Towertalk
[Top] [All Lists]

Re: [TowerTalk] Trifilar Winding Question

To: "RICHARD SOLOMON" <w1ksz@q.com>
Subject: Re: [TowerTalk] Trifilar Winding Question
From: "Rick Karlquist" <richard@karlquist.com>
Reply-to: richard@karlquist.com
Date: Tue, 19 May 2009 13:07:46 -0700 (PDT)
List-post: <towertalk@contesting.com">mailto:towertalk@contesting.com>
RICHARD SOLOMON wrote:
>
> In putting together the switch box for the K9AY Loop, I decided to read
>> vs "plaited". But one came up I never heard of. He suggested using three
> conductors from a computer ribbon cable and winding them flat on the
> toroid.
>
> Any thoughts on the merit of doing it this way ?
> 73, Dick, W1KSZ

I did a study of this about 10 years ago.  Garden variety ribbon
cable has a characteristic impedance of 150 ohms.  You might possibly
find some with 120 or 100 ohms.  It works very well, if you can
tolerate the 30 gauge wire you usually get.  I wound some 50 ohm
to 450 ohm transformers with this stuff.  Since 150 ohms is the
optimum characteristic impedance in this case, the ribbon cable
transformers were extremely accurate in terms of having 50 ohms
input impedance when terminated with 450 ohms.  The only thing not
to like is the small gauge wire, which might be an issue in
a transmitting situation (not K9AY).  The other issue is that 150
ohms may or may not be optimum for a K9AY loop.  You should model
it to see the effect.  If you want less than 150 ohms, you can use
2 or 3 pairs in parallel to get 75 or 50 ohms.  If you want more,
you can use wire 1 vs wire 3 to get 300 ohms, or wire 1 vs wire 4 to get
450 ohms.  Intermediate wires should left open.

For higher power, instead of ribbon cable, I use speaker wire.  The
impedance of this may be higher or lower than 150 ohms.  So called
Monster cable might be considerable lower, and will handle a lot of
power.  The vinyl dielectric is not great in terms of loss or
temperature capability, but in practice is good enough in most cases.
You can connect two speaker cables in "series" to double the
impedance ("series" as defined in network theory, as opposed to "cascade"
which is end to end, and obviously doesn't increase the impedance).

I did a very careful measurement to determine coupling into the
ferrite due to the pair laying right on top of the ferrite.  Turns
out there is no measurable coupling, so don't worry about it.

One more practical tip: if at all possible use the "rainbow" or
"spectra strip" stuff, instead of the all gray type.  It is way
easier to work with.


Rick N6RK


_______________________________________________



_______________________________________________
TowerTalk mailing list
TowerTalk@contesting.com
http://lists.contesting.com/mailman/listinfo/towertalk

<Prev in Thread] Current Thread [Next in Thread>