<div style="color:black;font: 10pt arial;">Excellent point Bob! People forget that lightning itself is a high-frequency phenomenon - most of its energy is up around 1 MHz. Further, the vast majority of damage is done not by the strike itself (there's little that can protect from the 10,000 to 150,000 A current in a direct strike) but by the "secondary voltages magnetically induced into power lines and long conductive structures of any kind, including premises power wiring and signal cables. Actually, those induced voltages are the basis for the up to 4 kV surge test waveforms with 1 us rise-time and 50 us exponential decay standardized by IEC. The rise-time is considered typical for spikes delivered by power lines (filtered by premises wiring). Incidentally, a typical AC outlet will internally arc-over somewhere around 5 kV peak.
<div><br>
<div><br>
</div>
<div>As Bob emphasized, the "ground" path for high-speed data I/O protection needs to be effective at UHF frequencies. There can be nano-second rise-time "tertiary" arcs that occur within inches of the equipment - with little cable to slow its rise-time, the spike can be delivered "full-force" to the IC inside. Since many modern versions of these ICs have internal ESD and spike protection - and since these lightning-induced spikes are common-mode, I'd think that a ferrite clamp-on "EMI suppressor" could also be an effective protection layer if applied as close as possible to the I/O port.</div>
<div><br>
</div>
<div><br>
</div>
<div>Bill Whitlock</div>
<div><br>
</div>
<div>Whitlock Consulting</div>
<div><br>
</div>
<div>Ventura, CA</div>
<div><br>
</div>
<div>Office: (805) 755-5018<br>
<br>
<br>
<div style="font-family:arial,helvetica;font-size:10pt;color:black">-----Original Message-----<br>
From: Bob Katz <bobkatz@digido.com><br>
To: proaudio <proaudio@bach.pgm.com><br>
Sent: Tue, Nov 5, 2019 10:07 am<br>
Subject: Re: [ProAudio] AES on the west coast, was Re: Ampex/ProAudio List Dinner<br>
<br>
<div id="yiv5811795502">
<div>
<div>An important aside: The key to lightning protection is that the
protector clamps the high voltage as closely as possible to the
same reference point (possibly called "ground") as the circuit
under protection. An actual earth ground is unnecessary and
frankly, useless.</div>
<div>I have some Furse Ethernet protectors. The key is to connect the
ground strap of the Ethernet protector to the circuit land of the
ethernet circuit. That's as close as possible to the actual
reference voltage ("ground") of the active ethernet chips. I've
soldered the ground strap directly to the metal ring around the
CAT-x female. Works perfectly for lightning. If you run anything
longer than a couple of inches over to a so-called "earth" or
"u-ground" you just won't get any protection.</div>
<div>Some people have screwed that strap to an outlet screw or even to
the chassis of an Ethernet Switch. FORGET ABOUT IT. It's no
protection at all for high voltage lightning spikes. <br clear="none">
</div>
<div>Just proving once again that "earth" is a meaningless concept.
How do you think they keep airplane electronics from burning out
in lightning storms!<br clear="none">
</div>
<div>Bob</div>
</div>
</div>
<div class="yqt3505685994" id="yqt79231"><br clear="none"></div>
</div>
</div>
</div>
</div>