[RFI] Earth isolation

[Cortland Richmond]

Jumping in late here... the green wire, AC safety ground must be able to carry enough current to open a fuse or circuit breaker if a hot-to-chassis AC power short develops, and to keep the chassis-to-ground voltage low enough to be safe meanwhile. It has no other purpose and cannot be depended upon to carry off high-frequency transients or RF as its impedance is uncontrolled at much above powerline frequencies.  

It *is* acceptable to place an RF impedance in series with the safety ground if -- and only if -- that does not adversely affect this protective function. If I remember correctly (I am too lazy to look it up) the standard for an equipment "green wire" ground is that it be able to carry 30 amps for five minutes. 

Transient and filter grounds must offer a low impedance to the transients and frequencies they are designed to discharge or bypass.  For this reason, protectors and filters often occupy a separate panel having its own, low-impedance ground.  However, by the time we get THIS serious, we are probably not talking home AC power sockets any more; people who have paid attention to grounding and transient/lighhtning protection often have indusctrial-strength installations. 

Cortland
KA5S

-----Original Message-----
From: "EDWARDS, EDDIE J" <eedwards at oppd.com>
Sent: Sep 9, 2004 8:21 AM
To: Jim Brown <jim at audiosystemsgroup.com>
Cc: RFI Reflector <rfi at contesting.com>
Subject: RE: [RFI] Earth isolation

You're correct Jim-- wrong word choice.  
"Characteristic Impedance" has a specific meaning in electrical theory
related to transmission lines.  My goof.

That's why peer review for papers is so important!  :-)

On the last part, I disagree however.  You've simply restated what I
said except in one case.  You have not disconnected anything when adding
a choke (and you actually state this by noting the DC connection); you
only change the impedance which is frequency dependant: Z=SQ
RT[(R*R)+(X*X)] & X=[-wfL*(1/wfC)]/[(WfL)-(1/wfC)].  Note the "f" in
each term.  Now, at certain frequencies, other circuits that were higher
in impedance before are now lower in impedance by comparison.  That can
lead to "ZAP!"

73, de ed -K0iL

-----Original Message-----
From: rfi- On Behalf Of Jim Brown

On Thu, 9 Sep 2004 08:48:49 -0500, EDWARDS, EDDIE J wrote:
>If you place the choke on the AC mains ground you WILL create a change
>of that circuit's characteristic impedance 

Leave out the word "characteristic" -- it's not a transmission line.
Just plain old 
series impedance. Now the sentence is right on! 

>and it will no longer
>function as the AC safety ground whenever you need it.  

Yes

>You also violate
>the single point grounding principle since that circuit now sees other
>grounds as lower impedances (like thru your equipment circuit boards or
>through humans in some cases).  

Again, not the right words, but the right problem. The single point
principle is 
violated because the choke disconnects the power system from the single 
point. And the system is unsafe because the choke blocks the required 
connection at the frequency of potential power faults (spikes, lightning
hits, 
etc.). You DO see the connection if you measure it with a DC ohmmeter
(or 
audio frequency bridge), but lightning and power spikes aren't
predominantly 
DC or audio, most of their energy is at much higher frequencies. 


Jim Brown  K9YC