dansawyeror wrote:
The combination of cable and antenna presents something
other then R = 50 ohms 0 reactance and the the transmission line see
discontinuities. The result is it radiates.

If the currents are balanced, a 50 ohm transmission line seeing
something other than a 50 ohm load does NOT cause it to radiate.
If a 50 ohm unbalanced transmission line sees a 50 ohm balanced
load and common-mode currents flow on the outside of the coax,
it will usually result in radiation from the feedline. Simply
knowing the magnitude of the feedpoint impedance doesn't tell us
anything about feedline radiation.
--
73, Cecil http://www.qsl.net/w5dxp


----== Posted via Newsfeeds.Com - Unlimited-Uncensored-Secure Usenet News==----
http://www.newsfeeds.com The #1 Newsgroup Service in the World! 100,000 Newsgroups
---= East/West-Coast Server Farms - Total Privacy via Encryption =---

Tom Ring wrote:

Cecil Moore wrote:
Why is it definite? What is the loss in 50 ft. of LMR-400 at the
frequency of interest when the SWR is 50/8 = 6.25:1?

Ok, I need teaching here. Why would the loss change? The loss on the
line is forced to what happens at the nominal 50 ohms doesn't it? The
SWR shouldn't be able to change it unless the voltage limits are hit I
would think. I need an explanation of why it wouldn't be so.

The loss specified on transmission line charts are usually matched-line
losses (SWR=1:1). When the SWR is higher than 1:1, additional losses
appear due to the higher SWR. There's a chart in my ARRL Handbook that
gives those additional losses. The matched-line loss for 50 ft. of LMR-400
at 3.5 MHz is about 0.1 dB. Wes calculated the total loss at 0.4 dB but I
think that must have been for 100 ft. of LMR-400 at 3.5 MHz with an
SWR of 6.25:1.

The higher SWR causes additional losses because the maximum current and
maximum voltage is higher for the same power delivered to the load than
for the matched-line case.
--
73, Cecil http://www.qsl.net/w5dxp


----== Posted via Newsfeeds.Com - Unlimited-Uncensored-Secure Usenet News==----
http://www.newsfeeds.com The #1 Newsgroup Service in the World! 100,000 Newsgroups
---= East/West-Coast Server Farms - Total Privacy via Encryption =---

The coil measures about 60 uH. The antenna is elevated about 3 feet on a
short tripod. The radials angle down the tripod legs and then out.

The coil is about 4 inchs in diameter, number 12, wound on a fiberglass
form. It is centerloaded. I am looking at it accross the yard, it is about
6 inches long. It is would with about a point .5 pitch. Calculations for a
1:1 pitch predict a Q of about 450.

Thanks,
Dan

Dan, I have just run a NEC 2 model of your antenna. I have used the
Sommerfeld/Norton ground, with average parameters of: Er = 13, and Sigma =
5 S/m. The top of the antenna is at 18 ft, and the base at 3 ft. I have
twelve 40 ft radials, spaced at 30 deg, and within the limitations of NEC 2,
placed them at 3" above ground. The first 5 ft of the radials drops from 3
ft to 3" at an angle of 45 deg. The monopole is center loaded with an
inductor of Q = 450. The model has 640 (6") segments and takes 3 minutes to
run (3.5 - 4.0 MHz in 50 kHz increments). What I notice is that I need 92
uH to resonate at 3.9 MHz. The input impedance is 12 ohms. I used a lumped
element model for the inductor. I may try a physical helix later. These
data do not seem to agree with your measured results. NEC 2, with the
Sommerfeld/Norton ground solution, is supposed to give a reasonable result
with wires at 10^(-3) wavelengths above ground (Basic Antenna Modeling,
Cebik p. 15-16 Nittany Scientific).

Gain and take-off angle are excellent, with max gain of -3 dBi at 28 deg.
elevation. The lower 3 dB point (8 deg elevation) gain is -6.6 dBi. The
NEC output file indicates an antenna efficiency of 54%. A free space model
shows an input impedance of 8 ohms, so your ground losses are not
significant (At least with my model).

Apart from adding horizontal wires, in "T" or inverted "L" fashion, I doubt
any antenna you could put up would match its performance at distances over
500 miles.

With 100 ft of LMR 400 the additional loss is about 0.45 dB. I would be very
interested to know if anybody has any ideas why my calculations appear to be
different from the measurements.

Regards,

Frank

On Sat, 27 Aug 2005 17:49:42 -0700, dansawyeror
wrote:

I am not sure what you mean by -always-. If you mean there is no such thing as a
perfect coax line then your statement is true but does not add any real value.
If you mean by -always- the feedline is a significant component in the antenna
system then I would have to disagree. When operated at their design point coax
transmission lines do not radiate and are not part of the radiating "antenna
system".


Well, now, we've just morphed "antenna system" to "radiating antenna
system."

But never mind, nothing's changed, you're still all wet. Of course
the feedline is a component in the antenna system regardless of
whether it radiates or not. BTW the line can be perfectly matched and
radiate or it can be highly mismatched and not radiate.


Coax is designed to work in a specific environment as a transmission line. These
transmission lines are designed not to radiate. When transmission lines are
operated significantly outside their design range the(y) radiate.

Squeak squeak---squeak squeak---squeak squeak, there I just screwed a
tee connector on the end of a run of 50 ohm coax and on the tee I
screwed on two 50 ohm loads. The line is now terminated in 25 ohm
making the SWR 2:1. 'Splain to me how well this line, "operating
outside its design range", radiates.

If that's not "significantly" far enough outside the "design range"
then allow me to remove the loads, add two more tees and terminate
them with four 50 ohm loads, making the SWR 4:1. How well does the
line radiate now? Should I continue to 8:1 or are you convinced?


Adding a tuner to
one end only controls the characteristics at that point. It does not 'clean up'
the mismatchs.

It can certainly "clean up" the mismatch at the input to the tuner,
which unless I've been deluding myself for 45 years or so, is the
point.

On Sat, 27 Aug 2005 16:53:39 -0700, dansawyeror
wrote:

Let's take the case of a 50 Ohm line and some mismatched antenna. The result is
a combination other then 50 Ohm with most likely a zero complex component.

Surely you don't believe this do you? It is -much- more likely that
the impedance is reactive than not. At one (fundamental) frequency
the reactance is zero. At every other frequency it is reactive.

All a
tuner does is match 50 Ohm at the radio to the complex impedance presented to it
at the source of the line.

Isn't that enough?


That the only place with 50 Ohms and zero inductance in the line - antenna
system. The combination of cable and antenna presents something other then R =
50 ohms 0 reactance and the the transmission line see discontinuities. The
result is it radiates.

Oh dear me.

Thanks Wes...done...btw the antenna is for 160 meters, not 80m, so the loss
is even less. It doesn't look to me like it's worth doing anything more than
tuning out the mismatch in the shack. 73

....hasan, N0AN

Dan,
I appreciate your reply, but it is filled with misinformation. Consider
reading some reference materials. The errors have been discussed in other
responses to your post.

I don't mean to insult you for trying to help, but the information you
provided is so incorrect, that I couldn't let it pass. I hope no one
attempts to make use of what you said...it will only compound the error.

I asked simply is it worth it? (in terms of loss). Wes had the calculator
software that showed the loss is insignificant, so it saves the
work/complexity at the base of the antenna. Radiation from the feedline is
not a function of mismatch. I'm surprised Reg didn't go apoplectic over that
one.

The tuner in the shack will do the job nicely.

....hasan, N0AN
"dansawyeror" wrote in message
...
Since you are talking about 50 Ohms I assume you are talking about a
transmission line. If that is the case you should definitely match the
feedline to antenna at the antenna feed point. Any attempt to match the
feedline with a tuner in the shack only turns the whole feedline into part
of the antenna system. By doing that you have lost any good work in
building the antenna.

Dan

hasan schiers wrote:
I found the error, I had to fix two conditions that I had changed in the
model:

Copper wire (for loss)
Ground characteristics

Now that both antennas have the same conditions, the T has ever so
slightly better gain at 20 degrees than the Inverted L. Not enough to
bother with the increased complexity, and the input Z is now down around
5 ohms for the T and 8 ohms for the L.

Now, is it worth matching the 8 ohms up to 50 at the feedpoint, or just
using the tuner in the shack to take care of it? (coax feed, LMR-400,
about 50')

...hasan, N0AN

"Roy Lewallen" wrote in message
...

John Ferrell wrote:

. . .
I am a perpetual antenna student!

And so are we all.

Roy Lewallen, W7EL

Tom Ring wrote:

Cecil Moore wrote:
Why is it definite? What is the loss in 50 ft. of LMR-400 at the
frequency of interest when the SWR is 50/8 = 6.25:1?

Ok, I need teaching here. Why would the loss change? The loss on the
line is forced to what happens at the nominal 50 ohms doesn't it? The
SWR shouldn't be able to change it unless the voltage limits are hit I
would think. I need an explanation of why it wouldn't be so.

For the same amount of power delivered to the load, transmission
line losses increase with increasing SWR. For a Z0-matched system:
(PLoad = Pfor - Pref) As the load mismatch is increased, energy
reflected from the mismatched load increases and is re-reflected
back toward the load at the tuner Z0-match point. As the load
mismatch is increased, more energy is stored in the transmission
line during steady-state. This is indirect proof that reflected
energy waves actually exist.
--
73, Cecil http://www.qsl.net/w5dxp

----== Posted via Newsfeeds.Com - Unlimited-Uncensored-Secure Usenet News==----
http://www.newsfeeds.com The #1 Newsgroup Service in the World! 120,000+ Newsgroups
----= East and West-Coast Server Farms - Total Privacy via Encryption =----

"Cecil Moore" wrote in message
...
dansawyeror wrote:
The combination of cable and antenna presents something other then R = 50
ohms 0 reactance and the the transmission line see discontinuities. The
result is it radiates.

If the currents are balanced, a 50 ohm transmission line seeing
something other than a 50 ohm load does NOT cause it to radiate.
If a 50 ohm unbalanced transmission line sees a 50 ohm balanced
load and common-mode currents flow on the outside of the coax,
it will usually result in radiation from the feedline. Simply
knowing the magnitude of the feedpoint impedance doesn't tell us
anything about feedline radiation.
--
73, Cecil http://www.qsl.net/w5dxp

Even grossly mismatched open wire transmission line does not radiate
significantly. For example a 66 ft length of 3" spaced, open wire line,
shorted at one end, radiates only 4 - 5% of the input power. 95% is
dissipated in the conductor losses.

73,

Frank

Frank,

Thank you. Is there any way you can forward the saved parameters. This is a
screwdriver antenna, I will remeasure the coil and double check.

My modeling of the free space antenna showed about 4 Ohms but it was with a much
simpler program. It was that program I used to measure Q.

Thanks,
Dan

Frank wrote:
The coil measures about 60 uH. The antenna is elevated about 3 feet on a
short tripod. The radials angle down the tripod legs and then out.

The coil is about 4 inchs in diameter, number 12, wound on a fiberglass
form. It is centerloaded. I am looking at it accross the yard, it is about
6 inches long. It is would with about a point .5 pitch. Calculations for a
1:1 pitch predict a Q of about 450.

Thanks,
Dan


Dan, I have just run a NEC 2 model of your antenna. I have used the
Sommerfeld/Norton ground, with average parameters of: Er = 13, and Sigma =
5 S/m. The top of the antenna is at 18 ft, and the base at 3 ft. I have
twelve 40 ft radials, spaced at 30 deg, and within the limitations of NEC 2,
placed them at 3" above ground. The first 5 ft of the radials drops from 3
ft to 3" at an angle of 45 deg. The monopole is center loaded with an
inductor of Q = 450. The model has 640 (6") segments and takes 3 minutes to
run (3.5 - 4.0 MHz in 50 kHz increments). What I notice is that I need 92
uH to resonate at 3.9 MHz. The input impedance is 12 ohms. I used a lumped
element model for the inductor. I may try a physical helix later. These
data do not seem to agree with your measured results. NEC 2, with the
Sommerfeld/Norton ground solution, is supposed to give a reasonable result
with wires at 10^(-3) wavelengths above ground (Basic Antenna Modeling,
Cebik p. 15-16 Nittany Scientific).

Gain and take-off angle are excellent, with max gain of -3 dBi at 28 deg.
elevation. The lower 3 dB point (8 deg elevation) gain is -6.6 dBi. The
NEC output file indicates an antenna efficiency of 54%. A free space model
shows an input impedance of 8 ohms, so your ground losses are not
significant (At least with my model).

Apart from adding horizontal wires, in "T" or inverted "L" fashion, I doubt
any antenna you could put up would match its performance at distances over
500 miles.

With 100 ft of LMR 400 the additional loss is about 0.45 dB. I would be very
interested to know if anybody has any ideas why my calculations appear to be
different from the measurements.

Regards,

Frank