Feeding Zepp-antenna
Olle Pettersson
all-band antenna with two coaxial lines. The antenna is a half wavelenght at
the lowest band and fed with two coaxial lines, the inner-conductors are
connected the legs, the braids are conected to each other at the the feeding
point and also connected to each other the bottom end. The inner conductors
are fed via a match-box, the braids are also connected to ground at the
match-box. The two coaxial-cables are of the same lenghts, 1/4 wavelenght +
about 20% at to lowest frequency.
SM0ESD/Olle
Dan Richardson
I wonder exactly what information you may be seeking. I can say that
to begin with that your losses would be about the same as using a
single run of coax and the cost is double.
73 Danny, K6MHE
John - KI4RO
Sounds like a LOT of trouble to me.
If you're concerned about RF in the shack, then just feed it with 450
ohm ladder line or 300 ohm ladder line which would then come down to an
external balun from which you could run some coax into the shack. I've had
tremendous success with such a configuration and very little RFI problems.
My Zepp is 136 feet long from end to end.
73 es DX!
John
Olle Pettersson wrote in message ...
Bob
I use that setup. I started after I melted a couple of single runs of
coax. I am running about 1100 watts. Works good for me. I have some
metal close to the cable and this works better than open ladder line.
It is just another sort of twin lead. You need that to opperate a G5RV
type antenna. It is more expensive though. If you feed it as ladder
line it should work for you.
Bob N5RRR
* Sent from RemarQ http://www.remarq.com The Internet's Discussion Network *
The fastest and easiest way to search and participate in Usenet - Free!
W6RCecilA
> I should like to have some information about feeding a flat, center fed
> all-band antenna with two coaxial lines.
Main problem is on the frequency where the dipole is one wavelength,
the SWR will be sky high. Assuming 50 ohm coax, the side-by-side ZØ
will be 100 ohms. The SWR will be greater than 50:1. Almost all the
power will be dissipated in the coax. This will happen on most of
the HF bands. OTOH, if the dipole is one wavelength, the SWR on
450 ohm ladder-line will be around 12:1, a very managable value.
--
73, Cecil, W6RCA http://www.bigfoot.com/~w6rca
Robert Lay (W9DMK)
>I should like to have some information about feeding a flat, center fed
>all-band antenna with two coaxial lines. The antenna is a half wavelenght at
>the lowest band and fed with two coaxial lines, the inner-conductors are
>connected the legs, the braids are conected to each other at the the feeding
>point and also connected to each other the bottom end. The inner conductors
>are fed via a match-box, the braids are also connected to ground at the
>match-box. The two coaxial-cables are of the same lenghts, 1/4 wavelenght +
>about 20% at to lowest frequency.
Dear Ole,
The method you are using will have less sensitivity to rain, I believe
- not sure because have never tried it.
If you are interested in isolation from nearby metal, it may be better
to use the twin coax segments only where necessary and use open wire
or window line elsewhere. The impedance of the line is of little
concern. The impedance bumps that occur with the transitions from one
Zo to another are also of little concern. I use the twin coax method
only for the last 6 feet in order to come from the outside through the
wall to the tuner.
Every foot of the twin coax will have significant loss in comparison
with the ladder line.
73 de Bob, W9DMK, King George, VA
Change "nobody" to "w9dmk" for e-mail
http://www.qsl.net/w9dmk
http://www.fortunecity.com/meltingpot/albania/784
W6RCecilA
> The impedance of the line is of little concern.
Maybe to you, Bob. But the lower the impedance of the line,
the higher the maximum SWR for non-resonant length antennas.
For instance, 100ft of RG-58 has a matched-line loss of about
2dB on 10m. With an SWR of 20:1 the losses are increased by
about 6dB, i.e. 3/4 of the power is lost due to SWR.
OTOH, 100ft of 450 ohm ladder-line has a matched-line loss of
about 0.25dB on 10m and a lower maximum SWR with close to
negligible losses. (Unless you live in a rain forest like
Roy does :-)
Reg Edwards
It is a simple matter to obtain a 50 ohm coax line which has a lower
loss than 450-ohm window-line or 600-ohm open wire.
Open-wire feeders are to be preferred to coax simply because they have
a lower loss for a given expenditure of copper or dollars.or pounds
weight per 100-feet.
Not much connection with mismatches and SWR.
Reg.
=================================
W6RCecilA wrote
ka4cnh
center fed antenna is the voltage handling capabilities as compared to
450-ohm ladder line. Please comment. Dave-KA4CNH
W6RCecilA
>
> Gents, my greatest concern with using dual coax in feeding an all band
> center fed antenna is the voltage handling capabilities as compared to
> 450-ohm ladder line. Please comment. Dave-KA4CNH
Reg seems to have one-inch hardline available. :-) Using six feet
of side-by-side RG-213 coax to get it into the house is OK. Using
100 feet of side-by-side coax to feed a non-resonant length dipole
is out of touch with reality, IMO. I have available (for free) high
quality 75 ohm coax and I still use 450 ohm window-line to feed my
non-resonant length dipoles. Why? Because 6k/150 is a 40:1 SWR on
the coax and 6k/450 is a 13:1 SWR on the window-line.
W6RCecilA
>
> Just a comment -
> It is a simple matter to obtain a 50 ohm coax line which has a lower
> loss than 450-ohm window-line or 600-ohm open wire.
For the same price as window-line? i.e. $12 per 100 ft?
We're talking hams here, Reg. :-)
ka4cnh
venture a guess as to the voltage handling capability of it? I do not find
any readily available specs. for it. Thanks, Dave-KA4CNH
W6RCecilA wrote:
W6RCecilA
> Cecil, as to your statement of Reg having one-inch hardline, would you
> venture a guess as to the voltage handling capability of it? I do not find
> any readily available specs. for it. Thanks, Dave-KA4CNH
I don't have hardline specs either but 1.2" dia LMR1200 has a maximum
voltage rating of 10 kva compared to RG213's 3700 volt rating.
W6RCecilA
>
> ka4cnh wrote:
> > Cecil, as to your statement of Reg having one-inch hardline, would you
> > venture a guess as to the voltage handling capability of it? I do not find
> > any readily available specs. for it. Thanks, Dave-KA4CNH
>
> I don't have hardline specs either but 1.2" dia LMR1200 has a maximum
> voltage rating of 10 kva compared to RG213's 3700 volt rating.
Make that a 10 kv rating. Don't know how the 'a' slipped in there.
Reg Edwards
I'm not quite in tune with this thread. But I don't think you need be
unduly worried about voltage rating on coax. At HF, with solid
polyethylene dielectric, the cable melts due to current overheating
long before the maximum allowed voltage is exceeded. Power input is
limited by current not volts.
With air insulated coax the differential between volts and current
rating may be less. But it is only at LF and VLF that a coax will
breakdown under high voltage before it overheats due to current and
power input..
Select your cable on the basis of the manufacturer's power rating at
the operating frequency. Do some amps, volts and Zo calulations and
you will very likely find the line voltage will not exceed the
manufacturers voltage rating. Select a cable power rating that is SWR
times the actual power you propose to run into it and calculate volts
on that basis.
For power and ratings of many RG cable types versus frequency, go to -
http://www.therfc.com/powerrat.htm
Unfortunately voltage ratings are not available directly from this
site and I am unable to suggest an alternative. The ARRL Handbook has
a list of voltage ratings of a dozen or so cable types.
Reg, G4FGQ
--
=================================
Dave wrote in ...
> Cecil, as to your statement of Reg having one-inch hardline, would
you
> venture a guess as to the voltage handling capability of it? I do
not find
> any readily available specs. for it. Thanks, Dave-KA4CNH
>
> W6RCecilA wrote:
>
> > ka4cnh wrote:
> > >
> > > Gents, my greatest concern with using dual coax in feeding an
all band
> > > center fed antenna is the voltage handling capabilities as
compared to
> > > 450-ohm ladder line. Please comment. Dave-KA4CNH
> >
> > Reg seems to have one-inch hardline available. :-) Using six feet
> > of side-by-side RG-213 coax to get it into the house is OK. Using
> > 100 feet of side-by-side coax to feed a non-resonant length dipole
> > is out of touch with reality, IMO. I have available (for free)
high
> > quality 75 ohm coax and I still use 450 ohm window-line to feed my
> > non-resonant length dipoles. Why? Because 6k/150 is a 40:1 SWR on
> > the coax and 6k/450 is a 13:1 SWR on the window-line.
Dennis C. O'Connor
10,000 amperes, transformer...
Denny
W6RCecilA wrote:
>
> W6RCecilA wrote:
> >
> > ka4cnh wrote:
> > > Cecil, as to your statement of Reg having one-inch hardline, would you
> > > venture a guess as to the voltage handling capability of it? I do not find
> > > any readily available specs. for it. Thanks, Dave-KA4CNH
> >
> > I don't have hardline specs either but 1.2" dia LMR1200 has a maximum
> > voltage rating of 10 kva compared to RG213's 3700 volt rating.
>
> Make that a 10 kv rating. Don't know how the 'a' slipped in there.
Robert Lay (W9DMK)
wrote:
>Robert Lay (W9DMK) wrote:
>> The impedance of the line is of little concern.
>
>Maybe to you, Bob. But the lower the impedance of the line,
>the higher the maximum SWR for non-resonant length antennas.
>For instance, 100ft of RG-58 has a matched-line loss of about
>2dB on 10m. With an SWR of 20:1 the losses are increased by
>about 6dB, i.e. 3/4 of the power is lost due to SWR.
>
>OTOH, 100ft of 450 ohm ladder-line has a matched-line loss of
>about 0.25dB on 10m and a lower maximum SWR with close to
>negligible losses. (Unless you live in a rain forest like
>Roy does :-)
I think the context of my remark about the impedance being of little
importance was clearly with regard to the ladder line. Is there
something about the way I expressed it that suggested that I was
talking about the coax?
W6RCecilA
>
> On Tue, 02 Nov 1999 08:06:58 -0600, W6RCecilA <Cecil....@IEEE.org>
> wrote:
>
> >Robert Lay (W9DMK) wrote:
> >> The impedance of the line is of little concern.
> >
> >Maybe to you, Bob. But the lower the impedance of the line,
> >the higher the maximum SWR for non-resonant length antennas.
> >For instance, 100ft of RG-58 has a matched-line loss of about
> >2dB on 10m. With an SWR of 20:1 the losses are increased by
> >about 6dB, i.e. 3/4 of the power is lost due to SWR.
> >
> >OTOH, 100ft of 450 ohm ladder-line has a matched-line loss of
> >about 0.25dB on 10m and a lower maximum SWR with close to
> >negligible losses. (Unless you live in a rain forest like
> >Roy does :-)
>
> I think the context of my remark about the impedance being of little
> importance was clearly with regard to the ladder line. Is there
> something about the way I expressed it that suggested that I was
> talking about the coax?
Yep, if I remember correctly, you said it's OK to use paralleled coax
where necessary and the impedance of the line is of little concern.
Dan Richardson
wrote:
>
>Yep, if I remember correctly, you said it's OK to use paralleled coax
>where necessary and the impedance of the line is of little concern.
For a little dinky run (getting outside the shack) of less 8 feet or
so don't you think Cecil that impedance is of little concern?
Danny
W6RCecilA
I'm not concerned about that. I'm concerned about the hams who might
think that "coax where necessary" means it's OK to use it all the way
to the antenna.
Wes Stewart
>Reg Edwards wrote:
>>
>> Just a comment -
>> It is a simple matter to obtain a 50 ohm coax line which has a lower
>> loss than 450-ohm window-line or 600-ohm open wire.
>
>For the same price as window-line? i.e. $12 per 100 ft?
>We're talking hams here, Reg. :-)
>--
>73, Cecil, W6RCA http://www.bigfoot.com/~w6rca
Oh, I don't know about that, Cecil. I once drove down a street and observed
some linemen cutting down CATV cable and rolling it up. I stopped and ask them
for some. They said sure how much do you want? Because the poles were about
200' apart, I just asked for one roll. They threw two into my truck an off I
went. Not only did I get 400' of 1/2" and 7/8" coax, I got 400' of 5/16" EHS
galvanized messenger cable too. I used it to guy my Rohn 45 foldover tower.
I finally traded the CATV coax away. I've gone to Heliax. As Reg says, the
loss of even *little* 1/2" Heliax is comparable to "450 ohm window line" and it
doesn't care about getting wet or being buried.
73,
Wes N7WS
Dan Richardson
[SNIP]
>
>I finally traded the CATV coax away. I've gone to Heliax. As Reg says, the
>loss of even *little* 1/2" Heliax is comparable to "450 ohm window line" and it
>doesn't care about getting wet or being buried.
>
>73,
>
>Wes N7WS
Wes,
That brings up a question. If you were to use your Heliax to feed a
center fed one-wavelength doublet as is done using open line. How
would you control the common mode current?
73,
Danny, K6MHE
Bart Rowlett
>
> On Sun, 07 Nov 1999 03:13:22 GMT, n7...@arrl.net (Wes Stewart) wrote:
>> >
> >I finally traded the CATV coax away. I've gone to Heliax. As Reg says, the
> >loss of even *little* 1/2" Heliax is comparable to "450 ohm window line" and it
> >doesn't care about getting wet or being buried.
> That brings up a question. If you were to use your Heliax to feed a
> center fed one-wavelength doublet as is done using open line. How
> would you control the common mode current?
Placing a common mode choke in the feedline .25 wavelengths (common
mode) from the feedpoint would be a good start.
bart
wb6hqk
W6RCecilA
> That brings up a question. If you were to use your Heliax to feed a
> center fed one-wavelength doublet as is done using open line. How
> would you control the common mode current?
Another question: What are the results of feeding a 5000 ohm antenna
with heliax (SWR ~ 100:1) Vs feeding the same antenna with 450 ohm
ladder-line (SWR ~ 11:1)?
Wes Stewart
>On Sun, 07 Nov 1999 03:13:22 GMT, n7...@arrl.net (Wes Stewart) wrote:
>
>[SNIP]
>>
>>I finally traded the CATV coax away. I've gone to Heliax. As Reg says, the
>>loss of even *little* 1/2" Heliax is comparable to "450 ohm window line" and it
>>doesn't care about getting wet or being buried.
>>
>>73,
>>
>>Wes N7WS
>
>Wes,
>
>That brings up a question. If you were to use your Heliax to feed a
>center fed one-wavelength doublet as is done using open line. How
>would you control the common mode current?
>
>73,
>Danny, K6MHE
The quick answer is that I wouldn't feed a one-wavelength antenna with anything.
I'm assuming that you refer to a single-wire, untrapped, multiband antenna.
Until I can convince myself that the overall system efficiency of this
arrangement, including transmission line, balun and tuner losses, is better than
a trapped dipole or one made of parallel half-wave-resonant sections, I won't
consider using one.
At this time, I eat what I cook. I'm using a combination of traps and parallel
dipoles with a ferrite common-mode choke at the feedpoint. No tuner to mess
with. Now if I could just afford a no-tune amplifier, I'd be really happy.
Forty plus years of twiddling knobs is enough.
And one more thing. For the record, a "Zepp" antenna is fed at the end anyway,
not the middle.
Your friend,
Wes
Wes Stewart
>Dan Richardson wrote:
>> That brings up a question. If you were to use your Heliax to feed a
>> center fed one-wavelength doublet as is done using open line. How
>> would you control the common mode current?
>
>Another question: What are the results of feeding a 5000 ohm antenna
>with heliax (SWR ~ 100:1) Vs feeding the same antenna with 450 ohm
>ladder-line (SWR ~ 11:1)?
>--
>73, Cecil, W6RCA http://www.bigfoot.com/~w6rca
First of all, where do you get a "5000 ohm" antenna?
I used EZNEC to calculate a typical situation: A copper wire, 124.6 foot
dipole, 60' above real ("good", hi-accuracy) ground.
At 3.8 MHz, Z = 81.4 + j0.
At 7.2 MHz, the same dipole Z = 2780 + j2400. Using this value, and my
measurements of Wireman type 554, the line Zo = 359 - j1.3 and the loss per
100' @ 7.2 MHz = 0.156 dB.
Calculating the total loss for 100 ft, I get: Lt = 0.97 dB.
Peforming the same calculations for LDF-4 Heliax, the Andrew catalog specifies
the line at 0.30 dB/100' @ 20 MHz. Calculating the matched loss at 7.2 MHz, I
get: L = 0.18 dB/100'. Using Andrew's specified velocity factor of 0.88, the
line Zo = 50 - j0.2.
Calculating the total loss for this situation yields: Lt = 5.01 dB. Before you
say Ah Ha, let's look at what happens when the Wireman line gets wet.
From my measurements, the line Zo = 343 - j17.6 and the loss for 100' at 7.2 MHz
is ~2.3 dB! Feeding the same antenna, the total loss is now: Lt = 7.27 dB.
I believe that the foregoing argues against both situations. The coax is
clearly not a good choice. But then, if you expect rain or other degradation
due to UV breakdown, etc. to the ladder line, then I don't believe it is a good
choice either. I know many will argue with the latter conclusion, but it is my
opinion.
(Ref: "Balanced Transmission Lines in Current Amateur Practice", ARRL Antenna
Compendium Vol 6.)
73,
Wes N7WS
Dan Richardson
>On Sun, 07 Nov 1999 15:27:00 GMT, k6...@arrl.net (Dan Richardson) wrote:
>
>>Wes,
>>
>>That brings up a question. If you were to use your Heliax to feed a
>>center fed one-wavelength doublet as is done using open line. How
>>would you control the common mode current?
>>
>>73,
>>Danny, K6MHE
>
>
>The quick answer is that I wouldn't feed a one-wavelength antenna with anything.
>I'm assuming that you refer to a single-wire, untrapped, multiband antenna.
>
>Until I can convince myself that the overall system efficiency of this
>arrangement, including transmission line, balun and tuner losses, is better than
>a trapped dipole or one made of parallel half-wave-resonant sections, I won't
>consider using one.
>
>At this time, I eat what I cook. I'm using a combination of traps and parallel
>dipoles with a ferrite common-mode choke at the feedpoint. No tuner to mess
>with. Now if I could just afford a no-tune amplifier, I'd be really happy.
>Forty plus years of twiddling knobs is enough.
>
>And one more thing. For the record, a "Zepp" antenna is fed at the end anyway,
>not the middle.
>
>Your friend,
>
>Wes
Wes, Wes,
Now, Now simmer down... I didn't say anything about a Zepp.
Danny
Dan Richardson
[snip]
>Until I can convince myself that the overall system efficiency of this
>arrangement, including transmission line, balun and tuner losses, is better than
>a trapped dipole or one made of parallel half-wave-resonant sections, I won't
>consider using one.
>
[snip]
>
>Your friend,
>
>Wes
Hi Friend,
Humm, let me see..... Using EZNEC and modeling a dipole which is
self-resonate at 3.75 and changing the frequency to 7.15 MHz. Here's
what I get
Impedance 4,206 +J1474.
Next using Dean Straw's program TL.EXE for calculating line and tuner
losses I get:
(for 100 feet of 600 Ohm open line) *
SWR at load 7.8:1
Matched line loss 0.047 dB
Additional loss from SWR .143 dB
Tuner losses (unloaded inductor Q 200 - unloaded Capacitor Q 1000)
"L" type tuner 0.02 dB
"T" type tuner 0.08 dB
"Pi" type tuner 0.17 dB
Best case "L" tuner total losses (less the antenna itself) -0.21 dB
Worse case Pi tuner -0.36 dB
Not too shabby for a all band/ broadband antenna system. Wouldn't you
agree?
73
Danny
* That's open line not window line... It rains a lot here!
W6RCecilA
> First of all, where do you get a "5000 ohm" antenna?
From Dan's one-wavelength doublet question above.
> At 3.8 MHz, Z = 81.4 + j0.
> At 7.2 MHz, the same dipole Z = 2780 + j2400.
If you make the 80m antenna resonant at 3.5-3.6 MHz the 7.2MHz
feedpoint impedance approaches the 5000 ohm value.
> The coax is clearly not a good choice.
Aha. :-)
> But then, if you expect rain or other degradation
> due to UV breakdown, etc. to the ladder line, then I don't believe it is a good
> choice either.
Around here I wouldn't dare operate while it's raining because rain
is almost always accompanied by lightning. And it only rains a
couple of days a month. I'm happy with the performance of my
ladder-line. Most of it is tucked under the eave of my house
out of the rain. Different strokes for different folks.
Wes Stewart
[SNIP]
>
>Wes, Wes,
>
>Now, Now simmer down... I didn't say anything about a Zepp.
>
>Danny
I know, I know. But look at the title of this thread.
Wes
>
Reg Edwards
Frequency = 7.1 MHz
Dipole end-to-end length = 132 feet. (ONE WAVELENGTH)
Dipole wire diameter = 14 Awg
Dipole height above ground = 33 feet
Feeder length = 83 feet
Feeder impedance = 450 ohms
Feeder wire diameter = 14 Awg
Feeder velocity factor = 0.8
SWR relative to 50 ohms at Tx = 1.4 : 1
Antenna input resistance = 6200 ohms.
Resistance presented to Tx = 36 ohms
Tuner not needed so no loss in tuner.
OVERALL RADIATING EFFICIENCY = 88 percent
Relative to ideal that is a loss less than 1 / 10th of an S-unit.
When raining there would no noticeable change in efficiency.
It would take an extremely large diameter 50-ohm coax to achieve the
same high efficiency.
Data obtained using program DIPOLE1.
=================================
Cheers, Reg, G4FGQ
For free technical radio software go to:-
http://www.btinternet.com/~g4fgq.regp
=================================
W6RCecilA <Cecil....@IEEE.org> wrote in message
news:38261307...@IEEE.org...
Dan Richardson
<ba...@wb6hqk.ampr.org> wrote:
>
>Placing a common mode choke in the feedline .25 wavelengths (common
>mode) from the feedpoint would be a good start.
>
>bart
>wb6hqk
Hi Bart,
Your suggestion seemed plausible so I decided to model such a setup
using EZNEC.
Well, it didn't work as hoped. What I ended up with was an unbalanced
antenna that still had fairly high common mode current on the
feedline.
Placing a choke (1,000 Z impedance - that's a lot of beads)
ź-wavelength down the coax from the feed point, as you suggested. I
essentially ended up with an antenna that had one ˝-wave element on
one side and two elements (one ˝-wave and one ź-wave) on the other.
Computer modeling revealed that I had fairly high common mode current
(above and below the choke) on the feed line plus quite a distorted
radiation pattern.
In essence what I ended up with was something like the Carolina Windom
type antenna except this one had a higher impedance (~ R970 -J1400).
Thanks for the thought Bart. It gave me something to do on a rainy
afternoon.
73
Danny, K6MHE
Dan Richardson
Damn Wes, This is fun.... Glad your now retired...
Danny
Wes Stewart
[snip]
>
>Thanks for the thought Bart. It gave me something to do on a rainy
>afternoon.
>
I rest my case. It's raining at Danny's place and he uses ladder line. He has
no choice but to QRT and get on the computer.
73,
Wes N7WS
Dan Richardson
Wes,
If you mean plastic cover window line that could be true, but I'm
using open parallel line and the rain is NO PROBLEM my friend.
Danny
Wes Stewart
>On Sun, 07 Nov 1999 19:39:51 GMT, n7...@arrl.net (Wes Stewart) wrote:
>
>[snip]
>
>>Until I can convince myself that the overall system efficiency of this
>>arrangement, including transmission line, balun and tuner losses, is better than
>>a trapped dipole or one made of parallel half-wave-resonant sections, I won't
>>consider using one.
>>
>[snip]
>>
>>Your friend,
>>
>>Wes
>
>Hi Friend,
>
>Humm, let me see..... Using EZNEC and modeling a dipole which is
>self-resonate at 3.75 and changing the frequency to 7.15 MHz. Here's
>what I get
>
>Impedance 4,206 +J1474.
Please expand on this and give all of the parameters of the antenna. Wire size,
length, height, ground type and conductivity.
>
>Next using Dean Straw's program TL.EXE for calculating line and tuner
>losses I get:
>
>(for 100 feet of 600 Ohm open line) *
This thread started with "450" ohm ladder line as I recall.
>
> SWR at load 7.8:1
>
>Matched line loss 0.047 dB
>
>Additional loss from SWR .143 dB
>
>Tuner losses (unloaded inductor Q 200 - unloaded Capacitor Q 1000)
Funny thing. No one ever seemed to consider tuner losses until I wrote a
critique of "The Lure of the Ladder Line" to QST's Technical Correspondence
(never published). After considerable back-and-forth with Dean Straw, tuner
loss suddenly became important and shortly after, Frank Witt's paper on
measuring tuner losses got published. Straw also immediately incorporated tuner
losses in his TL program. What a coincidence.
>
>"L" type tuner 0.02 dB
>
>"T" type tuner 0.08 dB
>
>"Pi" type tuner 0.17 dB
>
>Best case "L" tuner total losses (less the antenna itself) -0.21 dB
>Worse case Pi tuner -0.36 dB
Remember too, this is for *optimum* adjustment of the tuner. It is easy to
achieve a match (SWR 1:1) with an infinite number of combinations when there are
three or more reactances in the network. And if you have a roller inductor, the
unloaded Q can get down to 30 or 40. Then what happens?
>
>Not too shabby for a all band/ broadband antenna system. Wouldn't you
>agree?
No, I don't agree one bit. Just for giggles, change the line length to 65' and
use the pi-network. Report back here with the results.
(Clue: A 1.5:1 SWR bandwidth of 28 KHz ain't gonna be fun when you want a quick
QSY. But if you QSY is the right direction, the self-heating of the inductor
might make the network "track" your frequency change.)
Furthermore, nowhere have balun losses been considered. The networks described
are all unbalanced. W7EL, et.al. have clearly proved that the place for the
balun is at the input to a "balanced" tuner. Add another inductor to these
networks and see what happens.
If the balun is between the match point and the antenna, then its loss must (to
the first order) be added to the matched transmission line loss and *this* value
used to compute the additional loss due to mismatch.
I remain unconvinced.
73,
N7WS
Bart Rowlett
>
> Dan Richardson wrote:
> > That brings up a question. If you were to use your Heliax to feed a
> > center fed one-wavelength doublet as is done using open line. How
> > would you control the common mode current?
>
> with heliax (SWR ~ 100:1) Vs feeding the same antenna with 450 ohm
> ladder-line (SWR ~ 11:1)?
This might be a good topic for amateur experimentation.
The attenuation specifications given for most coaxial transmission line
is based on matched line conditions. For most dielectrics (esp.
Heliax), the attenuation increases in proportion with the square root of
the frequency. This implies the loss is dominated by the series
conductor resistance, which increases proportionally with the square
root of frequency due to the skin effect. At frequencies above a GHz or
so for most plastic dielectrics, manufacturers data usually shows the
loss increasing proportional to frequency indicating dielectric loss has
become the dominant loss mechanism.
Common practice at HF is to assume the loss is dominated by the series
resistance and the dielectric loss is ignored. Given the high impedance
levels which exist in the scenario being discussed, it might be worth
taking a second look at dielectric losses to verify they can be safely
neglected. Probably ok for Heliax but maybe not solid polyethylene.
bart
wb6hqk
Reg Edwards
comparison with conductor resistance loss to be negelected, it will
remain small enough regardless of the line's terminating impedances
and the swr.
Program COAXRATE gives an insight into coax loss and temperature rise
versus power input, frequency and swr.
--
Dan Richardson
>Please expand on this and give all of the parameters of the antenna. Wire size,
>length, height, ground type and conductivity.
I did not keep my original EZNEC file so have re-done it as follows:
Center fed doublet
Length 126.2
#14 copper wire
Height 30 feet
Ground type real (Conductivity: 0.005 S/m - Dielectric Const: 13.00)
(Note the 30' height was chosen to arrive at a feed point impedance
closely matching 50 Coax)
Modeling results
3.75 MHz. R51.85 -J0.01791
7.15 MHz. R3831 +J2683
Using TL.EXE
Transmission line 600-Ohm open-wire ladder line
Length 100'
Frequency 7.15 MHz.
SWR at load 9.58
SWR at lnput 9.11
Match line loss -0.047dB
Additional loss due to SWR -0.179 dB
Total line loss -0.226 dB
Tuner losses (Inductor unloaded Q = 200 / capacitor unloaded Q =1000)
"L" type = -0.02 dB (Total transmission line and tuner loss -0.246 dB)
"T" type = -0.10 dB (Total transmission line and tuner loss -0.326 dB)
Pi type = -0.14 dB (Total transmission line and tuner loss - 0.366 dB)
>
>Funny thing. No one ever seemed to consider tuner losses until I wrote a
>critique of "The Lure of the Ladder Line" to QST's Technical Correspondence
>(never published).
How come?
>After considerable back-and-forth with Dean Straw, tuner
>loss suddenly became important and shortly after, Frank Witt's paper on
>measuring tuner losses got published.
>
It would appear that the tuner losses shown above are running from a
high of about 30% to a low of 8% of total transmission line & tuner
losses.
> Straw also immediately incorporated tuner losses in his TL program.
>What a coincidence.
To Dean's credit. Glad he is updating his program by adding new
features.
>
>Remember too, this is for *optimum* adjustment of the tuner. It is easy to
>achieve a match (SWR 1:1) with an infinite number of combinations when there are
>three or more reactances in the network. And if you have a roller inductor, the
>unloaded Q can get down to 30 or 40. Then what happens?
>
Yes, a tuner like any piece of equipment can be misused, abused and
poorly built. But, it can also be built and operated correctly.
>
>No, I don't agree one bit. Just for giggles, change the line length to 65' and
>use the pi-network. Report back here with the results.
>
pi-network = -2.66 dB
Type L = -0.28 dB
Type T = = -0.36 dB
>Furthermore, nowhere have balun losses been considered. The networks described
>are all unbalanced. W7EL, et.al. have clearly proved that the place for the
>balun is at the input to a "balanced" tuner. Add another inductor to these
>networks and see what happens.
>
Why should I consider balun losses? The network I suggest and use is
balanced. In other words, I'm not using a balun between the tuner's
circuitry and the antenna.
>If the balun is between the match point and the antenna, then its loss must (to
>the first order) be added to the matched transmission line loss and *this* value
>used to compute the additional loss due to mismatch.
>
Again this is not a consideration for the system configuration I am
suggesting.
>I remain unconvinced.
>
I submit that you consider a *properly engineered* tuned feeder system
(open line/balanced tuner) and not seek to only consider contrived
situations that fits a preconceived belief [such as improperly
adjusting an antenna tuner]. Indeed, I could easily *contrive* a setup
showing how poorly coax *can* work. Just replace your Helix with
RG-174 and operate a dipole that is physically resonate at 3.75 MHz on
3.5 MHz.! Under such a circumstance the total feed line (no tuner)
loss would be about 5.75 dB.
You know Wes, a much smarter man than I once said "You can learn from
a fool - if you take time to listen." - - - - Are you listening?
73
Danny
Larry Benko
Dan Richardson wrote:
>
> On Mon, 08 Nov 1999 04:30:48 GMT, n7...@arrl.net (Wes Stewart) wrote:
>
> >Please expand on this and give all of the parameters of the antenna. Wire size,
> >length, height, ground type and conductivity.
>
> I did not keep my original EZNEC file so have re-done it as follows:
>
> Center fed doublet
> Length 126.2
> #14 copper wire
> Height 30 feet
> Ground type real (Conductivity: 0.005 S/m - Dielectric Const: 13.00)
>
> (Note the 30' height was chosen to arrive at a feed point impedance
> closely matching 50 Coax)
>
> Modeling results
>
> 3.75 MHz. R51.85 -J0.01791
> 7.15 MHz. R3831 +J2683
>
> Using TL.EXE
>
I cut off the rest of the message. Where is TL.EXE available from?
Thanks,
Larry, W0QE
Dan Richardson
wrote:
>I cut off the rest of the message. Where is TL.EXE available from?
>
>Thanks,
>Larry, W0QE
Larry,
You can find it at:
http://oak.oakland.edu/pub/hamradio/arrl/bbs/programs/
It is zipped as TL.ZIP
Danny
Wes Stewart
>On Mon, 08 Nov 1999 04:30:48 GMT, n7...@arrl.net (Wes Stewart) wrote:
>
>
>>Please expand on this and give all of the parameters of the antenna. Wire size,
>>length, height, ground type and conductivity.
>
>I did not keep my original EZNEC file so have re-done it as follows:
>
>Center fed doublet
>Length 126.2
> #14 copper wire
>Height 30 feet
>Ground type real (Conductivity: 0.005 S/m - Dielectric Const: 13.00)
>
>(Note the 30' height was chosen to arrive at a feed point impedance
>closely matching 50 Coax)
>
>Modeling results
>
>3.75 MHz. R51.85 -J0.01791
>7.15 MHz. R3831 +J2683
These are different from before but okay.
>
>Using TL.EXE
>
>Transmission line 600-Ohm open-wire ladder line
>Length 100'
>
>Frequency 7.15 MHz.
>
>SWR at load 9.58
>SWR at lnput 9.11
>
>Match line loss -0.047dB
>Additional loss due to SWR -0.179 dB
>Total line loss -0.226 dB
>
>Tuner losses (Inductor unloaded Q = 200 / capacitor unloaded Q =1000)
>
[two networks snipped]
>Pi type = -0.14 dB (Total transmission line and tuner loss - 0.366 dB)
>
>>
>>Funny thing. No one ever seemed to consider tuner losses until I wrote a
>>critique of "The Lure of the Ladder Line" to QST's Technical Correspondence
>>(never published).
>
>How come?
Not invented here I believe. Remember, the author is an ARRL staffer.
[snip]
>>
>>Remember too, this is for *optimum* adjustment of the tuner. It is easy to
>>achieve a match (SWR 1:1) with an infinite number of combinations when there are
>>three or more reactances in the network. And if you have a roller inductor, the
>>unloaded Q can get down to 30 or 40. Then what happens?
>>
>
>Yes, a tuner like any piece of equipment can be misused, abused and
>poorly built. But, it can also be built and operated correctly.
The operative word is "can." If you use the pi-network example below, I can
almost guarantee misadjustment.
>
>>
>>No, I don't agree one bit. Just for giggles, change the line length to 65' and
>>use the pi-network. Report back here with the results.
>>
>
>pi-network = -2.66 dB
[two networks snipped]
No comment on the bandwidth and the order of magnitude change in loss of this
combination, Danny?
A simple shortening of the transmission line by 35' (normally a good thing)
changes a usable system into one completely untractable. The average ham that
tries this will soon be posting to this newsgroup asking, "Why doesn't this
work?" The loaded Q of the pi-network is over 100 and this doesn't take into
account the Q of the antenna/feeder combination.
>
>>Furthermore, nowhere have balun losses been considered. The networks described
>>are all unbalanced. W7EL, et.al. have clearly proved that the place for the
>>balun is at the input to a "balanced" tuner. Add another inductor to these
>>networks and see what happens.
>>
>
>Why should I consider balun losses? The network I suggest and use is
>balanced. In other words, I'm not using a balun between the tuner's
>circuitry and the antenna.
>
>>If the balun is between the match point and the antenna, then its loss must (to
>>the first order) be added to the matched transmission line loss and *this* value
>>used to compute the additional loss due to mismatch.
>>
>
>Again this is not a consideration for the system configuration I am
>suggesting.
Maybe not, but the example you cite from "TL" is an unbalanced tuner. I have
had nothing to go on (until now) to know that you are using a balanced tuner.
>
>>I remain unconvinced.
>>
>
>I submit that you consider a *properly engineered* tuned feeder system
>(open line/balanced tuner) and not seek to only consider contrived
>situations that fits a preconceived belief [such as improperly
>adjusting an antenna tuner]. Indeed, I could easily *contrive* a setup
>showing how poorly coax *can* work. Just replace your Helix with
>RG-174 and operate a dipole that is physically resonate at 3.75 MHz on
>3.5 MHz.! Under such a circumstance the total feed line (no tuner)
>loss would be about 5.75 dB.
Lighten up, Danny. This is a hobby and is supposed to be fun. Before you
suggest that *I* am "contriving" a situation, you should re-examine your
beginning premise: "(Note the 30' height was chosen to arrive at a feed point
impedance closely matching 50 Coax)" Is that a contrived situation or what? Do
we usually adjust antenna height to facilitate a match? <g>
You know that I am speaking in generalities, as was the tone of my paper, that
address the common misconceptions of the average Joe Ham. There are certainly
others, such as yourself, that usually consider all of the factors of system
design.
You have gone to considerable effort to use open-wire line. Operationally, you
have two or three more knobs to fiddle with when you change bands. You have a
dipole with an almost unknown pattern on the higher bands. (I know, I know, in
some directions, you have "gain" over a 1/2 wave).
My system of a pair of traps and an electrically parallel dipole give me the
frequency coverage I desire with a match that is easily accounted for by the
autotuner in the TS-870 or the pi-net in the L4B.
You use your system, I'l ues mine. We will both be content with the
"correctness" or our respective approaches.
>
>You know Wes, a much smarter man than I once said "You can learn from
>a fool - if you take time to listen." - - - - Are you listening?
I'm listening to fools all of the time. I even read Chirp's and his trained
seal, Phil's posts.
73,
Wes
ka4cnh
with the apex at 37 ft. Each leg is 41 ft. long and is fed in the center with 450 ohm
ladder line for about the first 20ft. , then dual 8214 coax for about 25 ft. to a 4:1
voltage type balun mounted outside of the shack. I have used about 14ft. of 8267 into
the shack and the transmatch. Probably 7 ft. of this coax is coiled up outside the
shack entrance. I am using the Murch UT-2000A Ultimate transmatch. I am using a
Heath-Kit SB-200 amp.with about 500 watts out. My previous antenna was an 80-40
inverted-v using Telrex traps fed with coax. On the same feed point I was using 5-wire
flat ribbon rotor cable cut for 10-15-17-20-30 meters. The apex, etc. was the same. I
keep regular skeds on 75 and 40 meters and NOBODY has been able to tell the
difference! I also have not changed the ground conductivity, orientation of the
antenna, etc. And, this antenna works in the rain! Why, you may ask, did I use 41 ft.
per leg? Because I wanted to! I just picked a figure out of thin air and used it. I
have been having fun working DX on the higher bands and get overall good reports
anywhere, 10-80 meters.The ONLY difference I have seen is this antenna is a bit tricky
to get tuned to an swr of 1.1:1 on some frequencies. Damn the math and the computer
models, build it, hang it up and the DX will come! Isn't science wonderful?
Questions? 73, Dave-KA4CNH-Eyeball Engineer with Sledgehammer Tactics!
Bart Rowlett
>
> On Sun, 07 Nov 1999 09:16:51 -0800, Bart Rowlett
> <ba...@wb6hqk.ampr.org> wrote:
> >
> >Placing a common mode choke in the feedline .25 wavelengths (common
> >mode) from the feedpoint would be a good start.
>
> using EZNEC.
>
> Well, it didn't work as hoped. What I ended up with was an unbalanced
> antenna that still had fairly high common mode current on the
> feedline.
I thought about the problem a bit more and don't see any simple way to
maintain good balance when feeding a double zepp with an unbalanced coax
cable. The feedpoint impedance is simply too high achieve good balance
with anything short of a truly balanced feed system.
The classic approach is to feed the antenna through a quarter wave stub
made from balanced open wire. Such an arrangement will maintain
symmetry near the antenna and result in a minimum of common mode
coupling to the feeder. Coupling low impedance coaxial cable to the
stub is easily done by selecting the stub impedance such that the input
impedance to the stub will match that of the coax. Use of a common mode
choke at the coax to balanced line transition is recommended.
Another approach, which is of questionable practicality, is to place a
carefully balanced common mode choke of very high impedance at the
antenna feedpoint. The common mode impedance should be 50,000 ohms or
greater while maintaining good capacitive balance to the nearby
radiator. I don't know how to build such a device but I suspect an
airwound bifilar choke of solenoidal geometry could do the job if made
self resonant at the operating frequency. This approach has the
attraction of being usable at half the frequency where the feedpoint
impedance will be near that of the coax. Several parameters, such as
the characteristic impedance of the bifilar winding and the coil L/D
ratio are available for optimization.
A strictly monoband variant of the previous could be to mount a link
coupled tank at the feedpoint. The L/C ratio should be selected for the
minimum Q necessary to achieve the match.
If single band operation is acceptable and a coax feed is required I
would recommend a variant of the method described by Les Moxon in the
book "HF Antennas For All Locations". The method is an extension of the
Windom antenna concept and would be a good topic for another thread.
bart
wb6hqk
Dan Richardson
[snip a whole bunch of stuff]
>
>Maybe not, but the example you cite from "TL" is an unbalanced tuner. I have
>had nothing to go on (until now) to know that you are using a balanced tuner.
>
Didn't mean to mislead you. My tuner is an L type but if I draw out it
looks like a sideways "U" to maintain balance and is made with two
series coils that track one another. It's configuration is low pass
and was based upon the design of Rick Measure's which was publish in
QST (I think in Feb of 1990 but not real sure). I also have an old
Johnson Matchbox that can be pressed into service.
Thought I would this a step further and use a unloaded Q of 40, as
you suggested, for the inductance (all other things remaining the
same) and heres what I got.
"L" type (high pass) -0.05 dB
"L" type (low pass) -0.22 dB
"T" type -0.39 dB
Pi type -0.56 dB
Still not to bad.
Adding the total line loss of 0.226 dB from before. In my setup it
seems that over all my total losses would add up to under 1/2 dB. I
feel that is a very small price to pay for total freedom to move
around.
As far as knob twisting - that a very small price to pay. I should
point out that living up here in the mountains my vehicle is equipped
with a 5 speed manual transmission. At times a little more work but
the added control it gives me on these mountain roads plus the saving
in break linings and fuel expense it worth it to me. Guess I look at
things(twisting knobs and shifting gears) that way..
[snip]
>>You know Wes, a much smarter man than I once said "You can learn from
>>a fool - if you take time to listen." - - - - Are you listening?
>
>I'm listening to fools all of the time. I even read Chirp's and his trained
>seal, Phil's posts.
>
>73,
>
>Wes
Ohooooo! OUCH! The pain, the pain.
Now you did it! I going to go pout!
Danny
Reg Edwards
dipole of any overall length of any wire diameter, fed over twin
feeder of any length of any wire diameter, of any Zo or wire spacing,
at any velocity factor, at any height above an average sort of ground,
should download program DIPOLE1 from my website.
Obtain (if you are interested) feedpoint impedance of the dipole, and
(what is more important) the input impedance of the feedline presented
to the transmitter / receiver. Impedances are computed in terms of R
+ jX and also SWR.
If overall power radiating efficiency computed by this program exceeds
about 70 percent, equivalent to a negligible 1/4th of an S-unit
relative to perfection, stop worrying. You can spend a fortune
improving efficiency beyond 70 percent and no-one will ever notice any
difference. The program tells you when you have arrived there and
whether you will need an ATU or Z-match to keep your transmitter
happy.
Download in a few seconds. Not zipped-up. One small file only. Run
immediately after download. Input data on a "what-if ?" basis.
Get accustomed to International Units of measurements, ie., metres and
millimetres rather than feet and inches, and its all very user
friendly. After all, ever since radio began, we have become used to
thinking in terms of a wavelength being so many metres.
--
=================================
Regards to all, Reg, G4FGQ
For free technical radio software go to:-
http://www.btinternet.com/~g4fgq.regp
=================================
Wes Stewart <n7...@arrl.net> wrote in message
news:38273e1c...@news.azstarnet.com...
Wes Stewart
>On Mon, 08 Nov 1999 22:08:04 GMT, n7...@arrl.net (Wes Stewart) wrote:
>
>[snip a whole bunch of stuff]
>>
[summore snipped]
>As far as knob twisting - that a very small price to pay. I should
>point out that living up here in the mountains my vehicle is equipped
>with a 5 speed manual transmission. At times a little more work but
>the added control it gives me on these mountain roads plus the saving
>in break (sic) linings and fuel expense it worth it to me. Guess I look at
>things(twisting knobs and shifting gears) that way..
>
Jeeze, we can't agree on anything. I currently have three four-wheel drive
vehicles. Just got the newest one ('99 Chev 3/4 ton 454 cid) and the others
('86 Suburban and '85 F-250 460 cid) will be sold soon. All of them have
automatic transmissions. In my off-road experience, I find the automatic to be
far superior, particularly for its ability to slowly bring up the power until
the vehicle starts to move. Contrast this to the frantic cluch slipping, wheel
spinning or killed engines that a manual box features. (Yes, I can drive a
stick! I've had lots of them)
My hot wheels is a '99 Camaro SS. I auditioned both the six-speed (M6) and the
automatic four-speed (A4) and went with the auto. Many drag racers are finding
the elapsed times of the A4 are lower than the M6s, although the M6s have
slightly higher trap speeds. To me, rowing an M6 through traffic is about like
re-tweaking a tuner during a DX contest.
Well, enough of this, I better be Zepping along.
73,
Wes
Mark Keith
> >
> > Your suggestion seemed plausible so I decided to model such a setup
> > using EZNEC.
> >
> > Well, it didn't work as hoped. What I ended up with was an unbalanced
> > antenna that still had fairly high common mode current on the
> > feedline.
>
> I thought about the problem a bit more and don't see any simple way to
> maintain good balance when feeding a double zepp with an unbalanced coax
> cable. The feedpoint impedance is simply too high achieve good balance
> with anything short of a truly balanced feed system.
How about using small coils on each leg to match to 50 ohm coax. Seems
that would work, and you could use a normal coax choke for current on
the feedline. You could also feed an EDZ this way. The coils would be at
the feedpoint. One for each leg. MK
Bart Rowlett
>
> Bart Rowlett wrote:
>
> > I thought about the problem a bit more and don't see any simple way to
> > maintain good balance when feeding a double zepp with an unbalanced coax
> > cable. The feedpoint impedance is simply too high achieve good balance
> > with anything short of a truly balanced feed system.
>
> How about using small coils on each leg to match to 50 ohm coax. Seems
> that would work, and you could use a normal coax choke for current on
> the feedline.
I'm not sure how that would work. In order to achieve impedance
transformation you need both a series and shunt element. With a series
element alone, the length of the antenna elements would need to be
adjusted for a feedpoint resistance matching the coax. Then the antenna
would no longer be two halfwaves in phase.
A good compromise can be reached by using a 1:9 (or 1:16) balun and
adjusting the radiator length to match the 450 (or 800) ohm transformed
feedline resistance. A series reactor (usually capacitance) is then
used to cancel out the series reactance presented by the nonresonant
radiator. The resulting radiation pattern will be nearly the same as
for the double zepp and fairly broadband match will have been achieved
with a coax feed. The radiator impedance near the feedpoint is much
reduced and will significantly reduce the tendency for common mode
coupling to the feedline through stray capacitance. If feedline
radiation (and coupling to building wiring) is a serious problem, a
couple of common mode chokes in the feedline would probably be
beneficial, especially if a current probe isn't available.
bart
wb6hqk
K1BQT
>> that would work, and you could use a normal coax choke for current on
>> the feedline.
>I'm not sure how that would work.<
You're on the right track. Here's what does work. Use small series coils on
each leg at the feedpoint to tune out the reactance. This (in free space)
leaves you with a 100-Ohm feedpoint (roughly). Then, feed the element with a
1/4-wave transformer made from 75-Ohm cable to make the transition to 50 Ohms.
Finally, wind a portion of the 75-Ohm cable into a choke balun. That takes
care of the reactance, the impedance, and the balance problem. I know it works
because I've built several antennas this way--and published the results.
If your antenna is in the HF spectrum and fairly close to ground, you can
probably skip the 75-ohm transformer--the proximity will pull the impedance
down toward 50 Ohms anyway. Give it a try.
73
Rick K1BQT