An antenna experiment that Freecell once told me about.

[bob85]

i see a modified form of open sleeve or ssfm which according to respected sources does radiate from the sleeve in phase with the upper 1/2wave, no magic,

the current density plot in cst seems to suggest the current density is not as high as that in the upper 1/2wave, but what radiation is there is in phase with the upper 1/2wave,
yes im sure some of the gains over a shorter antenna are due to the height advantage of current maxima,

where does the sleeve radiation come from? we could look at what cebik says causes radiation from the short leg of wider spaced j-poles also look at transmissionlines with unbalanced currents or voltage differentials,

look at what conditions must be met in order for the sleeve to not radiate.

 

[Shockwave]

Don't be so quick to assume there is no "non apparent collinear" affect taking place here. I touched on this idea in another post but it may be useful here to. The typical collinear antenna utilizes a phase delay network to allow the undesired phase angles to pass through while radiating as little as possible through the phase delay line before driving the next section.

How is this so different in your mind then what is taking place in the Sigma? You know that vertical radiators longer then 1/2 wave produce inverted currents in the lower section of the radiator. One that's 3/4 wavelength will have the bottom 1/4 wave completely inverted. In this case, this is the location we want the phase shift to take place in.

I suggest to you that placing the cone around this lower 1/4 wave portion is forming a 90 degree phase delay inside the cone (transmission mode currents)before driving the top 1/2 wave. Since the currents on the 1/4 wave cone are 180 degrees out of phase with the first vertical 1/4 wave on the main radiator, it's now in phase with the top 1/2 wave.

We have a 1/4 wave phase delay inside the cone that feeds directly feeds a 1/2 wave above it. Because the cone is driven from the opposite side of the coax line, it automatically begins its first 1/4 wave of current 180 degrees out of phase from the center element. That's a wonderful thing since that puts it in the perfect phase to combine with the top 1/2 wave in the far field.

The effect of phase delay is present with a 1/2 wave and 1/4 wave radiator. While not apparent, I agree with Cebik in that the antenna has collinear properties and it sure can stir debate.

It would be very self informative for you to model the 3/4 wave ground plane I mentioned. Just take a standard 1/4 wave ground plane and extend the vertical radiator to 3/4 wave. Observe the field pattern with it's dominant 40 degree lobe at 1 wavelength in height. Then start bringing the angle of the radials upward more like the Sigma. That one change will start the collinear action and begin to reinforce the lobe on the horizon while it shrinks the 40 degree lobe. Try it.

 

[Marconi]

Marconi, the only question that comes to mind at the moment (at least until I see your email), is whether the point .625 and .75 that you refer to in the above post is the same as that to which SW referred. Are you both talking about these length in the overall lengths of the antennas, or from the cone ring upward.
When I built and mounted the Sigma antennas, whether 5/8, 3/4, or 7/8, I measured from the bottom of the antenna to the tip top fr those lengths.
Are you fellows on the same page here?

I was just guessing that you guys were talking about what would happen if we slanted radials up on a .625 wave radiator, and I thought he said it would not work.

I must have misread the words. Other than that, I don't know for sure if SW and I are ever on the same page. I like SW, and I don't know why that happens to be the case.

I was thinking 22.5' for the 5/8 wave with 4 x 105" or so, with slanted up radials.

To be honest Homer, things are going so fast between these two threads, I'm not sure what I thought SW was thinking or saying. I'll have to go back and re-read the posts.

The 5/8 wave with slanted up radials was not what my point was however, I was intending on trying to demonstrating more on my idea about how the S4 design might be working, and that the process may be as simple of what slanted up radials do, rather than this business with phasing and what have you that is so hard to understand.

No body is going to slant radials up on there 5/8 wave, so forget my idea, I wasn't trying to suggest that as an improvement. What I saw was not an improvement, in the sense it would be worthwhile to pursue. The process was also not destructive, like I thought SW was suggesting, as I recall.

I'll revisit the work and email you, since I think you asked for some info.

 

[Shockwave]

i see a modified form of open sleeve or ssfm which according to respected sources does radiate from the sleeve in phase with the upper 1/2wave, no magic,

the current density plot in cst seems to suggest the current density is not as high as that in the upper 1/2wave, but what radiation is there is in phase with the upper 1/2wave,
yes im sure some of the gains over a shorter antenna are due to the height advantage of current maxima,

where does the sleeve radiation come from? we could look at what cebik says causes radiation from the short leg of wider spaced j-poles also look at transmissionlines with unbalanced currents or voltage differentials,

look at what conditions must be met in order for the sleeve to not radiate.

With respect to the lower current density being radiated off the cone, there is little doubt this area produces less radiation then the upper 1/2 wave. However, we must not forget that current is divided into four radials and when we look at the CST model our eyes tend to focus right in on the radiator element itself.

It doesn't get any more near field then this. Look towards the outside edges and you can begin to get a sense of what will happen in the far field. The lighter shades of color further out begin to look more proportional with the top 1/2 wave then when looking at the currents directly on the individual radials. Granted, they are still weaker as they should be but it's not as bad as it looks at first glance.

 

[Shockwave]

I was just guessing that you guys were talking about what would happen if we slanted radials up on a .625 wave radiator, and I thought he said it would not work.

I must have misread the words. Other than that, I don't know for sure if SW and I are ever on the same page. I like SW, and I don't know why that happens to be the case.

I was thinking 22.5' for the 5/8 wave with 4 x 105" or so, with slanted up radials.

To be honest Homer, things are going so fast between these two threads, I'm not sure what I thought SW was thinking or saying. I'll have to go back and re-read the posts.

The 5/8 wave with slanted up radials was not what my point was however, I was intending on trying to demonstrating more on my idea about how the S4 design might be working, and that the process may be as simple of what slanted up radials do, rather than this business with phasing and what have you that is so hard to understand.

No body is going to slant radials up on there 5/8 wave, so forget my idea, I wasn't trying to suggest that as an improvement. What I saw was not an improvement, in the sense it would be worthwhile to pursue. The process was also not destructive, like I thought SW was suggesting, as I recall.

I'll revisit the work and email you, since I think you asked for some info.

Perhaps I was being pessimistic with the 5/8 wave Marconi. Because it is 1/8 wavelength over a 1/2 wave, there is a chance you will see some slight improvement slanting the radials a bit upward. Since the vertical portion on a 5/8 wave is only long enough to take advantage of a 1/8 wave delay at its base, the gain here will be much less then slanting the radials up on a 3/4 or 7/8 wave. This will also shift the drive impedance and require changes in matching that 5/8 wave.

 

[Marconi]

Don't be so quick to assume there is no "non apparent collinear" affect taking place here. I touched on this idea in another post but it may be useful here to. The typical collinear antenna utilizes a phase delay network to allow the undesired phase angles to pass through while radiating as little as possible through the phase delay line before driving the next section.

How is this so different in your mind then what is taking place in the Sigma? You know that vertical radiators longer then 1/2 wave produce inverted currents in the lower section of the radiator. One that's 3/4 wavelength will have the bottom 1/4 wave completely inverted. In this case, this is the location we want the phase shift to take place in.

I suggest to you that placing the cone around this lower 1/4 wave portion is forming a 90 degree phase delay inside the cone (transmission mode currents)before driving the top 1/2 wave. Since the currents on the 1/4 wave cone are 180 degrees out of phase with the first vertical 1/4 wave on the main radiator, it's now in phase with the top 1/2 wave.

We have a 1/4 wave phase delay inside the cone that feeds directly feeds a 1/2 wave above it. Because the cone is driven from the opposite side of the coax line, it automatically begins its first 1/4 wave of current 180 degrees out of phase from the center element. That's a wonderful thing since that puts it in the perfect phase to combine with the top 1/2 wave in the far field.

The effect of phase delay is present with a 1/2 wave and 1/4 wave radiator. While not apparent, I agree with Cebik in that the antenna has collinear properties and it sure can stir debate.

It would be very self informative for you to model the 3/4 wave ground plane I mentioned. Just take a standard 1/4 wave ground plane and extend the vertical radiator to 3/4 wave. Observe the field pattern with it's dominant 40 degree lobe at 1 wavelength in height. Then start bringing the angle of the radials upward more like the Sigma. That one change will start the collinear action and begin to reinforce the lobe on the horizon while it shrinks the 40 degree lobe. Try it.

I have plans to do as you suggest, in fact I have already done that idea, and what I saw is why I have this change of thought on the subject of phasing make the Sigma design work.

I can't exclude the idea you and Bob hold completely, because I tend to see what I think you and Bob see in the Sirio model, but I'm just not sure we are seeing phasing that is collinear in affect.

I tried to get an idea for NB to work the other day that was working with similar ideas, and it failed miserably as far as I could see, and it had real collinear phase shifting elements in the model. I realize just having such features in a model does not mean it will work right, so I didn't mean this statement that way.

The idea, in a practical way, is almost identical to what the Wolf .64 antenna does, and I can't get that model to work right either. I have perfect dimensions to go by with the Wolf, because I have the antenna to measure.

So, there is more work to be done and it is hard for this old man to keep up with all the ideas in the two threads working currently.

Maybe you can save me some time re-reading your words with Homer. Were you two talking about a standard .625 wave antenna with slanted up radials?
Just let me know if you can.

 

[Marconi]

Perhaps I was being pessimistic with the 5/8 wave Marconi. Because it is 1/8 wavelength over a 1/2 wave, there is a chance you will see some slight improvement slanting the radials a bit upward. Since the vertical portion on a 5/8 wave is only long enough to take advantage of a 1/8 wave delay at its base, the gain here will be much less then slanting the radials up on a 3/4 or 7/8 wave. This will also shift the drive impedance and require changes in matching that 5/8 wave.

Well, I don't understand this stuff nearly as well as you, so that is why I argue when I see things differently.

Again, you make statements that are open ended so you have a way out, I presume. I read these words as though you are indicating a negative effect will occur on the impedance when raising the radials on this .625 wave antenna. I find the very opposite in these models.

You post:

This will also shift the drive impedance and require changes in matching that 5/8 wave.

And this leaves the reader to guess what you mean and what result you are suggesting.

I'll tell you flat out, when I raised the radials on the 5/8 wave radiator I saw the natural match improve by about 50%, and that was without any matching device.

So, IMO the slanted down radial model would require 1/2 the transformational energy to match the antenna. I see that as a possible plus, and there was no difference in maximum gain angle, the gain dropped a very little bit, but the pattern was improved with the higher lobe generated by the 1/8 wave bottom portion (I presume) pushing the pattern down due to the slanted up radials, another presumption.

Here it is for you and Homer to mull over and pick apart. You will note that the radiator length is indicated by wire #2, for both models, and the caption for the horizontal model indicates the radial length. I used the same model to generate both.

View attachment Homer's .625 wave idea with slanted radials vs. horizontal.pdf

 

[Shockwave]

Well, I don't understand this stuff nearly as well as you, so that is why I argue when I see things differently.

Again, you make statements that are open ended so you have a way out, I presume. I read these words as though you are indicating a negative effect will occur on the impedance when raising the radials on this .625 wave antenna. I find the very opposite in these models.

You post:

And this leaves the reader to guess what you mean and what result you are suggesting.

I'll tell you flat out, when I raised the radials on the 5/8 wave radiator I saw the natural match improve by about 50%, and that was without any matching device.

So, IMO the slanted down radial model would require 1/2 the transformational energy to match the antenna. I see that as a possible plus, and there was no difference in maximum gain angle, the gain dropped a very little bit, but the pattern was improved with the higher lobe generated by the 1/8 wave bottom portion (I presume) pushing the pattern down due to the slanted up radials, another presumption.

Here it is for you and Homer to mull over and pick apart. You will note that the radiator length is indicated by wire #2, for both models, and the caption for the horizontal model indicates the radial length. I used the same model to generate both.

View attachment 8155

Sorry Marconi, I didn't mean to suggest this change in impedance was a bad thing. Just that the matching network on the 5/8 would have to be changed to accommodate the new lower impedance. It seems you may have swept the radials up too high in an attempt to match the impedance? Forget the impedance and try reducing the upward angle on the radials to around 45 degrees. I think you may find the 40 degree lobe is still reduced but it may also make a slight increase in gain if it behaves like the Sigma.

 

[Marconi]

Sorry Marconi, I didn't mean to suggest this change in impedance was a bad thing. Just that the matching network on the 5/8 would have to be changed to accommodate the new lower impedance. It seems you may have swept the radials up too high in an attempt to match the impedance? Forget the impedance and try reducing the upward angle on the radials to around 45 degrees. I think you may find the 40 degree lobe is still reduced but it may also make a slight increase in gain if it behaves like the Sigma.

I wasn't watching the match as I raised the radials. I didn't save all the iterations I did in the process, I was just looking to see what happened with the pattern and if it went to heck in a hand basket, due to the bottom being so small.

I didn't even check the degree of radial angle I used. I just adjusted the radials to make the 105" long and slanted up some. The first model adjustment I did to the .625 wave ground plane with horizontal radials was about 45* degrees. I saw it didn't destroy the pattern as I thought you might be suggesting. So, I advanced the radials up some more to see what happened, and I ended up with about 12* degrees. My New Vector model set to the dimensions that Bob gave us earlier, with the 107" radials and the 30" hoop, comes in at 8* degrees, so I was just trying to get close.

Here is the bottom of my New Vector showing the angle that it ends up with using the dimensions that Bob gave me. If it is not right, let me know.

View attachment Shockwave's look at my New Vector..pdf

I guess Homer was right, we were not on the same page. Sorry for the bad words I used.

Eddie

 

[Shockwave]

It does appear to match the physical dimensions of the new Vector. Nice work.