I've made two changes to the circuit, one of them rather important.
Goto ABSE & ABPR for the revised schematic & parts list.

As any practitioner of electrical theory will tell you, the current once

established in a conductor likes to keep on flowing. This is very
noticeable in an inductor such as the piece of Hammond iron I used
as a choke in the power supply. If 100 ma of load current is being
delivered to the load, what happens when the standby switch is opened?
I didn't think about things like that 40 years ago when I was building
the original version of this PS.

Aside from burnt contacts in the switch, the stored energy in the choke
magnetic field needs to be dissipated somewhere. No problem when
the L/C switch is set for C filter input during operation. A serious
problem
occurs when set for L filter input operation, since when the switch is
opened there is nowhere for the current to go. As a result a very large
negative going voltage occurs at the switch end of the choke as it is
wired in the diagram. The very fact that the choke insulation has not
yet
failed after more than 40 years operation is an attestation to the
reliability
of Hammond products. This same phenomena is used in your car's
ignition to provide the spark (unless you have compression ignition).

The fix is quite simple, that being a diode & resistor from the loose
end
of the choke to the common lead. Some of the stored energy is dissipated

in the choke resistance while the remainder shows up in the filter caps.

I've included a 10R resistor for my own interest in order to monitor
the transient at switch off.

For the other change I've moved the DC ON-OFF switch to the
transformer HV secondary, as I've also done on some other recent
projects. That makes life easier for the switch since it now has to
interrupt AC rather than DC. Since AC passes thru zero at twice the
line frequency any arc will be self extinguishing (we hope).

Cheers, John Stewart