Space-RHA
Tools / Latchup and burnout / Proton Proxy Risk Tool

Heavy-ion risk after a zero-event proton test

Enter the proton energy, fluence and events seen, the depth over which you assume the part collects charge, and the mission. You get how much of the heavy-ion behavior the proton test bounds, the worst case it leaves open, and how the result updates a historical SEL prior.

after Ladbury and Lauenstein, IEEE TNS 2016175 candidate Weibull behaviorsrecoil spectra, 50 to 500 MeVEHSP 2026-09 prior, Monte Carlo with a fixed seed95 parts tested both waysMethodHelp

Proton test

Device and SEE mode

The depth sets how much credit a passed proton test earns: a recoil that deposits a given energy has half the equivalent LET in a 20 µm volume that it has in a 10 µm one. For bulk CMOS whose process is unknown the tool assumes 20 µm, a precaution rather than a better estimate, and every run also repeats the analysis at 5, 10, 20 and 30 µm (below the result). Choosing the depth.

Mission

Where is this environment? See its orbit and a real mission in 3D

Advanced
Why the conventions behind this tool are what they are: LET 15, 37, 60 and 75, why 200 MeV and 63 MeV, the 1e11 proton fluence, low energy protons, 3.7 versus 2.3 for a null result.

Residual heavy-ion risk after proton test

What the paired data say: proton versus heavy-ion SEL in the combined test corpus

Every part in the corpus with both heavy-ion and proton SEL records (95 parts, 92 from published REDW/RADECS/GSFC sources; internal reports counted in aggregate only). Heavy-ion onset = LET@10⁻⁸ from the Poisson-Weibull fit where available, else the author-stated threshold. Full per-part table and case studies: Proton vs. Heavy-Ion Case Studies.

P(proton SEL observed | heavy-ion onset)

Bars: fraction of parts in each heavy-ion onset bin that showed proton SEL, counting proton-SEL parts plus parts with an adequate null test (≥ 50 MeV, ≥ 10¹⁰ p/cm²). Error bars: 90% credible interval on that fraction (Jeffreys/Beta method, which stays sensible for small counts and for 0-of-n or n-of-n bins). n below each bar; weaker nulls are not counted.

Heavy-ion onset of parts that did vs. did not latch under protons

Each dot is one part, placed at its heavy-ion onset (MeV·cm²/mg). Parts with proton SEL cluster below LET 5; parts with a null proton test span the whole range, and the low-onset ones are the residual risk this tool quantifies.

How it works

What a proton test can and cannot see

01

Protons act through recoils

A proton test reveals heavy-ion susceptibility only through the nuclear recoils the protons knock loose, about one per 289,000 protons at 200 MeV, and those recoils travel only microns.

02

Recoils reach only modest LET

Their equivalent LET falls with the depth of the sensitive volume, so the test probes low LET and hardly reaches high LET, while the mission's heavy ions go far beyond it.

03

Bound what it covers, cap the rest

Every candidate Weibull behavior the test would have revealed is bounded by the Poisson limit on the observed count. Behaviors it cannot constrain are carried at the die-area cap: the residual worst case.

How the numbers are computed recoils, the model grid, the mission rate

Residual heavy-ion SEE risk after a proton-only test campaign. Recoil LET-equivalent coverage and Weibull model-grid rate bounding after Ladbury & Lauenstein, IEEE TNS 2016, and Ladbury, Lauenstein & Hayes, IEEE TNS 2015; recoil spectra after Hiemstra & Blackmore, IEEE TNS 2003.

Method

A null (or N-event) proton test constrains heavy-ion susceptibility only through the nuclear-reaction recoils the protons generate, about 1 per 289,000 at 200 MeV. The recoil population is described as an integral fluence vs. equivalent LET (LETEQ = Edep/ρd), which collapses toward low LET as the sensitive-volume depth grows, because short-range recoils cannot deposit their full LET across a deep volume. For each candidate Weibull σ(LETEQ) model on the published grid (onset 0.5–6.5, width 5–25, shape 0.5–2.5), the expected test event count per unit limiting cross section is μ₁ = ∫F(>L)dW; the Poisson upper limit on the observed count then bounds σsat ≤ μUL/μ₁. Models with σUL < τ are "meaningfully bounded"; the bounded worst case is the max mission rate among them, computed by full CREME96 spectral integration. Models the test cannot bound are evaluated at the die-area geometric cap, the residual worst case. Mission exposure uses the same environment tables and episodic-flare model as the SEL Test-LET tool.

Emulator calibration and validation

Emulator calibration

The recoil fluence-vs-LETEQ family is a piecewise-linear emulator of the published CRÈME-MC results (2016 Fig 5 depth family; Hiemstra 2003 shallow spectra), with energy scaling (counts ×0.6–1.4, spectral hardness) and a maximum-recoil-energy cap Er,max/(ρd). Validation against the papers: worst-case bounded rate at ISS, 10-µm SV, 10¹⁰ 200-MeV p/cm² reproduced the published power-law to ×1.04 with the site's earlier ISS table; the September 2026 CREME96 ISS run is 5 to 37 percent higher at the relevant onsets, so the tool now sits about ×1.2 above the published curve, an environment difference rather than an emulator one; the %-unbounded matrix (2016 Fig 9) matches within ~±8 points across 50–400 MeV; at 10¹² p/cm² the bound runs up to ~×4 conservative. High-Z package option adds the p+Au fission component (~2,250 events/cm² per 10¹⁰ at 200 MeV, LETEQ to ~38 shallow, collapsing with depth per 2016 Fig 10).

References

R. Ladbury and J.-M. Lauenstein, "Evaluating Constraints on Heavy-Ion SEE Susceptibility Imposed by Proton SEE Testing and Other Mixed Environments," IEEE Trans. Nucl. Sci. vol. 64, no. 1, pp. 301–308, 2016.
R. Ladbury, J.-M. Lauenstein and K. P. Hayes, "Use of Proton SEE Data as a Proxy for Bounding Heavy-Ion SEE Susceptibility," IEEE Trans. Nucl. Sci. vol. 62, no. 6, pp. 2505–2510, 2015.
D. M. Hiemstra and E. W. Blackmore, "LET Spectra of Proton Energy Levels From 50 to 500 MeV and Their Effectiveness for Single Event Effects Characterization of Microelectronics," IEEE Trans. Nucl. Sci. vol. 50, no. 6, pp. 2245–2249, 2003.
T. L. Turflinger et al. "RHA Implications of Proton on Gold-Plated Package Structures in SEE Evaluations," IEEE Trans. Nucl. Sci. vol. 62, no. 6, pp. 2468–2475, 2015.

This tool is provided as a free community resource. Results should be verified independently. The help page covers the inputs, the model grid and the validation against paired data.See Help & User Guide