Does height above Earth effect the common mode current response?

[Marconi]

The two models below are center fed dipoles with a mast attached. One is at 36’ feet to the feed point, and the other at 40.5’ feet.

I also show these same antennas without a mast, because this is what we generally have presented in books and reports we see. This may be why we read W8JI always complaining about the shortcomings of modeling without a mast.

The importance of my approach here is done because the models without a mast show the antenna currents are almost perfectly balanced at all heights. However, in the real world when we add a mast, this perfect looking balance we see tends to go away at some heights. My Eznec models suggest to me that random heights above Earth show a far different CMC response in most cases than we see when installed at wavelength heights or multiples. So, I figure height may be important to consider for this issue.

Keep and eye on the red line current indicators.

I follow the assumptions noted in bold below:

We understand that the magnitude of currents, flowing on the outside of a mast or feed line, primarily develop due to a lack of current balance (imbalance) at the feed point.

So, as the difference in space between the red lines noted in the Eznec models and the elements increases, so is the magnitude of the common mode currents (CMC) increasing.

There are generally many segments in the wires of an Eznec model. Each segment is calculated as to its current value by the software, and these values/phase are noted in the tabular Currents Log.

For this report I make note of the space between the red lines and the elements as an indication of the magnitude. As the current magnitude differences increase at the feed point, so does the common mode current results. As the results increase, so can the issues associated with CMC, such as TVI, RFI, skewing of the antenna pattern, and changes in the performance noted. IMO however, common mode currents are not always a bad actor. Our antennas would not work without this RF characteristic, so in some cases such currents can be beneficial. This may be the reason many manufactures ignore the issue.

The goal of this thread is to suggest that antenna feed point height may largely be a deciding factor in how common mode currents manifest on the mast or feed line in our antenna installations, and thus the possible problems they can cause or not…to the pattern and performance.

So height should be considered if you see CMC problems, and this idea may suggest that a small change in height may help mitigate the problem somewhat. I have seen such changes regarding the CMC responses I saw in my shack, but unfortunately I have never tested the idea to a conclusion. I did have a feeling that something was afoot however.

View attachment Eznec and CMC's.pdf

I have also included two models of a 5/8 wave antenna to compare using this same idea about height.

 

[Shockwave]

It's not the height above ground that effects CMC. It's the length of the mast in your model. Take the resonant full wave mast and place it 40 feet above the ground and you'll see it's the mast length. Furthermore, there seems to be some imbalance in the location of the source that is making the CMC appear worse than expected.

 

[bob85]

unless im missing something we have covered this before several times,

1/2wave multiples of mast or coax connected to ground reflect the low end impedance giving you a low common mode impedance= high common mode current magnitude, this is w8ji's worst case scenario,

odd 1/4wave multiples of grounded mast/coax invert the low end impedance giving you a high common mode impedance= low common mode current magnitude, best case scenario,

if the mast/coax is not connected to ground the impedance and current magnitude are inverted,

1/2wave multiples reflect the high ungrounded end impedance giving you a high common mode impedance= low common mode current, best case scenario,

odd 1/4wave multiples invert the high ungrounded end impedance giving you a low common mode impedance= high common mode current, worst case scenario.

 

[BOOTY MONSTER]

maroni , do you get the same results irregardless of the dipole being insulated OR connected to the metal mast ?

 

[Marconi]

It's not the height above ground that effects CMC. It's the length of the mast in your model. Take the resonant full wave mast and place it 40 feet above the ground and you'll see it's the mast length. Furthermore, there seems to be some imbalance in the location of the source that is making the CMC appear worse than expected.

I hope I made it clear that I was using the mast as the feed line in these examples, just to prove the point. I get your point though, it is the feed line here.

In this case the mast is connected to the same point as Eznec considers the ground, and this does not typically happen in real world installs. In the models the situation is similar to when we use an analyzer to analyze the feed point, the wire we use is balanced, whatever length, because both sides of the coax terminate at the same point, the meter. I take the position this is similar to what happens using Eznec this way.

The point for me is, in any case the height using Eznec shows to have an effect on the magnitude of CMC's due to this small height change.

IMO, folks often just get lucky with their installs, and this sounds like what guys like W8JI and Cebik are referring to, as a worst case scenario, when they're talking about modeling issues?

I tried to keep my remarks brief, so I choose not to go into the fact that a real world install with a feed line is not likely to produce similar results...like I see here.

In real world installs we might know and be able to measure the feed line (center conductor) from the feed point to the input of the transmitter, but we likely will never know how long the shield side of the feed line actually is. I take this idea from Maxwell II, as he describes at 21-3. The shield may go to the utility pole ground, and that could be in your neighbors yard. Read this in second column, top paragraph, as I marked below.



Concerning the source. I can make the source for the dipoles split and get closer to the center, but we here discussions all the time about dipoles needing a choke of some kind at the feed point in order to balance the feed point currents, else we are likely to have CMC that ill-affect the pattern and performance. I will post the results if this change makes a noticeable difference.

I did not consider the source of the 5/8 wave models in this case. I just placed them as close to the base as I could. Changing the source will have an effect however.

 

[BOOTY MONSTER]

" ...... so I choose not to go into the fact that a real world install with a feed line is not likely to produce similar results...like I see here. ...... "

so you're trying to make a point about something that probally doesn't really happen ???

wassupwitdat ????????????????????????????????

 

[Marconi]

unless im missing something we have covered this before several times,

1/2wave multiples of mast or coax connected to ground reflect the low end impedance giving you a low common mode impedance= high common mode current magnitude, this is w8ji's worst case scenario,

odd 1/4wave multiples of grounded mast/coax invert the low end impedance giving you a high common mode impedance= low common mode current magnitude, best case scenario,

if the mast/coax is not connected to ground the impedance and current magnitude are inverted,

1/2wave multiples reflect the high ungrounded end impedance giving you a high common mode impedance= low common mode current, best case scenario,

odd 1/4wave multiples invert the high ungrounded end impedance giving you a low common mode impedance= high common mode current, worst case scenario.

Bob, we have discussed this before.

Shockwave made a remark the other day that reminded me about this issue. If your remarks above do not agree, then I can't explain why the models show this difference in results, where the 36' foot height to the feed point looks bad, and the random length at 40.5' mast setup looks much better?

Isn't what you are saying exactly what this project is showing for a grounded mast?

 

[bob85]

the 36ft looks bad because its close to 2x 1/2waves connected to ground, think of it as you would the input impedance of an antenna, even 1/2waves give high impedance odd 1/4waves give low input impedance, the situation reverses if the bottom is grounded.

 

[Marconi]

" ...... so I choose not to go into the fact that a real world install with a feed line is not likely to produce similar results...like I see here. ...... "

so you're trying to make a point about something that probaly doesn't really happen ???

wassupwitdat ????????????????????????????????

Yes and no, but you need to understand the distinction that Shockwave is making here.

Imagine that the coax connected to the feed point is terminated at a measuring device like an analyzer using a feed line 15' feet long. This means that the shield and center conductors are equal in length. The ground and the shield terminate at the same point in such a setup.

Then imagine, if your real world coax is 15' feet from the feed point to your transmitter. This means, that the center conductor is also equal to 15' feet, but your shield to ground can run out to your electric service pole ground.


Did you read the Maxwell II article? It explains the idea.

BM, you might not have enough faith in me to even consider this idea, but if you can understand the words...maybe you can believe what Maxwell says on the subject. Read it again, and see that the coax length as Maxwell describes it...does make some difference.

I didn't talk about this when presenting my thread, because I knew it would likely be misunderstood. Shockwave's remarks forced me to talk about this issue to him.

This could be a good lesson for us to try and understand when we see things happen that don't make sense at first glance. I did not know about this issue until I read Maxwell's remarks, but for a long time I wondered why so many things changed on me as I was trying to tune an antenna that was not close to being tuned already.

IMO, this business about CMC is not significant for us...just using our radios in most cases. You can see in the quote, in my signature below, there is a general explanation as to why.

In the real world we are looking at the big picture, but when guys like Maxwell, W8JI, Cebik, and other folks discuss this stuff, they're often looking at the little picture...like a scientist using a microscope. This is why we often have trouble understanding what they are saying.

I'm still praying for you though.

 

[Marconi]

maroni , do you get the same results irregardless of the dipole being insulated OR connected to the metal mast ?

No, of course not. If you checked out the dipole without a mast vs. the dipole with a mast, you could see and figure out the answer. You have to stop, just looking at the pictures.

BM, when the antenna is isolated from the mast the antenna acts pretty much like the antenna without a mast, at least in the models. There may be a little RF flowing on an isolated mast directly from the antenna radiation, but it likely won't be much unless the design is like a Starduster for example.

So, when we hear Bob and others talk about isolating their antennas, there is some difference to be noted, and I can see this in the models. But, on-the-other-hand these same folks will bemoan the same idea...when we read W8JI crying foul for not including the mast in our modeling.

So, Same O, Same O.

I've tried mast isolation, but I've really never seen a difference that I could tell just using my radio. That is not to say there is no difference, I say it might be a difference that is insignificant or without a distinction.

I believe Bob when he says he sees differences, big or small, but I have just never been able to duplicate his results.

So, no-way-no.

If I ever do however, you'll all hear it from me...good Lord willin' and the creek don't rise.

 

[northern35s]

Including the mast is important, it's the very reason why isolating the coax alone (RF choke) on a DC grounded vertical isn't enough to eliminate CMC, the mast will carry the CMC instead, it's worthy to note that unless the vertical antenna has resonant elevated radials then potentially the choke won't have enough impedance to eliminate the coax CMC completely

I've heard a number of chaps that won't use gamma matching on their beam antennas simply because of the current imbalance in such an installation, I can't say I've put much thought into it, maybe they have a point

 

[Marconi]

Including the mast is important, it's the very reason why isolating the coax alone (RF choke) on a DC grounded vertical isn't enough to eliminate CMC, the mast will carry the CMC instead, it's worthy to note that unless the vertical antenna has resonant elevated radials then potentially the choke won't have enough impedance to eliminate the coax CMC completely

I've heard a number of chaps that won't use gamma matching on their beam antennas simply because of the current imbalance in such an installation, I can't say I've put much thought into it, maybe they have a point

35, I tend to agree that including the mast in our models is important, but my point was to consider...what if a modest change in height were enough to mitigate the CMC issue. Or, like Shockwave correctly mentioned...it is actually coax length instead of height above Earth. It would seem to me this idea would be pretty simple to duplicate if someone has an antenna that produces some TVI that is easy to detect. Just change the length of your feed line 4'-9' and see if there is some change in the TVI. It would not be exactly the same as I noted with my models, but it might show something. I don't have an antenna up right now that I could check or I would add a short jumper to the transmitter end and see what happens.

I just now tried readings on my stationary field strength meter with a 14' coaxial antenna, and when adding 4 feed of coax to my Starduster shows a 100% increase in FS indicated vs. the regular feed line. So I know this 4' foot jumper adds field strength below the antenna, and I assume it is from the feed line/mast which is right outside the window of my shack. However the old SD'r is very good, and does not produce any detectable TVI that I can hear on my PC speakers. IMO, most vertical antennas make CMC, but depending on the magnitude the issue might not even be noticeable.

Regarding your gamma match comment. IMO anything we can do to place the feed point in a more symmetrical environment has to be technically more better.

However, with such comments I consider to ask...how much difference does it make? Is this technical difference enough that we can tell just operating our radios? If not, then it probably doesn't matter much. I haven't tested this idea, but I do have a HyGain Long John with a hairpin match and isolated driven element, and I thought my 4 element yagi with a gamma was better.



I did not exactly experiment with the idea, and I did not compare these two, because at the time I didn't realize the big controversy, so have you some experiences you can post?

 

[Marconi]

Shockwave you commented the other day that my models of the 1/2 wave dipoles source location was off a bit, and thus had a bad influence on the results I was presenting. I posted I would use the split function then and get the feed point right in the middle of the dipole, but I forgot that Eznec does not allow us to use the split function on a wire junction with three or more connections. I used two wires to make the dipole so I could connect the mast in the middle, and thus I end up with a three wire connection.

How much do you think the source results currents could be off due to the source point not being right in the middle of the dipole?

 

[BOOTY MONSTER]

" ...... I just now tried readings on my stationary field strength meter with a 14' coaxial antenna, and when adding 4 feed of coax to my Starduster shows a 100% increase in FS indicated vs. the regular feed line. ...... "




what is that %100 equal to in DB ? also , what were you measuring ? the signal strength from the antenna or a increase in CMC's ?

and what happens when using a vertical with proper 1/4WL ground elements ?

 

[Shockwave]

Shockwave you commented the other day that my models of the 1/2 wave dipoles source location was off a bit, and thus had a bad influence on the results I was presenting. I posted I would use the split function then and get the feed point right in the middle of the dipole, but I forgot that Eznec does not allow us to use the split function on a wire junction with three or more connections. I used two wires to make the dipole so I could connect the mast in the middle, and thus I end up with a three wire connection.

How much do you think the source results currents could be off due to the source point not being right in the middle of the dipole?

I think you would find just moving the source to favor the opposite side of the dipole would shift the entire curent imbalance to the other side of the dipole too. This is because a majority of these problems are the result of poor antenna design or poor matching that places standing waves on the coax line.

I know for example you can take a Sigma and tune it for a perfect 50 ohm match and experience very little if any CMC problems. If you detune that match the slightest bit you can clearly see the results of the standing waves on the coax with increased CMC problems.

This is demonstrated by the fact the position or length of the feedline shows no change in the resonant SWR when tuned correctly. When detuned even the slightest change in the position of the coax in relation to the mast can show measurable changes in reflected power. Proving we have now made the feedline a substantial part of the radiating antenna.

If the Yagi element were fed in a balanced manner and simply had its center grounded to the mast the maximum current would be in the center and trough the element with virtually no RF potential on the mast. Of course things like coax and real world matching devices don't typically allow this to work perfectly.