Solarcon A99 Experimental Results and Analysis of Ground Plane Radials

[The DB]

I redid the models using a mast that runs to the ground to show the differences it would make in relation to the models above. This is just one possible case for these antennas, and is only intended to explore some of the possible differences between these antenna designs.

Here we see the differences in feedpoint impedances. Both R and X changed more in the end fed design than in the Big Hair design. The change in reactance (X) in the end fed design was especially significant. Because of this it is much easier to design an efficient matching network that will function consistently for the Big Hair antenna than the end fed design.

Here we see fewer currents flowing in the simulated mast in the Big Hair design over the end fed design, although not much. This was expected and predicted above.

Here we see a slight improvement in gain for the Big Hair design over its equivalent with no mast, while the end fed design shows a loss. The difference between the two in this case is only 0.58 dB, which is not enough be noticed in the real world. Any change in the angle of maximum radiation is also minimal in this case.


The DB

 

[Marconi]

If you guys look close at DB's model with radials you will see that the radials are out of phase with the top 1/2 wave radiator, plus radials typically show cancellation due to the apposing radials producing currents that are flowing in the opposite direction, thus radials contribute very little to no RF into the far field.

The Big Hair's 1/2 wave over a 1/4 wave idea is nothing more than another way of adding slanted down radials to the radiator, very similar to what DB's model shows us.

BH's collinear stacked 5/8 wave idea with the shorted phasing stub setup attached between the two radiators may have some gain/angle advantage in the design however, but I doubt seriously in the real world...it will make the 7.5db gain claimed.

Driving around in a mobile to test a base antenna is fraught with many additional variable issues that must be considered, however you can get more dependable results if the mobile is stationary, and testing just one base locations.

It is not possible to consider all the things that might get in between the antennas at any point when moving around in a mobile. Even the Earth effects with a low profile mobile antenna can vary considerably within a few feet, and with vehicles and other type obstructions in the way as one passes by...the task can become fruitless.

Like the old say goes..."if the idea or results seem too good to be true, then.........."

 

[kaosfm]

What I wonder about that BigHair colinear 5/8 over 5/8 antenna is how the hell can that thing be self supporting? Just dealing with an 18' A99, I see the fiberglass elements flapping all around and transferring that energy down the mast. I can't imagine a 50' tall stacked 5/8 CB antenna that could possibly remain vertical in still air, let alone wind.

In regards to my testing methodology of mobile reception charted in each cardinal direction, you are correct, it is far from ideal. Yet, it is sufficient enough to display a statistically significant improvement. I wish I had the test gear required so that I could provide actual signal strength values in dBm at several different points, but I don't unfortunately. This leaves my results rather subjective. Regardless, I think I've proven that owners of this antenna would do well to add 1/4 wave radials to their own A99's if they want to maximize this antenna's potential.

Considering the low cost, ease of availability, proven reliability and reasonable broadbandness of the A99, I think it deserves a little more credit than the typical scorn I tend to observe in its regard, both online and even in discussions with other CBers on the air!

 

[Marconi]

What I wonder about that BigHair colinear 5/8 over 5/8 antenna is how the hell can that thing be self supporting? Just dealing with an 18' A99, I see the fiberglass elements flapping all around and transferring that energy down the mast. I can't imagine a 50' tall stacked 5/8 CB antenna that could possibly remain vertical in still air, let alone wind.

In regards to my testing methodology of mobile reception charted in each cardinal direction, you are correct, it is far from ideal. Yet, it is sufficient enough to display a statistically significant improvement. I wish I had the test gear required so that I could provide actual signal strength values in dBm at several different points, but I don't unfortunately. This leaves my results rather subjective. Regardless, I think I've proven that owners of this antenna would do well to add 1/4 wave radials to their own A99's if they want to maximize this antenna's potential.

Considering the low cost, ease of availability, proven reliability and reasonable broadbandness of the A99, I think it deserves a little more credit than the typical scorn I tend to observe in its regard, both online and even in discussions with other CBers on the air!

I agree there is a serious lack of feasibility in the BH stacked 5/8 wave collinear idea.

Like I said to you in my first post in this thread, I appreciate anybody who does some work to support their results.

I will look on YouTube for some of my own videos comparing my A99 to another antenna, but here is an operator named Penn in England comparing his Gain Master to his A99 using FM. This is pretty similar to what I see here.

Sirio Gain Master vs Antron 99 - YouTube

I hold my A99 in high regard, and at times if performed equal or better than others I've tested on the local scene and most of the locals I talk to here in my area use the A99.

 

[The DB]

If you guys look close at DB's model with radials you will see that the radials are out of phase with the top 1/2 wave radiator, plus radials typically show cancellation due to the apposing radials producing currents that are flowing in the opposite direction, thus radials contribute very little to no RF into the far field.

Just a correction, the colors are only showing magnitude, the radials in that picture are in phase with the vertical element. Here is that antenna again, shown left is currents, right is phase.

As we can see the currents flowing in the radials and the simulated mast are very small compared to the currents in the antenna itself.

On the phase picture above we can see a phase shift in the antenna near where the radials connect. This antenna is on the cusp of flipping the phase of the upper 1/2 wavelength radiator, and it wouldn't take much more length to cause that to happen, at which point the radials fight against the low angle radiation and changing the primary lobe to a lobe at near 45 degrees.


The DB

 

[The DB]

When it comes to the Big Hair antenna I'de love to set one up or even see one in person...

I'm guessing by the design that that upper radial is actually a phasing stub, but am not sure. I suppose they could run a coax to the upper section through the lower section as well... As I'm not sure how they are achieving the collinear between these methods it is difficult for me to model an antenna like it...

I would be curious as to the price but I'm sure I can't afford it at the moment, and I have yet to see directions for it, or anyone who has one with them putting it together on-line.

One thing I do know is they are not self supporting. If you look at the background images on their main page there are clearly guy wires attached...


The DB

 

[M0GVZ]

His design supposedly sports 3dB gain over a dipole alone by its lower angle of radiation.

Hmm. The cynic in me sees that as Market-twaddle.

Lowering the angle of radiation doesn't increase overall gain, just the angle of take off of the biggest lobe.

The cynic in me thinks that its one of two things:

1) He's comparing it to a vertical dipole so you lose the advantage of gain over ground and

2) what he is actually meaning is that at X degrees it has 3dB gain over a dipole. However as you raise a horizontal dipole its gain increases at lower take off angles.

 

[M0GVZ]

I would be curious as to the price but I'm sure I can't afford it at the moment, and I have yet to see directions for it, or anyone who has one with them putting it together on-line.

I wouldn't even bother. For local comms its not really going to be that much better and for DXing, for the cost its likely to be you can put up a beam. Even if the beam doesn't have much more gain the one thing it has which no vertical has is side and rear rejection so if you're trying to work a station in Asia for example, you don't have to try and listen through everyone else in CONUS to the sides and rear of you who are also calling.

With the UK being on the edge of the "wall of Europe" as we like to call the super loud Russian, Italian and baltic state amateurs in neighbouring countries who never seem to be anything other than at least 20 over, being able to drop their signals down 20 or 30dB when you're trying to work the USA is a godsend.

 

[Marconi]

Just a correction, the colors are only showing magnitude, the radials in that picture are in phase with the vertical element. Here is that antenna again, shown left is currents, right is phase.

As we can see the currents flowing in the radials and the simulated mast are very small compared to the currents in the antenna itself.

On the phase picture above we can see a phase shift in the antenna near where the radials connect. This antenna is on the cusp of flipping the phase of the upper 1/2 wavelength radiator, and it wouldn't take much more length to cause that to happen, at which point the radials fight against the low angle radiation and changing the primary lobe to a lobe at near 45 degrees.

The DB

Well DB it looks like I missed the mark on my earlier comments regarding what I saw with the radial currents and phase on your model and what it indicates.

However, my model indicates that 21 of the 24 segments assigned for each of the radials shows the opposite of your model in the radial area...approximately 80% of the segments in my model are out of phase with the 1/2 wave radiator.

Do you have any ideas as to why our models are so different?

Below are my A99 models at 32' where I tried to duplicate your models. I have attached my tabular Currents Data report to the model with radials, and I added some notes. BTW, I set the power for my model to 95 watts in order to try and get my current magnitudes close to what your model shows. They are close, but I don't think they always jive.

View attachment A99 with and without radials.pdf

I apologize to faosfm for getting of track in your thread.

 

[The DB]

Well DB it looks like I missed the mark on my earlier comments regarding what I saw with the radial currents and phase on your model and what it indicates.

However, my model indicates that 21 of the 24 segments assigned for each of the radials shows the opposite of your model in the radial area...approximately 80% of the segments in my model are out of phase with the 1/2 wave radiator.

Do you have any ideas as to why our models are so different?

Below are my A99 models at 32' where I tried to duplicate your models. I have attached my tabular Currents Data report to the model with radials, and I added some notes. BTW, I set the power for my model to 95 watts in order to try and get my current magnitudes close to what your model shows. They are close, but I don't think they always jive.

View attachment 12212

I apologize to faosfm for getting of track in your thread.

I mentioned earlier that if I lengthen the overall length of the 1/2 wavelength vertical that its polarization flips, and in my models I seem to be right on the edge of that flip. When playing around with my model any amount of extra length caused the upper half wavelength and the radials to be out of phase...

You might try shortening your vertical element some to see what happens... We may be teetering on opposite sides of said phase shift in our models...

One thing I noticed is my wire tends to be much thinner than yours, this may be the cause, at least partially, for the discrepancy between our models. My models run at 100 watts...

My .nec file if it is of any use to you...

CM 
CE
GW	1	25	0	0	11.03	0	0	16.545	.0001
GW	2	13	1.9498	0	9.0802	0	0	11.03	.0001
GW	3	13	-1.9498	0	9.0802	0	0	11.03	.0001
GW	4	13	0	1.9498	9.0802	0	0	11.03	.0001
GW	5	13	0	-1.9498	9.0802	0	0	11.03	.0001
GW	6	51	0	0	0	0	0	11.03	.0001
GE	1
GN	2	0	0	0	13	0.005
EK
EX	0	1	1	0	1	0	0	'Voltage source (1+j0) at wire 1 segment 5.
FR	0	0	0	0	27.185	0
EN

The DB

 

[Marconi]

I mentioned earlier that if I lengthen the overall length of the 1/2 wavelength vertical that its polarization flips, and in my models I seem to be right on the edge of that flip. When playing around with my model any amount of extra length caused the upper half wavelength and the radials to be out of phase...

You might try shortening your vertical element some to see what happens... We may be teetering on opposite sides of said phase shift in our models...

One thing I noticed is my wire tends to be much thinner than yours, this may be the cause, at least partially, for the discrepancy between our models. My models run at 100 watts...

The DB

I checked your idea of changing my length from 17.5' feet to 19' and the model did as you suggested.

I also made my radiator 16' feet and that made all the segments for the radials out of phase with the radiator like I was suggesting to the guys earlier. Making the radiator shorter also added .47dbi gain to the model. I wish I could check the effects on resonance, but the match is way too bad without a matching device, and I can't model that yet.

I recalled that you also tend to make you wire diameters very thin for your models. I made my model with .001 diameter wires, but I didn't see any change of note. The only thing it did was lower the amount of current magnitude for the radials a very little bit.

I can't remember if I've talked about your using a scan step rate of 5* degrees, but I use 1* degree. That too makes a notable difference in results as one might expect.

My radials are also 72" inches too. How long did you make your radiator in feet?

 

[The DB]

I checked your idea of changing my length from 17.5' feet to 19' and the model did as you suggested.

I also made my radiator 16' feet and that made all the segments for the radials out of phase with the radiator like I was suggesting to the guys earlier. Making the radiator shorter also added .47dbi gain to the model. I wish I could check the effects on resonance, but the match is way too bad without a matching device, and I can't model that yet.

I recalled that you also tend to make you wire diameters very thin for your models. I made my model with .001 diameter wires, but I didn't see any change of note. The only thing it did was lower the amount of current magnitude for the radials a very little bit.

I can't remember if I've talked about your using a scan step rate of 5* degrees, but I use 1* degree. That too makes a notable difference in results as one might expect.

My radials are also 72" inches too. How long did you make your radiator in feet?

1/2 wavelength radiator is at 18.093832 feet as translated from meters. The radials are very close to half of that length (very close to 108 inches), give or take hundredths of a foot or so in rounding... The base of the antenna is 36.187664 feet high, again converted from meters. If you calculate the lengths in meters they may also seem somewhat random, but were calculated from the 11.03 base wavelength of the frequency used, which was 27.19 MHZ. Ground is NEC's average ground. I have in the past, and will likely again use 27 MHZ in the future as it comes very close to the 11 meter even mark, which makes calculations easier to figure in meters, which seems to be my favored set of lengths when using 4NEC2.

I don't recall a mention of the scan steps, I have seen the settings for them, but have not changed them as of yet. Perhaps I should consider it in the future.


The DB

 

[Marconi]

1/2 wavelength radiator is at 18.093832 feet as translated from meters. The radials are very close to half of that length (very close to 108 inches), give or take hundredths of a foot or so in rounding... The base of the antenna is 36.187664 feet high, again converted from meters. If you calculate the lengths in meters they may also seem somewhat random, but were calculated from the 11.03 base wavelength of the frequency used, which was 27.19 MHZ. Ground is NEC's average ground. I have in the past, and will likely again use 27 MHZ in the future as it comes very close to the 11 meter even mark, which makes calculations easier to figure in meters, which seems to be my favored set of lengths when using 4NEC2.

I don't recall a mention of the scan steps, I have seen the settings for them, but have not changed them as of yet. Perhaps I should consider it in the future.


The DB

If you do lower your step rate it should allow for more smoothing of the lobe contours on the antenna pattern and maybe yield a bit more accuracy. It will also increase the data iterations, so the model may run a bit slower. You might not notice this however, with our CB antenna models not being complicated.

 

[kaosfm]

Hi Everyone!

I never could have predicted just how great the responses have been to this particular thread. Thanks so much for all of the valuable insights on the topic.

I wanted to share an update with you all that is part of another thread I recently posted with some relevance to this thread.

Since my A99 is presently tuned to the 11-meter band, it's performance on 10-meter is somewhat compromised, especially in the 29 MHz portion of the band. I also wanted to expand my antenna arsenal to include horizontal polarization. So I built a parallel (fan) dipole using 14 AWG insulated copper speaker wire soldered directly onto the center conductor and braid of some nice RG6 quad-shield. I also created a coax choke below the feedpoint to prevent the shield from radiating down the coax.

Because of installation constraints, I mounted the dipole onto the top wooden mast section immediately below the feedpoint of my A99. That puts the point of maximum radiation from the dipole up at ~28 feet. I used the inverted V configuration. I thought as an added bonus the inverted dipole sections would further add to my current ground plane configuration already proven effective with my A99 vertical.

Sure enough, the presence of the dipole had no negative de-tuning consequences for the A99. If anything, it slightly lowered SWR towards the top of the band. So, if anyone else wants to add a dipole or two to their current A99 installation, they can do so without consequence. A couple of pics are included below.

 

[Skynet 15]

well I know this post is years old, I recently just purchased an a-99 and I do love it. I have been debating on doing a radial kit to see if it benefit's. I know the 11m is pretty much dead at this time and ssb contacts are hard to get for me if not harder to get than my old homebrew. He's the pics, I'm around 28' or so.