NASA Office of Logic Design

NASA Office of Logic Design

A scientific study of the problems of digital engineering for space flight systems,
with a view to their practical solution.


Memories: Radiation Test Results

 

Title, Authors, Reference, Link Abstract, Summary, Conclusions


Radiation Issues and Applications of Floating Gate Memories

L. Z. Scheick,; D. N. Nguyen
NASA Jet Propulsion Laboratory
Non-volitile Memory Technology Symposium
January, 15, 2000

Abstract
The radiation effects that affect various systems that comprise floating gate memories are presented. The wear-out degradation results of unirradiated flash memories are compared to irradiated flash memories. The procedure analyzes the failure to write and erase caused by wear-out and degradation of internal charge pump circuits. A method is described for characterizing the radiation effects of the floating gate itself. The rate dependence, stopping power dependence, SEU susceptibility and applications of floating gate in radiation environment are presented. The ramifications for dosimetry and cell failure are discussed as well as for the long term use aspects of non-volatile memories.


The Effects of TID on Wear-Out of Advanced Flash Memories

D. Nguyen,; L. Scheick,; G. M. Swift,; S. M. Guertin
NASA Jet Propulsion Laboratory
Nuclear and Space Radiation Effects Conference , July 6, 2000

Outline
  • Introduction
  • TID Effects on Wear-Out of Erase/Write Capability
  • Dependence of Wear-Out on Charge Pump Degradation
  • Correlation of Charge Pump Degradation with Erase Failure
  • Mechanism of Wear-Out
  • Summary

Technologies: Intel Strata 28F128; Samsung KM29T128; Samsung KM29U128T


ATMEL AT40KEL040 re-programmable SRAM based FPGA Radiation Test Report:
 Total Dose (TID) and Single Event Effects (SEE)

 

Abstract (excerpt)
This chip has been designed on the ATMEL CMOS 0.35 mm radiation hardened process. This report gives a summary of the radiation test results obtained on this product. An other test report covers the total dose and Single Event Effects test results for the serial 1Mbit EEPROM AT17LV010-10DP to be used as a configuration memory in association with this FPGA.


AT40KEL040 Reprogrammable SRAM based FPGA Total Dose (TID) and Single Event Effects (SEE)

 

Abstract (excerpt)
Atmel has redesigned its existing commercial SRAM based reprogrammable FPGA, into a radiation hardened device: the AT40KEL040, while SEU (Single Event Upset) hardening all heavy ions susceptible structures. This chip has been designed on the Atmel CMOS 0.35 micron radiation hardened process. This report gives a summary of the radiation test results obtained on this product. In addition, some information are given on the 1Mbit Serial EEPROM AT17LV010-10DP to be used as a configuration memory in association with the AT40KEL040 FPGA.


SEE Measurements on the Aeroflex MCM Flash Memory

Document: D-20185
January 2001
mars_flash_final.pdf

Overview
This report describes the testing of a Aeroflex Flash Multi Chip Module (MCM) at Texas A&M cyclotron (TAM). The purpose of this test was to determine the SEE characteristics of a flash memory under heavy ion radiation. The cross section of each device, for both Single Event Upsets (SEU) and Single Event Latch-up (SEL), as a function of ion Linear Energy Transfer (LET) was the primary goal. The SEU and SEL threshold of these devices was also determined. Another observation was any long term or total dose effects from the ion radiation.  The part number is ACT-F2M32A-090F18C.  This module uses the AMD AM29F016 die.


SEE Measurement at Brookhaven National Laboratory for the SRAMs:

WMS128k8 128x8, MT5C2568 32Kx8, MT5C2564 64Kx4

Document: D-18677
December 1999
srams_18677.pdf

Overview
The purpose of this test was to determine the SEE characteristics of three different SRAMs under heavy ion radiation. The cross section of each device, for both Single Event Upsets (SEU) and Single Event Latch-up (SEL), as a function of ion Linear Energy Transfer (LET) was the primary goal. The SEU and SEL threshold of these devices was also determined. Another observation was any long term or total dose effects from the radiation. The three devices tested were all SRAMs.
  • WMS128K8, White Micro, 128kx8
  • MT5C2564, Austin, 64kx4
  • MT5C2568, Austin, 32kx8


SEE Measurement for the SDRAM and the Blackjack ASIC

December, 1999
sdram_lsi_asic_1999.pdf

Overview
The purpose of this test was to determine the SEE characteristics of an SDRAM  and an ASIC under heavy ion radiation. The cross section of each device, for both Single Event Upsets (SEU) and Single Event Latch-up (SEL), as a function of ion Linear Energy Transfer (LET) was the primary goal. The SEU and SEL threshold of these devices was also determined. Another observation was any long term or total dose effects from the radiation.
  • SDRAM, Micron, 64Mb
  • ASIC LSI


SEE Measurement of SDRAMs

December, 1999
sdrams_1999.pdf

Overview
The purpose of this test was to determine the SEE characteristics of various SDRAMs under heavy ion radiation. The cross section of each device, for both Single Event Upsets (SEU) and Single Event Latch-up (SEL), as a function of ion Linear Energy Transfer (LET) was the primary goal. The SEU and SEL threshold of these devices was also determined.   The devices tested were 256Mbit SDRAMs.  All devices are organized in a 8Mx4x8 configuration.
  • Hitachi
  • IBM
  • Hyundai


SEE and TID Extension Testing of the Xilinx XQR18V04 4Mbit Radiation Hardened Configuration PROM

Carl Carmichael1 , Joe Fabula1, Candice Yui2, and Gary Swift2

1Xilinx Inc., San Jose, CA
2Jet Propulsion Laboratory, California Institute of Technology, Pasadena, CA

Presented at the 2002 MAPLD International Conference, Laurel, MD, September 10-12, 2002.

p21_carmichael_p.pdf (Paper)
p21_carmichael_s.ppt (Poster)

Abstract
The XQR18V04 was evaluated for single event upset rates using proton and heavy ions. The PROM was demonstrated to be immune to latch-up, as well as to static upset in the flash memory cells, to an LET > 125 MeV/mg/cm 2 (effective). The PROM was also tested in a dynamic mode, which revealed three distinct error modes: Read Bit Errors, Address Errors, and a Single Event Functional Interrupt (SEFI) which affected the data output drivers. Saturation cross-sections, and onset thresholds, for these error modes were measured at the heavy ion facility at Texas A&M University, and the proton facility at UC Davis. Additional testing was performed at UC Davis and the Cobalt 60 source at McClellan Air Force Base to examine the effect to TID life as a function of power biasing. The PROM demonstrated a 100% improvement in total TID life with an 84% percent decrease in device usage.


Backside Device Irradiation for Single Event Upset Tests of Advanced Devices

Gary M. Swift
JPL/California Institute of Technology

Thirteenth Biennial Single Effects Symposium
Manhattan Beach, CA, April, 2002

swift_thin_seesymp02.ppt

Tales from the Cave

  • Texas A&M Wishlist:
    • Upstream Degrader (for uniformity)
    • LET and/or Energy Spectrum Detector
  • Backside thinning is NOT so easy
    • Yield Problems
    • Need Long-Range Ions
  • Backside Irradiation Requires Careful LET Assignments

Examples include SDRAMs and the Power PC (May 7, 2002)

HitachiEEProms Total dose test report on Hitachi 128k x 8 EEPROMs - packaged by Austin Semiconductor and Space Electronics, Inc.
Austin_EEPROM_TID1 June 30, 1998: Radiation Evaluation of the Austin Semiconductor 1 MEG EEPROM, August 5, 1997.  (.pdf 491 kbytes)
EEPROM_PPM-96-003.pdf Total dose test report (PPM-96-003) on 58C001 EEPROM.   The die are from Hitachi and are packaged by Austin. (.pdf 1.2 Mbytes).
28C256_PPM-95-187.pdf Total dose test report (PPM-95-187) on 28C256 EEPROM.   The die are packaged by SEI. (.pdf 636 kbytes).
280C010_PPM-95-182.pdf Total dose test report (PPM-95-182) on 28C010 EEPROM.   The Hitachi die are packaged by SEI. (.pdf 1.0 Mbytes).
28C256_PPM-95-175.pdf Total dose test report (PPM-95-175) on 28C256 SEEQ EEPROM. (.pdf 663 kbytes).
28C256_PPM-95-169.pdf Total dose test report (PPM-95-169) on 28C256 SEEQ EEPROM. (.pdf 320 kbytes).
28C256_PPM-95-141.pdf Total dose test report (PPM-95-141) on 28C256 SEEQ EEPROM. (.pdf 280 kbytes).
28C256_PPM-95-147.pdf Total dose test report (PPM-95-147) on 28C256 SEEQ EEPROM. (.pdf 1.0 Mbytes).
28C256_PPM-91-610.pdf Total dose test report (PPM-91-610) on 28C256 SEEQ EEPROM. (.pdf 600 kbytes).
28C256_PPM-93-052.pdf Total dose test report (PPM-93-052) on 28C256 SEEQ EEPROM. (.pdf 364 kbytes).
HM1-6617_PPM-91-065.pdf Total dose test report (PPM-91-065) on HM1-6617/883 2Kx8 PROM. (.pdf 367 kbytes)
RAMTRON_TID_NOBias.PDF Total dose test report on unbiased Ramtron FRAMs. (.pdf 833 kbytes).
FRAM1_TID.pdf Quick-look TID test on serial ferro-electric RAMs (FRAM).   FM24C16-P/D80851R and FM25160-P/BR3090.  Icc quiescient currents were <=   ~ 20 mA at 50 krads (Si) when the parts were removed. (.pdf 30 kbytes).
Serial FRAM_BNL1198.htm Two models of  Ramtron serial ferro-electric RAMs (FRAM) were subject to a heavy ion single event latchup (SEL) test at Brookhaven National Laboratory, November, 1998.  One unit each of the FM24C16 and the FM25160 were tested with no no latchup observed up to an LET of 74 MeV-cm2/mg.

TID_1608_Hoy10x_ResearchFab.pdf
FRAM_FM1680_TID1.htm
FM1608S (from R&D fab) were total dose tested to 60 krad(Si).
Device leakage currents were low but the parts were non-functional.
FM1608S (from R&D fab) total dose summary report, April 1999.
LM4kx9_SEE_BNL0798.htm Heavy Ion SEE Test of the Lockheed-Martin 4kx9 FIFO.   BNL, July, 1998.
FRM1608_Fujitsu_BNL0499.htm Heavy Ion SEE Test of Ramtron FM1608/Fujitsu.  BNL, April, 1999.
FM1608_BNL0999.htm Heavy Ion SEE Test of Ramtron FM1608/Fujitsu.  BNL, September, 1999.  All devices latched.
IntelFlash_TID.pdf Total Dose Test of the Intel Flash DA28F016SV.
FM1608_BNL1000.htm Heavy Ion test of FM1608 at BNL. (October 24, 2000).

Seu and Sel Response of the Westinghouse 64K E2PROM, Analog Devices AD7876 12-Bit Adc, and the Intel 82527 Serial Communications Controller

Sexton, F.W.; Hash, G.L.; Connors, M.P.; Murray, J.R.; Schwank, J.R.; Wlnokur, P.S.; Bradley, E.G.
Radiation Effects Data Workshop, 1994 IEEE, pp. 55 -63

Abstract (excerpt)
The Westinghouse SA3823 64K E2PROM radiation-hardened SONOS non-volatile memory exhibited a single-event-upset (SEU) threshold in the read mode of 60 MeV-cm2/mg and 40 MeV-cm2/mg for data latch errors.  The minimum threshold for address latch errors was 35 MeV-cm2/mg.  Hard errors were observed with Kr at VP = 8.5V and with Xe at programming voltages (VP) as low as 7.5 V.  No hard errors were observed with Cu at any angle up to VP=11V.   The system specification of no hard errors for Ar ions or lighter was exceeded.   No single-event latchup (SEL) was observed in these devices for the conditions examined.


Single-Event Upset Test Results for the Xilinx XQ1701L PROM

S. Guertin, G. Swift and D. Nguyen
Jet Propulsion Laboratory
California Institute of Technology
Pasadena, California

1999

Xilinx_1701_Report.pdf

Abstract
A 3.3V serial PROM, used to configure advanced Xilinx FPGAs, was tested for single event effects with heavy ions.  Device latchup was observed with an LET threshold of 55 MeV per mg/cm2 and a saturated cross-section of 10-5 cm2 .  Three types of upsets were measured: (1) address errors, (2) premature end-of-program signals, and (3) functional interrupt.


Single-Event Upset Test Results for the Xilinx R1701L PROM

S. M. Guertin
Radiation Effects Group, Office 514
Jet Propulsion Laboratory
California Institute of Technology
Pasadena, CA

August 24, 2000

XilinxEpiPROM.pdf
EOP_Fail.gif
Address_Fail.gif
Functional_Failure.gif

Summary
Radiation testing of the Xilinx R1701L 3.3V 1-Mb serial PROM took place on September 14 and 15, 1999 at SEE Test Facility, Brookhaven National Lab. The R1701L is a special version of the standard commercial XQ1701L PROM that is fabricated on an epitaxial substrate, 7 µm thick, in order to reduce susceptibility to latchup. The data taken here is similar to that taken for the 1999 NSREC workshop publication on the standard XQ17011 part which uses a bulk substrate [1]. Latchup was not observed in the R1701L part up to LET=120 MeV cm2/mg (sigmaLU<5x10-8cm2 ), indicating that the processing change was successful in improving the latchup hardness of the device. It is important to note, however, that latchup may not be the greatest concern for this device, depending on the application, because the special version of the part is still susceptible to single-event upset effects. Careful consideration must be given when using these devices for space applications to allow for the various single event upset effects and their impact on FPGA devices that are interfaced to the PROM.

Home - NASA Office of Logic Design
Last Revised: April 02, 2005
Digital Engineering Institute
Web Grunt: Richard Katz
NACA Seal