Booty Monster's Vector 40000 thread on Eham

[Marconi]

Marconi, I've tried to explain the inaccuracy I see with EZNEC and the Sigma design many times. It comes down to exactly what you see each time you experiment. The programs complete inability to identify significant radiation currents combining from the cone.

If you still think EZNEC is right, go build a collinear version with an additional 1/2 wave radiator on top. When the design peaks its gain in EZNEC, build the prototype and field test it. You will find there is no improvement in gain whatsoever. How can that be if we have added another 1/2 wave of in phase radiation?

It cannot be and the problem is EZNEC peaks its gain on the horizon when the phasing section is 100% too long. Not 10 or 20% off but 100%. EZNEC will guide you to construct a 180 degree phase delay between sections because it sees the cone as nothing more than transmission line.

The proof that the radiation from the cone is significant comes from understanding that EZNEC's phasing section places 50% of the added collinear 1/2 wave in a phase that is bucking the constructive radiation from the cone. If that radiation on the cone was weak, the entire top half wave would absolutely become dominant and there would be gain in the field using the 180 degree delay.

For there to be no improvement in gain with a 180 degree delay is very strong evidence that the cone radiates nearly as much energy as if it were a free standing 1/4 wave.

Henry, Thank you for your explanation after looking at the CST video. I think I'm now understanding what you mean as this crossover point in phase changes in step with the phase the antenna is driven in. I can see this effect in CST.

Donald, you're right!

Eznec does force a 1/2 wave reversing stub to make a collinear Vector model work. But, that is the way I thought such a setup would work when two 1/2 wave radiators are stacked vertically, end to end, with a small separation between them.

I've always thought we need two 1/4 wave wires, one across each end of the insulator space that are shorted together at the far ends, thus we see a 1/2 wave length wire connecting the ends of the two radiators. Am I wrong?

Note: The following models are at 36' feet. I also include a pattern overlay of my Vector that I posted above. I also used this Vector model to add the other 1/2 wave on top.

View attachment Collinear Vector.pdf

So, are you suggesting that a 1/4 wave stub is what should be used to force current reversing between closely spaced vertically stacked 1/2 wave radiators?

If you're saying such a setup, using a 1/2 wave stub, will not show any collinear gain advantage on the horizon, when tested in the real world, or are you saying that Eznec will not model such a setup correctly.

I'll have to take your word for it that this won't work, I have no way to test such an idea in the real world.

 

[Shockwave]

Donald, you're right!

Eznec does force a 1/2 wave reversing stub to make a collinear Vector model work. But, that is the way I thought such a setup would work when two 1/2 wave radiators are stacked vertically, end to end, with a small separation between them.

I've always thought we need two 1/4 wave wires, one across each end of the insulator space that are shorted together at the far ends, thus we see a 1/2 wave length wire connecting the ends of the two radiators. Am I wrong?

Note: The following models are at 36' feet. I also include a pattern overlay of my Vector that I posted above. I also used this Vector model to add the other 1/2 wave on top.

View attachment 9379

So, are you suggesting that a 1/4 wave stub is what should be used to force current reversing between closely spaced vertically stacked 1/2 wave radiators?

If you're saying such a setup, using a 1/2 wave stub, will not show any collinear gain advantage on the horizon, when tested in the real world, or are you saying that Eznec will not model such a setup correctly.

I'll have to take your word for it that this won't work, I have no way to test such an idea in the real world.

I think you're understanding what I'm saying now or at least very close. Using a phasing section that is an electrical 1/2 wavelength long provides a 180 degree phase shift or a complete inversion of phase. This is appropriate when stacking multiple 1/2 wave radiators.

The Sigma is a 270 degree 3/4 wavelength radiator that already has a crude method of phase correction taking place at the base with it's confined radiation inside the cone. Therefore, a collinear version with a 1/2 wavelength phase delay causes the top section to radiate out of phase with the cone.

The proof of this can only be found in the field when you ignore EZNEC and cut the phase delay in half. Then the collinear version will go from having no gain advantage and a messed up match to about 2 db over a the original with a perfect match.

The difference between 1/4 wave and a 1/2 wave phase delay is a very significant miscalculation that only becomes apparent when experimenting with collinear versions. More important is to understand that this error is not because we made the antenna into a collinear.

The root of the problem is that EZNEC cannot model the original design accurately and ignores beneficial radiation on the cone. If EZNEC were right and a 1/2 wavelength delay was required, it would also mean the Sigma was just a 1/2 wave. Because the gain peaks on the horizon in the field with a 1/4 wave or 90 degree delay, proves the original antenna is more than a 1/2 wave of in phase radiation.

 

[Marconi]

makes perfect sense to me

just out of curosity ...........
how does the toa of a 3/4 wgp with drooped ground elements compare to a 1/4 wgp with sloped ground elements at the same feed-point ? say ..... 18 - 27 ft. .

BM, here is an overlay of a 3/4 wave with slanted down radials, compared to one with slanted up radials, and a Starduster...all at 18' feet to the feed point.

 

[BOOTY MONSTER]

thanks OGP . i was curious about the 1/4wgp or the 3/4wgp having a lower toa . but it certainly shows slanting the ground elements upwards making a significant difference in the pattern/TOA in the blue/green 3/4 plots .

can you do two more ? a vector type 1/2 wave/18 ft vertical . and a 1/4 wave/9 ft vertical version with the vertical just little over the the top of the basket/up-swept ground elements . each compared with a starduster and a .82 vector . thanks

i honestly am lost in this conversation at this point , but i do like pretty pictures ! hehehe . the one you did above gave us a chance to see how the angle of the ground elements/basket effects the vertical . and this would allow seeing how vertical length effects the basket . i also wonder how just the basket and 9 ft vertical turned upside down (shades of the astroplane) does . ....... or maybe i'm just having too much fun today

thanks

hey shockwave , can you do the same with your CST program so we can compare the differences ?

thanks 2u2

 

[Marconi]

thanks OGP . i was curious about the 1/4wgp or the 3/4wgp having a lower toa . but it certainly shows slanting the ground elements upwards making a significant difference in the pattern/TOA in the blue/green 3/4 plots .

can you do two more ? a vector type 1/2 wave/18 ft vertical . and a 1/4 wave/9 ft vertical version with the vertical just little over the the top of the basket/up-swept ground elements . each compared with a starduster and a .82 vector . thanks

i honestly am lost in this conversation at this point , but i do like pretty pictures ! hehehe . the one you did above gave us a chance to see how the angle of the ground elements/basket effects the vertical . and this would allow seeing how vertical length effects the basket . i also wonder how just the basket and 9 ft vertical turned upside down (shades of the astroplane) does . ....... or maybe I'm just having too much fun today

thanks

hey shockwave , can you do the same with your CST program so we can compare the differences ?

thanks 2u2

I can't be sure I made this like you suggested, but here the Vector with 9' and 18' foot radiators.

I didn't turn the 9' model upside down however, because it shows no gain, and there is no radiation resistance in the model. The direction of the antenna would probably make no difference to anything.

I did add a Starduster at 36' feet to the pattern overlay.

View attachment Booty's .50w Vector idea.pdf