[TowerTalk] solid THHN vs jacketed copperweld

[Lux, Jim]

On 9/27/22 4:04 PM, Steve Maki wrote:
> Thanks Jim.
>
> I know this is a scenario of diminishing returns...how many skin 
> depths does one shoot for?

That's an interesting question.  It has to do with the fraction of 
current in the "skin".  If you've got 5 skin depths, then almost all of 
the current is in the copper. The current density goes as 
exp(-d/skindepth) and exp(-5) is 0.007 - so 99.3% is flowing in the top 
5 skin depths.

But as microwave people plating gold over nickel have found, if you're 
at the 1 skin depth, then you're only down to 37% at that level, so a 
fair amount of the current is flowing in whatever is underneath. And if 
"underneath" is ferromagnetic, then the skin depth in *that* is really 
tiny, so all the current is flowing in a tiny area, so the loss is high.

For 2 skin depths, you're down to 86.5%, and 13.5% is in the 
underneath.  That's like 0.6 dB, if you assume all the 13.5 is loss.



Note: this doesn't even get into "round" conductors. As long as skin 
depth << diameter, then you can treat it as a flat plate.






>
> On 9/27/2022 5:31 PM, Lux, Jim wrote:
>> On 9/27/22 2:22 PM, Steve Maki wrote:
>>> I seem to recall ordering Copperweld wire, but it was a long time ago.
>>>
>>> Just a quick note that after replacing the (insulated) #12 THHN with 
>>> the same length insulated #12 "copper clad steel tracer wire", we 
>>> could see no change in resonant frequency, so there seems to be 
>>> enough copper cladding for my purpose at 3.5 MHZ.
>>>
>>> There will only be an inch or so of wire exposed to the air at the 
>>> center insulators, so that will be easy to monitor for corrosion 
>>> over time (such as it is :-) Or I may coat the exposed wire with 
>>> liquid tape.
>>>
>>> 73, Steve K8LX
>>
>>
>> Skin depth in copper at 4 MHz is 1.3 mils, 32 microns.
>>
>>
>> Copperweld
>>
>> https://www.copperweld.com/application/files/4915/3833/2604/Welded_Copper-covered_Steel_CCS_21.pdf 
>>
>>
>>
>> 21% CCS AWG 12 has 0.0808" diamter (2.05mm) with a nominal thickness 
>> of 0.0616 mm ( so about 2 skin depths at 4 MHz)
>>
>>
>> https://www.copperweld.com/application/files/7115/3833/2604/Welded_Copper-covered_Steel_CCS_40.pdf 
>>
>>
>> 40% CCS  has copper that is 0.2052 mm thick ( 6 skin depths)
>>
>>
>>
>>>
>>>
>>>
>>> On 9/21/2022 2:07 PM, Donald Chester wrote:
>>>> Steve Maki lists at oakcom.org Tue Sep 20 15:19:37 EDT 2022 wrote:
>>>>
>>>>> I had sworn off copperweld many years ago after noting that the 
>>>>> copper
>>>>> plating did not last forever.
>>>>
>>>> Was that real copperweld, or was it copper-plated steel with only a 
>>>> minuscule layer of copper deposited on the surface, like the stuff 
>>>> used for electric fence wire?
>>>>
>>>> Real copperweld has a thick copper jacket that is practically 
>>>> indestructible.  I use #10, salvaged from an abandoned railway 
>>>> telegraph line of yesteryear, scored free for the taking many 
>>>> decades ago.  You can beat it with a hammer or leave it buried in 
>>>> the soil (as ground radials) for years and the jacket won't come 
>>>> off unless it's deliberately removed with a file or grinding wheel.
>>>>
>>>> That copper-plated junk begins to rust through in only a couple of 
>>>> years.  The copper is so thin that it probably has negligible 
>>>> benefit  from the skin effect; you are effectively using plain old 
>>>> steel wire.
>>>>
>>>> Don k4kyv
>>> _______________________________________________
>>>
>>>
>>>
>>> _______________________________________________
>>> TowerTalk mailing list
>>> TowerTalk at contesting.com
>>> http://lists.contesting.com/mailman/listinfo/towertalk
>>>
>>
>> _______________________________________________
>>
>>
>>
>> _______________________________________________
>> TowerTalk mailing list
>> TowerTalk at contesting.com
>> http://lists.contesting.com/mailman/listinfo/towertalk
> _______________________________________________
>
>
>
> _______________________________________________
> TowerTalk mailing list
> TowerTalk at contesting.com
> http://lists.contesting.com/mailman/listinfo/towertalk
>