SBAK043 April   2026 DAC39RF10-SP , DAC39RFS10-SP

 

  1.   1
  2.   2
  3.   Trademarks
  4. 1Introduction
  5. 2Single-Event Effects
  6. 3Device and Test Board Information
  7. 4Irradiation Facility and Setup
  8. 5Test Setup and Procedures
  9. 6Single-Event Latch-Up (SEL) Results
  10. 7Single-Event Functional Interrupt (SEFI) Results
    1. 7.1 Converter Performance and Digital (DUC + JESD204C Link) Hardness
    2. 7.2 Configuration Register Hardness
    3. 7.3 SPI Programming During Irradiation
  11. 8SEU Results
    1. 8.1 JESD204C Link Monitoring Results
    2. 8.2 Digital Up-Converter and NCO Upset Recovery
    3. 8.3 Estimating Upset Rates in Unprotected Data Paths
    4. 8.4 Event Rate Calculations
    5. 8.5 Summary of Radiation Hardness
  12. 9References
  13.   A Appendix: Recommendations for Hi-Rel Systems
    1.     A.1 Summary of Rad-Hard Design Features
    2.     A.2 SPI Programming
    3.     A.3 JESD204C Reliability
    4.     A.4 Equalizer Usage in Radiation Environments
    5.     A.5 NCO Reliability
    6.     A.6 NCO Frequency and Phase Correction (Strategy #1)
    7.     A.7 NCO Frequency Correction (Strategy #2)
    8.     A.8 NCO Self-Sync/Self-Coherent Mode (Strategy #3)

8.2 Digital Up-Converter and NCO Upset Recovery

During SEE testing, we employed “Strategy #1” for NCO upset recovery (see Appendix – NCO Reliability). With this strategy, we were unable to detect any NCO wandering which places NCO tones in the ‘wrong place’ or potentially ‘mute’ tones as seen in the FFT monitoring software (see Figure 8-1).

To prove the robustness of our SEU-recovery design methodology, we also experimented with the NCO running without using any hi-rel strategy. During these test runs, NCO wandering was observable and detectable.

Note: This data is presented to highlight the importance of following our hi-rel recommendations to enable automatic self-recovery from potential upsets.

Run With No NCO Upset Recovery Strategy Not Employed Figure 8-1 Run With No NCO Upset Recovery Strategy Not Employed
Note: In a test run where NCO upset recovery strategy was not employed, this capture with the beam on shows an upset on three of the NCOs (these upsets were random and recoverable with the next ion strike). This data is presented to highlight the importance of following our hi-riel recommendations to enable automatic self-recovery from potential upsets.