Question on the Vector

[Marconi]

7.67 dBi gain at 14 degrees, the antenna is a horizontal center fed dipole one wavelength high and has a second higher angle lobe.




I could tell they were pretty close to a match just from the pictures above... There was no reason that they shouldn't have been a match...




So you want a dipole made of two wires, I'm assuming equal length wires? The feedpoint won't be exactly centered so I doubt currents will match exactly, or should I end feed it?


The DB

I use an Eznec split feed point feature to get the feed point in the center of the model I posted. I think if you make the 4Nec2 model using two elements it automatically sets the split if it is in the center.

Setting the FP on one of the wires will make no difference to the gain or angle, but it will make a small difference in the current distribution...and you can see that in the currents log.

DB, my only hope here is to see if 4Nec2 handles the current distribution on a two wire dipole with two end 1's, or two end 2's, connected...just like my Eznec does. If it does then you can change the end of one leg and make an end 1, connected to an end 2, and check it out with your software and report.

 

[Lil'Yeshua]

Picture presentation
http://fmbroadcastantenna.com/images/Dominator NWE-34 in CST.gif

 

[The DB]

I'm curious, what is the point of posting that here now? Everyone that has posted in this thread has seen that image, and many of us have referred to it over the years...

Bob and Donald see it as evidence for common mode currents, however I see something completely different, I see the three dimensional fields generated by the currents on the different elements, including the vertical element. If the cause was common mode currents are as they say I think the animation would look very different.

I can simply ask, if common mode currents are causing the two different polarity fields to appear on either side of the radials, what is causing the same thing to happen on the vertical element where we have no common mode currents? This is the key to my concern and part of why the common mode currents explanation makes me uncomfortable.

That is also not to say I'm against entertaining the idea of common mode currents theory, if you will. I have no qualms with whatever comes out of that line of thinking, an advance is still and advance. On the same note, I also know that the first idea that someone comes up with isn't always correct, even if it explains something beautifly. I can point to many times that has happened since the birth of modern science.

The Newtonian Laws of Physics, for example. They work well as long as you stay in an environment that we have on the surface of the Earth, however, if you try and use them to explain how the earth revolves around the sun they fail miserably. I can point to countless examples from recent history that tell you the same thing.

I personally think that the theory of common mode currents on the Vector antenna design has its flaws, which opens me up for other possible explanations that others who treat it as absolute fact don't even consider. While I, personally, believe that this theory is wrong, that does not mean that the gain they think it is causing doesn't exist, I simply think the gain comes from other means.

As a hint of what I think is actually happening, the currents that people are so focused on, to me, are only indicators of the three dimensional charged fields that they generate. Those fields are evident on the animated picture shown, one side of any given element has one charge and the other side has the other. By themselves currents can only tell you so much It is in the study of the fields they generate that I think the truth of the additional gain this antenna design has lies.


The DB

 

[Lil'Yeshua]

I felt like posting it. I don't think that it's an accurate presentation. It would be nice to have an computer generated 3D animation that can be manipulated. I don't think that there is a equation that explains this type of propagation. Final thought. I really posted the animation for our guests and members as well for encouragement to rethink this model.

 

[Shockwave]

DB, the different polarity fields are not a left side and right side thing. That's just the way a model displays the phase. I think you know that. It's an inside the cone or outside the cone comparison. The only reason we have a phase shift anywhere is because the vertical element is 3/4 wave. You could remove the cone and the phase difference is still present at the base of the 3/4 wave.

It's only because it is wrapped with the cone around it that we get a phase delay without undesired radiation from the 1/4 wave inside the cone. That is what brings the vertical into phase with the cones normal "radial radiation" and any CMC that is present on the cone.

I don't suggest that all gain comes from CMC, just that any CMC adds to the gain and is mostly radiated from the cone. Offsetting the resonance of the cone can cause tremendous CMC issues to appear.

Lil'Yesusa, please feel free to elaborate on the evidence that led you to believe the CST model is inaccurate more so than any opinion. Hold your thoughts about any manipulation of a model until you can provide anything that contradicts what has already been provided. An "I think" is not appropriate at this point. Something starting with "I found" is required to dispute current models.

 

[The DB]

Perhaps I described that part poorly, but...

NOTE: There is a long version and a short version here. If your not Shockwave and want just the short, non-technical version then skip to the last paragraph.

1) I am aware that the apparent opposite phased fields on either side of the element is in fact the same field, but to say that is just how the model shows the field is simply wrong. There is a reason that the field is represented that way, and inductors, electro-magnets, and anything that uses a coil of wire use the exact same effect to work.

2) What you are using to explain the fields around the radials, namely common mode currents, is also happening with the central vertical element, but there that is just how it works? I'm sorry, but the exact same thing is shown in both cases and you are treating them inconsistently. If you want me to accept such an inconsistency you need to have one hell of a good explanation as to why. You are in essence saying that one of the apples is an orange, even though it looks, feels, and tastes like the other apples around it.

3) I'm well aware of the phase change that happens an electrical 1/2 wavelength from the outer tip of any given length of wire/tubing. The use of the additional near 1/4 wavelength section on the vertical element of this antenna is nothing more than a way to position the top half wavelength in an optimal spacial location to benefit the rest of the antenna while keeping the out of phase lower portion surrounded by the cone. The additional full 1/4 wavelength section beyond the phase change is not needed for the phase change to exist, for that you simply need an element that is longer than 1/2 electrical wavelength. There are, however, other implications to this phase change, and honestly, it is just as important to my version of how the antenna works as it is to yours.

For example, this phase change would be needed for upward pointing radials, assuming no common mode currents on said radials, to be in phase with the upper portion of the central vertical element.

In the case of common mode currents being present in the cone this phase change actually complicates things. You now need the inner and outer parts of the cone, and the mast, to add up to over 1/2 electrical length. The animated picture made by CST shows the radiation on the inside of the radials to be in phase with the radiation from the lower central element. To get that to happen you need that phase change to happen at a specific spot on the antenna, and that spot is very close to the feedpoint. This setup would drive the feedpoint impedance of the antenna up as you essentially have four 1/2 wavelength radials. This might also cause the radials to act as the dominant current handlers of the antenna as antenna mode currents seem to like 1/2 electrical wavelength multiple elements more than any other length.

You could use the lengths of the inner and outer cone and the mast added together to potentially put the phase change at the tip of the radials. This would have several advantages over the phase change being near the feedpoint, including a lower impedance at the feedpoint, and more gain, which would be needed to get anywhere close to the 2 dBd gain that Sirio claims. However, the phase change being at the tip of the radials would cause the CST image to look very different, actually making the currents on the inside of the radials in phase with the currents on the outside of the radials, and out of phase with the lower central element of the antenna. The CST plot disagrees with this entirely as it clearly shows the inside of the radials as being in phase with the lower vertical element.

Sorry for going so long Donald, but long story short, when I really look at what you are saying, there are inconsistencies with your common mode currents theory of what is happening and the CST model you present as proof. I also point out that Sirio didn't specifically claim that common mode currents were involved, or even present on the antenna design. From everything that I have read on the topic that is an idea that was conceived independently (if I'm wrong please link to the post that I missed). Also, the arguments (or perhaps they are better called observations), which I think you fell back on way to quickly in your post, such as the phase change that is happening in the vertical element, are not only not in disagreement with how I think the antenna as a whole works, but required.


The DB

 

[Shockwave]

Perhaps I described that part poorly, but...

NOTE: There is a long version and a short version here. If your not Shockwave and want just the short, non-technical version then skip to the last paragraph.

1) I am aware that the apparent opposite phased fields on either side of the element is in fact the same field, but to say that is just how the model shows the field is simply wrong. There is a reason that the field is represented that way, and inductors, electro-magnets, and anything that uses a coil of wire use the exact same effect to work.

Do you mean there is a difference in phase between one side of the vertical and the other? I understand phase as changing along the length of the radiator and not from one side to the other.

2) What you are using to explain the fields around the radials, namely common mode currents, is also happening with the central vertical element, but there that is just how it works? I'm sorry, but the exact same thing is shown in both cases and you are treating them inconsistently. If you want me to accept such an inconsistency you need to have one hell of a good explanation as to why. You are in essence saying that one of the apples is an orange, even though it looks, feels, and tastes like the other apples around it.

I must confess I don't totally understand how CMC affects the vertical element. I thought these currents flowed on the outside of transmission lines and not on the vertical radiator itself. I evidence in the CST model and from the engineer who built the model that CMC are in phase on the cone and transmission line. Since the cone is only 1/4 wavelength, most of the currents radiate there. That same phase continues along the first portion of the mast and coax. I'm not sure if I misspoke or made a mistake before but I'm not sure where the inconsistency is here.
The builder of the CST model explained that the small amount of current descending on the mast (or coax) would add another small benefit to the antennas gain. How does the phase shift taking place inside the cone complicate the any CMC current since it appears to be in the constructive phase. It appears the CMC current would start at the top of the cone since it has to fold over the outside of the "transmission line".
It looks like the path would be through the 1/4 wave radials and any residual CMC on the mast should follow this same constructive phase over the first 1/4 wavelength of mast and coax. That is a much better condition than having the first section of mast or coax bucking the phase of the antenna. Again, it seems the only thing responsible for making this change in phase is the fact the vertical has been phase delayed by 90 degrees. I see no change in phase of the radials in the cone or CMC currents, only in the main radiators phase.

3) I'm well aware of the phase change that happens an electrical 1/2 wavelength from the outer tip of any given length of wire/tubing. The use of the additional near 1/4 wavelength section on the vertical element of this antenna is nothing more than a way to position the top half wavelength in an optimal spacial location to benefit the rest of the antenna while keeping the out of phase lower portion surrounded by the cone. The additional full 1/4 wavelength section beyond the phase change is not needed for the phase change to exist, for that you simply need an element that is longer than 1/2 electrical wavelength. There are, however, other implications to this phase change, and honestly, it is just as important to my version of how the antenna works as it is to yours.

For example, this phase change would be needed for upward pointing radials, assuming no common mode currents on said radials, to be in phase with the upper portion of the central vertical element.

In the case of common mode currents being present in the cone this phase change actually complicates things. You now need the inner and outer parts of the cone, and the mast, to add up to over 1/2 electrical length. The animated picture made by CST shows the radiation on the inside of the radials to be in phase with the radiation from the lower central element.

I agree with your assessment based on the video CST produced. I was expecting to be able to find frames in the video that would show the inside surface of the cone to have the same phase as the outside, at least right up on these elements. It is not apparent there but I will elaborate more on this idea below.

To get that to happen you need that phase change to happen at a specific spot on the antenna, and that spot is very close to the feedpoint.

That is what you we happening inside the cone. The phase change takes place starting right at the feedpoint and continues through the inside of the cone.

This setup would drive the feedpoint impedance of the antenna up as you essentially have four 1/2 wavelength radials. This might also cause the radials to act as the dominant current handlers of the antenna as antenna mode currents seem to like 1/2 electrical wavelength multiple elements more than any other length.

You could use the lengths of the inner and outer cone and the mast added together to potentially put the phase change at the tip of the radials. This would have several advantages over the phase change being near the feedpoint, including a lower impedance at the feedpoint, and more gain, which would be needed to get anywhere close to the 2 dBd gain that Sirio claims.

You may be onto something now. I've been assuming the source of the CMC had to be the top of the cone. Where this current begins is very important to determine its phase with respect to the rest of the antenna. Let's go one step further and forget about CMC and just look at the normal radials currents with this same thought in mind.
Suppose the electrical length of the cone is not just 1/4 wave and the currents have to travel up the inside surface and down the outside surface? There is another 90 degree offset in phase occurring here in that case. Is this offset in the right direction to correct the collinear problem? Another thought I had was could the 3/4 wavelength radiator have an affect on the phase of any currents going on at the base cone like it changes the phase at the base of the main vertical?

However, the phase change being at the tip of the radials would cause the CST image to look very different, actually making the currents on the inside of the radials in phase with the currents on the outside of the radials, and out of phase with the lower central element of the antenna. The CST plot disagrees with this entirely as it clearly shows the inside of the radials as being in phase with the lower vertical element.

Now we are in the gray area a little because I can't see opposing currents inside the cone but at the same time this would have to be the case if part of the function is to act as a transmission line to the 1/2 wave. I can't be sure if that is due to the opposing fields having "pushed" the radiating currents to the outside as it would if the principles were simply magnetic or if the currents are truly that different on the two sides of the cone.

Sorry for going so long Donald, but long story short, when I really look at what you are saying, there are inconsistencies with your common mode currents theory of what is happening and the CST model you present as proof. I also point out that Sirio didn't specifically claim that common mode currents were involved, or even present on the antenna design. From everything that I have read on the topic that is an idea that was conceived independently (if I'm wrong please link to the post that I missed). Also, the arguments (or perhaps they are better called observations), which I think you fell back on way to quickly in your post, such as the phase change that is happening in the vertical element, are not only not in disagreement with how I think the antenna as a whole works, but required.

The DB

The engineer did comment that any currents descending on the first part of the mast would add a small gain as mentioned before. Currents on the mast would have to be CMC. Do you not agree? The only reason we don't have the J-pole CMC issues in this design is the cone. It's fairly logical to assume it must have a majority of those characteristic end fed currents radiated from the cone.
If you make enough error in the resonant length of the cone you'll get RF burns on your lip from your mic. Not to mention every change in coax length or movement of the cable can begin to affect the SWR. Properly tuned and drastic changes here make just about no change in the most sensitive measuring equipment. Going to the extent of removing the meter movement from the directional coupler and replacing it with a millivolt DVM.
As I recall Bob has also indicated CMC are responsible for some of the cones radiation. I think I'm guilty of putting too much emphisis on the CMC current as there is more than one current on the cone.

 

[Marconi]

I felt like posting it. I don't think that it's an accurate presentation. It would be nice to have an computer generated 3D animation that can be manipulated. I don't think that there is a equation that explains this type of propagation. Final thought. I really posted the animation for our guests and members as well for encouragement to rethink this model.

Lil'Yeshua, I think that the Sirio CST image of the New Vector 4K may be accurate, but I don't believe the interpretation that Donald and other's suggest (common mode currents on the outside of the cone) is correct either.

Such thinking is convenient for their idea, but IMO it does not fit the fact that there are only two out of phase currents flowing in the radials and the base of the monopole inside of the cone.

I do not believe the cone is coaxial (forming a tapered transmission line as suggested) and I don't believe the cone produces Common Mode Currents on the outside of the radials. As soon as the physical construction of the coax center conductor and the shield is altered in any way...these two wires will begin to radiate, and the longer the wires are the more radiation will occur. This cone is not coaxial. CMC's may be present on the feed line attached below this antenna, but the taper of the cone does not simulate coax above the feed point. Just do an Internet search for coaxial pig tails and you will see what happens if the proper construction ratio for coax is altered.

My Eznec model for my Sigma 4 shows only two out of phase currents flowing in the elements in the area of the cone, and these two currents are very close to having identical magnitudes. When two elements are close together and either in series or parallel and showing an out of phase current condition we will see far field cancellation. This is what we see between the radial elements and the center monopole of the S4.

We also see this cancellation in 5/8 wave ground plane antennas with horizontal radials. We see this cancellation in the S4 style radials on the S4. The only far field radiation will be the small difference that remains between these out of phase elements, and the benefit can be either destructive or constructive...depending on the phase with the strongest total magnitude of current. So, if the 1st segment in each radial shows (-) .33 amps per element for a combined total of (-) .99, and the 1st segment of the monopole shows (+) 1.05 amps...then the results will be (+) .06 amps difference. If the top 1/2 wave element is (-) phase then this (+) .06 amps will be destructive and thus substracted from the max current in the radiator above. Either way the difference is not much.

I see very little far field radiation from the cone on my Sigma 4 model. Depending on how I place the source (feed point) on the radiator, this difference can result in either destructive or constructive addition to the top 1/2 wave element. Either way this result will still be very small.

If you model this antenna with NEC type software you must add a radial hub at the base of the radiator, the segment count for both radials and monopole must be similar in size and count, and you cannot use the taper scheduled noted in the Manual for the elements...you must average the element diameters as close as possible for CB band if you are using the dimensions in the Manual.

If you do this correctly, your model should show a complex impedance that shows resonance with a very low reactance at 27.205 mhz...the middle of the CB band.

Below is my S4 model with the pattern expanded to better show the currents on the elements. You will see the phase difference in the base section at the bottom of the monopole vs. the top 1/2 wave radiator. The currents on the radials are in phase with the top 1/2 wave radiator. The current in the 1st segment of the monopole is a fraction greater, as noted above compared to the sum of the three radials.

Also note that the magnitude of currents note in red on the radials are greater at the base than at the top...and this is the way the current distribution for a 1/4 wave element will look. If these currents on the radials were CMC emanating from the top of the radial elements...these current patterns would be the exact opposite of what we see here.

 

[Shockwave]

Marconi, I patiently await the day for ANYONE to build a collinear version and field test it. When you find there is nothing you can do to make it work other than cut the phase delay in half to compensate for the "little" cone radiation, you'll instantly learn just how significant the radiation form the cone is. Little is a relative term you used to describe the radiation displayed from a faulty model. To many, 2dbd is a tremendous amount to achieve in a non directional.

Every field test you can imagine has been conducted to determine the cones impact on overall gain. The collinear test proves the 1/4 wave cone has virtually the same gain as any other 1/4 wave. Fields tests show when you use the 180 degree phase delay, 50% of the added 1/2 wave bucks the phase of the antenna below. It now takes the second half of the top collinear section just to return to the same gain as the stock antenna.

This shows an equal length of radiator is required to compensate for the 1/4 wavelength of cone radiation that has been ignored. The exact degree of phase EZNEZ has miscalculated in EVERY Sigma model we have seen thus far. Something to consider each time you complain the field tests are wrong in favor of a model that has easily been proven inaccurate.

 

[Marconi]

Marconi, I patiently await the day for ANYONE to build a collinear version and field test it. When you find there is nothing you can do to make it work other than cut the phase delay in half to compensate for the "little" cone radiation, you'll instantly learn just how significant the radiation form the cone is. Little is a relative term you used to describe the radiation displayed from a faulty model. To many, 2dbd is a tremendous amount to achieve in a non directional.

Every field test you can imagine has been conducted to determine the cones impact on overall gain. The collinear test proves the 1/4 wave cone has virtually the same gain as any other 1/4 wave. Fields tests show when you use the 180 degree phase delay, 50% of the added 1/2 wave bucks the phase of the antenna below. It now takes the second half of the top collinear section just to return to the same gain as the stock antenna.

This shows an equal length of radiator is required to compensate for the 1/4 wavelength of cone radiation that has been ignored. The exact degree of phase EZNEZ has miscalculated in EVERY Sigma model we have seen thus far. Something to consider each time you complain the field tests are wrong in favor of a model that has easily been proven inaccurate.

Donald, I await that day too. You have already claimed to have done such testing, but you have posted no evidence for what you speak about in that regard. I said nothing about your idea of doing a test of a NV4K with a 1/2 wave radiator above it. You can talk about your testing, but you can't prove any of it.

I do try and provide some support for my ideas however, but you have none...just words.

I responded to Lil'Yeshua and just gave my opinion and added my Eznec model...is that OK?

 

[Shockwave]

Donald, I await that day too. You have already claimed to have done such testing, but you have posted no evidence for what you speak about in that regard. I said nothing about your idea of doing a test of a NV4K with a 1/2 wave radiator above it. You can talk about your testing, but you can't prove any of it.

I do try and provide some support for my ideas however, but you have none...just words.

I responded to Lil'Yeshua and just gave my opinion and added my Eznec model...is that OK?

Marconi, that is sort of nonsense when you look back over the last few years. Not only have I proved the cone is an effective 1/4 wave radiator, I've detailed the specific test required to replicate the same results that show the 90 degree offset in phase between EZNEC and CST or real world tests.
Several hams have been sure it's just a 1/2 wave J-Pole and you know the same 4 wire test was posted in other forums where some claimed they would put it to the test. Notice we never hear back from them? I realize it may not be easy to do on 11 but it's simple on 2.

I don't want you to be confused that I'm opposed to sharing ideas or responding to other individuals. That's not the case at all. The issue is you consistently base and support your ideas with a model that is easily proven inaccurate in the field. The facts are there for anyone willing to put them to the test but you ignore this to accept the model that fails to simulate field tests.

Am I "OK" with that? Honestly, no but I think that's obvious by now.

 

[Marconi]

Marconi, that is sort of nonsense when you look back over the last few years. Not only have I proved the cone is an effective 1/4 wave radiator, I've detailed the specific test required to replicate the same results that show the 90 degree offset in phase between EZNEC and CST or real world tests.
Several hams have been sure it's just a 1/2 wave J-Pole and you know the same 4 wire test was posted in other forums where some claimed they would put it to the test. Notice we never hear back from them? I realize it may not be easy to do on 11 but it's simple on 2.

I don't want you to be confused that I'm opposed to sharing ideas or responding to other individuals. That's not the case at all. The issue is you consistently base and support your ideas with a model that is easily proven inaccurate in the field. The facts are there for anyone willing to put them to the test but you ignore this to accept the model that fails to simulate field tests.

Am I "OK" with that? Honestly, no but I think that's obvious by now.

Ok, Donald so your idea is the only one that counts. If your idea about this antenna design were true, like you claim, it would be the hottest antenna on the market, but hardly anybody ever talks about their NV4K, and the idea would seem to scale nicely into other bands as well...and I have never heard of this design outside of 10-11 meters and your FM product of course.

You suggest that even Cebik would have considered this antenna to be a far superior design...but I find nothing in his body of work on the subject.

W8JI, has recently updated his article on "End-fed Vertical and J-Pole," and he demonstrates ideas using slanted up, slanted down, and horizontal radials on a J-Pole. He doesn't say much, but he did not make any claims to increased gain except where so noted and it wasn't much gain anyway. He also may have added this idea of slanted up radials to the J-Pole as a result of the eHam thread that Booty Monster posted and you mentioned above. If you recall he did comment in that thread at one point.

He did comment that this idea could be make coaxial by using a pipe in a similar configuration, but did not show an example and gave no further details.

He also says nothing about this radial idea creating a collinear effect that increases gain on the J-Pole.

Here is the link: http://www.w8ji.com/end-fed_vertical.htm

You are right about one thing using these Antenna Software design programs...if you don't get the antenna design right as to the physical design we know works, and consider the product limitations...no model will likely give good results. I noted about this in my remarks above...so you will possibly know why Eznec did not work for you in your efforts to model your S4 idea. I'll bet you did not consider any of those design ideas back then when you made Eznec models.

 

[Shockwave]

It's likely that many people do not follow what I'm talking about when I say the model is off by 90 degrees or that the collinear EZNEZ model has two 1/4 wave sections of exposed radiator that have opposing phase as a result. However, there are other analogies that relate much the same way and are easier to understand.

Think of the 3/4 wave Sigma as 3 equal fans all blowing in the same direction to create a combined wind speed of 15 MPH. Each fan representing one 1/4 wavelength of the antenna. Now I have two more fans that represent the next 1/2 wave collinear section. I want to use these two fans to increase the MPH. What's the first thing I need to know in order to do this? The compass bearing or direction the original 3 fans are pointing in.

Think of that compass bearing being the phase delay required to add the top 1/2 wave in my collinear test. If I don't know what the other 3 fans are doing, I can't add two more in an effort to increase power. If I don't know the exact phase on each element of the Vector, I can't accurately predict the degree of phase shift required to stack the next collinear 1/2 wave either. That's the EZNEC problem.

Essentially the results show one of the two fans we added in the EZNEC model are blowing in the opposite direction and canceling one of the 3 below resulting in the same 15 MPH wind speed. Once we figure out the correct degree of phase delay on the antenna or compass bearing to point all 5 fans in, we get closer to 25 MPH in the field tests.

The concept is further complicated by the typical requirement of a 180 degree phase delay being needed to stack two 1/2 waves. Trouble is the Sigma is not a 1/2 wave and we didn't just stack one 1/2 wave on top of another. We are stacking one 1/2 wave onto an existing 3/4 wave that has already had the base currents phase shifted by 90 degrees inside the cone with respect to the source. Since the normal radial currents and the CMC one 1/4 wavelength from the source have had no change in phase, starting the center verticals radiation 90 degrees later creates the offset needed to make them constructive.

 

[Shockwave]

It's likely that many people do not follow what I'm talking about when I say the model is off by 90 degrees or that the collinear EZNEZ model has two 1/4 wave sections of exposed radiator that have opposing phase as a result. However, there are other analogies that relate much the same way and are easier to understand.

Think of the 3/4 wave Sigma as 3 equal fans all blowing in the same direction to create a combined wind speed of 15 MPH. Each fan representing one 1/4 wavelength of the antenna. Now I have two more fans that represent the next 1/2 wave collinear section. I want to use these two fans to increase the MPH. What's the first thing I need to know in order to do this? The compass bearing or direction the original 3 fans are pointing in.

Think of that compass bearing being the phase delay required to add the top 1/2 wave in my collinear test. If I don't know what the other 3 fans are doing, I can't add two more in an effort to increase power. If I don't know the exact phase on each element of the Vector, I can't accurately predict the degree of phase shift required to stack the next collinear 1/2 wave either. That's the EZNEC problem.

Essentially the results show one of the two fans we added in the EZNEC model are blowing in the opposite direction and canceling one of the 3 below resulting in the same 15 MPH wind speed. Once we figure out the correct degree of phase delay on the antenna or compass bearing to point all 5 fans in, we get closer to 25 MPH in the field tests.

The concept is further complicated by the typical requirement of a 180 degree phase delay being needed to stack two 1/2 waves. Trouble is the Sigma is not a 1/2 wave and we didn't just stack one 1/2 wave on top of another. We are stacking one 1/2 wave onto an existing 3/4 wave that has already had the base currents phase shifted by 90 degrees inside the cone with respect to the source. Since the normal radial currents and the CMC one 1/4 wavelength from the source have had no change in phase, starting the center verticals radiation 90 degrees later creates the offset needed to make them constructive.

 

[Marconi]

I was feeling good today, but I'll have to consider your last post tomorrow.

Good night Donald.