[TowerTalk] Back of desk grounding buss

[Jim Lux]

On 3/22/14 9:04 AM, Gary Schafer wrote:
>
>
>> -----Original Message-----
>> From: TowerTalk [mailto:towertalk-bounces at contesting.com] On Behalf Of
>> Jim Lux
>>
>> Wide flat strips have low AC *resistance* because of skin effect, but
>> the inductance isn't much different from a round wire.
>>
>> AC resistance is *very important* in applications like RF grounds for
>> commercial broadcast antennas, because resistance = heat = lost power =
>> lost money, so they use flat ribbon.
>
> But lightning energy peaks around 1 MHz where low AC resistance is
> important.

The problem is that the inductance reactance is orders of magnitude 
greater than the AC resistance, so when it comes to transient voltages, 
it's L you worry about not R.

Compare the AC resistance of a meter of AWG 10 wire with the inductive 
reactance..

AWG10 wire has a ac resistance at 1 MHz of 0.03 ohms
A one meter long conductor has an inductance of about 1 uH, and at 1 
MHz, that's 6.28 ohms.

So the transient voltage you are seeing is more determined by the 6 
ohms/meter from L than the 0.03 ohms/meter of R. Lowering R by going to 
a wide flat ribbon doesn't help much, say you drive the resistance down 
to 0.01 ohms/meter.  The 6.28 still dominates.





>
>> The inductance of two parallel inductors is:
>> (L+M)/2
>> where L is the inductance of a single inductor and M is the mutual
>> inductance.
>
> Yes flat strap has mutual inductance across its width but isn't mutual
> inductance considerably lower with a flat strap than separate parallel
> wires.
>

Not really.  If the current is flowing evenly along the strip it makes 
no difference whether it is separate wires or one continuous flat wire. 
  There's no current flow "cross ways" across the strip, so it makes no 
difference if there's insulator in the way.


there is a "proximity effect" in close wound coils which is akin to skin 
effect, where the current is carried in the exposed surface of the turn, 
as opposed to the sides of the turns. It's the same thing as skin 
effect: the current gets pushed to the outside.

But it's still tightly coupled by the magnetic fields, so the inductance 
doesn't change much.

If you go from a 1x1 cm bar to a 0.05 x 20 cm strap, the inductance is 
reduced by about 30%.  The AC resistance, though, drops by 90% for the 
same change.

So if you're in a steady state kind of scenario (AM broadcast 
transmitter), where the inductance is "tuned out", that 90% change in 
loss is worth it.


If you get up into microwave frequencies, the AC resistance starts to 
get larger, because skin depth goes as 1/sqrt(f)

go up to 100 MHz, and now that AWG 10 wire a meter long has an AC 
resistance of 0.3 ohms.  The L is still 1 uH, so the Z is now 628 Ohms


You have to go way down around a few hundred Hz to where the resistance 
starts to dominate over the L.