The way I read it is that you put in the feed impedance that a similar-
length plain-ol' dipole in the same situation would have. In a Yagi this
can be significantly lower than 72 ohms, which is the whole reason you're
considering a folded dipole.
Does the author have contact information? Perhaps you could ask the
source?
In what case would I alter the 72 ohms?
Mike
In what case would I alter the 72 ohms?
Mike
I've seen older Yagis that use folded dipoles, but the newer ones seem to
lean more toward gamma or T matches. You get more latitude for adjusting
impedance, in the gamma case you get a kinda sorta good match to coax
without a balun, and if you're employing plumber's delight construction
it's no more difficult to fabricate than a folded dipole.
So why a folded dipole, pray tell?
> Hi All,
> I'm having a disagreement about the use of this folded dipole calculator
>impedance calculator.
>http://www.k7mem.150m.com/Electronic_Notebook/antennas/folded_dipole.html
> We are using it to calculate the impedance of folded dipoles to drive a
>yagi.
Excuse me, but what does such simple dipole calculator have to do with
the calculations for a Yagi feed element ?
Even assuming a single element feed element antenna, playing with the
tube diameter (upper and lower bars) or the number of bars in a folded
dipole will radically change the feed impedance.
Paul
I think it all started with the MFJ-1800 that uses a folded dipole, but
there
are many yagis that use a folded dipole as the driven element.
But, I think (now) I understand enough to know why you ask the question.
I think your point is the impedance of dipole feed on a yagi is highly
modified
by the reflector and directors.
> Even assuming a single element feed element antenna, playing with the
> tube diameter (upper and lower bars) or the number of bars in a folded
> dipole will radically change the feed impedance.
>
> Paul
>
I excerpted this from my latest email to my friend I'm having the
disagreement with. (Might
not even be a disagreement, it could be a miscommunication.)
Excerpt;
"The online calculator takes the (Impedance of a dipole x Ratio). This
gives you the impedance
after folding the dipole.
We still have no idea what the impedance of a dipole is when surrounded by
the reflector and
director. We do know how to change the transformation ratio once we know
what the impedance
of dipole would be in that surrounding."
Mike
The nominal impedance of a dipole in free space is approximately 72
ohms. If it is part of a system its impedance will be influenced by
that system causing it to change from nominal.
Folding the dipole allows it to behave as a transformer ! Thus the
nominal dipole impedance can be altered to a value that can more easily
be matched to the feed line.
Commonly the impedance of a dipole that is part of a system. ie "Yagi"
falls to a much lower value. Under these conditions the folded dipole
can be used to raise the impedance seen at the feed point to a more
usable value.
Someone mentioned "Gama match". Whilst symmetrical gamma matches can be
used, single ended is popular because it is inherently unbalanced and
can be easily matched to a co-axial feedline.
--
Best Regards:
Baron.
Due to the transformer effect of folding it.
--
Best Regards:
Baron.
Most of the bandwidth enhancement on a folded dipole comes from the
fact that it's a "fatter" radiator rather than because it's folded.
Section Bii) here:
http://www.karinya.net/g3txq/folded_dipole/
Steve G3TXQ
Of course, the 'transformer effect' is unlikely to increase the bandwidth by
itself, and when the common type of half-wave-line balun (and 4:1 impedance
transformer) is used to connect the dipole to co-axial cable the resonant
nature of this additional half wavelength is likely to impair the bandwidth
to some extent (by introducing additional reactance at frequencies for which
its length isn't a true half wavelength).
Certainly the additional fatness of the folded-dipole structure must
contribute to improved bandwidth by reducing the slopes of the terminal
resistance and reactance, but the treatment given in the link provided
above - suggesting that the susceptance added at the drive point by the
'parasitic' parallel wire or rod, joined to the driven element at its ends,
has insignificant effect - is a rather broad generalisation which may not be
true for some geometries (i.e. length/thickness ratios of the wires/rods and
their spacing). Generally, it's dangerous to generalise!
There's quite a lot going on in a folded dipole. Some further insight (on
allied, but not the same, topic) is offered in
http://www.physics.princeton.edu/~mcdonald/examples/cutantenna.pdf.
Chris