best antenna

[abcd567]

I'll pop out this weekend and get some coax of known VF.

The simulation results of two cocos, one with VF=0.66 and other with VF=0.83, show that coco built with coax of higher VF, and hence longer element length, gave higher gain & better radiation pattern.

Higher VF means longer element length, which in turn means the shield is closer to 1/2 wavelength (shield VF=1 as it is outside the insulation).

[abcd567]

........The SAT 100 cable's core and screen is non-magnetic (so pure copper?) I do have a short length of RG6 U which is magnetic so must have some iron in it? .......

Coax cable Core comes in 2 varieties:
(a) Copper
(b) Copper-Covered Steel.

The Shield also comes in two varieties:
(a) Copper foil & Copper braid
(b) Aluminum foil & Aluminum braid.

RG6 almost always has FPE insulation & Copper-Covered Steel core and aluminum shield (aluminum foil+aluminum braid).
RG58 almost always has PE insulation & copper core and copper shield.
RG59 is manufactured in both the above varieties.

But being magnetic or non-magnetic does not affect performance of CoCo.

[trigger]

The simulation results of two cocos, one with VF=0.66 and other with VF=0.83, show that coco built with coax of higher VF, and hence longer element length, gave higher gain & better radiation pattern.

Higher VF means longer element length, which in turn means the shield is closer to 1/2 wavelength (shield VF=1 as it is outside the insulation).

This may be a silly question but here goes anyway:

The insulation slows down the radio waves so the better the insulation the lower the Velocity Factor. To get coax whose VF is closer to 1, it needs poor insulation. I have a length of coax fly-lead which has PE (translucent) insulation of ~3.2mm diameter. Should I try to make a CoCo from this? Any suggestions for the element length?

[abcd567]

This may be a silly question but here goes anyway:

The insulation slows down the radio waves so the better the insulation the lower the Velocity Factor. To get coax whose VF is closer to 1, it needs poor insulation. I have a length of coax fly-lead which has PE (translucent) insulation of ~3.2mm diameter. Should I try to make a CoCo from this? Any suggestions for the element length?

(1) The conclusion in my post was drawn from comparison of results of two simulations by 4nec2 software. After failure of experiment of twin lead, the accuracy of complex simulations by 4nec2 has become doubtful, unless confirmed by experiment.

(2) Answer to your question (which is not silly): I think by "better the insulation" you mean insulation with higher dielectric constant, since the velocity of radio waves decreases when the insulation's dielectric constant increases. So poor insulation (= lower Dielectric constant) results in higher velocity of propagation & higher VF. Air should therefore be categorized as "poorest insulation" as it has lowest dielectric constant =1, and highest VF=100%.

Below is the quote from document by Belden (one of best coax manufacturers) :

VELOCITY OF PROPAGATION
• The transmission speed of an electrical signal through a length of cable compared to the speed of light in a vacuum.
• Expressed as a percentage of the speed of light.
• High Velocity % = Lower losses.


DIELECTRIC CONSTANT
• Electrical property used to determine Capacitance, Velocity of Propagation, Impedance and relative performance of the insulating material.
• Symbol: Er
• Very important for electronic cables
• Lower Er = Lower capacitance, Higher impedance, Lower attenuation
• Air is best dielectric (Er = 1)
• Adding air to (foaming) any material lowers its dielectric constant.

Please see Belden's document at http://www.belden.com/docs/upload/In...ns-Jackets.pdf
The above quote is from Pages 10 & 11 of the document.

[trigger]

(2) Answer to your question (which is not silly): I think by "better the insulation" you mean insulation with higher dielectric constant, since the velocity of radio waves decreases when the insulation's dielectric constant increases. So poor insulation (= lower Dielectric constant) results in higher velocity of propagation & higher VF. Air should therefore be categorized as "poorest insulation" as it has lowest dielectric constant =1, and highest VF=100%.

Yes, that was what I was trying to say

[abcd567]

(2) Answer to your question (which is not silly): I think by "better the insulation" you mean insulation with higher dielectric constant, since the velocity of radio waves decreases when the insulation's dielectric constant increases. So poor insulation (= lower Dielectric constant) results in higher velocity of propagation & higher VF. Air should therefore be categorized as "poorest insulation" as it has lowest dielectric constant =1, and highest VF=100%.

Yes, that was what I was trying to say

The quote from Belden's document says exactly the opposite i.e. "Air is the best Insulation".
Actually all your assumption were right, only you categorized them wrongly, by thinking that "the insulation which slows down more is better". Belden says the insulation which slows down more, has more losses/attenuation, hence poorer.

I think you categorized good & poor based on "dielectric strength" i.e. the breakdown voltage per mm thickness. Yes, PE is better than FPE, which is better than Air when it comes to insulation breakdown. But "dielectric strength" has nothing to do with Velocity & attenuation. It is another characteristic "dielectric constant" (also known as "permittivity") which plays a role here.

[abcd567]

@ trigger:

Hi Dave!
As you did not accurately know VF of your coax, you adopted trial-and-error method by making 7 cocos of different element lengths, homing-in to 91mm, which gave very good results, almost same as your precisely built, impedance matched Franklin.

Your strategy, though tedious, was scientific & practical.

I decided to do the same thing with simulation of SAT100 coco.
I ran computer simulation of SAT100 8-element coco for various element lengths between 85mm & 105mm, and homed-in to one length which gave excellent results:
Gain = 8.2 dB, SWR = 1.27, radiation pattern = horizontal lobes ∞ (lazy 8).

So your experimental result that "8-element coco made of SAT100 PE gives excellent result" tallies with my simulation results.
BUT there is a problem:
The element length you arrived at by experiment is 91mm, and the element length I arrived at by simulation is 100mm.

This shows that the simulation of complex antenna, like 8-element insulated conductor coco, by 4nec2 software gives results which are partially correct & partially wrong. What a pity! The 4nec software gives very accurate results for wire antennas like Franklin, dipoles, Yagis etc, but seems to go erratic with antennas involving cylindrical surfaces, like coco.

simulation results:

100mm element length, 8 element SAT100 coco
Daves 8 Element-SAT100 PE-100mm long.PNG


91mm element length, 8 element SAT100 coco
Daves 8 Element-SAT100 PE-91mm long.PNG

[trigger]

Hi ab cd,
Many thanks for your results, very interesting! As the weather today is thick mist, I have an excuse not to do any gardening so I'll make another 100mm CoCo and compare results with my 91mm.

[abcd567]

Hi ab cd,
Many thanks for your results, very interesting! As the weather today is thick mist, I have an excuse not to do any gardening so I'll make another 100mm CoCo and compare results with my 91mm.

Thanks Dave,
The comparison of results of 100mm with 91mm will help in clarifying weather or not 4nec2 software gives erratic results for antennas involving surfaces, insulation & large number of elements.

[trigger]

I've done several back to back short tests today using the frame counter in adsbScope. There is a definite drop in frame count with the 100mm compared to the 91mm. All tests carried out without an amplifier. I also checked a 107mm 8 element CoCo and this was worse than both the others. I'm going to cut the 107mm down to 100mm and check the repeatability of construction.

I'll be back .....

[abcd567]

I've done several back to back short tests today using the frame counter in adsbScope. There is a definite drop in frame count with the 100mm compared to the 91mm. All tests carried out without an amplifier. I also checked a 107mm 8 element CoCo and this was worse than both the others. I'm going to cut the 107mm down to 100mm and check the repeatability of construction.

I'll be back .....

Thanks...... Your initial results are as I expected...The 4nec2 results for CoCo do have an error (about 10% as 91mm x 110% = 100mm)

[abcd567]

I have now done another test .... I pot on trial run the coco with twin-lead for some time, then collapsed the twin-lead to almost zero, and there was no noticeable difference in reception. This experiment plus Dave's experiment with his new two cocos, shows that the results given by simulation software 4nec2 are erroneous for:
(1) optimum length of 4 & 8 element coco with insulated core, has about 10% error.
(2) the conclusion that excellent impedance match is achieved by twin-lead, is wrong.

I will remove the posts for twin lead, or put a note there so others don't get allured into trying it and then get frustrated by its failure.

Now we are back to square one!

If
(1) One knows the VF of his coax
and
(2) calculates element length correctly
and
(3) cuts and assembles it accurately

he gets a very good coco.

(2) & (3) are not so difficult. The worst part is (1), i.e. accurately knowing the VF.
Most will already have coax length in their house, but no specs.
Most store will sell good coax, but they don't have specs either.
and the coax are available in VAST variety with vast difference in Velocity Factors.

So most of coco builders end up in a situation where they work as a blind watch maker.

[abcd567]

Unless some other solution is found, "Rooster"s advice below is good one.

.......buy some new cable of known spec !

...or just "suck it and see" and wonder what all the fuss was about ;-) lol

But how many novice understand importance of using a coax of known VF? They will just make a coco with whatever coax they have or buy, and will cry when they get poor reception.

[abcd567]

I know of only two novice DIY coco makers who started with a coax of unknown VF, and when got poor reception, did not cry or give up. Instead they tried several cocos with slightly different element lengths, till they homed in to correct length.

These two noviceI know are trigger & me. We both had to make and try over half a dozen cocos to reach correct length (91mm for trigger's coax & 114mm for my coax).

But how many novice will have patience to go through this tedious process?

[abcd567]

A recent example of failure of DIY coco by a novice:

The gentleman found this design, and was very confident & excited about it:

http://www.arrl.org/files/file/QST/This Month in QST/January 2014/VirtualRadarJan2013QST.pdf

The instructions clearly mentioned RG6 coax to be used with element length 4.5 inches (114 mm). He did not realize the importance of VF and cable specs. He had a large quantity of RG59, so he built his coco using RG59, but used element length specified in the instructions, which was for RG6. He ended up in a poor coco.

RG59 is manufactured both with PloyEthylene & FoamedPolyEthylene. I gave hime photos of both versions & explained the difference between the two. He confirmed his RG59 insulation is PolyEthylene. I told him he should then try 91mm element length, as PE has VF=0.66.

But he was so much discouraged by the 1st coco's failure, he did not bother to built another coco with different element length. Instead he straight away made a Cantenna, enclosed it inside an empty 100 DVD case, put it up, and is very happy with it. He has now ordered an amplifier also!