[R-390] Grounds

[2002tii]

Dave wrote:

>So,.... my question is: if we have additional grounds for our
>equipment, are we safer or more in danger if we do not connect them to
>the service entrance ground. Perhaps this is beyond the scope of this
>reflector but it seems like we have some engineers lurking in the
>background and I would really appreciate the science of this anomaly
>when talking about grounds to the entire group.

Because lightning can draw tens of thousands of amps of current 
(albeit for a very brief time), it can develop substantial voltages 
across any resistance in its current path.  So, for lightning 
protection, everything that you count on to be at "ground" potential 
should be connected with very low resistance paths.  Then, the 
difference between the potential of a "ground" that is in the 
lightning's current path will be close to the potential of all other 
"grounds," and the risk of getting a shock between the two "ground" 
points should be low even during a lightning strike.  Note that 
lightning current is often conducted laterally through the soil, so 
"ground" stakes separated by some distance can be at different 
potentials during a lightning strike even if the lightning current is 
flowing in neither of them, due to the resistance of the soil itself.

Otherwise -- that is, at all other times besides during a lightning 
strike -- there should be very little current flowing into any 
"ground" (or through the ground past one "ground" point and 
another).  Therefore, you should be able to count on the various 
"grounds" being at close enough potentials not to cause a risk of 
shock if you touch more than one at a time.  If you get a shock when 
touching two different "grounds," something is wrong with at least one of them.

RF grounding is another, separate issue.  In the RF case, it is not 
only the resistance of the path to earth ground that matters, but the 
inductance as well.  If the path is any longer than 1/10 wavelength 
or so at the frequency of interest, it will have significant 
inductive reactance and, therefore, you can develop significant 
voltage down the "ground" wire even if it has low resistance.  You 
can minimize this by making the ground conductor out of wide strap or 
braid rather than a round wire, but only to a certain extent.

So:  Best practice is to establish a robust "station ground."  Bond 
all equipment securely together with 2" or 3" wide copper strap, then 
connect this to a good, solid earth ground (my suggestion is a 12 
foot length of 1" copper pipe driven into the ground) with the 
shortest possible length of copper strap.  If it is unavoidable that 
this length will be more than 20 feet, use wider strap (6") or 
multiple straps.  If your station ground is near the service entrance 
ground, great.  If not, connect the two ground stakes (at ground 
level, or buried) with the shortest possible length of copper 
strap.  Use the widest strap you can find.

Best regards,

Don