PSU switch on moment

[Haible Juergen]

It can be a problem indeed.
I didn't fry anything, but I got things latched up.

	>Off: well there is energy storage in the electrolytics, we don't
really
	>know which supply goes down first. Same situation as ON, the
heavier
	>load goes down first.

I had made a later addition to my FS-1 frequency shifter, and I had quite
some over dimensioned RC filtering for certain parts. It would sometimes
lock up on pwer on. I went a brute force empiric way to cure the problem
(rather than doing a clean analysis): I put in a relais that would power on
the newly added circuit half a second after the main circuit. Works, but
not much to learn from it.

	>4. Is a tracking regulator recommended?

Told this story a few times already, so just a short version now:

JH-3 (my big Modular) has +/-18V pre-regulation in the PSU (3 pairs
of 7818 / 7918), and I intended to use 7815 / 7915 or 78L15 / 79L15
for on-board regulation on each module. This would *often* latch up
at power on, even with the usual quadruple of backward diodes
(they gladly prevented damage).
This problem was completely cured when I switched to RC4195
tracking regulators for the on-board stuff. They are affordable, (DM 5,-),
small (DIP 8), and the can carry approx. 100mA if you have an
18V pre-regulation.

JH.