MFJ Moxon

[vkrules]

I find the easiest thing to do is to change the element lengths
What I like to do is start with a higher frequency on the calculator and make the moxon a little short.I then use corner tails to load it down.
To tune I mount the ant high enough so I can reach the tails standing on a ladder. I then set up my analyzer as a sig gen and mount it on a step ladder at the other end of my yard, ( yes I know this is less than ideal, but hey you work with what you have) You could use a known signal source Beacon, freind up the road ect. Check forward signal on the s meter . I like to set my gain control so the meter is reading about s7 or 8 as a reference ( the 746 pro gets less linear on higher sig readings) then reverse the ant . Dont worry about swr at this time.
Start with the reflector tails folding them over or cut them about a cm at a time and recheck for minimal signal .When done you should have a front to back of about 7 s points.Check forward direction again, just to make sure all is well.
Now check resonate frequency if its low, fold or cut the driven element tails.If it's high add more length to the tails.There is a little interaction between elements so just check F to b again in case you have to fine tune the reflector .

 

[Marconi]

@Marconi When you get the time, and are feeling well enough, if you could take the posted dimensions in my graphic I provided and see what a model reveals I would really appreciate it. As you can see in the photo, I use a coaxial choke about 24" back along the coax feedline from the feed point.

Homer I re-did your model earlier and posted it above in post #25. If this is not the model you're talking about above...then let me know.

Just to be clear do you want me to add a mast or a feed line, and maybe a choke? Is this what you are interested in checking out to see if the Moxon produces currents on the feed line or mast below the antenna?

I went ahead and modify the model and added the mast. Check this out and if you think a feed line added will show something else, I can also add a feed line, just let me know. I don't know how to effectively add a coaxial choke to the feed line, but with this said, I think adding a mast will help show us, in-part, what you want to see, but I'm just guessing.

In the antenna view there are some red lines and they represent the currents flowing on the adjacent wire they cover. The farther the red line is from the wire...the more current magnitude there is. I hope this explains something useful about currents in modeling.

Now you will note there is very little current indicated on the mast...so I claim this model of a Moxon does not indicate a choke is necessary.

My theory here is: an effective coaxial choke will not ill-effect the performance of any vertical CB antenna, but it is said that some chokes (bad placement, design, or construction) are know to cause problems.

Again Homer, I'm guessing a bit here, so look this over and let me know if this info is not precisely what you are looking for.

I have more to add about Vkrules use of loading tails idea, but I will make another post to show how the idea works with an Eznec model.

I added some notes to your model so you can check the dimensions. If they are not legible I think the PDF file will allow you to zoom in on the image a bit...try that.

 

[Marconi]

Homer, here is your Moxon model with a feed line added from the ground to the feed point. I set the model with the feed line in two locations, one is running along the mast and out to the feed point along the support wire #10 at the antenna feed point. The other feed line is hung down a little bit...sort like your images show above. This feed line has an icon with a little box with a big 'T' in the middle. When you see this line with no icon, it is just a supporting mast instead, and shows up as just another wire in the model.

You will see that nothing here is changed much if any, even the match compared to the model I fixed without a mast or a feed line. For me this is another indicator of how well balanced this design is.

Again, you can see that the mast/feed line still does not indicate any significant currents, even after I added all this extra metal to the antenna including the mast.

With this said however, there is a consideration that needs to be understood here, and that is regarding how models handle the Earth ground vs. your antenna at home. The model sees the ground for both the Earth and the antenna circuit at the small circle you see with a dot in it. In several ways this is not what you have at your location, but this is difficult to explain. So, consider this a difference that might possibly cause your antenna to work differently than this model...and you could have CMC that this model does not indicate.

If your antenna works like you want it...then I recommend you keep the choke.

 

[Marconi]

Here are a few examples of what Eznec shows using the tails idea on a Moxon antenna that Vkrules posted.

I start with tails 20" inches, and this length is noted in the model's antenna views details, for wire #12 for each model. Then I cut these two wires to 10", 5", and then deleted the tails.

 

[HomerBB]

Thanks, Marconi. For some reason I wasn't seeing your posts. Now I see them ok.
It seems you're right, adding mast and coax has little to no effect on the model. Also, adding those tails was a disaster to the model.

The model showing the radiation pattern of the antenna I posted may explain a curiosity of this particular Moxon. I have found the F/B to be very good at times, and then less good at others. According to the model it is likely because of the angle at which I am receiving the propagation. The is a very small lower back end lobe that shows realyl good rejection, and a rather large higher up lobe off the back, too. I can completely, or nearly so, cut out signals at the back with this antenna, and dramatically improve them from the front. Other times the prop apparently favors the the higher lobe in the back and I do better from the side rejection.

@Marconi What are the chances you could do a model of this typical two element Yagi for comparing to the Moxon?

 

[Marconi]

You were reading my mind.

Just this morning I opened up a 2 element yagi to compare to your Moxon. But I was considering the ease of building and tuning a yagi.

I wasn't impressed with the model as it was. I tried a few changes to try and re-tune the model for better rejection, but nothing I did made a positive change. I was surprised that a two element looks to be easy to build as a direct feed, too. I was thinking this direct feed and a good match idea maybe due the the Moxon design, but maybe it is due to there being only 2 elements.

The reflection does not look near as good as the Moxon, but the gain is a bit better. Here is the model.

I will try your dimensions.

 

[Marconi]

What are the chances you could do a model of this typical two element Yagi for comparing to the Moxon?

Here is your 2 element yagi model.

 

[Marconi]

Homer, let me check this yagi stuff out tomorrow. I'm too tired. I noticed a miscue in the first model and I think it is in your model too. The model shows to be isolated (ISO) and I did not realize that when I posted it. ISO can and will make a difference, and I don't know how that effects what I just did.

I don't expect the performance values to change much, but the match could be effected, and the model does not show any feed line radiation, so my over-site could effect CMC's showing up or not.

 

[HomerBB]

ok. thanks. the match is off compared to the Moxon, but the gain is a little better. F/B is not so good.

 

[Marconi]

Sorry Homer, but I noticed that I also used 27.205 on your models last night, and you asked for 27.385. I don't think that will make a big difference in the long run however, but the isolation I had included in the model that I started with...will made a difference with CMC shown on the models. So, I just messed your stuff up.

 

[HomerBB]

Sorry Homer, but I noticed that I also used 27.205 on your models last night, and you asked for 27.385. I don't think that will make a big difference in the long run however, but the isolation I had included in the model that I started with...will made a difference with CMC shown on the models. So, I just messed your stuff up.

I noticed it was 27.205, but I didn't think it would make much difference either way.
What 2 element Yagi model(s) did you use that gave you a direct 50 Ohm feed?

 

[Marconi]

I noticed it was 27.205, but I didn't think it would make much difference either way.
What 2 element Yagi model(s) did you use that gave you a direct 50 Ohm feed?

Homer, I fixed the yagi model to 27.385. I added the big X markers to indicate when I have an opened wire, so it will indicate the open ends of wires and isolation (ISO) better. The connected wires have a small black square at the junction. You will note the differences are due to isolation. Also note that wire 1 and 2 are the same length at 36" inches.

Also note the red lines for the currents in the antenna view. Where the currents don't line up well the feed line current is increased. IMO, this is what balance is all about, and it is pretty obvious in these examples that the connection to Earth has an ill-effect on the balance.

Both models suggest this one can be setup as direct feed with a very good natural match at the feed point, but there is nothing around this antenna...so your real world experience my differ.

Sorry I messed up this project.

I also added an overlay of the patterns. I added notes for the dimensions.

 

[vkrules]

For the Yagi .If you take the element spacing out to .2 or.22 of a wavelength you can get a good match for 50 ohm and good gain but the front to back is not the best.

 

[Marconi]

Vk's suggestion is right.

 

[tba02]

I've attached some updated dimensions, theory and real world results for the MFJ-1890, at my location, in my environment.