Mark says;
>Skin depth in copper at 50MHz is 2.5micrometers.
Not in my books (Terman, and the ITT Reference Data for Radio Engineers)
Skin depth for copper is 6.61/sq.rt. f where f is in Hertz and the depth is
in cm, or 2.6/sq.rt f, where f is in MHz and depth in mil (0.001 inch). I get a
skin depth of .000368 inches or 9.34 micrometres.
A 6mm tube has a periphery of 18.8mm. A 1mm wire has a periphery of 3.14mm, so 6
of them should have a total periphery of 18.8mm - no difference. The skin depth
in the case of the wire is still about 1/100 of the wire diameter. It's normally
considered that the current depth is 5 skin depths - that's for 99.993% of the
current to be in that depth of copper, so theoretically, in either case, a tube
of 1.84 mil ( 'thou' in English) or 47 micrometres wall thickness should be
adequate.
I suspect that the answer is comes from Terman; '....as the frequency becomes
higher the benefits disappear. This is because the irregularities of stranding,
and the capacitance between the strands, cause a failure to realize the ideal
condition at very high frequencies.' Also, there's proximity effect to
consider, which reduces the benefits of Litz or bunched conductors.
73
Peter G3RZP
--
FAQ on WWW: http://www.contesting.com/FAQ/amps
Submissions: amps@contesting.com
Administrative requests: amps-REQUEST@contesting.com
Problems: owner-amps@contesting.com
|