Base Should I stick it?

[JAF0]

Glad to see you back JAF0. Sometimes there are lulls in these discussions, but sooner or late they get active again. All you have to do is check in from time to time and respond to kick start the thread again.

I think the drift of my models of the S4 vs. New Vector 4K on adding horizontal radials suggest the NV4K does not respond as well in such cases. I think DB suggested the same, saying he did not see much difference adding radials to his Vector or a model he made.

Even so, I don't know for sure if I considered all that is necessary when I modeled your idea and since the delay in the discussion...I have gone on to other projects.

However, only real world testing this idea can say for sure, and even then the location and the installation can show different results in many cases. The only thing I can advise is keep your attention on the possible common mode currents problems that might come up with this antenna design, and try and keep the installation in the clear above and all around the install, it is reported by some to potentially have CMC issues.

If you decide to add horizontal radials...I would set the antenna up without them and get the antenna working right first...and then make your modifications, one step at a time. It is a lot of work but it might just pay off if you really what answers to your questions.

Over the years I have tested and compared my Sigma 4, but I've never had a NV4K to work with.

I will also add that I've watched closely on this and other forums...and I seldom see anyone with one of these type of antennas that gives any comprehensive performance reports that I thought reliable. Just saying it is the best antenna I ever owned or it stomped a hole in my Imax...is not reliable.

If you do this project, adding horizontal radials, please give us some good performance reports at points around your location, and if you do some comparisons to another antenna you have working...let us know how they compare in signals, noise, and maybe match.

Keep us posted on your progress and modifications.

JAF0, there is usually someone around that might be interested in what is said about these two antenna designs. I encourage you not to give up hope.

Marconi I like what I see and am impressed with your model showing an excellent pattern for the Sigma 4 with radials in your post #32 above.

I like the energy it appears to add over the NV4K in all the angles above 8-9 degrees. Looks like close to 3db in some angles while not losing anything in the lower 8 degree range to the NV4K.

That was part of why I was wondering what the shorter a99 radial kit would do because I've seen one of your models or one of the DBs models showing improvement on the Imax only with the shorter downward radials of the Imax kit.

I wonder if I should trust your model in post #32 to show what I'd actually get if I changed the NV4K to Sigma 4 spec with a longer radiator and a 90" cone with added 1/4 wave radials and hub from a Maco v58?

I don't mean I'm wondering if I can trust you & your modeling, I mean, can I trust the program to show me what I'd really be getting, enough to be worth the effort of pulling it down and adjusting?

It wouldn't be hard to extend the top section 8"-10" and slide the upper cone tubes down into the lower tubes about 15"-16".

Marconi you really have me thinking here. Might it be worth the effort?

 

[Marconi]

I like the energy it appears to add over the NV4K in all the angles above 8-9 degrees. Looks like close to 3db in some angles while not losing anything in the lower 8 degree range to the NV4K.

I agree with what you see in the nice pattern noted for my Sigma model, and you also notice the Vector is much different and IMO is lacking. However, if you go back and look at DB's pattern results for his Vector model in his post #18 in this link, Should I stick it?, you will see he gets similar patterns for his Vector to my Sigma4 model.

My Vector model is set to the specification for the Sirio New Vector 4K, and it is probably very different from DB's model. He also looks to direct feed his model at the base of the antenna, and maybe that allows his models to show noticeably more gain.

JAF0, the only way you will ever know for sure...is to experiment with your ideas on adding horizontal radials. Since you likely have a New V4K I would follow DB's suggestions. However, he did report seeing no difference with his Vector and that is similar to my claim that the Vector does not appear to respond to radials like the Sigma4 does with it smaller radial sleeve design. My model of the NV4K indicates some difference between adding horizontal radials vs. no radials, but the model with no radials shows me more gain.

The opposite looks to be true with my Sigma 4 models, but I cannot explain that at this point.

I would also recommend you first set-up your Vector to specifications and get it working with a good match and showing the bandwidth that is close to what Siro reports for this antenna.

And then modify the antenna by adding horizontal radials. Just keep us posted.

I have some old models of my Sigma4 and my NV4k set to specs and back then, 2015, I still see similar differences in the patterns for these two similar designs. If they check out close to accurate, have no glaring errors, and the Average Gain results in Free Space are good...I may post them tomorrow just so you have more for comparison.

 

[The DB]

My Vector model is set to the specification for the Sirio New Vector 4K, and it is probably very different from DB's model. He also looks to direct feed his model at the base of the antenna, and maybe that allows his models to show noticeably more gain.

Just for clarification...

My vector models are based on the dimensions that HomerBB gave me from his homemade version. I am not sure how close those specs are to the production version of the Vector antenna that I believe Marconi said he used. While not perfect, the HomerBB version of my models is one of the better set of dimensions I have used for said antenna. The only way from that model to get a significant increase in gain is to use a larger ring diameter, which most people don't tend to use.

My model above also includes a gamma match and is fully tuned. In my experience with modeling, adding such a matching network *can* change, at least to some extent, the current distribution on the model itself.

I'm not sure if the difference in dimensions or the matching system is the cause of the differences in our results. I would lean towards the potential difference in dimensions that we used being the largest cause of the differences we see as typically, when we model similar antennas we get fairly similar results.


The DB

 

[Marconi]

Just for clarification...

My vector models are based on the dimensions that HomerBB gave me from his homemade version. I am not sure how close those specs are to the production version of the Vector antenna that I believe Marconi said he used. While not perfect, the HomerBB version of my models is one of the better set of dimensions I have used for said antenna.

DB, I found a post (#226365) using the "search feature" where Homer talks about his using a 28.2' foot radiator (338" inches) and 13.75" inch mounting bracket below the hub.

He also suggested in the post that he started his build using the dimensions that Bob posted some time back. On that sheet I get the overall, from the bottom of the hub to the tip below the top hat ball, at roughly 317.50" inches when tuned near 27.205 MHz. These are the dimensions I use for my NV4K radiator models, and I think these are to specs. I also use 107" radials, set with a 30" inch diameter loop at the top. I also typically include the small top hat as well.

I figure that Homer probably started the project and later tuned his antenna and ended up with 338" inches in length using the materials, the construction, the gamma he used, and the height for his setup at the time. I also figure he is not including the top hat in his Vector. This said however, I'm not sure this post is referring to the dimensions you tell us you used.

In the post he also does not mention the radial length or the diameter for the loop he used...if he used a loop.

The only way from that model to get a significant increase in gain is to use a larger ring diameter, which most people don't tend to use.

Have you modeled and confirmed this claim?

Back in the days Bob or Donald one suggested that a larger loop would increase the gain and a discussion is noted in the patent on column #4 at lines 23 - 36 about the radial loop.

Long ago I made some models and I saw no increase in the gain. I just repeated this work with newer models and the models showed a very modest improvement with each 10" increase in the diameter of the loop up to 50" inches. The thing I also noticed was the match got better the wider the loop. The model with the 50" inch loop showed a near perfect feed point match, but that did not relate to much difference in the gain. What the patent warned about with increasing the loop diameter was the added footprint it caused.

My model above also includes a gamma match and is fully tuned. In my experience with modeling, adding such a matching network *can* change, at least to some extent, the current distribution on the model itself.

DB, I can't disagree and I don't use a gamma on this model. However I do set the feed point about 30" inches up the base of the radiator where the gamma connects.

This said, I see lots of antenna notables on the Internet showing us their model results and most often there are no indications that matching is included, and I think we should see a device inserted in the model somewhere near the feed point.

I'm not sure if the difference in dimensions or the matching system is the cause of the differences in our results. I would lean towards the potential difference in dimensions that we used being the largest cause of the differences we see as typically, when we model similar antennas we get fairly similar results.

DB, since you can tune your Vector model at the gamma or not tune it, maybe you could confirm if there is a notable difference in gain to be noted with and without the gamma?

I will default to your claim, but at this time and with this particular antenna design I don't believe the match is bad enough to make a big difference in the gain...whether the antenna model shows to be matched or not.

I too can make a wild guess that a noticeable difference in dimensions might be the problem with our models not showing more similarity in results, and IMO the match is not an issue.

 

[The DB]

DB, I found a post (#226365) using the "search feature" where Homer talks about his using a 28.2' foot radiator (338" inches) and 13.75" inch mounting bracket below the hub.

I don't remember if it was a private chat or in a public thread somewhere, but he posted everything including the radial length to the ring and the ring diameter. I would have to dig to find it...

Have you modeled and confirmed this claim?

Back in the days Bob or Donald one suggested that a larger loop would increase the gain and a discussion is noted in the patent on column #4 at lines 23 - 36 about the radial loop.

Long ago I made some models and I saw no increase in the gain. I just repeated this work with newer models and the models showed a very modest improvement with each 10" increase in the diameter of the loop up to 50" inches. The thing I also noticed was the match got better the wider the loop. The model with the 50" inch loop showed a near perfect feed point match, but that did not relate to much difference in the gain. What the patent warned about with increasing the loop diameter was the added footprint it caused.

Yes, multiple times with different versions of the antenna. You can't just make the ring bigger, you also have to adjust the length of the radial ring supports. The larger the ring the smaller the ring supports. When I did it before I had to make several small changed in succession to keep all the changes showing peak gain. I posted one of said models one that pushed the concept to its limit, but I think it was you (I could be mistaken) that then stated that it was not a Vector...

DB, I can't disagree and I don't use a gamma on this model. However I do set the feed point about 30" inches up the base of the radiator where the gamma connects.

This said, I see lots of antenna notables on the Internet showing us their model results and most often there are no indications that matching is included, and I think we should see a device inserted in the model somewhere near the feed point.

Aside from what the two of us have done when modeling various matching systems, such as both of our versions of the I-10k/Shockwave models, I haven't seen any models that includes any form of matching anywhere. I'm not sure why that is, perhaps people are happy with just the basic antenna layouts? I can't say that is good enough. I have noted, for example, that when I modeled a Maco V58, for example, when I include the matching system (or in that case my best attempt as that design is difficult for modeling to properly match) that the optimal antenna length that provides the most low angle gain is very close to the antenna's actual length. The model shows that extending the vertical element to a full 5/8 wavelength, the higher angle lobe has already become the dominant lobe at the cost of the lower angle lobe.

If others aren't even making the attempt, in my opinion that is their loss.

DB, since you can tune your Vector model at the gamma or not tune it, maybe you could confirm if there is a notable difference in gain to be noted with and without the gamma?

I will default to your claim, but at this time and with this particular antenna design I don't believe the match is bad enough to make a big difference in the gain...whether the antenna model shows to be matched or not.

I too can make a wild guess that a noticeable difference in dimensions might be the problem with our models not showing more similarity in results, and IMO the match is not an issue.

In most cases adding or removing a matching system does not effect gain that much. The antennas I've made with matching systems do tend to have slightly lower gain compared to their unmatched counterparts. This makes sense as antenna matching always adds some loss to the antenna. That being said, the losses so far are measured in 1/10'ths and 1/100's of a dB, so its not a major concern. The strange thing is, such matching systems sometimes seem to have more of an effect on the current distribution of an antenna than they do the antenna's gain, which is the primary reason I brought it up above when discussing adding radials to a Vector and seeing changes.


The DB

 

[Marconi]

I don't remember if it was a private chat or in a public thread somewhere, but he posted everything including the radial length to the ring and the ring diameter. I would have to dig to find it...

DB, I think Homer has told us he has built at least 3 versions of the S4/NV4K. So, I don't really need Homer's dimensions, but I was curious what dimensions you used in the model you are talking about here. So, don't bother trying to dig Homer's dimensions out of all this record. I think you said you modified his results anyway, so that gives us another set of possible dimensions.

Above you explained your observations quite well and I tend to agree. However, I did not change the radial lengths as I made the loop larger in diameter. I will try that later. Can you give me the loop diameter and the radials lengths where you saw the peak in gain?

Yes, multiple times with different versions of the antenna. You can't just make the ring bigger, you also have to adjust the length of the radial ring supports. The larger the ring the smaller the ring supports. When I did it before I had to make several small changed in succession to keep all the changes showing peak gain

You are right, because the loop adds inductance to the supporting elements (radial cone) and this allows for these radials to be made resonant using shorter lengths. This is also noted in the patent. So to test this idea, both the loop needs to be made larger in diameter while the radials are made shorter.

I agree that would be a lot of work for me using Eznec.

I posted one of said models one that pushed the concept to its limit, but I think it was you (I could be mistaken) that then stated that it was not a Vector...

I'm one to try and stick by these CB antennas modeled to the specs, but I don't recall the subject or the comment from the past...so I can't confirm if I said it or not.

This reminds me of Bob's recent thread on the new modified Vector idea by Vortex...where they extended the radial cone to 13' feet. Have you modeled that one yet? If so, maybe you could post your results and ideas in Bob's earlier thread.

Aside from what the two of us have done when modeling various matching systems, such as both of our versions of the I-10k/Shockwave models, I haven't seen any models that includes any form of matching anywhere. I'm not sure why that is, perhaps people are happy with just the basic antenna layouts

I have claimed that matching made little difference in performance results. I know that eliminating Common Mode Currents (CMC) from a model sometimes makes a lot of difference.

W8JI makes it pretty clear on his Website how CMC, not using a mast, feed line, or radials on a model can make a big difference in results. I'm just not sure about matching...mainly because I'm not completely sure about my efforts to match properly and very little is written or discussed on the issues of matching.

In most cases adding or removing a matching system does not effect gain that much.

The antennas I've made with matching systems do tend to have slightly lower gain compared to their unmatched counterparts. This makes sense as antenna matching always adds some loss to the antenna.

DB, I have see the same thing happen. I've also seen the gain sometimes go up and sometimes go down in models as I attempted to fix small Eznec errors. The same is often true when I fix errors made in dimensions.

Thanks for your responses.

 

[Marconi]

DB, here is Bob85's image of the New Vector 4000 that I use for the specs on my NV4K models.

Bandwidth is 1.60 MHz at 27.205 Range 26.400 - 28.000 MHz <2.00:1 SWR.

In case the dimensions are hard to read in the image below.
Radiator at 27.205 MHz is 317.5' x 0.75" with the Top Hat...or 322.5" x 0.75" without the Top Hat.
Radials are 106.5" x 0.50".
Radial Loop is 29.9" x 0.375" in diameter or 94" x 0.375 inches in circumference.
The mounting area is 13.75" x 1.456" in the model below the radial bracket.

If you do not include the Top Hat basket on top in your model I find adding 5" inches works well to restore resonance.

There may be better dimensions for this antenna, but I prefer to stick to specs. If I find better performance in a model that I modify or is suggested...then I try and make sure to note that in my post.

Models below.

1. Vector specs nM 122417 FS this model shows the Average Gain results in Free Space. I used this model to make the model below that is over Real Earth.

2. Vector specs wM 36' ISO 122417 this model has no Top Hat and is 5" inches longer at 322.5 inches. The model is over Real Earth using Average Ground.There is no gamma match included but is direct fed at some 50 ohm point above the bottom of the radiator. IMO, this is just like Henry tells us is possible in his article on " The Avanti Sigma 4."

3. Image of the New Vector 4000 with dimensions as Bob posted a while back. I added some notes for the dimensions I got from Bob's work.