If its that easy to fake models what do we look for to check the model has been done correctly,
if there's no indicator its been fudged people can claim anything they like.
Here is an example of an accurate model, and two manipulated counterparts. One has an AGT of 1 which is accurate (green line below). An AGT of 2 which is far outside of being accurate (red line). The software is also telling me for the AGT 2 output that it is over reporting gain by 3.01 dB, and in this case that number appears to be dead on. Also, I created a model with an AGT of 3, (blue line) which is even further outside of being accurate. The software is telling me that it is over reporting gain by 4.77 dB, and if you look below you will see it has 4.77 dB of gain over the accurate model. I can go on here and further skew the results, but I think this is far enough to show what I, and Marconi are talking about when it comes to AGT and manipulating models to show whatever I want it to show.
I want to point out that I made only one change between these models, and that is a trick that ghz24 taught me to control AGT. Like I stated above it is easy to show whatever gain result I want to show. The elements and such are the same length, and except for one 2 inch segment, are all the same diameter as well. Seriously, changing only the diameter of a 2 inch length portion of the antenna model was able to cause this much of a change in the output of the model, and I didn't even push it to its limit.
So how can you, the person sitting at home in front of a computer screen, tell how accurate a model is? Unfortunately, just by looking at the model you really can't. The only thing you have to rely on is the saying "if it looks to good to be true it probably is". If it is on a forum like this one, ask the modeler questions. For example, don't be afraid to ask what the AGT of a model is, that in and of itself should tell you if the software considers the model accurate. If the modeler doesn't know what AGT is they are either inexperienced, or they don't care to check, or they know and are hiding it from you. I'll have to dig up the numbers for what the range that is considered accurate is, unless Marconi knows off the top of his head. With my models you can assume that they all have an AGT of 1, which is perfect accuracy, unless I tell you otherwise and that is something I will tell you when I post such a model (I've only had one model since I started using this method that I couldn't get the AGT to 1). Marconi's models aren't always 1, but I am pretty sure he keeps them in the accurate range that I mentioned above, so his models should be reliable.
AGT isn't the be all and end all to this. Ground types can also affect reported gain, although it is pretty easy to tell if they pushed a model to its extreme by using "perfect ground" as for those models, for the most part, peak gain is at the 90 degree mark (or 0 or 1 degree radiation angle) or very close to it. Modeling the antenna at excessive heights will also show a higher gain and a lower angle of radiation, although if you have some experience, those are fairly easy to pick out as every half wavelength of height to the antennas current node adds another lobe to the radiation pattern, so as we can see with the models I posted above, with the second lobe formed we are about 1/2 wavelength in height to said current node, which is 1/4 wavelength below the antennas tip.
Aside from that, ask to have someone you trust model the antenna for you and compare, or even sometimes just look at the model and get a second opinion. After playing with literally hundreds of models in all types of conditions, I will generally be able to pick out more from a given model than someone who hasn't done the same.
I wish I had more to tell you, and I wish their was an easy way, but in some cases their really isn't...
DB
i think the article was W4RNL "some j-poles that i have known"
for some reason i can't post a link
I've read that article before.
The DB
