Message-ID: <3617AC3245C2D1118A840000F805359C01752858@PDW-MERCURY-1>
From: "Talbot Andrew" <ACTALBOT@dera.gov.uk>
To: "'LF Group'" <rsgb_lf_group@blacksheep.org>
Subject: LF: Coil winding conumdrum
Date: Thu, 13 Apr 2000 09:39:30 +0100
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1; format=flowed
Content-Transfer-Encoding: 8bit
Precedence: bulk
Reply-To: rsgb_lf_group@blacksheep.org
Sender: <majordom@post.thorcom.com>

Well, Gamal and Paul between them hit on the solution.  It definitely
appears to be proximity effect causing extra losses for the Litz wound
coil.  I re-measured both at 80kHz and the Litz one came out
significantly better at 42 as opposed to 30 for the single layer one.
The losses caused by proximity effect obviously dominating the tiny
improvement from not very efficient Litz winding with each strand still
4 - 5 skin depths in diameter.
Dielectric loses were insignificant, the former material barely got warm
in a Microwave oven.

G4MD :
>Would not increasing the Q 
>lead to higher circulating currents at resonance, thus greater 
>heating? Unless of course the effective x-sectional area of the 
>conductor was increased in square-law relationship to the increase in 
>current!

Q is only the relationship between loss resistance and reactance.  A
higher Q means lower series R and must mean lower loss.  It all falls
out in the maths :-(

ZL2CA:
>You did not report in your findings if the Litz wire coil ran cooler
>than the single wire coil, but I am presuming it did.

I did not even bother loading up, it was obvious from the Q measurement
it would dissipate more so not worth trying.

--------
I wound a new coil as Gamal suggested.  Using 68mm diameter drain pipe
(I bought a 2m length for the grand sum of Ł2-60 - anyone want some coil
formers :-)    42 turns of 0.8mm tinned wire spaced at 3mm over a length
of 130mm.  Q now twice that for the original single layer coil at 250.
I couldn't use the spacing to conductor diameter ratio of 1.4 to 2.2
that Peter suggested as it was difficult winding the bare wire onto the
smooth former with a spacing much less.

It was very instructive comparing the measured loss resistance from Q
value with that calculated from conductor length, diameter and skin
depth.  For both single layer coils the actual loss resistance was only
about twice that calculated.  Considering what other loss mechanisms are
present, this result is very satisfyingly close.

Now I suppose I'd better actually have a QSO on topband although I've
still got to measure the antenna properties on that band and compare
them with the values measured when it was first built for 73/137.
Anything to avoid going on air !

All the fun is in making things work rather than using them when they do
:-)

Andy  G4JNT


-- 
The Information contained in this E-Mail and any subsequent correspondence
is private and is intended solely for the intended recipient(s).
For those other than the recipient any disclosure, copying, distribution, 
or any action taken or omitted to be taken in reliance on such information is
prohibited and may be unlawful.