SINGLE EVENT EFFECTS TEST REPORT. Heavy Ion Test Report DAC5675A. Rad-hard 14-bit 400MSPS D/A converter. Texas Instruments. RADEF/JYFL, Finland

Transcription

1 SINGLE EVENT EFFECTS TEST REPORT Heavy Ion Test Report Part Type DAC5675A Technology - Description Chip manufacturer Test facility Rad-hard 14-bit 400MSPS D/A converter Texas Instruments RADEF/JYFL, Finland Test Date April, 2010 Tesat Spacecom, PO No U dated 13/01/2010 Tesat Spacecom Technical Officer: Hartwig Storm Hirex reference : HRX/SEE/0293 Date : July 26th, 2010 Written by : F. Lochon Design Engineer Authorized by: F.X. Guerre Study Manager HIREX Engineering SAS au capital de RCS Toulouse B Siège social: 2 rue des Satellites Toulouse

2 RESULTS SUMMARY Facility: RADEF, Jyvaskyla (Finland) Test date: April 2010 Device description: DAC5675A - Rad-tolerant class V, 14 bit, 400 MSPS digital-to-analog converter Heavy ions results: SEL: o No SEL occurred during all the tests up to a LET of MeV/(mg/cm 2 ) with a fluence up to 1E+07 #/cm 2 at 125 C. SET on reference voltage output: o No SET bigger than 50 mv have been detected up to a LET of MeV/(mg/cm 2 ) with a fluence up to 4E+06 #/cm2 at ambient temperature (about 60 C). o SET smaller than 50 mv (detection threshold) may still occur SET/SEU on DAC output: o events have been detected over the whole LET range from 1.87 to MeV/(mg/cm 2 ) o Several event populations have been identified on the DAC output o SEU occurred at any time and lasted until a new conversion clock active edge occurred on the DUT o Positive-only, negative-only and bipolar SET have been observed o o o o o SEFI: o One SEU population consists in 1 DUT clock period (1 µs) long event with an arbitrary amplitude (up to 1.73 V), and in rare case 2 DUT clock period (2 µs) events have been detected One SEU population consists in preferred amplitude events with an arbitrary duration (up to 1 µs) One SEU population consists in arbitrary amplitude (up to 1 V) and arbitrary duration (up to 0.74 µs) events One SET population consists in single sample events whose duration is 1 ADC sample (10 ns) with amplitude up to 0.27 V. One SET population consists in arbitrary amplitude (up to 0.2 V) and arbitrary duration (up to 0.45 µs) No SEFI occurred during all the tests up to a LET of MeV/(mg/cm2) with a fluence up to 1E+07 #/cm2 at 125 C. Error cross-section: o Error cross-section has been determined for both full scale current (5 ma and 20 ma), and Weibull parameters have been extracted. o The DUT is more sensitive at 5 ma full scale current than at 20 ma full scale current. HRX/SEE/0293 Issue 02 Page 2

3 DOCUMENTATION CHANGE NOTICE Issue Date Page Change Item 01 04/05/2008 All Original issue 02 26/07/2010 4, Added distribution graphs for Neon, Argon, Iron and Krypton. Contributors to this work: Frédéric Lochon Hirex Engineering HRX/SEE/0293 Issue 02 Page 3

4 SEE TEST REPORT On DAC5675A TABLE OF CONTENTS 1 INTRODUCTION APPLICABLE AND REFERENCE DOCUMENTS APPLICABLE DOCUMENTS REFERENCE DOCUMENTS DEVICE INFORMATION DEVICE DESCRIPTION SAMPLE IDENTIFICATION TEST SET-UP HIREX TEST SYSTEM TEST PRINCIPLE TEST CONDITIONS DUT Bias Chain calibration Test run conditions Reference voltage monitoring Temperature monitoring TEST FACILITY SEE TEST RESULTS SEL & SEFI SET ON THE REFERENCE VOLTAGE SEU & SET AT DUT OUTPUT GLOSSARY PRESENTED EVENTS DETAILED INFORMATION DETAILED RESULTS PER RUN DISTRIBUTION GRAPHS AT NEON DISTRIBUTION GRAPHS AT ARGON DISTRIBUTION GRAPHS AT IRON DISTRIBUTION GRAPHS AT KRYPTON...32 HRX/SEE/0293 Issue 02 Page 4

5 LIST OF FIGURES Figure 1 Device identification... 7 Figure 2 DUT Bias description... 8 Figure 3 Example of initial calibration results... 9 Figure 4 DAC5675A, Single event cross-section per device Figure 5 Typical events Figure 6 Amplitude vs duration for A events at Nitrogen Figure 7 Amplitude vs duration for B events at Nitrogen Figure 8 Amplitude distribution at Nitrogen Figure 9 Duration distribution at Nitrogen Figure 10 Typical B events at Nitrogen Figure 11 Worst case B events at Nitrogen Figure 12 Worst case A events at Nitrogen Figure 13 Amplitude vs duration for A events at Xenon Figure 14 Amplitude vs duration for B events at Xenon Figure 15 Amplitude distribution at Xenon Figure 16 Duration distribution at Xenon Figure 17 Typical short A events at Xenon Figure 18 Typical A- events at Xenon Figure 19 Typical A+ events at Xenon Figure 20 Typical B event at Xenon Figure 21 Worst case B events at Xenon Figure 22 Worst case A events at Xenon Figure 23 Amplitude distribution at Neon Figure 24 Duration distribution at Neon Figure 25 Amplitude vs duration for A events at Neon Figure 26 Amplitude vs duration for B events at Neon Figure 27 Amplitude distribution at Argon Figure 28 Duration distribution at Argon Figure 29 Amplitude vs duration for A events at Argon Figure 30 Amplitude vs duration for B events at Argon Figure 31 Amplitude distribution at Iron Figure 32 Duration distribution at Iron Figure 33 Amplitude vs duration for A events at Iron Figure 34 Amplitude vs duration for B events at Iron Figure 35 Amplitude distribution at Krypton Figure 36 Duration distribution at Krypton Figure 37 Amplitude vs duration for A events at Krypton Figure 38 Amplitude vs duration for B events at Krypton LIST OF TABLES Table 1 Used ions and features thereof Table 2 Weibull parameters for full-scale current of 5mA and 20mA Table 3 Event type proportions with respect to total events (all conditions and all DUT taken together) Table 4 Event detailed information Table 5 ADS5463, detailed run results HRX/SEE/0293 Issue 02 Page 5

6 1 Introduction This report presents results of heavy ions test program carried out on Texas Instruments 14-bit DAC DAC5675MHFG-V ( VXC). 6 samples were delivered by TESAT. 3 samples were exposed using RADEF facility cyclotron at University of Jyvaskyla (JYFL) in Finland. This work was performed under the Purchase Order reference No U dated 13/01/2010. Test set-up allowed detection of Single Event Upsets (SEU), Single Event Transients (SET) Single Event Latchups (SEL) and Single Event Functional Interrupt (SEFI). 2 Applicable and Reference Documents 2.1 Applicable Documents AD-1 SMD No , MICROCIRCUIT, DIGITAL-LINEAR, 14 BIT, 400 MSPS DIGITAL TO ANALOG CONVERTER, MONOLITHIC SILICON. AD-2 Hirex proposal, HRX/PRO/2911 Issue 2, December 16, AD-3 Testplan for SEE Testing of DAC 5675, TES-09/57/STO Issue A, Reference Documents RD-1 Single Event Effects Test method and Guidelines ESA/SCC basic specification No DEVICE INFORMATION 3.1 Device description Part Description: Rad-hard 14-bit 500MSPS D/A converter Package: 52-leads Quad flatpack with non-conductive tie bar Technology: - Marking: logo serial VXC DAC5675AMHFG-V THA 7ACR 0808A Q Samples Used: 212, 209, 211, 213, 215 and 217 HRX/SEE/0293 Issue 02 Page 6

7 3.2 Sample identification Package Marking Full die view Opened device Figure 1 Device identification Die marking Die area: 3.6 mm x 3.2 mm HRX/SEE/0293 Issue 02 Page 7

8 4 Test Set-up 4.1 Hirex Test system The test system is based on a Virtex5 FPGA which runs at 200MHz. The test board has 168 I/Os which can be configured using several I/O standards. The test system also features as a standard feature: latchup detection and voltage/current monitoring up to 24 channels, a voltage reference monitor and a DDR2 memory to store data (DDR2 not used in the present test). The communication between the test system and the computer is done thanks to a 100 Mbit/s Ethernet link which enables high speed data downloading and uploading. 4.2 Test principle In order to test the DUT, a 16-bit ADC was used to convert the differential output voltage. The chain formed by the DUT followed by the ADC can then be calibrated for each value of the word presented to the DAC by observing the word at the ADC output and doing simple statistics (min and max values) during a given sample count. This statistic can then be used during beam exposition to detect conversion errors. The conversion clock for the DUT was 1 MHz while the conversion clock for the 16-bit ADC was 100 MHz. The full scale current could be selected using a multiplexor to connect the BiasJ pin to a resistor (either 960 ohms or 3840 ohms). 4.3 Test conditions DUT Bias Figure 2 DUT Bias description HRX/SEE/0293 Issue 02 Page 8

9 4.3.2 Chain calibration For each DUT, preliminary to the test runs and whenever needed, a calibration was performed in the actual test conditions (under vacuum and at the test temperature). This chain calibration consisted in acquiring for each input step (16384 steps of the 14-bits DAC), the 16-bit ADC output value. For each step, 2E+6 conversions were performed and min and max ADC output values were recorded. Figure 3 shows a calibration example which was used in Run111. Min value, max value and delta (Max-Min) is plotted for each DAC word. 100 LSB for ADC corresponds to a differential output voltage at the DUT of about 3.05mV (about 25 LSB for the DAC) Test run conditions Figure 3 Example of initial calibration results Each run consisted then in continuously applying a ramp at the DAC input, each step being converted 10E+6 times at a conversion sampling frequency of 100MSPS, which corresponds to 10ms. For each step, the ADC output is compared with the min and max values recorded at calibration stage (increased by 15 LSB at 5mA full scale current and 45 LSB at 20 ma to prevent noisy events) and when these values are exceeded, an error is detected and counted. The actual settings are the following: Each time a conversion error is detected, the 31 previous samples and the subsequent conversions are recorded until 70 successive conversions are within the min and max values for this particular step. If the number of conversion errors exceeds 120, the conversions values are no more recorded but the total number of conversions in error is counted. Each run consisted in sweeping the whole DAC input range at least twice, resulting in a minimum run duration of about 6 minutes (2 * * 10ms + dead times) Reference voltage monitoring The internal reference voltage can be measured at EXTIO output pin. The signal is digitized with a 10-bit converter (1LSB~ 4.7mV) at a 400 MHz sampling frequency and 2 thresholds (low and high) are programmed Temperature monitoring Water cooling and thermal drain were used in conjunction with DUT package thermal pad. Package temperature was then maintained constant under vacuum for each test run. Temperature was measured using a thermocouple glued to the DUT package. HRX/SEE/0293 Issue 02 Page 9

10 5 Test Facility Test at the cyclotron accelerator was performed at University of Jyvaskyla (JYFL) (Finland) under HIREX Engineering responsibility. The facility includes a special beam line dedicated to irradiation studies of semiconductor components and devices. It consists of a vacuum chamber including component movement apparatus and the necessary diagnostic equipment required for the beam quality and intensity analysis. The cyclotron is a versatile, sector-focused accelerator of beams from hydrogen to xenon equipped with three external ion sources: two electron cyclotron resonance (ECR) ion sources designed for high-charge-state heavy ions, and a multicusp ion source for intense beams of protons. The ECR's are especially valuable in the study of single event effects (SEE) in semiconductor devices. For heavy ions, the maximum energy attainable can be determined using the formula, 130 Q 2 /M, where Q is the ion charge state and M is the mass in Atomic Mass Units. Test chamber Irradiation of components is performed in a vacuum chamber with an inside diameter of 75 cm and a height of 81 cm. The vacuum in the chamber is achieved after 15 minutes of pumping, and the inflation takes only a few minutes. The position of the components installed in the linear movement apparatus inside the chamber can be adjusted in the X, Y and Z directions. The possibility of rotation around the Y-axis is provided by a round table. The free movement area reserved for the components is 25 cm x 25 cm, which allows one to perform several consecutive irradiations for several different components without breaking the vacuum. The assembly is equipped with a standard mounting fixture. The adapters required to accommodate the special board configurations and the vacuum feed-throughs can also be made in the laboratory s workshops. The chamber has an entrance door, which allows rapid changing of the circuit board or individual components. A CCD camera with a magnifying telescope is located at the other end of the beam line to determine accurate positioning of the components. The coordinates are stored in the computer s memory allowing fast positioning of various targets during the test. Beam quality control For measuring beam uniformity at low intensity, a CsI(Tl) scintillator with a PIN-type photodiode readout is fixed in the mounting fixture. The uniformity is measured automatically before component irradiation and the results can be plotted immediately for more detailed analysis. A set of four collimated PIN-CsI(Tl) detectors is located in front of the beam entrance. The detectors are operated with step motors and are located at 90 degrees with respect to each other. During the irradiation and uniformity scan they are set to the outer edge of the beam in order to monitor the stability of the homogeneity and flux. Two beam wobblers and/or a 0.5 microns diffusion Gold foil can be used to achieve good beam homogeneity. The foil is placed 3 m in front of the chamber. The wobbler-coils vibrate the beam horizontally and vertically, the proper sweeping area being attained with the adjustable coil-currents. Dosimetry The flux and intensity dosimeter system contains a Faraday cup, several collimators, a scintillation counter and four PIN-CsI(Tl) detectors. Three collimators of different size and shape are placed 25 cm in front of the device under test. They can be used to limit the beam to the active area to be studied. At low fluxes a plastic scintillator with a photomultiplier tube is used as an absolute particle counter. It is located behind the vacuum chamber and is used before the irradiation to normalize the count rates of the four PIN-CsI(Tl) detectors. The RADEF ions used are listed in the table below. Range Ion Energy [MeV] [MeV/mg/cm2] [microns] 15N Ne Ar Fe Kr Xe Table 1 Used ions and features thereof HRX/SEE/0293 Issue 02 Page 10

11 6 SEE Test Results 3 samples s/n212, s/n216 and s/n217 have been exposed over an LET range from 1.87 to MeV/(mg/cm²) at ambient temperature (about 60 C) and at MeV(mg/cm 2 ) at 80 C and 125 C. Detailed results per run are presented in section 9 while the corresponding Single event crosssection per device is shown in Figure 4 and Weibull parameters are presented in Table SEL & SEFI Figure 4 DAC5675A, Single event cross-section per device Full scale current A x0 W s 5mA 3.8E mA 2.5E Table 2 Weibull parameters for full-scale current of 5mA and 20mA No SEL nor SEFI was detected on three samples up to an effective LET of MeV/(mg/cm2) and with a fluence up to 1E+07 #/cm2 at 125 C. 6.2 SET on the reference voltage A threshold of +/- 50 mv was used when monitoring the reference voltage output. In this configuration, no SET were observed on the reference voltage output, but smaller SET may exists. 6.3 SEU & SET at DUT output Conversion events were detected over the entire LET range, resulting in a total of events. To simplify the analysis, events have been assigned a unique number and classified in two categories depending on the ADC samples amplitude distribution. Category A corresponds to events which have less than 4% of samples between 10% and 90% relative to minimum and maximum sample in the event. Category B corresponds to events which are not in category A. Moreover, events have been tagged with an attribute polarity to determine if the event was: only above the maximum threshold ( + as positive only events), only below the minimum threshold ( - as negative only events), above the maximum threshold and below the minimum threshold ( 0 as bipolar events). All presented events are summarized in Table 4 in section 8 which shows event information. HRX/SEE/0293 Issue 02 Page 11

12 Hirex Engineering Figure 5 shows two typical events to illustrate A- and B- events. The A- event (set_68422) looks like an SEU while the B- event (set_58791) looks like an SET. More generally: B events are typically transients A events are typically either SEU which are only corrected after a new DAC input clock pulse, or very short transients with very few samples in error (typically 1 or 2). Figure 5 Typical events Table 3 shows event type proportion for each ion and tilt. It is possible to observe that: A0 events are quite uncommon which is because A events are mostly SEU, B0 events are more common with high LET ions compared to low LET ions, while B events population is comparable to A events population at low LET, B events are more common at high LET which is because transients are also becoming bigger in terms of amplitude and finally get detected when going over detection thresholds. Ion A+ A0 A- B+ B0 B- total A total B total events N 23.0% 0.0% 27.2% 20.8% 0.0% 29.0% 50.2% 49.8% 1909 Ne 25.2% 0.2% 26.8% 21.9% 0.5% 25.5% 52.2% 47.8% 3911 Ne@ % 0.0% 21.7% 23.1% 1.2% 33.6% 42.0% 58.0% 4643 Ar 19.5% 0.0% 20.6% 28.4% 2.2% 29.3% 40.1% 59.9% Fe 19.2% 0.0% 21.8% 29.0% 2.6% 27.4% 41.0% 59.0% Kr 18.1% 0.0% 19.2% 30.5% 3.3% 28.9% 37.3% 62.7% Xe 17.5% 0.1% 20.2% 27.7% 5.0% 29.4% 37.9% 62.1% Xe@60 - sleep 9.7% 0.9% 10.3% 35.1% 0.8% 43.2% 20.9% 79.1% Table 3 Event type proportions with respect to total events (all conditions and all DUT taken together) HRX/SEE/0293 Issue 02 Page 12

13 Hirex Engineering Figure 6 and Figure 7 represents amplitude versus duration graphs for A events and B events at Nitrogen (LET = 1.87). It is possible to observe that most A events are bigger in terms of amplitude than B events which, again, is because A events are mostly SEU and thus amplitude should be arbitrary. Figure 6 Amplitude vs duration for A events at Nitrogen It is possible to observe that some B events are 100 samples long (1µs), those events are in fact SEU which have not been categorized as A events because of their small size (the criteria does not work well because of the noise). Others long (more than 10 samples) B events may still be SEU (see Figure 10). Figure 7 Amplitude vs duration for B events at Nitrogen HRX/SEE/0293 Issue 02 Page 13

14 Hirex Engineering To have a more precise idea of the distribution of the events, Figure 8 and Figure 9 shows the distribution of the events at Nitrogen in terms of amplitude and duration. Amplitude distribution has peaks for 4mV to 8mV and 31mV to 63mV because of A events which, for some reason, have some preferred SEU amplitudes as seen on Figure 6. One possibility would be because of an internal pipeline which would be sensitive between each stage. In a similar way, duration distribution has peaks because 1µs-long A events are important, possibly because the input stage of the DAC is SEE sensitive. Figure 8 Amplitude distribution at Nitrogen Figure 9 Duration distribution at Nitrogen HRX/SEE/0293 Issue 02 Page 14

15 Figure 10 Typical B events at Nitrogen Figure 11 Worst case B events at Nitrogen Figure 12 Worst case A events at Nitrogen HRX/SEE/0293 Issue 02 Page 15

16 Similar graphs can be built for Xenon (LET = 54.95). Figure 13 and Figure 14 represents amplitude versus duration graphs for A events and B events at Xenon (LET = 54.95) Figure 13 Amplitude vs duration for A events at Xenon Figure 14 Amplitude vs duration for B events at Xenon HRX/SEE/0293 Issue 02 Page 16

17 Hirex Engineering Figure 15 and Figure 16 show amplitude and duration distribution at Xenon. Typical A events are shown in Figure 17, Figure 18, Figure 19 and Figure 20. Some worst cases (depending on the definition) for both A events and B events are shown in Figure 21 and Figure 22. Figure 15 Amplitude distribution at Xenon Figure 16 Duration distribution at Xenon HRX/SEE/0293 Issue 02 Page 17

18 Figure 17 Typical short A events at Xenon Figure 18 Typical A- events at Xenon Figure 19 Typical A+ events at Xenon HRX/SEE/0293 Issue 02 Page 18

19 Figure 20 Typical B event at Xenon Figure 21 Worst case B events at Xenon Figure 22 Worst case A events at Xenon HRX/SEE/0293 Issue 02 Page 19

20 7 Glossary Most of the definitions here below are from JEDEC standard JESD89A DUT: Device under test. Fluence (of particle radiation incident on a surface): The total amount of particle radiant energy incident on a surface in a given period of time, divided by the area of the surface. In this document, Fluence is expressed in ions per cm2. Flux: The time rate of flow of particle radiant energy incident on a surface, divided by the area of that surface. In this document, Flux is expressed in ions per cm2*s. Single-Event Effect (SEE): Any measurable or observable change in state or performance of a microelectronic device, component, subsystem, or system (digital or analog) resulting from a single energetic particle strike. Single-event effects include single-event upset (SEU), multiple-bit upset (MBU), multiple-cell upset (MCU), single-event functional interrupt (SEFI), single-event latch-up (SEL), single-event snap-back (SESB), single-event hard error (SHE) and single-event transient (SET), single-event burnout (SEB), and single-event gate rupture (SEGR). Single-Event Upset (SEU): A soft error caused by the transient signal induced by a single energetic particle strike. Single-Event Functional Interrupt (SEFI): A soft error that causes the component to reset, lock-up, or otherwise malfunction in a detectable way, but does not require power cycling of the device (off and back on) to restore operability. A SEFI is often associated with an upset in a control bit or register. Single-Event Latch-up (SEL): An abnormal high-current state in a device caused by the passage of a single energetic particle through sensitive regions of the device structure and resulting in the loss of device functionality. SEL may cause permanent damage to the device. If the device is not permanently damaged, power cycling of the device (off and back on) is necessary to restore normal operation. An example of SEL in a CMOS device is when the passage of a single particle induces the creation of parasitic bipolar (p-n-p-n) shorting of power to ground. Single-Event Latch-up (SEL) cross-section: the number of events per unit fluence. For chip SEL cross-section, the dimensions are cm2 per chip. Single Event Transient (SET): A momentary voltage excursion (voltage spike) at a node in an integrated circuit caused by a single energetic particle strike. Error cross-section: the number of errors per unit fluence. For device error cross-section, the dimensions are cm2 per device. For bit error cross-section, the dimensions are cm2 per bit. Tilt and Roll angle: tilt angle, rotation axis of the DUT board is perpendicular to the beam axis; roll angle, board rotation axis is parallel to the beam axis Weibull Function: F(x) = A (1- exp{-[(x-x 0 )/W] s }) x = effective LET in MeV-cm 2 /milligram; F(x) = SEE cross-section in square-cm2/bit; A = limiting or plateau cross-section; x 0 = onset parameter, such that F(x) = 0 for x < x 0 ; W = width parameter; s = a dimensionless exponent. HRX/SEE/0293 Issue 02 Page 20

21 8 Presented events detailed information Event Run DAC input Effective DUT Ion Sleep Current Voltage Temperature Tilt number Hirex word LET N RT N RT N RT N RT N RT N RT Xe RT Xe RT Xe RT Xe RT Xe RT Xe RT Xe RT Xe RT Xe RT Xe RT Xe RT Xe RT Xe RT Xe Table 4 Event detailed information HRX/SEE/0293 Issue 02 Page 21

22 9 Detailed results per run Run # Hirex Run # JYFL Irradiated DUT Tested DUT Voltage Sleep temperature current Ion Energy LET Range TILT SEL Fluence TIME Mean flux Run Dose Eff, LET SET SET Fluence SET X-SECTION ,2 0 RT 20 Ar , ,21E , ,16E+06 7,07E ,2 0 RT 5 Ar , ,29E , ,01E+06 1,00E ,5 0 RT 5 Ar , ,68E , ,39E+06 1,04E ,5 0 RT 20 Ar , ,48E , ,19E+06 9,65E ,2 0 RT 20 Ar , ,85E , ,54E+06 5,83E ,2 0 RT 5 Ar , ,40E , ,24E+06 8,60E ,5 0 RT 5 Ar , ,08E , ,75E+05 9,14E ,5 0 RT 20 Ar , ,26E , ,12E+06 5,68E ,5 0 RT 20 Ar , ,12E , ,63E+05 5,18E ,5 0 RT 5 Ar , ,20E , ,07E+06 8,34E ,2 0 RT 5 Ar , ,20E , ,07E+06 8,25E ,2 0 RT 20 Ar , ,23E , ,08E+06 5,17E ,2 0 RT 20 Ne 186 3, ,06E , ,58E+05 2,40E ,2 0 RT 5 Ne 186 3, ,28E , ,08E+06 2,46E ,5 0 RT 5 Ne 186 3, ,36E , ,20E+06 2,25E ,5 0 RT 20 Ne 186 3, ,22E , ,08E+06 2,52E ,5 0 RT 20 Ne 186 3, ,14E , ,03E+06 5,04E ,5 0 RT 5 Ne 186 3, ,34E , ,19E+06 4,66E ,2 0 RT 5 Ne 186 3, ,29E , ,16E+06 5,34E ,2 0 RT 20 Ne 186 3, ,30E , ,16E+06 4,58E ,2 0 RT 20 Ne 186 3, ,21E , ,08E+06 3,04E ,2 0 RT 5 Ne 186 3, ,56E , ,40E+06 5,43E ,5 0 RT 5 Ne 186 3, ,75E , ,55E+06 5,55E ,5 0 RT 20 Ne 186 3, ,60E , ,42E+06 3,32E-04 HRX/SEE/0293 Issue 02 Page 22

23 Run # Hirex Run # JYFL Irradiated DUT Tested DUT Voltage Sleep temperature current Ion Energy LET Range TILT SEL Fluence TIME Mean flux Run Dose Eff, LET SET SET Fluence SET X-SECTION ,5 0 RT 20 Ne 186 3, ,31E , ,16E+06 2,16E ,5 0 RT 5 Ne 186 3, ,30E , ,16E+06 3,14E ,2 0 RT 5 Ne 186 3, ,23E , ,09E+06 3,64E ,2 0 RT 20 Ne 186 3, ,22E , ,11E+06 2,11E ,2 0 RT 20 Ne 186 3, ,53E , ,35E+06 2,73E ,2 0 RT 5 Ne 186 3, ,38E , ,26E+06 2,96E ,5 0 RT 5 Ne 186 3, ,68E , ,50E+06 2,93E ,5 0 RT 20 Ne 186 3, ,91E , ,71E+06 2,63E ,5 0 RT 20 N 139 1, ,68E , ,49E+06 1,26E ,5 0 RT 5 N 139 1, ,77E , ,58E+06 1,52E ,2 0 RT 5 N 139 1, ,83E , ,62E+06 1,57E ,2 0 RT 20 N 139 1, ,63E , ,44E+06 1,14E ,2 0 RT 20 N 139 1, ,58E , ,40E+06 2,19E ,2 0 RT 5 N 139 1, ,98E , ,76E+06 1,74E ,5 0 RT 5 N 139 1, ,94E , ,73E+06 1,58E ,5 0 RT 20 N 139 1, ,72E , ,52E+06 1,16E ,5 0 RT 20 Fe , ,20E , ,07E+06 1,15E ,5 0 RT 5 Fe , ,21E , ,07E+06 1,77E ,2 0 RT 5 Fe , ,16E , ,02E+06 1,60E ,2 0 RT 20 Fe , ,33E , ,18E+06 9,38E ,2 0 RT 20 Fe , ,09E , ,84E+06 1,02E ,2 0 RT 5 Fe , ,60E , ,15E+05 1,16E ,5 0 RT 5 Fe , ,86E , ,58E+06 1,80E ,5 0 RT 20 Fe , ,38E , ,22E+06 1,08E-03 HRX/SEE/0293 Issue 02 Page 23

24 Run # Hirex Run # JYFL Irradiated DUT Tested DUT Voltage Sleep temperature current Ion Energy LET Range TILT SEL Fluence TIME Mean flux Run Dose Eff, LET SET SET Fluence SET X-SECTION ,5 0 RT 20 Kr , ,22E , ,07E+06 1,65E ,5 0 RT 5 Kr , ,10E , ,40E+05 2,73E ,2 0 RT 5 Kr , ,06E , ,28E+05 2,60E ,2 0 RT 20 Kr , ,05E , ,15E+05 1,56E ,2 0 RT 20 Kr , ,06E , ,26E+05 1,52E ,2 0 RT 5 Kr , ,23E , ,08E+06 2,36E ,5 0 RT 5 Kr , ,01E , ,78E+05 2,62E ,5 0 RT 20 Kr , ,13E , ,89E+05 1,63E ,5 0 RT 20 Xe , ,68E , ,47E+06 1,93E ,5 0 RT 5 Xe , ,12E , ,93E+05 3,17E ,2 0 RT 5 Xe , ,28E , ,50E+05 3,02E ,2 0 RT 20 Xe , ,54E , ,34E+05 2,00E ,2 0 RT 20 Xe , ,36E , ,24E+05 2,07E ,2 0 RT 5 Xe , ,14E , ,11E+05 3,13E ,5 0 RT 5 Xe , ,28E , ,12E+06 3,26E ,5 0 RT 20 Xe , ,25E , ,05E+06 2,03E , Xe , ,01E , , Xe , ,00E , , Xe , ,00E , , Xe , ,00E , , Xe , ,00E , ,02E+05 4,96E , Xe , ,00E , ,80E+05 5,08E , Xe , ,00E , ,77E+05 7,48E , Xe , ,00E , ,80E+05 7,41E-04 HRX/SEE/0293 Issue 02 Page 24

25 Run # Hirex Run # JYFL Irradiated DUT Tested DUT Voltage Sleep temperature current Ion Energy LET Range TILT SEL Fluence TIME Mean flux Run Dose Eff, LET SET SET Fluence SET X-SECTION , Xe , ,00E , , Xe , ,00E , , Xe , ,00E , ,93E+06 5,12E , Xe , ,00E , ,50E+06 5,02E-04 Table 5 ADS5463, detailed run results HRX/SEE/0293 Issue 02 Page 25

26 10 Distribution graphs at Neon Figure 23 Amplitude distribution at Neon Figure 24 Duration distribution at Neon HRX/SEE/0293 Issue 02 Page 26

27 Figure 25 Amplitude vs duration for A events at Neon Figure 26 Amplitude vs duration for B events at Neon HRX/SEE/0293 Issue 02 Page 27

28 11 Distribution graphs at Argon Figure 27 Amplitude distribution at Argon Figure 28 Duration distribution at Argon HRX/SEE/0293 Issue 02 Page 28

29 Figure 29 Amplitude vs duration for A events at Argon Figure 30 Amplitude vs duration for B events at Argon HRX/SEE/0293 Issue 02 Page 29

30 12 Distribution graphs at Iron Figure 31 Amplitude distribution at Iron Figure 32 Duration distribution at Iron HRX/SEE/0293 Issue 02 Page 30

31 Figure 33 Amplitude vs duration for A events at Iron Figure 34 Amplitude vs duration for B events at Iron HRX/SEE/0293 Issue 02 Page 31

32 13 Distribution graphs at Krypton Figure 35 Amplitude distribution at Krypton Figure 36 Duration distribution at Krypton HRX/SEE/0293 Issue 02 Page 32

33 Figure 37 Amplitude vs duration for A events at Krypton Figure 38 Amplitude vs duration for B events at Krypton HRX/SEE/0293 Issue 02 Page 33