1 / 104

Radiation Effects on Programmable Technologies: Real-life Data and Testing

Radiation Effects on Programmable Technologies: Real-life Data and Testing. Presented by Ken LaBel, Rich Katz, and Igor Kleyner. Total Ionizing Dose (TID). Generic TID Testing. Initial Measurement. x Krad Irradiation. Measurements can be made either in-situ (during irradiation)

eron
Download Presentation

Radiation Effects on Programmable Technologies: Real-life Data and Testing

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Radiation Effects on Programmable Technologies:Real-life Data and Testing Presented by Ken LaBel, Rich Katz, and Igor Kleyner

  2. Total Ionizing Dose (TID)

  3. Generic TID Testing Initial Measurement x Krad Irradiation Measurements can be made either in-situ (during irradiation) or after biased irradiations Interim Measurement Dose rates are a function of technology Annealing Typical sources are Co-60 or X-ray (proton is sometimes used) Final Measurement Presented by Ken LaBel, Igor Kleyner, and Rich Katz at MAPLD 2002 Sep 9 2002

  4. Lab Instrument 1 Lab Instrument 2 Lab Instrument 3 Lab Instrument 4 GSFC Second Generation TID Data Collection System - Overview WWW TID Chamber Building TID Chamber DUT “K-labs” GPIB GSFC network “rk” Server “Stupid” PC Test Control PC Presented by Ken LaBel, Igor Kleyner, and Rich Katz at MAPLD 2002 Sep 9 2002

  5. GSFC Radiation Effects Facility (REF) Shielding for test circuits Co-60 Source Bias board Presented by Ken LaBel, Igor Kleyner, and Rich Katz at MAPLD 2002 Sep 9 2002

  6. Sample TID Dynamic BiasBoard w/ Lead Shield PbAl boxes are used to reduce backscatter in a room source Presented by Ken LaBel, Igor Kleyner, and Rich Katz at MAPLD 2002 Sep 9 2002

  7. Sample TID Static Bias Board Under static bias in-situ monitoring can be performed on parameters such as supply current\ or used for step irradiations Presented by Ken LaBel, Igor Kleyner, and Rich Katz at MAPLD 2002 Sep 9 2002

  8. A1425A/MEC TID TEST D/C 9819 - UCJ014X 6 kRads (Si) / Day NASA/GSFC August 14, 1998 0 5 10 15 20 25 krads (Si) Typical TID Run Device Information 15 S/N LAN202 S/N LAN203 S/N LAN204 S/N LAN205 Parameter degrading (no anneal) 10 Measured Parameter (mA) I CC 5 0 Dose deposited in material Presented by Ken LaBel, Igor Kleyner, and Rich Katz at MAPLD 2002 Sep 9 2002

  9. TID Degradation • Parameters can degrade in many ways • Gracefully (slow increase) • Previous slide is an example • Abruptly (fast increase) • Other • Functional Degradation • Sometimes occurs prior to parametric Presented by Ken LaBel, Igor Kleyner, and Rich Katz at MAPLD 2002 Sep 9 2002

  10. QL3025-2 ESPQ208R TID TEST D/C 9913CA 18.6 krad (Si) / Day NASA/GSFC April 14, 1999 0 5 10 15 20 25 30 35 40 krad (Si) TID Run – Abrupt Runaway 1000 S/N WONE001 I CC5.0V S/N WONE001 I CC3.3V 800 Power Supply Limit = 800 mA 600 (mA) CC 400 Delta I 200 Notes: 1. DUT in Pb-Al box per 1019.5 2. Experimental Device 0 Presented by Ken LaBel, Igor Kleyner, and Rich Katz at MAPLD 2002 Sep 9 2002

  11. TID Run - Extended Presented by Ken LaBel, Igor Kleyner, and Rich Katz at MAPLD 2002 Sep 9 2002

  12. RT54SX16/MEC TID TEST D/C 9849 - P05 12.95 rad (Si) / minute NASA/GSFC March 31, 1999 0 10 20 30 40 50 60 70 80 90 100 krads (Si) In Situ Functional Testing 50 S/N LAN1200 Functional Failure 40 Note: Some cases showed failure at less than 20 mA current with current jumps of 6-8 mA. 30 (mA) CC 20 Notes: Delta I 1. Devices in Pb/Al Box per 1019.5 2. Devices Packaged in CQFP256 10 3. 5V Bias Current Change was Neglible 4. Have seen failures with smaller jumps (a few mA). 0 Presented by Ken LaBel, Igor Kleyner, and Rich Katz at MAPLD 2002 Sep 9 2002

  13. In Situ Functional Testing • Traditional approach for antifuse-based FPGAs relies exclusively on bias current monitoring • Earlier families of Actel FPGAs (Act1, Act2, Act3) usually show ICC increasing exponentially to relatively high levels (>100 mA) before any irregularities (i.e. high current spikes) are observed; such “spikes” are usually a reliable indication of a functional failure • Initial testing of devices from SX family showed presence of “suspicious” small jumps in ICC current at relatively low current levels ( 10 to 50 mA) Presented by Ken LaBel, Igor Kleyner, and Rich Katz at MAPLD 2002 Sep 9 2002

  14. In Situ Functional Testing Example 1 • Actel antifuse-based FPGA device (RT54SX16) • Traditional test configuration (ICC strip-chart) is expanded to include in situ short functional test • Signal Generator is utilized to send stimulus signal(s) to the DUT • Output Voltage(s) are measured by Multimeter • The sequence is executed automatically at programmed time intervals and/or manually as desired by experimenters; functional test results are posted on the web site for convenient monitoring along with ICC strip-chart data • The test, and other subsequent tests of SX family devices, showed that the first small jump in the ICC level usually occurs immediately after the functional failure of a DUT Presented by Ken LaBel, Igor Kleyner, and Rich Katz at MAPLD 2002 Sep 9 2002

  15. Actel SX Family Device in situ Functional Failure Functional failure occurred at ~15mA; followed by immediate short current spike Presented by Ken LaBel, Igor Kleyner, and Rich Katz at MAPLD 2002 Sep 9 2002

  16. In Situ Functional Testing Example 2 • Two Actel SX family devices (same lot, dose rate) irradiated simultaneously. • S/N LAN3403 configured traditionally (static, basic functional test once per hour) - “Static” configuration • S/N LAN3404 is clocked continuously at 1kHz -“Dynamic” configuration • Dynamic device fails at ~8% lower total accumulated dose level than static device “Dynamic” device failure “Static” device failure Presented by Ken LaBel, Igor Kleyner, and Rich Katz at MAPLD 2002 Sep 9 2002

  17. Parametric In Situ Testing Example 1 • Flash-based FPGA device (Actel A500K050) • Initial testing involved standard elements • ICC monitoring • In situ basic functional test • Post-irradiation parametric testing showed significant increase in Propagation Delay (tPD) • Consequently, in situ measurement of tPD was added to the test configuration • Signal Generator supplies input pulse for DUT • Waveforms are captured by Digitizing Oscilloscope and tPD is measured Presented by Ken LaBel, Igor Kleyner, and Rich Katz at MAPLD 2002 Sep 9 2002

  18. In Situ measurement of Propagation Delay Real-time Digitized Input and Output Waveforms Before irradiation : tPD = 135ns After accumulating 90 krad : tPD = 260ns Presented by Ken LaBel, Igor Kleyner, and Rich Katz at MAPLD 2002 Sep 9 2002

  19. A14100A/MEC TID TEST - Room Temp Anneal D/C 9849 - UCL064 Lot Split Post 20 krad(Si) Exposure @ 2.45 rad (Si) / Minute NASA/GSFC Anneal Started August 3, 1999 150 S/N LAN1901/RB4 - UCL064 S/N LAN1902/RB5 - UCL064 125 100 75 (mA) I CC 50 25 0 0 10 20 30 40 50 60 Annealing Time (Days) Effect of Annealing on TID Results S/N LAN1903/RBA4 - UCL064A S/N LAN1904/RBA5 - UCL064A Presented by Ken LaBel, Igor Kleyner, and Rich Katz at MAPLD 2002 Sep 9 2002

  20. TID Failure Modes in Programmable Technologies • How a programmable device degrades is a function of • Technology • Design • Parameters and failure modes vary • Flash may be different than anti-fuse • The following charts provide examples of real data on sample devices • Acknowledgements to Fabula and Wang Presented by Ken LaBel, Igor Kleyner, and Rich Katz at MAPLD 2002 Sep 9 2002

  21. Submicron FPGA TID Tolerance m m 0.35 m to 0.6 m 50 40 30 (mA) 20 I CC 10 0 10 20 30 40 50 60 70 80 90 100 kRads (Si) General Trend:TID Capability vs. Feature Size m RT54SX16 Proto, 0.6 m, 3.3V, MEC m A54SX16 Proto, 0.35 m, 3.3V, CSM m A42MX09, 0.45 m, 5.0V, CSM m QL3025, 0.35 m, 3.3V, TSMC m XQR4000XL Proto, 0.35 m, 3.3V, 60 kRads (Si) m RH54SX16 Proto, 0.6 m, 3.3V, > 200 kRads (Si) A42MX09 QL3025 A54SX16 RT54SX16 Presented by Ken LaBel, Igor Kleyner, and Rich Katz at MAPLD 2002 Sep 9 2002

  22. ICC Transient - Peak Current A1280A/MEC I Transient Test CC D/C 9826 - U1H466 - LAN400 1 krad(Si) / Hour NASA/GSFC March 15, 1999 800 S/N LAN403 700 S/N LAN404 600 500 400 Note: Power supply rise time ~ 1.2 msec Peak Current (mA) 300 200 100 0 4k 6k 100C Activity (1 Day/tick except for irradiation) Presented by Ken LaBel, Igor Kleyner, and Rich Katz at MAPLD 2002 Sep 9 2002

  23. ICC Transient - Charge A1280A/MEC I Transient Test CC D/C 9826 - U1H466 - LAN400 1 krad(Si) / Hour NASA/GSFC March 15, 1999 100 S/N LAN403 90 S/N LAN404 80 70 60 C) m 50 40 Note: Power supply rise time ~ 1.2 msec 30 Charge ( 20 10 0 4k 6k 100C Activity (1 Day/tick except for irradiation) Presented by Ken LaBel, Igor Kleyner, and Rich Katz at MAPLD 2002 Sep 9 2002

  24. ICC Transient - Trigger Voltage A1280A/MEC I Transient Test CC D/C 9826 - U1H466 - LAN400 1 krad(Si) / Hour NASA/GSFC March 15, 1999 4.0 S/N LAN403 3.9 S/N LAN404 3.8 3.7 3.6 3.5 Supply Voltage @ Transient Start (volts) 3.4 3.3 3.2 3.1 Note: Power supply rise time ~ 1.2 msec 3.0 4k 6k 100C Activity (1 Day/tick except for irradiation) Presented by Ken LaBel, Igor Kleyner, and Rich Katz at MAPLD 2002 Sep 9 2002

  25. ICC Transient - Pulse Width A1280A/MEC I Transient Test CC D/C 9826 - U1H466 - LAN400 1 krad(Si) / Hour NASA/GSFC March 15, 1999 160 S/N LAN403 S/N LAN404 140 120 Transient Width ( 100 80 sec) m Notes: 60 1. Transient width measurement is full width at half-max. 2. Power supply rise time ~ 1.2 msec. 40 4k 6k 100C Activity (1 Day/tick except for irradiation) Presented by Ken LaBel, Igor Kleyner, and Rich Katz at MAPLD 2002 Sep 9 2002

  26. ICC Transient - “New” Part A1280A, 1 V/DIV, 100 mA/DIV Presented by Ken LaBel, Igor Kleyner, and Rich Katz at MAPLD 2002 Sep 9 2002

  27. ICC Transient - 6 krad (Si) A1280A, 1 V/DIV, 100 mA/DIV Presented by Ken LaBel, Igor Kleyner, and Rich Katz at MAPLD 2002 Sep 9 2002

  28. ICC Transient - Post Anneal A1280A, 1 V/DIV, 100 mA/DIV Presented by Ken LaBel, Igor Kleyner, and Rich Katz at MAPLD 2002 Sep 9 2002

  29. FRAM Memory Functionality Loss During Total Dose Test In situ static current measurements of two serial and one parallel FRAM device types. This initial study showed that Rohm (serial) and Ramtron research fab (parallel) devices could withstand moderate doses without significant leakage currents. Post irradiation testing of the FM1608 showed that all devices catastrophically failed. Annealling did not help. In situ functional tests or a step irradiation method are needed for determination of the functional limit. The base CMOS process is not the limiting factor for the FM1608. Presented by Ken LaBel, Igor Kleyner, and Rich Katz at MAPLD 2002 Sep 9 2002

  30. Process Mods - 0.25 m Presented by Ken LaBel, Igor Kleyner, and Rich Katz at MAPLD 2002 Sep 9 2002

  31. Post TID Testing Div by 2 Output Input Clock Failure is intermittent. Scope was set to trigger on glitch. Presented by Ken LaBel, Igor Kleyner, and Rich Katz at MAPLD 2002 Sep 9 2002

  32. Post TID Testing Div by 2 Output Input Clock Note: Some devices will fail only in localized portions. High fault coverage is needed. Presented by Ken LaBel, Igor Kleyner, and Rich Katz at MAPLD 2002 Sep 9 2002

  33. SRAM-based FPGATID Testing Results • 0.18µM Technology • TID evaluation performed on XQVR300E • parametric shifts were negligible to >80krads • no leakage at trench isolation to >80krads • device fully functional at end of dose • slow increase in Tilo and other timings • 100°C anneal fully restored device • room temp anneal showed no rebound Presented by Ken LaBel, Igor Kleyner, and Rich Katz at MAPLD 2002 Sep 9 2002

  34. Total Ionizing Dose Effect on 0.18µM Technology for SRAM-based FPGA Presented by Ken LaBel, Igor Kleyner, and Rich Katz at MAPLD 2002 Sep 9 2002

  35. PROM Response to TID Presented by Ken LaBel, Igor Kleyner, and Rich Katz at MAPLD 2002 Sep 9 2002

  36. Flash-based FPGATotal Ionizing Dose Test Results Presented by Ken LaBel, Igor Kleyner, and Rich Katz at MAPLD 2002 Sep 9 2002

  37. VCC Ionizing Radiation Control Gate ONO Tunnel Oxide Floating Gate Source Drain Data Path Total Dose Effects on FLASH Switch • Ionizing radiation discharge the floating gate • Increase ON-state NMOS transistor resistance, increase RC delay in the data path • Increase OFF-state NMOS sub-threshold leakage, increase ICC Presented by Ken LaBel, Igor Kleyner, and Rich Katz at MAPLD 2002 Sep 9 2002

  38. T1 T2 T3 T = T1 · T2 · T3 Total Dose Effects on FLASH Switch Tunnel Oxide Radiation-Induced Traps • Radiation-Induced Leakage Current (RILC) in tunnel oxide • Similar to Stress-Induced Leakage Current (SILC) cause discharge of the floating gate • Charge retention cause long term reliability issue Floating Gate Silicon Presented by Ken LaBel, Igor Kleyner, and Rich Katz at MAPLD 2002 Sep 9 2002

  39. Protection from TIDExample: Shielding of 46 MeV Protons Shielding can help mitigate TID effects, but beware of dose enhancement Presented by Ken LaBel, Igor Kleyner, and Rich Katz at MAPLD 2002 Sep 9 2002

  40. Single Event Effects (SEE)Testing at the Facilities

  41. Heavy Ion Testing at Brookhaven National Laboratories (BNL) Beam Diopter Vacuum Chamber Test Stage Test Equipment Beam Control Computer (can be interlocked with test computer) Presented by Ken LaBel, Igor Kleyner, and Rich Katz at MAPLD 2002 Sep 9 2002

  42. BNL - A little bit closer Assembly will lower, slide under vacuum chamber, raise up, then seal. Connector feedthrus are available - cable length to user is ~6 feet Presented by Ken LaBel, Igor Kleyner, and Rich Katz at MAPLD 2002 Sep 9 2002

  43. Proton Testing at UCDavis Crocker Nuclear Laboratory (CNL) User room at CNL: >50feet of cable needed from test area. Includes beam control computer. Video monitor of test area available. User end of beam line at CNL (test area) This wall has been removed, enabling larger test fixtures Presented by Ken LaBel, Igor Kleyner, and Rich Katz at MAPLD 2002 Sep 9 2002

  44. Proton Test Hints Beware of stray neutrons on your test equipment. Here, Borax is shown on top of a power suppiy. We have seen failures in test equipment from stray neutrons. GSFC’s 4-axis remote-controlled test stage. Allows devices under test to be re-positioned without entering the chamber Presented by Ken LaBel, Igor Kleyner, and Rich Katz at MAPLD 2002 Sep 9 2002

  45. Actel SX Test Controllers UTMC PAL RS-422 Interface Quicklogic Chip Express Example SEE DUT Card Presented by Ken LaBel, Igor Kleyner, and Rich Katz at MAPLD 2002 Sep 9 2002

  46. Cross Section versus LET Curve N = X-section x F Courtesy of Aerospace Corp for Berkeley Facility Presented by Ken LaBel, Igor Kleyner, and Rich Katz at MAPLD 2002 Sep 9 2002

  47. Single Event Effects (SEE)Category 1: Simple SEUs in Programmable Technologies In this section, we will cover data related to programmable technologies and examples of SEU hardening methods (circuit design)

  48. Act 2 SEU Flip-Flop Data An SEU in a flip-flop is perhaps the most traditional form of SEU Presented by Ken LaBel, Igor Kleyner, and Rich Katz at MAPLD 2002 Sep 9 2002

  49. SEU X-Section vs. Feature SizeSubmicron Technology Presented by Ken LaBel, Igor Kleyner, and Rich Katz at MAPLD 2002 Sep 9 2002

  50. SEU in PALs - Large X-Section Presented by Ken LaBel, Igor Kleyner, and Rich Katz at MAPLD 2002 Sep 9 2002

More Related