Is the Grazioli G-Max really a 10/11m collinear? how does a coil delay the signal 180 degrees

[The DB]

The above models were over Average "Real Ground", this one is over Moderate "Real Ground". Simply changing the quality of the earth makes this much of a difference...

It is 1.3 db of a difference... Lol, a higher amount of gain than what was in the manufacturers model... This is likely the difference...

It is understandable that they used Moderate ground over the Average Ground that I used. I'm actually the anomaly here, not them, as Moderate Ground is the default option in 4nec2. The reason I used Average Ground is more or less Eddie (Marconi) may he rest in peace, and him claiming anyone who used anything other than the Average Ground he used was trying to manipulate the data, in spite of the fact that Average Ground was shown to be anything but average... Long story, no need to revisit now...

Anyway, my model got a higher gain, does that mean I win?

With this information, I would call their model legit.


The DB

 

[bob85]

Looks like my S-METER is not lubricated with snake oil,

@ over 35ft tall it should outperform other antennas on the same mast unless there is a problem,
The concept of 5/8 over 14/wave+ 27mhz collinear has been proven in the past with BIG-MAC & HPSD Highlander

CEBIK got great gain figures at low angles & a direct feed to 50ohm coax in his article "THE CASE OF THE CURLY COLLINEAR"

we tested a Grazioli 1/2wave on 34ft pole vs Grazioli G-MAX on 18ft pole @ 22miles & the result was inline with the tests shown on Youtube,

That been said neither antenna was isolated from the pole & the coax choked so cmc could have been skewing the results, my G-MAX will be well isolated,

Thank you for taking the time to make the model DB

 

[The DB]

Its not a problem. Its not very often that an antenna strikes my interest much anymore. Most of the real interesting ones we have discussed to death already... We also don't see much innovation in the space anymore...

Although I could say the 5/8 over 1/4 does kind of interest me, the numbers Cebik posted in that article don't come across as particularly striking to me.

So here I'm going to post some thoughts.

I think the difference in the distance for the inductor length is a big factor that they should add to their model, although as it is already posted on the web I don't think they will. I would just call it a minor oversight on their part. Such things makes a difference.

I feel it needs to be pointed out, again, that my model is over-reporting gain to some small degree. The model estimates that the difference is 0.34 dB which would bring us to 6.33 dB of gain. This isn't a hard and fast number, it is an estimate, so it is not a for sure thing. Any which way, even though the assumed modified gain figure is lower than theirs, it is close enough to confirm that their model is legitimate. Hey a manufacturer posting a legitimate model, how often does that happen? This being said, I have no information on their AGT (Average Gain Test) data, so I have nothing to compare this to. I don't know if their model is over-reporting or under-reporting gain. This is something that can be manipulated, and I have demonstrated the effects of this in the past on this forum, although it doesn't appear that they have done that.

I'm also surprised, or perhaps I shouldn't be, by how similar the gain pattern results are. The two models look pretty much identical. This is a sign that our models are very close.

So where do the differences come from? For one the length of the inductor. I also had no data on the radials, so I used something close to 1/4 wavelength, with a commonly used diameter. I also didn't have diameter data for the upper and lower vertical section (above and below the inductor), so again I used something that I figured would be typical for such an antenna. I'm also not sure if they are using aluminum elements or not, I am using aluminum. In the end, I don't think this matters much and don't see the need to try and drag the data out of them.

When you do remote tests between two antennas, point to point, this type of model and what you are measuring really aren't the same thing. Apples and oranges really. Sometimes you will get similar results, and that is more luck than anything. I remember Marconi with his models and Shockwave with his point to point measurements... Both were right in their own context, but the contexts were treated as the same even though they were not.

Modeling this antenna in free space, i see the primary lobe is actually angled 5 degrees below the horizon. If the antenna could be modified in such a way that said lobes could be angled up higher, say closer to the angels shown in the models with the earth, I wonder how much this would help add gain... This is going to take some thought and playing... but not right now...

Actually, because of this slight downward angle, keep this antenna to that one wavelength mounting height, and I would say two wavelengths would be even better. When it comes to local distance contacts, your really putting the most power in the best direction in these setups. I would actually rate this as more important than a high gain figure when modeling, which really only applies to DX contacts...


The DB

 

[Henry HPSD]

Good job DB !

I dont think you are the anomaly at all, average ground conditions is the "unwritten" standard for real ground conditions to my understanding.
Though it should be mentioned each time for clarification.

Have you used NEC 5 DB ?
(My models show an average gain of 1 while using NEC5 with these kind of antennas)

@ Bob, yes it is certainly worth to illuminate CMC for 1/2 waves.
You might want to add a couple small radials and insert a RF choke say 1 meter below the feedpoint.
You could measure the CMC if you google RF current meter or probe lots of simple designs will show up.

Kind regards, Henry

Edit: I just checked software like Antenna Model, EZNEC(pro) and Mmana(pro) they use average as default. Ansof uses "costum" as default.

 

[The DB]

Have you used NEC 5 DB ?
(My models show an average gain of 1 while using NEC5 with these kind of antennas)

I have not used NEC5. I can easily afford it (even with its new three year license bs), but I can't really justify it, especially with how rarely I model anymore. Before this model I can't even tell you when the last model I made was? Was it within three years? If I do something I need it for I'll buy it, or if something changes and I start making a lot of models again, but the high level discussions on interesting antennas that I mostly made models for have disappeared... I guess we have explored all the interesting antennas already...

If I find myself working on a model where NEC5 will greatly benefit me, and by that I mean something far more than AGT which I can manipulate and adjust as I care enough to bother, I'll get it.

On the plus side, I do still seem to have it when it comes to antenna modeling, I'm actually surprised at how quickly I was able to put the models above together, even from scratch, although most of those methods are tied to 4nec2, which also doesn't support NEC5...

I guess we will have to see what the future brings...


The DB

 

[Henry HPSD]

Sounds logical
if u dont use it...there is no need to have it, I wasnt aware you rarely used it.

Just to have it said:
You CAN use NEC5 within 4NEC2 that is no problem.
I believe surface modeling is an issue, but that is also the case with EZNEC. NEC5 does come with its own tools for that.

NEC4/5 is highly recommended when you plan to model wires with (large) diameter differences and things like closed spaced wires or wires in acute angles etc. Although in some of those cases NEC2 migth provide a average gain of 1 the results will still be `not accurate`. Actually in many cases MININEC beats NEC2 in those situations.

Keep up the good work !

Kind regards,

Henry

 

[Henry HPSD]

Above an example of NEC2 versus NEC5 with the collinear I made in the past.
NEC2 shows 7.04 dBi in gain where NEC5 indicates 6.58. (Using that moderate ground conditions). This while the AGT tells us: it should be down with only 0.14dB. PS i didnt optimize for lowest NEC2 AGT....this is just to get an idea.

Just for fun...
Here is a MININEC example using "Antenna Model" by teri software,

Kind regards,

Henry

 

[The DB]

You CAN use NEC5 within 4NEC2 that is no problem.

You can use NEC5 with 4nec2? When the author gave his final update he specifically said he was not going to include NEC5 support, and I haven't looked into it since then, so this is news to me. As I look around the web for this, some people are saying that it appears to work, but the data is off. I would have to confirm with other software, like EZnecPro, which is supposed to work with NEC5 before confirming this as working.

I've run Mmana-Gal before which is mininec based. That program didn't have an AGT, I'm guessing it didn't need it. Further, NEC5 handles segments very similar to mininec, which is why I said above that it is much more forgiving when it comes to AGT.

The way I read the models above, the 4nec2 model, AGT 1.033 (0.14), this is over-reporting gain by about 0.14, making the actual gain the model has at 6.9 dBi. If I use any of my correction methods this is generally what I will get. This puts it at 0.32 dB higher then the NEC5 model. While closer I have to admit that this is a significant difference, at least as far as modeling goes (real world you wouldn't notice the difference). When it comes to the mininec model, it is higher than both of them, and while there are places where mininec is superior, we have to remember its design requirements, and the shortcuts Logan and Rockway (I think were their names) had to make to get there. When I use it I tend to use it as a reference more than anything.


The DB

 

[Henry HPSD]

Hello DB,

Yes, you can use NEC5 with 4NEC2.
The fact that Arie doesnt support the engine doesnt mean you cant use it.
(ps the same is true for example for NEC4mp etc)

You can either use Dan his trick https://forums.qrz.com/index.php?threads/using-nec-5-with-4nec2.817589/
Or simply upload the NEC5 engines into the .exe files in 4NEC2 and select NEC4 engine and (re-)write the engine name in the section "special engine".
as NEC5CL_x13 and your good to go.

The NEC5CL_x13 engine, which is debugged by Roy Lewallen works fine.
And results will be the same as with EZNEC.
You dont need to compare them, as the engine is the same, they will produce similair if not identical results.

Perhaps there lies the origine of your thougths.. as the initial NEC5 engine had quite an amount of bugs, especially when the model wasnt in "free space".
(But those bugs were also related to EZNEC)

There still are a few known hick-ups like, You have to "un-check" the "use the extended Kernel box in the "other" screen when using it with 4nec2.
And I believe the source placement isnt accurate displayed in the graphics.
(obviously you are aware of the the different source aspects of NEC5 and NEC2 etc). i also recall the gui interface on windows 11 failed... but I never use that, besides if needed..NEC5 comes with its own interface for that.

With this said.. im sure there still are a couple other bugs, (as is the case with other engines) but for the average user (Verticals / beams) it works perfect !

For MININEC I either use MMANA-GALpro or Antenna Model, that last one does have an built in average gain calculator.
And it indeed..MMANA-GAL does gives warnings, but not by means of an average gain factor.

You mentioned..The MININEC model isnt higher then both of them... im not sure I understand what u meant.
It displays 6,8 dBi which is closer to NEC5 then NEC2 is. (NEC2 7,04-0,14 = 6.9)
And although I dont really consider this to be a case where MININEC stands out from NEC2, it does show it is closer (to NEC5).
Delta Loops with alu and wire is for example one of those.
And yes.. for sure MININEC still has its limitations there, but the differences with NEC2 are much larger.
As you I am not interested in 0,1 dB gain difference, but I am in a shift of 500 KHz of resonant (which is also the case).

Your sentence regarding where MININEC is superior and to remember the design requirements, yes absolutly agree !
But I also would like to point out... exactly the same is true for NEC2
Both have limitations and advantages.
In many cases MININEC is actually a not a bad option when it comes to "freeware".
Situations with larger element differences, acute angels, closed spaced wires MININEC most likely has "the edge".

And yes...it certainly has limitations, so have I hihi....banging headache at the moment... early to bed tonight

 

[The DB]

I actually started the process to purchase this software right after my last post... This is still nuts. At this point all I have are complaints. At the moment I have to wait one or two business days for an email to create an account to get access to software that I have already paid for. This is already after waiting for hours just to get approval to purchase this software... If Amazon tried something like this they would be run out of business in days... I'm sure they would claim "security", but in reality its all just a BS process, and is this software really needing a security seal? Can't people do most of this what this software is capable of with an older freely available version? I think I knew something like this would happen, and that is part of why I was so hesitant...

I guess we will see what happens. How many more hoops am I going to have to jump through?

[/rant]


The DB

 

[Henry HPSD]

Yup...that is as it is...
Im glad you changed your mind ;-)

Kind regards, Henry

 

[The DB]

I had trouble getting NEC5 working with 4nec2. I had various settings set in 4nec2 that got in the way, I had to clear out and put a default install in, and after that NEC5 worked. Here is the g-max from above run with the NEC5 engine.

I don't have to worry about AGT anymore because NEC5, every model i checked came up with an AGT of 1. The NEC2 model shows 4.67 dBi, and was over-reporting gain, the model gave a correction factor, which put the gain to 6.33 dBi I tested and confirmed this). NEC5 shows 6.29 dBi, a whole 0.04 dB difference...

On the minus side, while variables still works, the optimizer doesn't run anymore, and in fact hard locks the model in question, to the point I have to remove the input and output files to get 4nec2 to run again. Yep, manual optimization... Fun fun.


The DB

 

[Henry HPSD]

Hello DB,

Good to see you have got it up and running.
You still need to worry about the average gain factor though it wont be an issue as often.
I assume you are referring to that not commonly used "moderate ground" ?

I did an analyse on the G-max:
Which can be found here:

(Click)
Grazioli G-max EZNEC REVIEW



I get 6,43 dBI ( using that moderate ground)
and 4,24dBi FREESPACE GAIN.

Freespace gain.

The manufacturer "moderate ground" ( 0,05 dB off from claim Grazioli).

Above average ground.


And the SWR is pretty much spot on...

Kind regards,

Henry

Edit: PS Just to have it confirmed for you. Did you "uncheck" the "use external kernel" box as i previously mentioned? that could have caused some of the issues. I have no issues running the optimizer.

 

[The DB]

I used the file from Dan Maguire, it wasn't in those instructions, and its been long enough that I didn't remember. I unchecked it, and it worked, but every time I go back in after saving it unchecked, it is checked again...


The DB