Hi,
I'm going to tackle you with a hard one.
I'm preparing a lab in hardware security in which the students needs to do a power analysis on an AES module.
In order to do it, I use an implementation of the AES module on the PL, but in order to do the power analysis I need to sense the current the core uses. Moreover, I need to sense it over a small resistor (1 Ohm or less) in order to sense the changes in the rounds of the AES module.
I looked in the schematics of the Zedboard and saw that the core gets a 1V supply from a power supply with no resistor on the line. The only thing on the line is an inductor.
The same power supply (the MAX15021ATI+) also supplies a 1.5V voltage to other functions of the zed.
So in order to measure the current, I disassembled the inducter (named L5) and thus disconnected the 1V supply line, and instead connected an external power supply, with a small resistor on the path (the zedboard got 1V after the resistor).
That actually worked, and the zed worked great. But as I left it on for over an hour, it stopped working, and I concluded that the MAX15021ATI+ power supply got fried (all the other voltages was ok except the 1.5V). I guess the reason is that because I disconnected it, it worked harder to supply the voltage until it died (although the external supply was connected to its FB loop).
So the next thing I decided to do (because I already practically ruined the zed), is to disconnect the second inductor (L4), and supply the 1.5V externally as well. But when I disassembled L4, its contact (the one thats connects to the core) got shorted to the ground! (I measured a resistance of only 1 Ohm, while in other, good zed, it is few K Ohms).
And so I come to you,
Do you have any better offer to measure the current that goes to the core without having the zed fried after an hour?
I had another idea, and that is to disassemble L5, and instead solder an inductor and a resistor in series, but by that it means the power supply needs to work harder to supply the 1V.
Thank you in advance for your help,
and sorry for the long post!
David.