On Wed, 20 Aug 1997 15:37:58 EDT k6se@juno.com (Earl W Cunningham)
writes:
>You wrote:
>>Earl, would the model show a difference if you tried placing the
>shunt
>>rod directly opposite the 12" face and also standing off from a leg
>so
>>that it effectively "sees" a taper? I found that placing off the
>leg
>>was the better match. This was evidenced by better 2:1 bandwidth and
>>FS measurements.
>
>=====================
>Interesting finding, Carl. The way I shunt feed my towers is to
>horizontally attach (via SS hose clamps) a 1" OD piece of aluminum
>tubing about 3' long to one of the horizontal rungs of the tower's
>zig-zag bracing. I attach the gamma wire (I use old discarded RG-58
>coax) to the outer end of the tubing. If needed, it's easy to loosen
>the clamps and slide the tubing in or out to adjust the spacing. For
>standoff insulators every 15' or 20' below that, I use 3' lengths of
>1" ID schedule 40 PVC with a 1/4" hole in their outer end through
>which passes the gamma wire. These are similarly attached to the
>tower for easy adjustment. Anyway, it looks like this from above:
>
> |o
> 1" tube -----> | o <------ tower legs
> |o
> |
> |x <------- gamma wire
>
>When you say "shunt feed directly opposite the 12" face of the tower",
>I assume you mean like this:
>
> o
> <------- tower legs
> o o
>
> x <------- gamma wire
>
>When you say "standing off from a leg so that it effectively 'sees' a
>taper", I think you mean like this:
>
> o o
> <------- tower legs
> o
>
> x <------- gamma wire
Both of your assumptions above are correct Earl.
I also started those tests (April-May 1984) with #12 for the gamma. Even
with 20 or so radials the bandwidth was narrow. I attributed this to the
D1/D2 ratios. The original 1949 CQ Magazine studies on the Gamma match
(The first in Ham Literature I believe and I can mail you a copy)
stressed that for efficient performance that the diameter ratios not
exceed 4X. Granted this is 1997 but I have yet to read anything (worth
reading) challenging it.
I finally wound up with .75" CATV cable as the gamma rod with it looking
at the taper.
This was selected since it easily fits in the thru side of standard .75"
PVC T fittings. I used regular .75" PVC Schedule 40 to go from the T to
the tower legs. I rested that horizontal PVC alongside a horizontal cross
brace and just taped it on over the 12" length. For that 100' tower and
a rod of ~ 65' I used 5 supports. I also drilled a hole in each T to
secure the hardline with a 8-32 self tapping screw. Once tuned I never
touched it for 5 years...when I moved up here. Same town, 4 miles
horizontal and 500' vertical.
The hardline never hung perfectly straight and looked a bit ugly but it
sure worked.
I had tried a 4 wire 8" diameter cage but I could never get it stable in
the wind....VSWR and amp loading was always changing. The cage had the
best FS readings but was a mechanical nightmare. The difference between
the cage and the hardline was very small and the difference between the
hardline and the #12 was about 1 dB as measured 2.5 miles away using
professional equipment.
I cant say if all of the loss was the wire or the Omega matching
network...probably a bit of both.
Keep the info flowing Earl. It is not too long before snow in the Rust
Belt and I suspect a lot of guys are planning new shunt feds for the
coming season.
73....Carl KM1H
>When I have modeled shunt feeds with a triangular structure, I've
>always place the gamma wire as in my first illustration above, with
>the 1" tubing. I shall try the other two scenarios and let you know
>the results. Of course, when using a circular "wire" to emulate the
>triangular tower, this doesn't make any difference.
>
>BTW, preliminary results of comparing a 12" triangle structure in
>EZNEC vs. a circular "wire" show that a 10.352" OD "wire" is very
>close to the same as a 12" triangle. This was not with a shunt feed,
>but merely base-feeding the equivalent of a 13-section (126' 9-1/2")
>Rohn 25 tower and comparing it with a 126' 9-1/2" round "wire". I
>used #8 "wires" to model the tower. When I changed all of these
>"wires" to 1" OD, I expected a large change, but the delta was hardly
>perceptible, which encourages me.
>
>Anyway that structure was 1/4-wave resonant at 1850.83 kHz, with an
>input Z of 35.95 + j0 ohms. A round "wire" 10.352" OD was resonant at
>the same frequency with an input Z of 35.96 + j0 ohms.
>
>===============
>
>>25G is actually 12.5" on a face and the legs are 1.25". There are 8
>>zig-zag braces per 10' regular section. Measured overlap is 3.5" per
>>section.
>
>===========
>Thanks for the 12.5" info. I'll use it instead of 12".
>===========
>
>>Top sections may be 8' which is the flat top version for use with a
>>thrust bearing.Part # 25AG4.
>>
>>There are numerous tapered top sections with the 25AG2 and 25AG3 both
>
>>having a 2.25" top tube, one extends 12" above the apex of the top
>>rails and the other 18". The Rohn catalog does not give the total
>>length of these two but they are under 10'. They are the most popular
>
>>since they accept a 2" mast. Other tapered models have smaller tubes.
>
>===================
>No one has come up with any info on exact lengths of such top sections
>yet, so I'll continue to model Rohn 25 with regular 10' (9' 8-1/2")
>sections (using 10.783" OD "wires" if the same ratio applies for a
>12.5" face).
>===================
>
>>Then you can start over again on 45G Earl!
>
>===========
>Yes, I will, Carl.
>===========
>
>>Thanks for your effort.
>>73...Carl KM1H
--
FAQ on WWW: http://www.contesting.com/topband.html
Submissions: topband@contesting.com
Administrative requests: topband-REQUEST@contesting.com
Problems: owner-topband@contesting.com
|