PROGRAMMABLE TECHNOLOGIES WEB SITE

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

LAN48xx - RT54SX32S

D/C 0049 – L/C T25JSP03A

Total Dose Experiment

Test and DUT breakdown

LAN4801 – kamikaze

LAN4802 - kamikaze, "4STRINGS" pattern. [all others older TD pattern]

LAN4803 – kamikaze with "mountJJ" - Spare global clock input not terminated.

LAN4804, LAN4805 – production:

• "parafunctested"
• irradiated to ~100 krad @10krad/day
• "parafunctested"

LAN4806, LAN4807 – production-turned-kamikaze:

• "parafunctested"
• irradiated to ~65 krad @24krad/day
• "parafunctested"
• irradiated again (by accident!) for ~90 krad until functional failure

All devices while irradiated are subjected to basic functional and parametric (propagation delay) test approximately once per hour

DUT Patterns

All devices, except for LAN4802 used the TDSX32CQ256_DUT pattern.   LAN4802 used the TD_4Strings_SX32_DUT   pattern.

• Safe power-up/power-down of the I/O's could not be verfied.   The verification effort failed.
• Some of the parts had a DC current on  V_CCI of over 10 mA, until power was applied to V_CCA.  These parts were not used for testing of LAN4804 through LAN4807; "cherry-picked" parts were used, with the V_CCI currents of a few mA, prior to the application of V_CCA.

• LAN4804 and LAN4805 lost functionality as a result of power-up/power-down sequence after irradiation, with I_CCA on both DUTs exceeding 500 mA. After more than 24 hours of "annealing" and repeated applications of clock functionality was restored to both DUTs. I_CCA values were also reduced to significantly lower values. Even after basic functionality retuned the DUTs still continued to exhibit irregular power-up/power-down behaviour, i.e. significantly varying I_CCA values for every power cycling sequence.
• LAN4806 and LAN4807 were also affected by power-up/power-down sequence at the end of first irradiation period. I_CCA increased significantly on both DUTs, however, functionality was not lost.
• MIL-STD-883 Method 1019.5 does not allow for any loss of functionality.  Only parametric failures can be "annealed."

LAN4801_sc.pdf – strip chart

LAN4801_bw.pdf – best/worst case I_CCA

LAN4803_sc1.pdf – strip chart (vs. time)

LAN4803_sc2.pdf – strip chart (vs. dose)

LAN4803_sc3.pdf – strip chart (zoomed in)

LAN4803_bw1.pdf – best/worst case (I_CCA)

LAN4803_bw2.pdf – best/worst case (I_CCI)

LAN4803_pd.pdf – prop delay

LAN4804_05_sc.pdf – strip chart

LAN4804_05_ft.pdf – functional

LAN4804_05_bw1.pdf – best/worst case (I_CCI)

LAN4804_05_bw2.pdf – best/worst case (I_CCA)

LAN4804_05_pd.pdf – prop delay

LAN4806_07_sc1.pdf – strip chart

LAN4806_07_sc2.pdf – zoomed in

LAN4806_07_ft.pdf – functional

LAN4806_07_bw1.pdf – best/worst case (I_CCI)

LAN4806_07_bw2.pdf – best/worst case (I_CCA)

LAN4806_07_bw3.pdf – best/worst case (I_CCA) zoomed in

LAN4806_07_pd.pdf – prop delay

LAN4802 Charts

LAN4802_03_Vcci_Current.pdf

LAN4802_03_Array_Current_vs_Time.pdf

LAN4802_03_Array_Current_vs_rad.pdf

LAN4802_03_Array_Current_vs_rad _zoomed.pdf

LAN4802_Icci_per_state.pdf

LAN4802_Icca_per_state.pdf

LAN4802_Output_Voltage.pdf

LAN4802_Output_Voltage_zoomed.pdf

Part #	LAN4804	LAN4805	LAN4806	LAN4807
pre-rad	1.59	1.56	1.55	1.56
post-rad	1.56	1.56	1.53	1.50

Startup Transients

Because of the parts failures, we were unable to obtain startup transient data.

Home
Last Revised: January 09, 2002
Digital Engineering Institute
Web Grunt: Richard Katz