Ladder line variability questions
Pete Soper
As I grease up my VISA card to buy another few hundred feet of
ladder line I'm bothered by W8JI's posting some months back about
some ladder line he got that was way off the expected impedance.
For a Lazy-H project I need the highest impedance I can get for
the phasing lines, but also absolute minimum weight and good durability,
as they will be dancing a lot in the wind. So I'm leaning toward
Wireman #553, which is listed as 450 ohm line with #18 stranded
conductors. Has anybody measured a sample of this? Likewise, for the
feed from the junction of the phasing lines I'm going with 300 ohm
line, but need to know ahead of time that it really is 300 ohms.
Here's my basic question. It seems that the various flavors
of line, with different conductor sizes, must be different widths
to maintain the target impedance. For instance, it seems impossible
that if "vanilla" 300 ohm ladder line has #20 conductors, the
type with #18 conductors could be the same width. But I've never
seen more than one size each of 450 and 300 ohm slotted lines, no
matter what the conductor size. Likewise, I see 450 ohm line with
everything from #18 to #14 conductors. The little bit I (think) I know
about how balanced line impedance relates to conductor size and
spacing would suggest either these different lines are different
widths, or else the "450" and "300" designations are bogus. What's the
real story here?
Incidently, plugging the numbers into the EZNEC transmission line
modeler, I can't come up with a plausible 300 ohm ladder line, based
on the dimensions and wire gauges I'm familiar with. That is, #20
conductors spaced around 1/2 inch comes out way, way over 300 ohms.
The model predicts 303 ohms for two #20 conductors spaced .2 inches
apart: much closer spacing than the conductors in the 300 ohm slotted
line I'm using for my tuner project . I must be missing something
fundamental about all of this. And I suddenly have an urge to
measure that "300" ohm line when I get home :-)
Regards,
Pete
KS4XG
Kevin AstirCS 1U KO0B
You are taking the dielectric into account right?
even though it is _mostly_ air, the isulation does increase the
capacitance/length, which lowers the impedance.
ko0b
Pete Soper
kfer...@aquilagroup.com (Kevin AstirCS "1U" KO0B) writes:
>You are taking the dielectric into account right?
I'd thought about it, but dismissed it. So this is the missing factor? Ladder
line with thicker conductors have different insulation characteristics to get
the impedance back to the target?
I guess this explains why the EZNEC transmission line modeler for user-defined
lines assumes air insulation?
-Pete
KS4XG
Roy Lewallen
In article <Dt5MM...@encore.com>, pso...@encore.com (Pete Soper)
wrote:
>Here's my basic question. It seems that the various flavors
>of line, with different conductor sizes, must be different widths
>to maintain the target impedance. For instance, it seems impossible
>that if "vanilla" 300 ohm ladder line has #20 conductors, the
>type with #18 conductors could be the same width. But I've never
>seen more than one size each of 450 and 300 ohm slotted lines, no
>matter what the conductor size. Likewise, I see 450 ohm line with
>everything from #18 to #14 conductors. The little bit I (think) I know
>about how balanced line impedance relates to conductor size and
>spacing would suggest either these different lines are different
>widths, or else the "450" and "300" designations are bogus. What's the
>real story here?
As you surmise, if the conductor sizes are different but all else is
equal,
the impedances have to be different.
>Incidently, plugging the numbers into the EZNEC transmission line
>modeler, I can't come up with a plausible 300 ohm ladder line, based
>on the dimensions and wire gauges I'm familiar with. That is, #20
>conductors spaced around 1/2 inch comes out way, way over 300 ohms.
>The model predicts 303 ohms for two #20 conductors spaced .2 inches
>apart: much closer spacing than the conductors in the 300 ohm slotted
>line I'm using for my tuner project . I must be missing something
>fundamental about all of this. And I suddenly have an urge to
>measure that "300" ohm line when I get home :-)
Good idea. EZNEC uses the equation for air dielectric, so for the same
dimensions, ladder line will always have a lower characteristic
impedance
than the Z calculated by EZNEC.
Roy Lewallen, W7EL
CHARLES J. MICHAELS
Pete said about ladder line -
I'd thought about it, but dismissed it. So this is the missing factor? Ladder
line with thicker conductors have different insulation characteristics to get
the impedance back to the target?
No, Pete;
The insulation is to hold the line together , the conductor
diameter is then selected to provide the specific impedance in the
environment of the insulation.
Charlie, W7XC
--
W8JI Tom
In article <4qfs2e$f...@news.asu.edu>, ha...@aztec.asu.edu (CHARLES J.
MICHAELS) writes:
Hi Charlie,
The 450 ohm ladder line I bought is around 375 ohms.
It doesn't matter much to me, because in my applications it's unmatched
anyway.
////
73 Tom
FORREST GEHRKE
MS> outperforms your 135 foot high dipole into Europe. I use a 4square
MS> on 80 and while it's hard to beat into Asia, I have found many
MS> nights when my horizontal loop would beat it into Europe.
Merv,
I find this very hard to believe. Are you sure you have
your 4square working correctly? What do you have for
a ground screen? More to the point: what are the self
impedances of your verticals? This will tell me more
about your ground screen than a description, unless
you are using radials significantly longer than a
quarter wave length.
//
k2bt
* RM 1.3 02583 * Wading bird wearing lingerie: Victoria's Egret.
at...@imap1.asu.edu
W8JI Tom (w8j...@aol.com) wrote:
>The 450 ohm ladder line I bought is around 375 ohms.
>
>It doesn't matter much to me, because in my applications it's unmatched
>anyway.
>
I believe Tom's measurements, but I also find this sort of difference
and others I have heard reported to be amazing. I assume that this is
450 window line, with a polyethylene insulation. If not, ignore
everything after this. I don't own any 450 ohm line, but maybe I'll
buy some and measure it out of curiosity. :-)
Parallel conductor lines with insulation where the dielectric constant
changes from one value in the insulation to another in air do not
operate as exactly TEM mode lines. In addition having the windows
always seemed to me to be really stupid, (since the line parameters
shift back and forth between two values every few inches), but still,
quasi-TEM mode analysis should be good at HF.
In that case, the characteristic impedance of the line should
be Z0=sqrt(L/C) where L is the inductance per unit length, and
C is the capacitance per unit length. The velocity factor
of the line is vf = 1/(c*sqrt(LC)), where c is the velocity of light.
Combining these, I get:
Z0 = vf*c*L
Since the polyethylene is not magnetic, the inductance per unit length
is the same as for air insulated lines:
L = mu_0/pi ln(b/a)
where b is the wire spacing and a is the radius of a wire, with the
usual result that (in ohms)
Z0 = vf*120 ln (b/a)
This gives a simple way to measure Z0 even without an rf bridge. You
only need to measure the wire radius, spacing, and the velocity
factor.
The log term is very insensitive to changes. If I assume that the
velocity factor is 1, I get for Z0 = 450 ohms, b/a = 42.5 while for 375
ohms, b/a = 22.8. So you have to make a mistake of a factor of 2 in one
of the dimensions to get 375 ohms. Nobody's quality control should be
that bad.
For pure polyethylene filling all space, vf = .66, which gives for Z0 =
450, b/a = 293, and for Z0 = 375, b/a = 113, and you need a mistake
of a factor of almost 3 to get to 375 ohms.
An upper bound on the capacitance per unit length is given by assuming
that the dielectric fills all space, so the true velocity factor has
to be between .66 and 1, unless the plastic is not polyethylene.
I assume that the conductor geometry is kept reasonably constant (and
as seen above, the spacing or the wire radius has to vary by at least a
factor of 2 to make the characterisitic impedance off by 20 percent.)
The ARRL antenna book says that 450 window line is made from 18 gauge
wire. From the copper wire table this is .02015 inch radius, separated
by an inch, which gives:
Z0 = vf*120*ln(1/.02015) = vf*469
The velocity factors I have seen reported for window line are around
0.95 which gives Z0 = 446. If the line has a Z0 of 375, it would again
seem that something strange is happening to get Z0 to change by 20
percent. The velocity factor has to change to around .8 to lower the
impedance to 375 ohms, indicating that the 0.95 is wildly wrong.
Alternatively, perhaps the manufacturer misread the formulas and
thought that the radius needed to be twice as large. For example, 12
gauge wire has almost twice the radius of 18 gauge, and with 1 inch
spacing and .95 velocity factor would give a characteristic impedance
of around 366 ohms.
Is it the geometry or the velocity factor that is wrong?
73 Kevin w9...@ptolemy.la.asu.edu
Pete Soper
w8j...@aol.com (W8JI Tom) writes:
>In article <4qfs2e$f...@news.asu.edu>, ha...@aztec.asu.edu (CHARLES J.
>MICHAELS) writes:
>>
>>Pete said about ladder line -
>>I'd thought about it, but dismissed it. So this is the missing factor?
>Ladder
>>line with thicker conductors have different insulation characteristics to
>get
>>the impedance back to the target?
>>
>>No, Pete;
>> The insulation is to hold the line together , the conductor
>>diameter is then selected to provide the specific impedance in the
>>environment of the insulation.
>Hi Charlie,
>The 450 ohm ladder line I bought is around 375 ohms.
>It doesn't matter much to me, because in my applications it's unmatched
>anyway.
I was lent a copy of Moxon's book "HF Antennas For All Locations" and
chapter four, "Feeding the Antenna" had exactly the information I was
after to understand these issues properly. This formula is the key:
Zo = sqrt( L/C )
Where the L and C can be thought of as series inductors and shunt
capacitors running the length of the line.
Next he says that Zo is inversely proportional to sqrt(K), K being the
dielectric constant of the line's insulation. So as the insulation
goes from air to thick plastic the impedance goes down, all other
things being equal.
My conclusion then is that there is likely to be a lot of variability
among same-sized ladder lines based on their conductors going from
18 to 14 gauge and a smaller variability with differences in
insulation thickness and proportion of the line open via its windows.
I should go for both the smallest conductor and thinest insulation and/or
largest amount of window cutouts to get the highest possible Zo and
lowest weight (my original goals). Making my own open wire line is of
course the guaranteed way to get what I want but I'm not convinced
I can make something that will last well with the periodic hula dancing
my trees do.
What gauge conductors are in your 375 ohm ladder line, Tom? My guess is
they are #16 instead of the more common #18 gauge.
(I did see Roy Lewallen's remark that Zo goes down when C goes up but
my Usenet server then went on a 5 day coffee break. And I fully expect
that the formula above will jump out and slap me the next time I crack
open the ARRL Antenna book.)
Regards,
Pete
KS4XG
W8JI Tom
>
>The velocity factors I have seen reported for window line are around
>0.95 which gives Z0 = 446. If the line has a Z0 of 375, it would again
>seem that something strange is happening to get Z0 to change by 20
>percent. The velocity factor has to change to around .8 to lower the
>impedance to 375 ohms, indicating that the 0.95 is wildly wrong.
>Alternatively, perhaps the manufacturer misread the formulas and
>thought that the radius needed to be twice as large. For example, 12
>gauge wire has almost twice the radius of 18 gauge, and with 1 inch
>spacing and .95 velocity factor would give a characteristic impedance
>of around 366 ohms.
>
>Is it the geometry or the velocity factor that is wrong?
>
>73 Kevin w9...@ptolemy.la.asu.edu
>
>
Hi Kevin,
The line I bought was advertised as heavy duty stranded 450 ohm line.
The conductors measure a "little" on the fat side of .08 inches and the
spacing is .81 inches c/c average.
Another 450 ohm line I have uses .0395 inch diameter conductors spaced .85
inches center to center.
Looks like they used the same insulation extrusion die with a different
gauge center conductor. I think the real problem is no one ever even knew
the formulas, they must have thought two wires spaced about .85 inch is
always 450 ohms.
At least it wasn't # 6 wire!
73 Tom
W8JI Tom
Hi Pete,
>
>My conclusion then is that there is likely to be a lot of variability
>among same-sized ladder lines based on their conductors going from
>18 to 14 gauge and a smaller variability with differences in
>insulation thickness and proportion of the line open via its windows.
>I should go for both the smallest conductor and thinest insulation and/or
>largest amount of window cutouts to get the highest possible Zo and
>lowest weight (my original goals). Making my own open wire line is of
>course the guaranteed way to get what I want but I'm not convinced
>I can make something that will last well with the periodic hula dancing
>my trees do.
I am "tickled pink" to get lower than 450 ohm line. A lower impedance is
better in my applications.
The low Z line will have lower losse, be less susceptable to weather
effects, last longer, and handle more power.
With the antennas I usually use, it gives me less impedance excursions
with frequency...and keeps the feedpoint impedance more constant. I'm glad
the manufacturer had no idea what he was doing. I just hope he makes "300
ohm" line soon, it'll make a good 75 ohm feedline, hi..
73 Tom
at...@imap1.asu.edu
W8JI Tom (w8j...@aol.com) wrote:
>Looks like they used the same insulation extrusion die with a different
>gauge center conductor. I think the real problem is no one ever even knew
>the formulas, they must have thought two wires spaced about .85 inch is
>always 450 ohms.
>
>At least it wasn't # 6 wire!
>better in my applications.
>
>The low Z line will have lower losse, be less susceptable to weather
>effects, last longer, and handle more power.
>
>With the antennas I usually use, it gives me less impedance excursions
>with frequency...and keeps the feedpoint impedance more constant. I'm glad
>the manufacturer had no idea what he was doing. I just hope he makes "300
>ohm" line soon, it'll make a good 75 ohm feedline, hi..
is accurate even for large diameter wires in parallel conductor lines:
Z0 = 60*vf*ln[(1+sqrt(1-(d/s)**2))/(1-sqrt(1-(d/s)**2))]
where s = spacing, and d = diameter of the wires, and vf= velocity
factor. The rf resistance is proportional to 1/d.
If you fix the spacing between the wires, the ratio of the power loss,
I**2*R, to the power flowing, I**2*Z0, is minimized when d/s is about
0.55. This corresponds to a characteristic impedance of about vf*145
ohms or for vf=.95 about 138 ohms. Therefore if they use a spacing of
0.85 inches, the minimum loss occurs with a wire diameter of 0.47
inches, and a characteristic impedance (assuming vf = .95) of about 138
ohms. Since 1 gauge wire is "only" 0.2893 inches in diameter, even it
would give lower matched loss (by decreasing the resistance more than
the characteristic impedance is reduced) than any of the smaller
diameter wires at this spacing.
73 Kevin w9...@ptolemy.la.asu.edu