is there any difference in a typical regulator or it is just the same?? and how doed it work?
I think I know what you're asking about, but I'm not sure so if none of this makes sense simply disregard. A floating
supply is somewhat of a separate problem if I remember correctly.
From National's LM317 page:
Since the regulator is “floating” and sees only the input-to-output differential voltage, supplies of several hundred volts can be regulated as long as the maximum input to output differential is not exceeded, i.e., avoid short-circuiting the output.
Floating in this sense simply means that the regulator's ground pin is not at the same voltage as the load current return path (usually the rest of the circuit's ground). Take as an example a 50VDC unregulated supply with a few volts of ripple and regulating it to 40VDC. You can use a standard LM317 (Max input of 37VDC) if you 'float' the ground up by putting a (lets say) 30V zener diode in series to the circuit ground. This tricks the ground pin into thinking ground is 30V higher, so now the supply looks like 20VDC with a few volts of ripple going down to 10VDC. The only problem you have now is ensuring that you don't short the supply and show the regulator enough voltage too cook it.
NEC (which was bought out by someone else) has
an excellent short guide on three terminal regulators, and there's a couple of examples on page 18 of high input/high output voltage regulation. There's a lot of information in this guide on various protection and high current designs as well, and it might be a good place to start.
Hope that helps.