I put this thing on EZNEC and with the information you have provided so
far. I can see no correlation to anything you have measured. The only
way I could get anything close was to assume a bad piece of coax. This
antenna should be a good match to 50 ohm coax at around 1.9 MHz (43 ohms
at 1.89 MHz). Two to 1 SWR bandwidth was 1.78 to 2.01 MHz. I calculate
the equivalent ground loss resistance to be 8 ohms for your 40 radials.
We are missing some information, maybe something you think is not
significant.
Your first post said the tower was 124 ft 9 inches tall, fed at the 4 ft
3 in point. The last post said the antenna was 120.5 ft. Assuming you
haven't changed the tower, the antenna is actually 124 ft 9 in. That
4.25 ft at the bottom is part of the antenna. Feeding it at this point
is OK, and it raises the feedpoint impedance by 3 to 4 ohms versus
feeding it at the bottom. Please describe how the feedline is connected
at this point. The center should be connected to the top part of the
antenna and shield to the bottom section at the 4.25 ft point, and the
shield of the coax should also be connected at the bottom at the point
where the radials are connected. This won't change the impedance any,
but it will remove some of the RF from the coax shield. If the
connection at the bottom is not made, the shield of the coax will also
act as part of the antenna, but in this case the contribution should be
small and I don't think it will account for the measurements you are
getting.
The information K1TTT gave you about the feedline transforming the
impedance is correct, but this antenna should be close to 50 ohms for
the configuration you have, and so there should be little transformation
of the resonant frequency (assuming good coax), certainly not enough to
account for the measurements you are getting. The fact that you get
something very different means that the measured data now has little
meaning because there WILL be frequency translation, apparently from
some other problem. More information might discover it. Got any
pictures? Possibilities: Bad coax, improper connection of the coax,
other close objects also resonant on 160, faulty SWR meter, an extremely
strong local AM station. If you will provide the information as to how
the guy wires are broken up, I will add that to the simulation to look
for any problems.
Jerry, K4SAV
AB5MM wrote:
>Gentlemen,
>
>I would like to extend a hardy "Thanks" to each of you that responded to
>my plea for help with matching the 120'-6" 160m Rohn 25 Vertical. It's
>still not working like I think it should, so let me tell you what we've
>managed to accomplish to date.
>
>1. We now have 40ea. 125' long ground radials installed. Another 20 will
>be added this weekend for a total of 60. I think I'm going to say that's
>enough for now. NOTE: Going from 22 to 40 ground radials made little
>difference!
>
>2. Both the MFJ-249 and MFJ-259B are virtually useless at these
>frequencies. Too much rf from the BC band. This alone would have saved
>us hours in trying various matching networks, not to mention the number
>of "Pepcid Complete" ingested.
>
>3. In sweeping the tower from 1.600 MHz to 2.500 MHz, we found it's
>natural (no matching other than the 75' of LDF4-50A) resonance at 1.788
>MHz. The lowest swr observed at this frequency was 2.6/1. At 1.800 it is
>3.0/1 and slowly rising to 4.0/1 at 2.000 Mhz. Conventional wisdom tells
>us that the thing is too long (tall). This still brings up the
>question, "How can this be the case when the vertical is only 120 ft.-6
>in. tall to begin with?" What are we missing here guys? The center of
>the band (1.900 mhz) is what we've been shooting for all along.
>
>4. A high pass "T" network brings the match down flat, but it's Q is
>high enough to only realize a practical band width of ~33kc.
>
>5. A low pass "L" network works as well and exhibits a little larger
>band width of ~60 kc. Our goal is ~100 kc wide... is this reasonable?
>
>We really don't know what to do or try next, if anything. Should we be
>satisfied with what we have?
>
>Press on, nothing is as persistent as persistence itself.
>
>Tnx & 73,
>Steve AB5MM
>Lloyd K5ZO
>_______________________________________________
>
>
>
>_______________________________________________
>TowerTalk mailing list
>TowerTalk@contesting.com
>http://lists.contesting.com/mailman/listinfo/towertalk
>
>
>
_______________________________________________
_______________________________________________
TowerTalk mailing list
TowerTalk@contesting.com
http://lists.contesting.com/mailman/listinfo/towertalk
|