Modified Vector 4000

[bob85]

I wish I could help Eddie,

in my minds eye there seems to be a conflict of interest in using a tl model to simulate coax with is losses & matching stub and having to add a 3rd wire to simulate the outside of the braid so it carries current & radiates,

if you were terminating the coax with a resistor the coax braid would not radiate so no need for an extra wire, you would just see the coax loss & matching loss due to loss in the stub & high vswr between top end of coax and the stub,

but that's not how a coax dipole works, its not terminated with a resistor,
its highly unbalanced causing the outside braid to carry all the return current

the choke only inserts reactance in the outside of the braid & there is no outside to carry current without a 3rd wire when using a tl model to simulate coax,

imho if you add the 3rd wire from feedpoint down to just above the choke to simulate the radiating part above the choke you won't see any current on the choke or the reactance the choke inserts in series with the braid,

that extra wire will terminate in air & the choke won't be doing anything but make the coax longer,

I could be wrong but it seems you have a task on your hands.

Good luck figuring it out.

 

[Alexis Mercado]

By adding a line isolator, resonant frequency shifted down a little .

For example, without isolator, SWR is 1.1
( R=54, X= 6 ) on 27,095.

With isolator, SWR is 1.2 ( R= 50, X=12) on 26,915.

I want it to resonate close to channel 20 - 25.
Should I shorten the last section or should I extend the gamma ? Which the best method ?

What can I expect when I install it 35 feet above ground, about 4 feet on top of a 14 feet boom yagi ( Mosley TA-33 ) , SWR will go up or down ?

I am testing my Vector on a 10 feet pipe above ground. Pole is not grounded.

 

[Marconi]

Well Bob, I don't understand much of this business in such technical terms, but you could be right.

Even an end fed model with no radials is not the best idea for locating the Feed Point at the very bottom, another 2 wire insertion object.

So, with the GM model I set the feed point in the middle of the radiator not below the coil, and that will blow the minds of some...saying just more BS modeling.

I can't argue with that either.

I have no doubt the GM can't be modeled...I just don't think I can't do it. That is because it may take the type of understanding of the details like you have.

Even here you pushed me to look closer in my thinking and I will always be in your debt for that

 

[Marconi]

For example, without isolator, SWR is 1.1
( R=54, X= 6 ) on 27,095.

With isolator, SWR is 1.2 ( R= 50, X=12) on 26,915.

Alexis, you analyzer probably shows you either a + or a - in front of the reactance that is reported, right?

That (+ -) sign is important to know, report, and understand.

 

[Alexis Mercado]

I am using a MFJ- 269C Pro Analayzer. I don’t think it can read negative reactance. It didn’t show any + or - symbol. Just show a number.

 

[bob85]

Alexis, the 269 won't show the sign of X meaning if its capacitive reactance or inductive reactance,
you can work it out by adjusting frequency,

if increasing frequency a little reduces your X the antenna is capacitive.
if lowering frequency increases your X its capacitive

if increasing frequency increases X the antenna is inductive
if lowering frequency reduces X its inductive

if you are measuring the antenna through a length of coax you are not measuring the antennas resonance,
you are measuring the system resonace, coax & antenna, not just the antenna

unless the coax is an exact multiple of electrical 1/4waves at the frequency where the antenna is resonant the coax will add reactance that can cancel reactance in the antenna moving where you see X=0 to a frequency other than where the antenna is resonant,

tuning for resonance through coax is misusing the analyser, the mfj user manual tells you the same thing i posted above,

some people have stopped using good vswr meters in favour of analyzers & tune their antennas looking for x=0 through coax thinking the antenna is resonant & its better than than using a good vswr meter but its not better,
doing it that way its worse,

IF you tune through coax tune for lowest vswr on the frequency that you want,

or put the analyser at the feed-point with no coax or the shortest coax possible that still allows you to connect the analyser so the coax is not shifting X=0 away from where the antenna is resonant..

hope this helps.

 

[Marconi]

I am using a MFJ- 269C Pro Analayzer. I don’t think it can read negative reactance. It didn’t show any + or - symbol. Just show a number.

I checked that out and I noticed the MFJ and some other analyzers do not indicate whether the reactance is + or - in value. Then if the frequency is too low you need to shorten the antenna and visa versa,

Sorry! Keep us posted.

 

[bob85]

Eddie,
if you make a model of a 5/8 dipole then putting the source in the middle is where it should be,

on the other hand if you want to model a coax dipole like the GM to include coax loss & matching loss i think it has to go at the bottom under the choke.

 

[Alexis Mercado]

Thanks Bob. By chance do you know the formula to calculate the coax for an exact multiple of electrical 1/4waves ?

 

[Marconi]

You said it Bob, better than anybody else I know.

All this time and not one soul has asked me one thing about the GM model I made that did not look, even close to the real thing.

 

[bob85]

Alexis you can use the analyser to tune a piece of coax to 1/2wave better than using a formula that must include the coax velocity factor,

problem is it won't help with shifting resonace since you don't know where the antenna is resonant you can't know what frequency to tune your coax on.

 

[Marconi]

Eddie,
if you make a model of a 5/8 dipole then putting the source in the middle is where it should be,

on the other hand if you want to model a coax dipole like the GM to include coax loss & matching loss i think it has to go at the bottom under the choke.

Well Bob, I made the coil with 3 sides per turn and now I getting a non-corruptible segment-check error message, saying the extended ends of the helix axis I just created...are too short. It is not very much difference and I can fix it, but what happens with other buried and maybe sequestered type errors in such a model.

I've seen this type of error message before in a model of a Hex multi-band antenna that M0GVZ posted a while back. I asked him if he could explain the error, and he said he was surprised at the error, saying the model came from some popular modeling Website. So, I have no idea what it all means. This is the kind of problems that I worry about trying to do the GM with all the bells and whistles.

Here is my GM model, in progress of testing with the 8 wire per turn idea with the FL noted at the bottom 2 wires. These 2 top images also show the details for the new 3 wire coil setup with the TL on the bottom 2 wires.

It also shows the antenna image with the new coil I added lower down below with only 3 wires per turn and the error message it shows with the new coil. This coil is also asymmetrical in the wire lengths and that also might cause some problems with Eznec.

I also recall seeing this symmetry problem before with some model of the NV4k/Sigma4, trying to reduce the number of wires for the model.

 

[Alexis Mercado]

Alexis you can use the analyser to tune a piece of coax to 1/2wave better than using a formula that must include the coax velocity factor,

problem is it won't help with shifting resonace since you don't know where the antenna is resonant you can't know what frequency to tune your coax on.

Thanks Bob. I think I am going to use a very short coax jumper, like 2-3 feet long. But I have a question,

Me, being so close to the antenna, will it affect resonance while taking readings from the
analizer ?

Jeje, just asking.

Hey Marconi, have you modeled with Eznec something like this ?

 

[Dr_DX]

Alexis you can use the analyser to tune a piece of coax to 1/2wave better than using a formula that must include the coax velocity factor,

problem is it won't help with shifting resonace since you don't know where the antenna is resonant you can't know what frequency to tune your coax on.

???

That's easy. Pick the frequency that you want to tune your antenna to and use your analyzer to create a 1/2 wave multiple (resonant @ freq.) piece of coax that will reach to the bottom of the antenna. Screw it in to the antenna and then you can use an analyzer at the other end to tune your antenna to the freq. you chose knowing the coax won't be reactive and add to the system resonance.

"unless the coax is an exact multiple of electrical 1/4waves at the frequency where the antenna is resonant the coax will add reactance that can cancel reactance in the antenna moving where you see X=0 to a frequency other than where the antenna is resonant"

I am pretty sure it is 1/2 wave length not 1/4 wave length...

 

[Marconi]

Hey Marconi, have you modeled with Eznec something like this ?

Alexis, I don't know if a person with their hand holding a hand-help antenna analyzer will produce bad results in readings up close to the antenna or not. I've heard lots of claims made that tend to render these older analyzers, as useless as tits on a boar hog. So, when I hear any such claim, I do like you and try and check it out.

I've heard that a tuned 1/2 wave coax line appears invisible to the analyzer and the antenna when used. If that was true, then why don't we just make our feed lines odd multiples of a 1/2 wavelength. Oh! I've heard that too.

Here I put a 5' x 1.25" vertical pipe 5' feet to one side of an A99 with the GPK radials close to the Feed Point, This is to try and simulate your concerns noted above to Bob. This shows little effect on the match and gain.

1. is an A99 with NO pipe beside the feed point.
2. is an A99 with a pipe, as describe above, beside the feed point.