Updated: September 13, 2000.
This test of the SX-A, 0.22 µm, UMCMEC die. This test was run with varying supply voltages. Nominal supply voltages are 5.0/2.5. Worst-case voltages are ±10% for the high voltage I/O supply and ±8% (2.7 VDC) for the core voltage supply. The antifuses are biased by the core supply and test conditions exceeded the 2.7 VDC maximum level, with some runs biased at 3.0 VDC.
The TMRSX32_Pattern logical pattern was used.
The device was packaged in a PQFP208.
Test frequency was 1 MHz.
Here's some schematics showing how the important part of the DUT is designed. TMRSX32_Logic.ppt. SEU's were
not an object of these test runs.
Ions used were 282.5 MeV, Br-81 and 345.0 MeV I-127 ions were used for all runs.
The LET of the Bromine ion was 37.4 MeV-cm2/mg with a range of 36.5 µm.
The LET of the Iodine ion was 59.9 MeV-cm2/mg with a range of 32.8 µm.
For all runs, the beam was normal to the surface of the part.
SEL and antifuse rupture were not detected.
Fluctuations in the current, from an unknown mechanism, were detected. These are visible on the ICC strip charts in the table below.
Loss of control, i.e., JTAG upset, was detected.
Two devices were tested, S/N LAN3701 and S/N LAN3702.
BNL |
Ion |
Fluence |
I/O Voltage (V) |
Core Voltage (V) |
Current Strip |
Loss of Control |
| 29 | Br | 10 | 5.0 |
2.5 | lan3701b1.pdf | YES |
| 30 | Br | 6.0 | 5.0 | 2.5 | lan3701b2.pdf | YES |
| 31 | Br | 10 | 5.0 | 2.8 | lan3701b3.pdf | YES |
| 32 | Br | 10 | 5.0 | 3.0 | lan3701b4.pdf | NO |
| 38 | I | 6.1 | 5.0 | 3.0 | lan3701i1.pdf | YES |
BNL |
Ion |
Fluence |
I/O Voltage (V) |
Core Voltage (V) |
Current Strip |
Loss of Control |
| 33 | Br | 10 | 5.0 |
2.8 | lan3702b1.pdf | YES |
| 34 | Br | 10 | 5.0 | 3.0 | lan3702b2.pdf | NO |
| 35 | I | 5.2 | 5.0 | 2.8 | lan3702i1.pdf | YES |
| 36 | I | 8.7 | 5.0 | 3.0 | lan3702i2.pdf | YES |
| 37 | I | 7.6 | 5.0 | 3.0 | lan3702i3.pdf | NO |
Home
Last Revised: January 09, 2002
Digital Engineering Institute
Web Grunt: Richard Katz