Undervoltaging a voltage regulator - what happens?

[sfr]

Oops.  Despite all my best colour-coding and separate cables and what have you, I managed to swap the power plugs for my 18V and 9V multi units.  the 9V seems to be fine, and should be, as I believe everything in there used parts rated for high enough voltages.  Only thing I'm not sure about is the Scrambler, which was an older build, but it seems to be okay.

The 18V unit seems to be behaving fine (although the A/DA Flange seems to be acting up, this was occuring before this happened) but the thing I'm curious about - with all these voltage regulators, can I damage them by underpowering them?  I know you're supposed to have a certain voltage difference of some amount between the input and desired output voltages, but I believe that is just to make sure that there's enough voltage left after the regulator does it's thing?  Is running 9V into 12 or 15V regulator bad for it? 

[Sir H C]

Undervoltage should be no problem.  The regulator will either go into drop-out where it acts more like a resistor from input to output or shuts down (rarely do they do this though).  If you think about it, they have to work from when power is applied, and this means going from 0 volts to whatever.  That can happen pretty slowly, so it has to handle all voltages to the maximum.

[PerroGrande]

I agree with Sir H C.  In general, for most of the 78xx-style regulators, you need to have input about 2-3v over Vout to get good regulation.  Below that, you tend to get Vout somewhere around Vin with no regulation.  In general, I'd worry more about over-voltage on low-current/low-voltage devices.  (I worry about undervoltage more on inductive motors and the like).

[runmikeyrun]

i cooked a 78LO5 in my 24vdc delay with a 12vdc ac supply.     plugs were the same size.

[PerroGrande]

The 78L05 (the "L" is important) is only a 100ma version of the regulator.  With the lower voltage into what is typically a current-thirsty class of device (delays, loopers, and the like tend to draw a lot of current) probably pushed it over the limit that its protection circuit could handle.