DB, attached below are two pdf files.
1. This is my old mobile objects with 102" - 108" inch whip on the center of the tailgate for the PU truck titled "Bronco Model 102" on TG" and is 180" long with a 102" inch whip. I just removed the cover over the back of my Bronco model and mounted the antenna on the center of the tailgate. This model uses 47 wires using 133 segments.
2. This is my New model object titled "Bronco 14' object" and is 168" long with a 108" whip on the top rear center at the back. I did this one today hoping the idea you you just posted above would work...your saying more wires is more likely to show better results and I hope it also shows a better Average Gain result in Free Space too. Captain Kilowatt, Joe, and I have been discussing the topic of the natural theoretical match for a 1/4 wave whip. This model uses 227 wires using 365 segments and is very busy so I turned the wire numbers off just to show the model.
I think you will be surprised at these results. These models are still showing bad results with the Average Gain test, and I'm not vouching for them, but nothing I've done in the last couple of days has lowered the end impedance for this 1/4 wave whip on a mobile body down anywhere near 35 ohms either. As you will note both of my models show a higher feed point impedance. The match is not bad and no matching was done on either model.
I realized this from making these mobile models too. The articles I've read from guys modeling and also doing real life testing with an HP vector analyzer...caution about the same thing concerning using grid patterns larger than 0.1% wavelength among other things. My grid patter here is 12" inches but my wire size is 5.75" inches...so there is very little empty space in the model.
I did my model subject similar to yours without the curves however, and made the grid everywhere mostly so I could move the antenna around on the vehicle hoping for some consistency in effects.
That still doesn't help us explain how my object with few wires did nearly as good as my more complicated object with far more wires.
I can't make a grid per se with my version of Eznec, but I agree with your comments here. The article in ARRL Compendium #7, also claims the same as you note above in addition to other factors.
Are you telling us here that your mobile object shows the theoretical match for a 1/4 wave at or near 35 ohms when mounted on you mobile object, and that adding the curves to the object did that?
Does you new object also show you a good AG test results?
If so, I sure would like to know how you did that.
