I hope your chip is still fine. I don't have a lot of background on your specific problem. However a couple of years ago I read through the the guide lines for designing a carrier board for the PicoZed. I seem to remember that there was a very specific sequence for applying all the different power rails to the PicoZed. Did you ensure that your carrier board respects the power sequencing requirements?
Yes I have a regulator for the VCCIO that only gets enabled once the power good signal comes from the module so power up should be respected, the rest of the boot up sequence is handled by the SOM. As I stated I was able to power up, it was detected, and I programed the board, the FPGA DONE light would light up correctly once booted. It is definitely dead as it gets too hot to touch very quickly.
"Is it possible for the chip to overheat and kill itself?, I don't have any heatsink or fan on the chip.."
I have used 7z020 in several designs, some rather large ones, without a heatsink and never had any issues so far. In vivado you can see the power report to get an estimate on your design's power consumption. My designs have had up to 2W of estimated power consumption as per the power report, and while the chip got hot it didn't get damaged. If your design is small and yet the actual power consumption is very high, then perhaps something else went wrong.
I was trying to stay hopeful but you might be correct in assuming that the device was damaged. So strange that things seemed to go south during the software load. I am not really an AMD/Xilinx guy so I can't really pin point what could have gone wrong here.
Thanks for the info, I didn't know about this in Vivado, It says 1.8W for my design, When I first plugged in the board I was drawing around 600 mA at 5V so about 3W.
After burning out the chip its drawing around 1.5A (although that might be the limit of the power supply). I will have to check the power draw of the other peripherals on the 5V rail to see if 3W is suspicious.
I do have one of my connectors to the board bodged on so I will double check the traces I have cut and rerouted to make sure I didn't miss anything, the only thing that could be an issue are the VCCO and 5V rail connections as everything else is generic gpio for the PL or the programming lines and I know those work for sure.
Here is a quick answer for Zynq FPGA power rails. The power rail table was from TI design for Zyqn 7015(tidua66.pdf):
Zynq 7015 is similar to 7020 except the high speed bus at PS side.
One tip to check whether the FPGA is dead or not if you cannot use JTAG: check FPGA temperature after powering it on. If FPGA temperature is blow 40~50C after power on and the PS does have proper clock input/power rails are fine, FPGA most likely is dead.