED ELEMENT.
C
20 CONTINUE
DEDXAL-F* (RN(K,J)-RN(K;J-1))+RN(K,J-1)
ZNE=ZN (K)
C
C IF THE ION IS THE ONE TABULATED,. YOU ARE DONE.
C
IF (Z.EQ.ZNE) RETURN
ZE:Z
EC=1.+2.*Z
C
C IF THE ENERGY IS HIGH ENOUGH, THE ION IS FULLY STRIPPED.
C
IF(E.GT.EC) GO TO 21
C
C COMPUTE THE EQUILIBRUM CHARGE STATE.
C
B=SQRT(l.-(E*.001073927+1 )*(-2.))
ZE =Zm (I. -EXP (- 130. *B*( Z**(-. 6666 ))) )
ZNE-ZNE*(1.-EXP(-130.*B*(ZNEh*(-.6666))))
C
C INTERPOLATE TO THE ION IN QUESTION.
C
21 CONTINUE
DEDXAL- (ZE*ZE )/( ZNE*ZNE )'DEDXAL
RETURN
END

94
 FUNCTION RAL(Z,A,E)

C THIS ROUTINE RETURNS THE RESIDUAL RANGE IN ALUMINUM IN G/CM**2.

C Z IS THE ION ATOMIC NUMBER.
C E IS THE ION ENERGY IN MEV/AMU.
C A IS THE ION ATOMIC MASS.

C ZN ARE THE ATOMIC NUMBERS OF THE TABULATED ELEMENTS.

C KK ASSOCIATES EACH ELEMENT WITH A TABULATED ELEMENT.
C AN ARE THE MEAN ATOMIC MASSES OF THE TABULATED ELEMENTS.
C DEDX ARE STOPPING POWER DATA IN (MEV/AMU)/(G/CM**2).
C
DIMENSION EN(66),RN(9,66),ZN(9),AN(9),KK(92),DEDX(9)
DATA IST/O/,ZN/1. ,2.,6. ,8. ,18.,26.,36. ,54.,92/
DATA IZS,J/O,3/,KK/1,2"2, 33,7"4,8"5,9*6
1 ,14'7,28*8,20*9/
DATA AN/1.008, 4.0026,12.01115,15.99994,39.948,55.847
1 ,83.8,131. 3,238.03/
DATA DEDX/1.706,I.720,5.174,6.913,14.12,21.22,27.34
1 ,40.05,68.69/
C
C SET ENTRY FLAG FOR RANGE.
C
NTRY=0
C
C READ IN THE TABLES UPON THE FIRST ENTRY.
C
IF(IST.NE.O) GO TO 5
C
C FILE ALUMNM.DT2 CONTAINS THE RANGE-ENERGY TABLES.
C
1 CONTINUE
OPEN(UNIT-60,READONLY,SHARED,STATUS-'OLD',FILE:'ALUMNM.DT2')
IST=l
DO 3 I=1,66
READ(60,2) EN(I),(RN(J,I),J=1,9)
2 FORMAT (6X,F9. 3,9 (3X,E9.3))
3 CONTINUE
C
C IF ENERGY FLAG SET, RETURN TO ENERGY PART.
C
IF(NTRY.EQ.1) GO TO 100
5 CONTINUE
IZ-Z+.5
C
C FIND THE NEAREST TABULATED ION.
C
K-KK(IZ)
C
C COMPUTE THE APPROXIMATE INDEX IN THE TABLE.
C

95
 IJ-66
IF(E.GT.8000.) GO TO 10
IJ=(ALOG (E)+4. 605) *4.85567-1.0
IJ=MAX0(IJ, 1)
IJ=MINO (IJ,66)
C
C FIND THE TABULATED ENERGY JUST ABOVE E.
C
10 CONTINUE
DO 11 I-IJ,66
IF(EN(I).LT.E) GO TO 11
J :I
GO TO 13
11 CONTINUE
C
C EXTRAPOLATE THE RANGE ACCORDING TO THE PLATEAU
C VALUE OF THE STOPPING POWER.
C
RNK-RN(K, 66)+(E-EN(66))/DEDX(K)
GO TO 14
C
C COMPUTE THE INTERPOLATION FACTOR.
C
13 CONTINUE
F=(E-EN(J-1 ))/(EN(J )-EN(J-I )
F-AMIN1 (F, 1.0)
C
C INTERPOLATE RANGE IN THE TABLE.
C
RNK-F*(RN(K,J)-RN(K,J-1 ))+RN(K,J-1)
C
C INTERPOLATE TO THE ION IN QUESTION.
C
14 CONTINUE
RAL-(A/(Z'Z) )*(ZN(K)*ZNU(K)/AN(K ))*RNK
RETURN
ENTRY EAL(Z,A,R)
C
C THIS ENTRY RETURNS THE ENERGY (IN MEV/AMU) OF AN ION HAVING
C A RESIDUAL RANGE R, IN G/CM**SQ.
C Z IS THE ION ATOMIC NUMBER.
C A IS THE ION ATOMIC MASS.
C
C SET ENTRY FLAG FOR ENERGY.
C
IZ-Z+.5
NTRY-I
C
C IF THIS IS THE FIRST CALL, READ IN THE TABLES.
C
IF(IST.EQ.O) GO TO 1
C

96
 C FIND THE NEAREST TABULATED ION.
C
100 CONTINUE
K=KK(IZ)
C
C COMPUTE THE ION INTERPOLATION FACTOR.
C
110 CONTINUE
FK:(A/(Z*Z))*(ZN(K)*ZN(K)/AN(K))
C
C FIND THE TABULATED RANGE JUST ABOVE R.
C
DO 120 I=1,66
RT =FK*RN (K, I )
IF(RT.LT.R) GO TO 120
J=I
GO TO 130
120 CONTINUE
C
C EXTRAPOLATE THE ENERGY.
C
RAT=DEDX(K)/FK
EAL=EN (66)+RAT* (R-FK*RN (K,66))
RETURN
C
C INTERPOLATE THE ENERGY.
C
130 CONTINUE
F= (R-FK*RN (K, J-1 ))/(RT-FK*RN (K, J-1 ))
EAL=F*(EN(J)-EN(J-1 ))+EN(J-1)
RETURN
END

1
A

S~97
 FUNCTION CUT (Z,E,Y,M)
C
C THIS ROUTINE OBTAINS DIFFERENTIAL PARTICLE FLUXES FROM CRF,
C APPLIES THE GEOMAGNETIC CUTOFF TRANSMISSION FUNCTION
C IN THE GTRANS.DAT FILE, AND RETURNS
C THE RESULTING FLUX, MODULATED TO THE ORBIT-AVERAGE CUTOFF.
C THIS ROUTINE ALSO ADDS SINGLY-IONIZED ANOMALOUS COMPONENT NUCLEI
C TO THE COSMIC RAY SPECTRA (WHEN M=4).
C Z = ION ATOMIC NUMBER.
C E = ION ENERGY IN MEV/AMU.
C Y = YEAR (1 9 7 5 .144 = SOLAR MIN.; 1980.598 . SOLAR MAX.).
C M = INTERPLANETARY MEDIUM WEATHER INDEX.
C
COMMON/MASS/A(92)
DIMENSION R(200),T(200)
DATA ITS/-l/
IF(ITS.EQ.1) GO TO 6
OPEN(UNIT=15, READONLY,SHARED, STATUS='OLD' ,FILE='GTRANS.DAT')
ITS-
DO 5 I-1,200
READ(15,1) R(1),T(I)
1 FORMAT (5X, F6.3, 5X, F8. 6)
5 CONTINUE
SHADOW=T (200)
6 IZ-Z+.5
CUT=0.0
ES =E
C
C DON'T CALL CRF WITH ENERGIES BELOW 10 MEV/U.
C
IF(E.LT. 10.) ES=10.
ME =M
MQ=M
c
C M=4 MEANS A SINGLY-IONIZED ANOMALOUS COMPONENT + GALACTIC COSMIC
C RAYS.
C GET THE GALACTIC COSMIC RAYS HERE, THE ANOMALOUS COMPONENT WILL
C BE ADDED BELOW.
C
IF(M.EQ. 4) MQ-l
C
C GET THE DIFFERENTIAL FLUX.
C
F--CRF( Z,ES, Y,MQ)
C
C COMPUTE THE MAGNETIC RIGIDITY.
C
P=(A(IZ)/Z)*(ES*ES+1862.324"ES)**.5/1 000
C
C LOOK UP THE TABULATED MAGNETIC RIGIDITY JUST ABOVE P.
C

98
 4 DO 2 I=2,200
IF(P.GT.R(I))GOTO 2
ISAV=I
GOTO 3
2 CONTINUE
TR-:SHADOW
GO TO 10
C
C INTERPOLATE THE TRANSMISSION FACTOR (AVERAGED FOR THE ORBIT).
C
3 TR=(T(ISAV)-T(ISAV-1))*(P-R(ISAV-1))/(R(ISAV)-R(ISAV-1))
1 +T(ISAV-1)
C
C APPLY THE TRANSMISSION FACTOR TO MODULATE THE FLUX.
C
10 CUT=CUT+F*TR
C
C ADD IN THE DIRECT CONTRIBUTION FROM THE TRAPPED PROTONS.
C
IF(IZ.EQ.1) CUT=CUT+PROTON(ES)
C
C UNLESS A SINGLY-IONIZED ANOMALOUS COMPONENT MUST BE ADDED, RETURN.
C
IF(MK.NE.4) GO TO 12
C
C ADD SINGLY-IONIZED HELIUM.
C
IF(IZ.NE.2) GO TO 40
F--(I. 54E+4 )*ES** (-2)
IF(ES.LT.195.) F=.4
P-4.*(ES*ES+1862.324 1*ES)**.5/1000
MK=I
GO TO 4
C
C ADD SINGLY-IONIZED CARBON.
C
40 IF(IZ.NE.6) GO TO 45
IF(ES.LT.1O.) GO TO 43
F=O. 27'ES** (-2)
GO TO 44
43 F:(4.OE-3)*EXP(-(ALOG(ES)-•.79)*'2/.7)
IF(F.LT.0.) FzO.
44 P=12.*(ES*ES+1862.324*ES)**.5/1000.
MK :
GO TO 4
C
C ADD SINGLY-IONIZED NITROGEN.
C
45 IF(IZ.NE.7) GO TO 50
IF(ES.LT.20.) GO TO 41
F=. 773*ES**((-2)
GO TO 42

99
 141 F: (1.54E-2)*EXP (-(ALOG(ES)-l.79)**2/.7)
IF(F.LT.0.) F=O.
42 P:114.*(ES*ES+1862Q 32~4 ES )**.5/1000.
MK:1
GO TO 14
C
C ADD SINGLY-IONIZED OXYGEN.
C
50 IF(IZ.NE.8) GO TO 60
IF(ES.LT.30.) GO TO 51
F=1.32*ES**(-2.)
GO TO 52
51 F:(6.0E-2)*EXP(-(ALOG(ES)-l.79)**2/.7)
IF(F.LT.O.) F=O.
52 P=16*(ES*ES+i1862.32L1*ES)**.,.5/1000.
MK :1
GO TO 14
C
C ADD SINGLY-IONIZED NEON.
C
60 IF(IZ.NE. 10.) GO TO 70
IF(ES.LT.20.) GO TO 61

GO TO 62
61 F=(8.OE-3)*EXP(-(ALOG(ES)-l.79)**2/.7)
IF(F.LT.0.) F=O.
62 P=20*(ES*ES+1862.3214ES)**.5/1000.
* ~MK =1
GO TO 4
C
C ADD SINGLY-IONIZED MAGNESIUM.
C
70 IFCIZ.NE. 12) GO TO 80
IF(ES.LT.20.) GO TO 71
F=O. 16*ES**(-2)
GO TO 72
71 Fz(8.OE-4)*EXP(-(ALOG(ES)-2. 3)**2/.7)
IF(F.LT.0.) F=O.
72 P=24. 3*(ES*ES+1 862. 3214*ES)**.5/10O0.
MK :1
GO TO 14
C
C ADD SINGLY-IONIZED SILICON.
* ., C
80 IFCIZ.NE.114) GO TO 85
IF(ES.LT.1O.) GO TO 81
F:O. 1*ES*(-2)
GO TO 82
81 Fz(1.OE-3)*EXP(-(ALOG(ES)-2.2)**2/0.4)
IFCF.LT.O.) F=O.
82 P=28.*(ES*ES+1862.3214ES)**.5/1000.
MK :1

100
 GO TO 4
C
C ADD SINGLY-IONIZED ARGON.
C
85 IF(IZ.NE.18) GO TO 90
IF(ESLT.20.) GO TO 86
F=O. 028*ES**(-2)
GO TO 87
86 F=(5.4E-4)*EXP(-(ALOG(ES)-I.79)**2/.7)
IF(F. LT. 0.) F=O.
87 P:=40.*(ES*ES+1862.324"ES)**.5/1000.
MK :
GO TO 4
c
C ADD SINGLY-IONIZED IRON.
C
90 IF(IZ.NE.26) GO TO 12
IF(ES.LT.30.) GO TO 91
F=0. 35*ES**((-2)
GO TO 92
91 F=(6.0E-4)*EXP(-(ALOG(ES)-2.48)*'2/2.)
IF(F. LT. 0.) F=O.
92 P:56.*(ES*ES÷1 862. 324*ES)**.5/1 000.
MK-1
GOTO 4
C
C THERE PROBABLY 4RE ANOMALOUS COMPONENTS IN THE SPECTRA OF
C THE NUCLEI HEAVIER THAN IRON, BUT THERE ARE NO DATA ON THEM
C AT THIS TIME, SO THEY CANNOT BE INCLUDED IN THE MODEL.
C
12 RETURN
END

101
 PROGRAM BENDEL
C
C THIS PROGRAM USES BENDEL'S FORMULA ("Proton Upsets in Orbit", by
C W. L. Bendel and E. L. Petersen, IEEE Trans in Nucl. Sci., Vol. NS-30,
C p. 4481-5, 1983) TO CALCULATE THE UPSET RATE CAUSED BY PROTONS. THE
C PARAMETER A MUST BE INPUT. THIS PROGRAM WAS CREATED BY MODIFYING THE
C SPEC PROGRAM. SUBROUTINES: INSIDE, CUT, CRF, DEDXAL, RAL.
C
REAL INSIDE
Z=1.
WRITE (6,103)
103 FORMAT(' ENTER THE NUMBER OF ENERGY STEPS USED IN THE'!
1' CALCULATION (MAXIMUM: 1000): ',$)
ACCEPT *,NDIF
WRITE (6,110)
110 FORMAT(' ENTER THE YEAR FOR WHICH YOU WANT THE PROTON-INDUCED 'I/
1' UPSET RATE (1975.144 FOR SOLAR MIN.; 1980.598 FOR SOLAR MAX.)',
2': ',$)
ACCEPT *,Y
WRITE (6,111)
111 FORMAT(' ENTER THE ORBITAL INDEX: 0 FOR NO CUTOFF AND NO ',
1 'TRAPPED PROTONS;'/' 1 FOR CUTOFF AND TRAPPED PROTONS',
2' (NOTE: STASS.DAT AND GTRANS.DAT'/
3' MUST BE SUPPLIED WHEN 1 IS ENTERED HERE): ',$)
ACCEPT *,IT
WRITE (6,112)
112 FORMAT( ' ENTER THE INTERPLANETARY WEATHER INDEX: ',$)
ACCEPT *,M
WRITE (6,113)
113 FORMAT(' ENTER THE SHIELDING THICKNESS IN INCHES OF 'I/
1' ALUMINUM (OR EQUIVALENT): ',$)
ACCEPT *,THK
WRITE (6,114)
114 FORMAT(' ENTER THE PARAMETER A (MEV). (ENTER 0'/
1' IF YOU DON''T KNOW IT): ',$)
ACCEPT *,A
IF (A.GT.O.) GOTO 150
C
C IF NOTHING WAS ENTERED, FIGURE OUT THE VALUE OF "A" FROM
C EXPERIMENTAL DATA.
C
WRITE (6,115)
115 FORMAT(' ENTER AN EXPERIMENTALLY MEASURED UPSET CROSS-SECTION'/
1' IN UNITS OF (UPSETS/BIT)/(PROTON/CM**2): ',$)
ACCEPT *,XSECT
WRITE (6,116)
116 FORMAT(' ENTER THE ENERGY (MEV) AT WHICH THIS CROSS-SECTION'/
1' MEASUREMENT WAS MADE: ',$)
ACCEPT #, ENERGY
ALOW 1.
AHIGH 200.

102
 XSECT = 1.El2*XSECT
DO 130 J=1,16
A = .5*(AHIGH+ALOW)
X = SIGMA(A,ENERGY)
C
C SIGMA IS A DECREASING FUNCTION OF A.
C
IF (X.GT.XSECT) ALOW=A
IF (X.LE.XSECT) AHIGH-A
130 CONTINUE
150 CONTINUE
C
C CONVERT THE THICKNESS TO G/CM**2 (ALUMINUM EQUIVALENT).
C
TH=THK*6.8528
C
C TYPE OUT THE INPUTTED DATA.
C
WRITE (6,117) Y
117 FORMAT(//' THE UPSET RATE WILL BE FOR THE YEAR ',F8.3,'.')
IF(IT.EQ.O) WRITE (6,118)
118 FORMAT(' THE COSMIC RAYS WILL BE UNSHIELDED BY THE',/,
1 ' EARTH''S MAGNETIC FIELD, AND THE CONTRIBUTION FROM TRAPPED',/,
2' PROTONS WILL NOT BE INCLUDED.')
IF(IT.EQ.1) WRITE (6,119)
119 FORMAT(' THE UPSET RATE WILL INCLUDE THE EFFECTS OF',!,
1 ' GEOMAGNETIC SHIELDING ACCORDING TO THE GEOMAGNETIC CUTOFF',/,
2 ' TRANSMITTANCE FUNCTION TABULATED IN GTRANS.DAT. THE',!,
3 ' CONTRIBUTION FROM TRAPPED PROTONS WILL BE INCLUDED',!,
4 ' ACCORDING TO THE ORBIT-AVERAGED TRAPPED PROTON SPECTRA',/,
5 ' TABULATED IN STASS.DAT.')
WRITE (6,120) M
120 FORMATC' THE INTERPLANETARY WEATHER INDEX IS ',12,t.')
WRITE (6,121) TH
121 FORMAT(' THE SHIELDING THICKNESS IS ',F7.3,' GM/CM**2
1 ALUMINUM (OR EQUIVALENT).')
WRITE (6,122) A
122 FORMAT(' THE UPSET THRESHOLD PARAMETER IS ',F6.2,' MEV.')
C
C SET UP THE ENERGY RANGE AND DIVIDE IT INTO EQUAL LOGARITHMIC STEPS.
C
EMAX-32000.
EMIN-1O.
ELNX=ALOG (EMIN)
DELN=(ALOG(EMAX )-ELNX)/NDIF
UPSETS=O.
C
C COMPUTE THE DIFFERENTIAL FLUX AT EACH ENERGY. THE SUBPROGRAMS RETURN
C FLUX IN PARTICLES/( (M**2)*STER*SEC)I(MEV/(G/CM**2)).
C
DO 200 I:z,NDIF
EN=EXP (ELNX)

103
 ELNX =ELNX+DELN
F-INSIDE(Z,EN,Y,M,IT,TH)
C
C COMPUTE BENDEL'S FLUX. E AND A ARE IN MEV, Y IS DIMENSIONLESS,
C SIGMA IS IN 1E-12 (UPSETS/BIT)/(PROTON/CM**2), UPSETS IS IN
C UPSETS/BIT*DAY.
C
C THE FACTORS IN THE CONVERSION CONSTANT ARE:
C 1E-12 FROM THE DEFINITION OF SIGMA;
C 86400 TO CONVERT FROM SECONDS TO DAYS;
C .01**2 TO CONVERT FROM M**2 TO CM**2;
C 4*PI TO GET RID OF THE STEPADIANS.
C
CONST-1.E-12*(.01**2.)*(4.*3. 1416)
UPSETS = UPSETS + CONST*SIGMA(A,EN)*F*(EXP(ELNX)-EN)
200 CONTINUE
WRITE (6,14 )UPSETS
14 FORMAT(' ERROR RATE = ',1PE12.5,' IN UPSETS/BIT*SEC.')
UPSETS =UPSETS 861400.
WRITE (6,15)UPSETS
15 FORMAT(' ERROR RATE = ',1PE12.5,' IN UPSETS/BIT*DAY.')
END

104

NNN+••%N.•'\%'k.'%+ . •• i Y -, 'r" or W-•+.

I.4.•.~tN'CJ Jt.•. .- %%%•. %'.''.- -•.. I "• -X+.e eil + . %
 F ~~ ~ ~ ~
1N&ý ~ ~'flA*~Xfe~ ~
U WlW--I ~-NI
-A-i~ w ~.- - W-

FUNCTION SIGMA (A,E)

C
C THIS IS THE SIGMA FROM PAGE 4484 OF THE BENDEL PAPER.
C
Y = (SQRT(18./A))*(E-A)
IF (Y.LT.O.) Y=O.
SIGMA = ((24./A)**14.)*((1.-EXP(-.18*SQRT(Y)))**4.)
RETURN
END

105
 PROGRAM SFEC

C THIS PROGRAM USES THE SUBPROGRAMS INSIDE, CUT, CRF, DEDXAL,

C AND RAL TO TABULATE THE DIFFERENTIAL AND INTEGRAL ENERGY SPECTRA OF
C COSMIC RAYS OF A SPECIFIED CHARGE.
C
REAL INSIDE
COMMON /MASS/AMSTAE(92)
CHARACTER* 12 ALPHA,BETA, GAMMA
DIMENSION F(1000),E(1001)
DATA NDIF/O/,NINT/O/,GAMMA/' 'I/
WRITE (6,101)
101 FORMAT(' ENTER THE NAME OF THE DIFFERENTIAL ENERGY SPECTRUM' ,/,
1' (ENTER NOTHING IF YOU DO NOT WANT ONE): ',$)
ACCEPT 102,ALPHA
102 FORMAT(A12)
IF(ALPHA.EQ.GAMMA) GO TO 104
OPEN (UNIT =4O,FILE =ALPHA)
WRITE (6,103)
103 FORMAT(' ENTER THE NUMBER OF POINTS IN THE TABULATION OF',/,
1' THE DIFFERENTIAL ENERGY SPECTRUM (MAXIMUM: 1000): ',$)
ACCEPT *,NDIF
104 WRITE (6,105)
105 FORMAT(' ENTER THE NAME OF THE INTEGRAL ENERGY SPECTRUM',/,
1' (ENTER NOTHING IF YOU DO NOT WANT ONE): ',$)
ACCEPT 102,BETA
IF(BETA.EQ.GAMMA) GO TO 107
OPEN (UNIT:45,FILE:BETA)
WRITE (6,106)
106 FORMATC' ENTER THE NUMBER OF POINTS IN THE TABULATION OF',/,
1' THE INTEGRAL ENERGY SPECTRUM (MAXIMUM: 1000): ',$)
ACCEPT *,NINT
107 NPT MAXO (NDIF, NINT)
WRITE (6,108)
108 FORMAT(' ENTER THE ATOMIC NUMBER OF THE ELEMENT FOR WHICH',/,
I' YOU WANT THE ENERGY SPECTRUM: ',$)
ACCEPT *,IELM
Z"IELM
WRITE (6,110)
110 FORMAT(' ENTER THE YEAR FOR WHICH YOU WANT THE SPECTRUM',/,
1' (1975.144 FOR SOLAR MIN.; 1980.598 FOR SOLAR MAXI): ',$)
ACCEPT *,Y
WRITE (6,111)
ill FORMAT(' ENTER THE ORBITAL INDEX: 0 FOR NO CUTOFF AND NO 't
1 'TRAPPED PROTONS;'/' 1 FOR CUTOFF AND TRAPPED PROTONS',
2' (NOTE: STASS.DAT AND GTRANS.DAT'/
3' MUST BE SUPPLIED WHEN 1 IS ENTERED HERE): ',$)
ACCEPT #,IT
WRITE (6,112)
112 FORMAT(' ENTER THE INTERPLANETARY WEATHER INDEX: ',$)
ACCEPF *,M

106
 WRITE (6,113)
113 FORMAT(' ENTER THE SHIELDING THICKNESS IN INCHES OF ',/,
1' ALUMINUM (OR EQUIVALENT): ',$)
ACCEPT *,THK
C
C CONVERT THE THICKNESS TO G/CM**2 (ALUMINUM EQUIVALENT).
C
TH=THK*6. 8528
WRITE (6,116) IELM
116 FORMAT(//' THE ENERGY SPECTRA WILL INCLUDE THE CONTRIBUTION'/
1 ' OF ELEMENT ',12,' .')
IF(NDIF.GT.O) WRITE (6,114) ALPHA,NPT
IF(NINT.GT.O) WRITE (6,115) BETA,NPT
114 FORMAT(' THE FILE ',A12,' WILL CONTAIN THE DIFFERENTIAL',/
I ,' ENERGY SPECTRUM, TABULATED IN 1,14,1 DATA POINTS.V)
115 FORMAT(' THE FILE ',A12,' WILL CONTAIN THE INTEGRAL',/
I ENERGY SPECTRUM, TABULATED IN ',14,' DATA POINTS.')
,'
WRITE (6,117) Y
117 FORMAT(' THESE ENERGY SPECTRA WILL BE FOR THE YEAR ',F8.3,'.')
IF(IT.EQ.O) WRITE (6,118)
118 FORMAT(' THESE ENERGY SPECTRA WILL BE UNSHIELDED BY THE',/,
1 ' EARTH''S MAGNETIC FIELD, AND THE CONTRIBUTION FROM TRAPPED',/,
2 ' PROTONS WILL NOT BE INCLUDED.')
IF(IT.EQ.1) WRITE (6,119)
119 FORMAT(' THESE ENERGY SPECTRA WILL INCLUDE THE EFFECTS OF',!,
1 ' GEOMAGNETIC SHIELDING ACCORDING TO THE GEOMAGNETIC CUTOFF',/,
2 ' TRANSMITTANCE FUNCTION TABULATED IN GTRANS.DAT. THE',/,
3 ' CONTRIBUTION FROM TRAPPED PROTONS WILL BE INCLUDED' ,/,
4 ' ACCORDING TO THE ORBIT-AVERAGED TRAPPED PROTON SPECTRA',/,
5 ' TABULATED IN STASS.DAT.')
WRITE (6,120) M
120 FORMAT(' THE INTERPLANETARY WEATHER INDEX IS ',12,'.')
WRITE (6,121) TH
121 FORMAT(' THE SHIELDING THICKNESS IS ',F7.3,' GM/CM**2
1 ALUMINUM (OR EQUIVALENT).' ,//)
C
C SET UP THE ENERGY RANGE AND DIVIDE IT INTO EQUAL LOGARITHMIC STEPS.
C
EMAX=32000.
EMIN=IO.
ELNX=ALOG (EM IN)
DELN=(ALOG (EMAX)-ELNX)/NPT
c
C COMPUTE THE DIFFERENTIAL FLUX AT EACH ENERGY.
C
DO 200 I=INPT
EN=EXP(ELNX)
ELNX=ELNX+DELN
E(I)=EN
F(I )=INSIDE(Z,EN,Y,M,IT,TH)
IF(NDIF.GT.O) WRITE(40,201) F(I),E(I)
201 FORMAT(2X,E12.5,2X,E12.5)

107
 200 CONTINUE
C
C SUM THE DIFFERENTIAL ENERGY SPECTRUM TO GET THE INTEGRAL ENERGY
C SPECTRUM (BOX INTEGRATION),
C
SUM-. 6666*INSIDE (Z,EN, Y,M, IT, TH )*EN
DO 300 I=NPT, 2,-i
SUM-=SUM+O. 5* MFI )+F (I-I1 ))* (F (I )-E (1-1)
IF(NINT. GT.O) WRITE (45,201) SUM,E(I-1-1
300 CONTINUE
END

108
 PROGRAM UPSET

C THIS PROGRAM COMPUTES THE UPSET RATE DUE TO THE

C DIRECT IONIZATION OF INDIVIDUAL PARTICLES. IT ASSUMES
C THAT FOR EACH BIT THERE EXISTS A SENSITIVE VOLUME.
C IF AN AMOUNT OF ELECTRICAL CHARGE (>QCRIT) IS CREATED
C WITHIN THIS VOLUME BY THE IONIZING PARTICLE, THEN AN
C UPSET WILL OCCUR. THE SENSITIVE VOLUME IS IDEALIZED AS
C A PARALLELEPIPED WITH DIMENSIONS X, Y, AND Z.
C THE INPUTS ARE:
C THE CRITICAL CHARGE QCRIT IN PICOCOULOMBS,
C THE DIMENSIONS OF THE SENSITIVE VOLUME IN MICROMETERS,
C THE NAME OF THE FILE CONTAINING THE INTEGRAL LET
C SPECTRUM (SPEC.INT IS ASSUMED IF NO NAME IS ENTERED).
C THE INTEGRAL LET SPECTRUM IS ASSUMED TO BE IN UNITS OF
C PARTICLES/(SQUARE METER *STERADIAN*SEC) VERSUS
C MEV*(SQUARE CENTIMETER)/GRAM.
C THE OUTPUT IS GIVEN IN UPSETS/(BIT*SECOND) AND UPSETS/(BIT*DAY).

C THIS CALCULATION USES THE METHOD DESCRIBED IN "THE VARIABILITY

C OF SINGLE EVENT UPSET RATES IN THE NATURAL ENVIRONMENT,"
C JAMES H. ADAMS, JR., IEEE TRANS. ON NUCL. SCI., NS-30, 4475-
C 4480, DEC., 1983.
C IT IS OPERATIONALLY EQUIVALENT TO ROCKWELL'S CRIER PROGRAM
C (PICKEL AND BLANDFORD, IEEE TRANS. ON NUCL. SCI. NS-26, DEC.
C 1979, PP.4735-4739) WHEN USED WITH HEINRICH'S LET SPECTRUM
C (W. HEINRICH, RADIATION EFFECTS, VOL. 34, PP. 1I3-8, 1977).

C THIS PROGRAM CALLS A SUBPROGRAM, DIFPLD, THAT RETURNS

C THE DIFFERENTAL PATHLENGTH DISTRIBUTION IN THE SENSTIVE
C VOLUME.
C
CHARACTER*1 2 ALPHA,DEFAULT, BLANK
REAL L(1000),LMIN
DIMENSION FLUX(1000)
DATA DSI/2.321/,BLANK/' '/,DEFAULT/'SPEC.INTI/
C
C GET THE INPUT DATA.
C
WRITE(6,1)
1 FORMAT(' ENTER THE CRITICAL CHARGE IN PICOCOULOMBS: t,$)
ACCEPT , QCRIT
WRITE(6,2)
2 FORMAT(' ENTER THE DIMENSIONS OF THE SENSITIVE VOLUME,
1 X, Y, Z, IN MICROMETERS: ',$)
ACCEPT *, X,Y,Z
C
C CONVERT FROM MICROMETERS TO CENTIMETERS.
C
x-x*.oool

109
 Z=ZZ.0001
WRITE (6, 3)
3 FORMAT(' ENTER THE NAME OF THE INTEGRAL LET FILE: l,$)
ACCEPT 4, ALPHA
4 FORMAT(A12)
C
C IF NO FILE NAME WAS ENTERED, ASSUME 'SPEC.INT'.
C
IF (ALPHA. EQ. BLANK) ALPHA =DEFAULT
OPEN(UNIT=1,READONLY,SHARED,STATUS='OLD' ,FILE =ALPHA)
C
C READ THE INTEGRAL LET SPECTRUM.
C FLUX IS IN PARTICLES/(SQUARE METER*STERADIAN*SEC).
C L IS IN MEV*(SQUARE CENTIMETER)/GRAM.
C
N=I
5 CONTINUE
READ (1,11,END=6) FLUX(N),L(N)
11 FORMAT (2X,E12.5,2X,E12.5)
N=N÷I
C
C DON'T OVERFILL THE ARRAYS.
C
IF(N.LE. 1000) GOTO 5
C
C COMPUTE THE NUMBER OF POINTS IN THE LET SPECTRUM.
C
6 CONTINUE
N=N-1
C
C COMPUTE THE SURFACE AREA OF THE SENSITIVE VOLUME.
C
S=(2.*X*Yi2.*X*Z+2.*Y*Z)
C
C CONVERT FROM SQUARE CENTIMETERS TO SQUARE METERS.
C
S=S.l0001
C
C CONVERT THE DIMENSIONS OF THE SENSITIVE VOLUME TO G/CM**2.
C
X=XDSI
Y=YXDSI
Z=Z*DSI
C
C COMPUTE THE MAJOR DIAMETER OF THE SENSITIVE VOLUME.
: C
PMAX=SQRT (X *X÷Y *Y÷ZZ)
C
C COMPUTE THE ENERGY (IN MEV) REQUIRED TO PRODUCE QCRIT(IN PC)
C HOLE-ELECTRON PAIRS IN SILICON.
C
ENERGY=22.5*QCRIT

110
 C
C COMPUTE THE MINIMUM LET THAT CAN PRODUCE AN UPSET.
C
LMIN=ENERGY/PMAX
SUM=O.O
Q=LMIN
C
C INTEGRATE FROM LMIN TO THE LARGEST LET IN THE SPECTRUM.
C
DO 10 InN,1,-1
IF(L(I).LT.LMIN) GO TO 10
C
C COMPUTE THE PATHLENGTH CORRESPONDING TO L(I).

DuENERGY/L (I)
C
C CARRY OUT THE INTEGRAL.
C
SUM-SUM+ (L(I)-Q)*DIFPLD(D,X,Y,Z)FLUX(I)/(L(I)"*2)
Q=L(I)
10 CONTINUE
C
C COMPUTE THE ERROR RATE.
C
12 CONTINUE
ERR =ENERGY *S*3.1416 *SUM
WRITE(6,1)ERR
14 FORMAT(' ERROR RATE a ',1PE12.5,' IN UPSETS/BITOSEC.')
C
C CONVERT TO UPSETS/BIT'DAY.
:::• C
ERR=ERR*86,400.
WRITE (6, 15 )ERR
15 FORMAT(' ERROR RATE =,1PE12.5,t IN UPSETS/BIT'DAY.')
END

I.l
 FUNCTION DIFPLD(SL,W,H)
C
C THIS FUNCTION RETURNS THE PROBABILITY DENSITY FOR PATHS
C OF LENGTH S THROUGH A PARALLELEPIPED OF DIMENSIONS
C L, W, AND H. S, L, W, AND H MUST BE IN THE SAME UNITS.
C THIS IS AN EXACT SOLUTION, DUE TO M. D. PETROFF OF
C ROCKWELL INTERNATIONAL (SEE J. C. PICKEL AND J. T. BLANDFORD,
C IEEE TRANS. ON NUCL. SCI. NS-27, 1006(1980)) WITH
C SIMPLIFICATIONS DUE TO WARREN BENDEL OF NRL (PRIVATE
C COMMUNICATION). THE EQUATION NUMBERS REFER TO THE APPENDIX
C OF PICKEL AND BLANDFORD'S PAPER.
C
REAL L
C
C EQUATION (A-7)
c
AP33. (H W4i L+L4W)
C
C EQUATION (A-8)
C
DIFPLD=(G(S,L,W,H).G(SW,L,H)+G(SL,H,W)+G(S,W,H,L)+
1 G(S,H,W,L).G(SHLW))/(3°1416'AP)
RETURN
END

112
 FUNCTITON G(S,X,YýZ)
REAL KSQ
C
C PRELIMINARY DEFINITIONS
C
KS Q:X *X.Y 'Y
TSQ%:X 'XeZ*Z
T=SQRT (TSQ)
RSQ :KSQ+Z* Z
R-SQFT (RSQ)
V=:12. *X*Y*Z' Z
PSQ=S*S-Z*Z
QSQZS'S-X*X-Z*Z
IF((S.GE.O.O).AND.(S.LT.Z)) GO TO 10
IF((S.GE.Z) .AND.(S.LT.T))GO TM 20
IF((S.GE.T) .AND.(S.LE.R))GO To 30
G=O.O
RETURN

C EQUATION (A-9)

c
10GS3.*YY*/KSQRT*(KSQ))#24(3.'XY/(TR))

RETURN
C EQUATION (A-10)
C
30 G=S(3.X'Z/ST(RSQR(S)))*-*3.XY(2R)*
2 .YZ**(SQRT(QSQ)/S)*(8./TSQ+M./(SUS))
3 -(V/AA(SS'S))'Ac*s(/SQRT(pSQ))*2/S*)
RETURN

c EQUTION
A-11