less hum: switching power supply?

[Heiko van der Linden]

Hi All,

When I had problems with 7805 or 7905 circuits and the voltage would be too
high the problems were caused by oscillations.
A discussion about this problem is given in:
http://www.national.com/an/AN/AN-1148.pdf

This discusses the use of linear regulators and focusses on LDO (low drop
out) regulators.
The item to look for is on 5: ESR and stability.
When you use a capacitor on the output that has a too high ESR you can get
oscillations.
See figure 17 for this.
For the output cap tantalum capacitors are recommended.
As for the input cap, this is only necessary if the regulator is far away
from the main (unregulated) power supply.
This cap also suppresses oscillations and should have good HF
characteristics. So use a disc cap for this one
if possible. By the way I always place this cap since it's not very
expensive to do so.

As for the spike story:
I remember a story by Bob Pease about these regulators. I think his talk was
about the LM317 but I believe this
is related enough to the 7805 to make the story applicable to the 7805.
His story was that the thermal shutdown in the regulator makes the chip
oscillate at a frequency of something like 100Hz
(don't kill me if this figure is a bit wrong). This feature was left in the
design because it allows the regulator to start "heavy"
loads like big caps that would eat too much current at startup. By making
the output oscillate these kinds of circuits will still
be working as normal and if the regulator would have a different protection
it would refuse to load this kind of circuit.
So maybe this will explain the spikes. I don't know what these spikes look
like maybe someone can show an oscillogram or
something like that.

I hope this info helps,
Heiko