There were several subtle variations in the P250/CC line: 4.5-V alkaline or 3.6-V lithium batteries and presence/absence of a CUDA switch being chief amongst them. If none of yours has a CUDA switch (although the solder-pads for one may be evident near the rear right of the MLB) it only obliges you to remove all power to the MLB for 10-20min to reset the MLB, which is as safe a way of resetting as anything that the CUDA can achieve.
As for the harder question: whaffor the need to soak a CC on the mains? before it will start up from the keyboard, the plain truth is that I dunno. It works in the vast majority of cases in CCs that have been off the mains for years. (One that I acquired had not been used for ten years or so, going by the dates of System files, but it responded to the SOTMFAD treatment, and still goes three years later.) Only a complete circuit diagram could support a guess at the cause of the behaviour, but it is common to all the pseudo-soft power Macs as far forward as the 5500. It definitely has nothing to do with the CPU, because all three of my 5500s have G3 upgrades. I could lean to the guess at large filter caps in the power supply needing to repair/be repaired, but caps that have grown shorts through the dielectric usually need a massive instantaneous discharge to burn off the dendrites, and it is still a risky 'repair'. The signal that such power supplies as the Macs have will need palliative care before use is the absence of the klunk when mains-power is switched on. What klunks I know not, and will not have time to explore this side of Armageddon, I fear.
de
As for the harder question: whaffor the need to soak a CC on the mains? before it will start up from the keyboard, the plain truth is that I dunno. It works in the vast majority of cases in CCs that have been off the mains for years. (One that I acquired had not been used for ten years or so, going by the dates of System files, but it responded to the SOTMFAD treatment, and still goes three years later.) Only a complete circuit diagram could support a guess at the cause of the behaviour, but it is common to all the pseudo-soft power Macs as far forward as the 5500. It definitely has nothing to do with the CPU, because all three of my 5500s have G3 upgrades. I could lean to the guess at large filter caps in the power supply needing to repair/be repaired, but caps that have grown shorts through the dielectric usually need a massive instantaneous discharge to burn off the dendrites, and it is still a risky 'repair'. The signal that such power supplies as the Macs have will need palliative care before use is the absence of the klunk when mains-power is switched on. What klunks I know not, and will not have time to explore this side of Armageddon, I fear.
de