Thicknet/10base5 Test Segment: The Cable is In!

[Peter Coghlan]

On 2018-06-28 at 13:20:25 -07:00, Chuck Guzis via cctalk wrote:

>
> So, on a TIA/EIA/RS-232C DB-25 connector, what's the official position
> on pin 1?  The standards calls it PGND = Protective ground and most
> reference seem to indicate that this is chassis/earth ground at both
> ends of a cable.
>

In my opinion, this pin 1 protective ground thing is one of the most poorly
thought out, useless, confusing ideas that this industry ever came up with
(in hardware anyway, there are probably better contenders on the software
side of the house...).

Nobody has ever been able to convince me that it is capable of providing any
protection to anyone or anything.  I have never had given to me a remotely
plausable explaination of any mechanism that it could be in any way useful
in any circumstance.

A "protective ground" conductor can be an appropriate and useful addition to
a set of conductors which are supplying electrical power to something.  There
are standards defined all over the world (except maybe in the Isle of Man in
the 1980s) that specify how to select an appropriate type of conductor to use
for this purpose, how to connect it and so on.  Anyone who knows a bit of
electrical theory and has a bit of common sense can figure out how they
provide protection and see that it is useful.

A "protective ground" conductor is also a useful thing to use to connect a
lightning conductor to a grounding system.

A "protective ground" conductor is not a useful thing to route with signal
conductors when there is no way of knowing what current and voltage it might
be expected to carry over what duty cycle and therefore no way of knowing how
to select a suitable conductor.  Even if a suitable conductor size could be
determined based on the capacity of the power supply circuits to the equipment
at each end of the serial link in question (on the assumption that they will
be responsible for delivering any fault current), I'm willing to bet that the
size of conductor involved will in many cases not be in any way appropriate
for connecting to a pin of a D connector.

Years ago, when making up serial cables, I dedided to not ignore pin 1 like I
usually did and this time I would figure out how to deal with it properly.
After a asking a lot of questions, looking for references and considering
theories of how it might provide protection, I came to the conclusion that I
had been right all along and the very best thing I could do with pin 1 was to
leave it unconnected.

Regards,
Peter Coghlan.