Ron wrote:
OK that is one case where your need two failures to cause a hot chassis.
Now for Doofus to get electrocuted he will also have to have one hand
on this deaf radio and one hand on the radio that has a metal chassis
that is plugged into a socket with a good ground.

He won't know the radio is "deaf" until he trys to turn it on... and you
know Doofus, he broke the on/off knob years ago, and he always turns the
bare aluminum shaft while he is standing barefoot on the cement floor.

The problem you
could also have a similar problem even if the neutral was wired
correctly (not tied to ground). If the ground wire broke in the cord or
inside the radio and then a capacitor or a transformer had an internal
short to the chassis or there was a resistor from hot to chassis ground
then the chassis would become hot. The so called safety ground is not a
100% sure thing in the case of a failure.

Nothing is ever 100% perfectly safe. But you do try and improve the odds
as much as you can comfortably afford. Adding a safety wire system gives
you much more bang for the buck in terms of safety than the simple addition
of 33% more wire would appear to offer.

Because safety grounds are as
prone to fail as anything a third external wire tying all metal cabinet
radios together is the only safety measure if one has a table full of
these old AC powered beasts.

Yes it is not a good policy to tie the ground and neutral together. I
just felt that a good explanation of why not to do so was required
instead of just a blanket statement stating "The safety ground must
NEVER be connected to the neutral at the load !NEVER! "

The problem here is I could write for the next week describing all of the
failure modes I know about that can and do happen with power distribution
grounds, and grounding. But I really don't have that kind of time. It
is far easier for me to tell you that doing such a thing is unsafe, and
that you should NEVER connect the safety ground to the neutral at the load.

If I tell you that E = I * R are you going to make me prove it? Or can I
just tell you from a position of assumed authority that this is a true
relationship? You can go and find a book that will tell you E = I*R, and
likewise, you can find a book, or do a google search, to find out why you
should NEVER connect the safety ground to the neutral at the load.

-Chuck