Base Maco V58 model

[Marconi]

The DB, I think we briefly talked about the problems with modeling this antenna before.

Do you have a model of the Maco V58 that includes the physical coil and shows a good match for this 5/8 wave antenna?

If not, have you modeled it using the 4Nec2 matching feature?

Either way can you give me some tips on modeling this design?

My model below that I think is set to the specs shows a good impedance but the reactance shows capacitive at -131.4 ohms, so the 5.5" inch radius inductor coil is not working as expected. I'm also not sure how to model the capacitance that is created in this design with the bottom of the radiator inside a larger tube and insulated using PVC or something similar as an insulator, but that may be the problem with my model...just not sure.

I added the main screen to show the Free Space Average Gain results at the bottom as well as the wires descriptions.

Can you help me to better understand this design?

Word edit: radius

 

[tuner]

Just a quick Hello Eddie , hope all is good . Interesting thread

 

[Marconi]

Just a quick Hello Eddie , hope all is good . Interesting thread

Thanks tuner. This was my first model to try and add a physical matching device back in 2013, and since then I have not improved that result one iota...as long as I keep the V58 radiator short like the specs call far. If I make the radiator closer to 271" inches I have a coil that makes the antenna show a perfect match...just using the same design with a coil about double the diameter.

IMO, this design problem for my modeling simply has to do with a few inches of the bottom of the radiator being inside the mast to mounting bracket. I don't think I can model a pipe within a pipe. Plus I don't think Eznec provides a non-conductive element or feature that duplicates an insulator. As far as I know, the only way to suggest an insulator is to use an empty space...about the same size as the insulator.

Here is a image of another similar setup. Note the black insulator where the radiator enters the mounting bracket. The difference here is this antenna allows one to adjust the radiator length a little bit for fine tuning only. The advantage is this can be done near the mounting point of the black PVC insulator just below the coil connection to the radiator making it is easy to access...if you can get to the mounting point for the antenna as installed.

Note: I find this particular design is not durable however. When the antenna is setup and tuned and the wind gets to working on this bolt connection point, that secures the setting, the set-screw buggers up the insulator inside the base mounting bracket. This makes subsequent adjustments near to impossible.

This adjustment point is also very sensitive to even small adjustments =< than 1/16" inch. Again this feature is for fine tuning only.

 

[The DB]

The DB, I think we briefly talked about the problems with modeling this antenna before.

Do you have a model of the Maco V58 that includes the physical coil and shows a good match for this 5/8 wave antenna?

If not, have you modeled it using the 4Nec2 matching feature?

Either way can you give me some tips on modeling this design?

My model below that I think is set to the specs shows a good impedance but the reactance shows capacitive at -131.4 ohms, so the 5.5" inch radius inductor coil is not working as expected. I'm also not sure how to model the capacitance that is created in this design with the bottom of the radiator inside a larger tube and insulated using PVC or something similar as an insulator, but that may be the problem with my model...just not sure.

I added the main screen to show the Free Space Average Gain results at the bottom as well as the wires descriptions.

Can you help me to better understand this design?

Word edit: radius

I don't have a model that has a great match, I was able to match it to an SWR of under 2.0, although that model had only 16 elements in the ring, not the 30 or so in the one you posted. The biggest issue I've had with modeling that antenna is on the actual antenna you can tap the ring at any point on said ring, however on the model I made I only have 16 points. I have thought about using more elements for the ring, however, at some point that method will also break down, as the segment size for said ring and the elements next to it can only have so much of a difference before 4Nec2 puts up errors or warnings.

You are getting a capacitive reactance reading, which doesn't surprise me as much as it would some other people. The two solutions for this would be to make the ring bigger, essentially adding series inductance, or add a parallel capacitor which would add parallel capacitance which is what Maco did. Both will have similar effects on tuning, although not identical effects.

The matching feature simply adds a matching network, I wouldn't call it the same as using an actual matching network, it is more a theoretical add on than anything.

I haven't looked beyond my 16 element ring as of yet, although I do have an idea as to how to compensate for the difficult to simulate infinite possible connection points we have on the ring. I am hesitant to mention it as it heavily relies on math. If I revisit that model it is something I will have to do.

Does anyone know what the actual capacitance rating of the hidden built in capacitor is? A few people have thrown out numbers in the past, however I don't think any of them are actual measurements.


The DB

 

[Needle Bender]

Eddie, the V58 or V5000 are both right about 20' of radiator, +/- .3"- 4".

On several occasions I've erected one in place of an Imax, Penetrator-500 and other full 5/8 waves, and have always seen about a 1 - 1.5 S-unit loss in performance.

I devised a new matching network of 1.75T x ~11.5"diameter ring and a lengthened ~22' 9" radiator, but lifted the radiator up & out of the base using 1" solid fiberglass rod & a 4' length of 1.125" tubing to remove unwanted capacitance.

 

[Marconi]

I don't have a model that has a great match, I was able to match it to an SWR of under 2.0, although that model had only 16 elements in the ring, not the 30 or so in the one you posted.

DB my question, whether or not you have a model of the Maco V58 that shows a good match, was intended to ask if you had a model that was set to specs and showed a good match.

My MV58 model done several years ago also used 16 elements for the coil, but shows a >9.00:1 SWR reading at the feed point. So, if your model shows less than 2.00:1 SWR...then you either made the radiator much longer and/or you did something else to fix the capacitance or add inductance in some other way.

I can make the radiator longer and get my model down into the >2 SWR range using a 11" - 12" coil with 16 segments or 32 segments too. But, no matter what else I do the model will not go below 2.00:1 SWR without making the radiator longer. My model with a 32 segment coil did not produce a critical geometry error, so I figure 4Nec2 can do that too.

The biggest issue I've had with modeling that antenna is on the actual antenna you can tap the ring at any point on said ring, however on the model I made I only have 16 points. I have thought about using more elements for the ring, however, at some point that method will also break down, as the segment size for said ring and the elements next to it can only have so much of a difference before 4Nec2 puts up errors or warnings.

I noted above that my model with 32 segments for the Maco V58 coil did not produce any error reporting using 1 or 2 segments per wire. However the model did break on using 3 segments per wire.

You are getting a capacitive reactance reading, which doesn't surprise me as much as it would some other people.

DB, I agree the model does show capacitance, but there was a time that I didn't know, so just lump me in with the "some" group you suggest here.

The two solutions for this would be to make the ring bigger, essentially adding series inductance, or add a parallel capacitor which would add parallel capacitance which is what Maco did.

Making the ring bigger is not a solution. My goal here is to model the Maco V58 wave to the specs in the manual.

I don't know how to add parallel capacitance to the model...like you suggest. Are you saying, parallel capacitance adds inductance?

By the way, several years ago Camaro1 posted his dimensions for the Maco V58 for another member and there he indicates the bottom of the radiator only goes into the mounting bracket, BA1P, 3.5" inches and this element in the top of the base assembly is fix and is not tunable. IMO the only capacitance created here is probably very small. The instructions say if you want to added inductance and reduce the capacitance, make the radiator longer, make the adjustment at the next element up above and that adjustment area is not very much either. So IMO, any adjustment for the Maco V58 at this point is intended for fine tuning only. The tubing overlap here is at best 6" inches as instructions suggest to start and that is not much to play with and maintain superior structural integrity.

Both will have similar effects on tuning, although not identical effects.

I am

I have thought about using more elements for the ring, however, at some point that method will also break down, as the segment size for said ring and the elements next to it can only have so much of a difference before 4Nec2 puts up errors or warnings.

You are right DB, but with the Eznec helix generator all the segments are the same length when you create a helix, a loop, or radials...there is no difference if you enter your dimensions correctly. And you don't have to do any math, it will generate the coil to specs if you understand the process necessary.

You can mess up with accuracy however. So, I would check my math if your coils have segments that have different lengths. The same is true with loops and radials.

Using Eznec, the error reporting problem associated with too many elements is simply
the fact that segments get too small to work, and are beyond the scope for the NEC engine. Eznec determines these geometry and segmentation properties before it performs each scan. Are you doing all your models using math computations by hand?

I am hesitant to mention it as it heavily relies on math. If I revisit that model it is something I will have to do.

Check your 4Nec2 Manual, maybe you will find a system feature that will allow you to automatically generate your coils, radials, and loops without having to do everything by hand. This way all you have to do is enter the dimensions in the templet screen for the function you wish to generate, and then click the OK button. At least this is the way Eznec does it. No wonder I've had trouble mastering 4Nec2.

Here is a screen shot of my Eznec Helix generator that makes coils easy without having to do it by hand with math. Of course you have to understand all the dimensions the elements need. When I first checked out this feature years ago...I sat for a month pondering what was next.


Thanks for your replies.

 

[The DB]

DB my question, whether or not you have a model of the Maco V58 that shows a good match, was intended to ask if you had a model that was set to specs and showed a good match.

My MV58 model done several years ago also used 16 elements for the coil, but shows a >9.00:1 SWR reading at the feed point. So, if your model shows less than 2.00:1 SWR...then you either made the radiator much longer and/or you did something else to fix the capacitance or add inductance in some other way.

I can make the radiator longer and get my model down into the >2 SWR range using a 11" - 12" coil with 16 segments or 32 segments too. But, no matter what else I do the model will not go below 2.00:1 SWR without making the radiator longer. My model with a 32 segment coil did not produce a critical geometry error, so I figure 4Nec2 can do that too.

It wasn't specifically that. What I was playing with was a so called Maco style matching ring. It had no capacitor, and the antenna was longer as I was playing with a full length 5/8 wavelength antenna.

I noted above that my model with 32 segments for the Maco V58 coil did not produce any error reporting using 1 or 2 segments per wire. However the model did break on using 3 segments per wire.

Yea, I knew it would happen at some point. Nec is picky about some of those things.

Making the ring bigger is not a solution. My goal here is to model the Maco V58 wave to the specs in the manual.

That wasn't what I was trying to do at the time, but it is the closest I have at the moment.

I don't know how to add parallel capacitance to the model...like you suggest. Are you saying, parallel capacitance adds inductance?

No, parallel capacitance does not add inductance, however, parallel capacitance has a similar effect on the antenna to series inductance. They are not the same, but their effects are close enough that in most cases one can be substituted for the other.

By the way, several years ago Camaro1 posted his dimensions for the Maco V58 for another member and there he indicates the bottom of the radiator only goes into the mounting bracket, BA1P, 3.5" inches and this element in the top of the base assembly is fix and is not tunable. IMO the only capacitance created here is probably very small. The instructions say if you want to added inductance and reduce the capacitance, make the radiator longer, make the adjustment at the next element up above and that adjustment area is not very much either. So IMO, any adjustment for the Maco V58 at this point is intended for fine tuning only. The tubing overlap here is at best 6" inches as instructions suggest to start and that is not much to play with and maintain superior structural integrity.

Lets see what math tells us about this. I just took some measurements on my V5000 with a micrometer.

Diameter of the inside of the insulation .996 inch
Diameter of the outside of the insulation 1.39 inch
Just over 4.5 inches from the edge of the outer metal tube to a screw that goes all the way through the large tube. Looking inside the teflon insulation and metal inside tube almost touch this screw. Estimating this at 4.5 inches, unfortunately at this point this estimation is the best I can do.

To get the ring circumference we take π multiplied by the diameter.

Inside insulation circumference = .996 * π = 3.2190263...
Outside insulation circumference = 1.39 * π = 4.3668138...

To get the area of each we multiply these by the length of the tube overlap.

The inside tube has 14.080618... square inches of surface area.
The outside tube has 19.650662... square inches of surface area.

Here we run into a problem. Generally when capacitance is calculated both plates of the capacitor are the same size, and that is not the case here. For the purposes of this I am going to take the average of the surface areas as that will get us close if not right where we want to be, that number is 16.86564...

Now that we have the surface area, we need the dielectric constant and thickness of the insulator, which is teflon. The dielectric constant for teflon is 2.1. The thickness of the teflon is (1.39 - .996) ÷ 2 = .197

When I punch this data into a calculator found here I get a capacitance of 40.3 picofarads. Assuming I didn't make a mistake somewhere, this should at least be close...

You are right DB, but with the Eznec helix generator all the segments are the same length when you create a helix, a loop, or radials...there is no difference if you enter your dimensions correctly. And you don't have to do any math, it will generate the coil to specs if you understand the process necessary.

You can mess up with accuracy however. So, I would check my math if your coils have segments that have different lengths. The same is true with loops and radials.

Using Eznec, the error reporting problem associated with too many elements is simply
the fact that segments get too small to work, and are beyond the scope for the NEC engine. Eznec determines these geometry and segmentation properties before it performs each scan. Are you doing all your models using math computations by hand?

Check your 4Nec2 Manual, maybe you will find a system feature that will allow you to automatically generate your coils, radials, and loops without having to do everything by hand. This way all you have to do is enter the dimensions in the templet screen for the function you wish to generate, and then click the OK button. At least this is the way Eznec does it. No wonder I've had trouble mastering 4Nec2.

I want to be able to go beyond the given design and easily modify the parameters to see what happens with other applications. For example, if you have no capacitance, how big of a ring do you actually need? If I want a longer element what effect does that have on the ring? Are their any benefits for using some other capacitance/antenna length/ring size combinations, perhaps efficiency? Yes the way I do things is much more work of front, but it is less later on when making variation models experimenting with the design. I think we have very different approaches to making our models, as you are just trying to model the antenna itself. It is amazing how often two entirely different concepts for modeling can have the similar results we consistently see.

Thanks for your replies.

Not a problem.

You were kind enough to put your data in, so I may play with your model some in 4nec2 and see what happens. When I get some time for it that is.


The DB

 

[Marconi]

Eddie, the V58 or V5000 are both right about 20' of radiator, +/- .3"- 4".

On several occasions I've erected one in place of an Imax, Penetrator-500 and other full 5/8 waves, and have always seen about a 1 - 1.5 S-unit loss in performance.

I devised a new matching network of 1.75T x ~11.5"diameter ring and a lengthened ~22' 9" radiator, but lifted the radiator up & out of the base using 1" solid fiberglass rod & a 4' length of 1.125" tubing to remove unwanted capacitance.

NB, you have a very responsive and sensitive meter in your radio...if you see 1 - 1.5 S-units improvement over the V58 compared to the two you compared.

Bob85 use to claim similar big differences in some of his reports, but I think he claimed his radio(s) had some kind of special meter. In my experience I never saw such differences, nor have I seen any such differences in my real world experiences.

I have at least 5 different radios that I've compared my antennas with over the years and even though I did see some minor difference in the scale from 1 - 9 S-units I never saw such a large difference as you guys report. I've seen a few others on the Internet comparing receive signals on their radio and I don't recall ever see that much of a difference between different antennas either.

Do you have any video links that shows the kind of responses you note here?
I would really be interested in seeing such real life reports.

Yep, again this model is about the same as my model with an almost perfect match and a raised up real 5/8 wave radiator. The radiator is longer at 267.29" inches and 104" radials, but I used a single coil 1" apart at the legs that has a 16" inch diameter to get such results.

I tried to duplicate your dimensions as I understand them, and the best impedance is very close to 50 ohms but the reactance now is +44.58 ohms at the feed point.

Below are two models of your current idea. The other model is a fix for the reactance problem noted in your model as best I understand your dimensions. I modified the coil by making it 3 segments shorter at the ground side of the coil and that fixed the inductive reactance your model shows me in the model.

I also have some overlays of several similar antennas I have model. I will post them if you would like to see more bad news.

 

[Marconi]

DB, I read your comments and I will make some response later after a more through look.

But for now what data are you referring to when you say the following?

You were kind enough to put your data in, so I may play with your model some in 4nec2 and see what happens. When I get some time for it that is.

 

[The DB]

DB, I read your comments and I will make some response later. But for now what data are you referring to when you say the following.

Page 4 of the .pdf in the first post of this thread.


The DB

 

[bob85]

No special s-meters here eddie,
i use a large external meter that i used for foxhunting & my fluke dmm if i want to see smaller changes,
What i report is typical of what people in this area experience.

I can use my lp-100 on direct input as a field strength meter if i want more accurate readings but im interested in my signal at a distance not at the bottom of the yard 100mtrs away.

s-meters on CB & HF sets new and old are all mickeymouse and not linear across the meter scale.
equating s-units to db's is a waste of brain power.

 

[Marconi]

No special s-meters here eddie,
i use a large external meter that i used for foxhunting & my fluke dmm if i want to see smaller changes,
What i report is typical of what people in this area experience.

I can use my lp-100 on direct input as a field strength meter if i want more accurate readings but im interested in my signal at a distance not at the bottom of the yard 100mtrs away.

s-meters on CB & HF sets new and old are all mickeymouse and not linear across the meter scale.
equating s-units to db's is a waste of brain power.

Bob, I think you have explained your setup to me before and I don't doubt what you reports.

My question is then, how does that explain the results the NB is claiming?

Do you think maybe NB has a similar device that shows him the correct gain differences he is reporting?

Or, do you think he was using a radio with a meter like you describe...as mickeymouse?

Most of us don't have lab quality test equipment to do our signal comparisons...so we have to use our radios and report what we see on the meter. I consider my radio meter to be a relative meter that is
not precision.

I can't honestly question what anybody sees and reports using their radio, so I guess I've been wasting my brain power comparing my signal reports and S-units. That said, I have never seen any of my CB vertical antennas, mounted with the current maximums set close to similar heights, ever show such differences among my locals RX signals either.

 

[Marconi]

It wasn't specifically that. What I was playing with was a so called Maco style matching ring. It had no capacitor, and the antenna was longer as I was playing with a full length 5/8 wavelength antenna.

That is all I have been able to do as well, but that is not the Maco V58 and I wanted to try and compare this shorter antenna to others.

I hear lots of good reviews from users of the Maco V58 and I have friends that use them and I know they talk. I've also heard folks that rave about the traditional .625 and for sure folks that claim the .64 is superior...suggesting the shorter Maco shows them far less gain like Needle Bender posted above.

DB, several years ago the Maco V58 was the first antenna that I tried to add a physical matcher to the model. I was never satisfied I did the V58 right...if I kept the model to specs. The coil inductor did not make a good match for the model, so I tried the idea on my I-10K instead. I think I have posted that model here on WWDX. I saw that by adding such a large matching network it did something else that I did not expect to see, noticeable horizontal radiation in the pattern and a suggestion that could be beneficial to overall performance.

Just for information I think my I-10K always worked well for me back at a time during very good skip and when WWDX was having regular group contact contest. But at the same time locally I did not think it was anything special on the local scene. At some point a bit later on I heard that Bob85 had an I-10K and I think he was reporting almost the opposite results in England.

When I punch this data into a calculator found here I get a capacitance of 40.3 picofarads. Assuming I didn't make a mistake somewhere, this should at least be close...

Ok, now that you have determined 40.3 picofarads might be close...what do you suggest I do with the information? How would I generally apply this number to Eznec? Do you have any ideas?

Thanks for the information and the math.

 

[bob85]

IF NB is using a radio's s-meter and using it to measure the difference in gain in db id call that precision guesswork,
now if he made a chart using a calibrated sig gen and made sure the signal was below the agc threshold he could get much closer to the truth but far from dead balls on.

he could also use a precision constant impedance attenuator like mine and adjust it for same signal fora more accurate reading,
or spend tens of thousands on propper calibrated test gear to discover what we already know,

i don't doubt that NB sees what he claims, it would be arrogant to believe that only my own and locals tests show differences in antennas,
it would also be foolish to think that any difference on the s-meter was due ONLY to the difference in gain between the antennas,
local topology and things around antennas effect different style antennas differently,

i don't doubt what you claim Eddie, put a low loss 1/2wave at the same tip height as a 5/8wave & they perform at least as well as the 5/8wave.

 

[bob85]

DB the sigma2 also has a similar value cap across the hoop.