More on Litz Wire
Sat, 25 Apr 1998 13:24:04 -0400
On the orgin of the 'Goliath' antenna design, from an excellent article
"The Monster Antennas", Communications Quarterly, Spring 1996:
"Late in Fall of 1941 the German Government designed and built... a
station near Kalbe. ...the effeciency, which reached 90 percent at
60kHz.....driven by a 1MW transmitter. ... the (entire) Goliath
installation was dismantled by the Russians at the end of WW II
and transported to the Soviet Union where it was reconstructed
near Gorki. "
Several VLF antennas are described in 'VLF Radio Engineering' by
A.D. Watt. The book was written in 1967 and I believe the NSS
antenna was rebuilt to 'goliath' configuration in 1969.
Litz wire for the VLF inductors is covered
also in Watt's book and in "Antenna Engineering Handbook" by
Richard Johnson (1993). The details of Litz on the
antenna tuning variometer are described and says copper tubing
can be used if the losses can be tolerated and connections welded.
Details of the aluminum clad and combination aluminum/steel
antenna conductors are also described. Litz seems to be essential
for the very high power VLF rigs.
The example on the New England Electric Wire page shows
an example calculation at 100Khz where the AC resistance is
reduced from 51.3 ohms to 2.79 ohms/1000ft using Litz.
From: Bob Bruhns <email@example.com>
To: Andre Kesteloot <firstname.lastname@example.org>
Cc: Robert E. Seastrom <email@example.com>;
firstname.lastname@example.org <email@example.com>; firstname.lastname@example.org
Date: 24 April 1998 02:04
Subject: Re: More on Litz Wire
The NSS antenna was reported to be a "modified Goliath" system
copied from (gasp) a Russian system. There were a few odd things
about it... For example, the brute-force parallel splitting of
the RF output to three feedlines of different lengths, fanning
from the top of the Frankensteinian tuner building to the 300,
600 and 900 foot levels of the 1200 foot tower, was strange. It
seems to me that at 21.4 KHz it would resemble an end loaded,
crooked half-bowtie monopole leaning over toward the south.
Maybe I ought to model it on MiniNEC...
I think the use of so many large bundles of such finely divided
Litz wire as feedline to a galvanized steel tower is
questionable. The magnetic properties of steel cause a severe
skin effect, and the electrical resistance of steel and zinc are
high compared to copper and silver... Seems to me that if
efficiency was that much of an issue, heavy copper lines should
also have been run along each tower leg, and that the top hat,
which was probably also galvanized steel, should have been
heavily copper-clad. But possibly the loss resistance was useful
for bandwidth purposes, and simple galvanization was less
expensive. Could the total resistance budget of the design have
been used up in the tower and tophat, requiring the use of a very
low-loss feedline? I wonder how this antenna's overall
efficiency (840 000W in, 30 000W radiated) and bandwidth (very
careful tuning required to accommodate 50 BPS, 50 Hz shift FSK)
compared to the efficiencies and bandwidths of other VLF antenna
systems around the world.
Bob Bruhns, WA3WDR
The transmit tuning arrangement was also strange, with so much
back and forth interactive tuning going on. But with 16% power
loss, maybe it was better than a broadband iron-core transformer.
Andre Kesteloot wrote:
> One of my own --possibly blasé-- theories is that, this
> being a huge Government contract, the fellow who wrote the
> original specs (against which the various manufacturers had to
> bid) did not know all about Litz wire, and decided that Litz
> was better than no-Litz (based on the sound bureaucratic
> approach that "one never knows"), and that none of the bidders
> dared to question the Government document's wisdom. And
> anyway, it increased the cost of the contract... ;-)
> Also, the equipment, as I recall, was built circa 1937. hence
> the specs were probably written i 1935 or '36. The quote I
> recently posted from the RCA book (and reprinted later in
> Langford-Smith) is dated 1941. Possibly they learned something
> between 1935 and 1941...
> André N4ICK