I've designed a board with TI DSP and Altera FPGA, power management, and memory ICs. I put the TPS25944A eFuse on the front end of the board so that everything downstream is protected from hot swap issues when plugging the board into +12V DC.
The eFuse is doing a great job protecting all ICs on the board as the +12V is connected. However, there is nothing to protect the input to the eFuse when connecting it to a hot +12V DC. An oscilloscope shows a +25V transient ringing when the +12V DC is connected to the board. This has destroyed the eFuse IC on all 6 of my prototype boards since the eFuse has an absolute maximum rating of only +20V. The inductance of the +12V DC power cable and the small input capacitance on the input side of the eFuse circuit has formed an L-C circuit that rings when +12V DC is first connected.
I implemented a simple R-C snubber on the input and this has fixed the issue. However, I will need to respin my board to make this a part of the design. I didn't see anything in the datasheet that warned or discussed this issue and I want to make sure this doesn't happen to others. Also I want to verify that an R-C snubber is the best approach for this problem, or should I use a zener diode to clamp the transient below +20V, or is there another better solution that TI recommends?
Thanks,
Paul Barker
(Digital Light Processing Group, TI)