TL;DR: board SI sims show ns-long overshoot on AD9364 CMOS logic inputs beyond the Table 11 abs-max (-0.3 V / VDD_INTERFACE+0.3 V), while the clamp current stays well below the +-10 mA rating. ADALM-PLUTO Rev D shows the same with the same method. Is this acceptable?
Setup: AD9364BBCZ, VDD_INTERFACE 1.8 V, CMOS interface, driven by Zynq-7010 LVCMOS18 (same topology as ADALM-PLUTO Rev D).
Sim (IBIS, PowerSI + SPDSIM, max corner):
- Zynq DRIVE 12 (default): 14 of 23 inputs exceed; worst 2.62 V / -0.83 V; clamp current 1.9 / -3.6 mA.
- Zynq DRIVE 4 SLEW SLOW: 3 of 23 exceed; worst -0.52 V.
- ADALM-PLUTO Rev D, same method: all inputs exceed; worst 2.50 V / -0.74 V.
Questions:
- Is a few-ns overshoot beyond -0.3 V / VDD_INTERFACE+0.3 V acceptable when the input current stays within +-10 mA? Is there any duration rule?
- For the CMOS interface, do you recommend DRIVE 4 SLEW SLOW (or series resistors) on the FPGA side, or is the Pluto default considered safe?
Related: thread 594451 (static levels only). Thanks!