Last month my SCSI slipped off the power-supply on to my x86-MotherBoard and everything stopped -- dead. Since the MOBO is just lying on the table [a protection against the local thieves, who prefer something in a nice box] it was easy to check for physical damage and try to restart. After all attempts at susbstituting a tested PSU failed, I accepted that the MOBO was finished, and immediately turned to my rPi to finish my pending jobs. That's when I realised the rPi *IS* a substitute PC; especially for the tricky job that needed multiple visible Windows/Frames plus coloring capability, via ALO replacing the x86-LEO [Linux-based ETHOberon].... and wily: the magical emacs/ETHO substitute.
Since MOBOs with both IDE & SCSI, are no longer made I was worried about the destroyed MOBO, and also wanted to KNOW, what killed it. Is the SCSI case floating or grounded, so that it hit some 5V pin? While investigating by substituting various old MOBO and PSU, the dead MOBO woke-up-as-if-nothing-had-happened !!
Do I correctly remember reading that a 'poly-fuse' is activated, which needs some time [hours] to recrystalise, for the device to recover? But the MOBO is necessarily old, to have IDE, so is the self-healing fuse technology also old?