IMAX 2000 continuity query

[JAF0]

View attachment 22346

I measured 42" to the bottom of the metal outer capacitor body, plus 96" of the top whip equals 138" from the top to the bottom of the capacitor, or about 135" up from the top of the matching network in the bottom section.

 

[Marconi]

We know most end fed 5/8 wave radiators show us the bottom 1/8 wave portion is out of phase with the top 1/2 wave, so this small area does not add constructively to performance.

I agree with DB that the capacitor in the middle section of the Imax looks to allow this 5/8 wavelength radiator to use more of the full radiator length, which has to be good. I see two distinctly high current peaks develop in the pattern as well, and IMO this suggest a very wide bandwidth or an improved BW. End fed 5/8 waves are generally considered as narrow banded vertical ground plan antennas, but it is generally reported that the Imax shows a very broad bandwidth. So, I think the capacitor in the center element works primarily to broaden the bandwidth.

I can't test or compare the BW idea using Eznec, because I cannot match the feed point and run a SWR scan producing any useful scan results. That said however, my real Imax shows me a very wide working bandwidth, so I don't have any doubts about this same characteristic showing up in a well tuned Imax model.

IMO this situation, adding the capacitor to the middle section does not appear to show a direct benefit in gain. My models with the capacitor shows the value of R for the match is much worse...compared to a 5/8 wave radiator without the capacitor added. IMO, this high resistance would seem to increase the required transformation for the match. Maybe this accounts for the lack of an increase in gain. DB suggested the pattern is improved with a proper capacitor added, and I agree.

I also note that adding the capacitor, in this case, looks to raise the maximum TO Angle by 1* degree at 36' feet of elevation for the feed point...while the gain drops a fraction. My model also shows maximum gain using 45pF vs. any other pF value I used. My radiator wire diameter for my Imax model is 0.125" inches, and a different wire diameter will likely require a different value for this capacitor.

 

[The DB]

I agree with DB that the capacitor in the middle section looks to allow this 5/8 wavelength radiator to use more of full radiator length, which has to be good. But I also see two distinctly high current peaks develop in the pattern as well, and IMO this suggest a very wide bandwidth or and improved BW.

End fed 5/8 waves are generally considered as narrow banded, but we see the Imax showing a very broad bandwidth and this is what I think the capacitor in the center element does for this antenna.

I can't test or compare the BW idea using Eznec, because I cannot match the feed point and run a SWR scan producing any useful scan results. That said however, my real Imax shows me a very wide working bandwidth, so I don't have any doubts about this same characteristic showing up in a well tuned Imax model.

I agree with this for the most part, I only want to caution that the matching system used for various antennas also has an effect on the antenna's bandwidth as well, so if it is a matched antenna, and we cannot model the actual matching system, it is hard to give accurate information on bandwidth on such antennas. That being said, I can see, as Marconi put forth, that a half wavelength antenna will generally have a wider bandwidth than a 5/8 wavelength antenna, for the reason Marconi gave, and another as well.

IMO this situation, adding the capacitor to the middle section does not appear to show a direct benefit in gain. My models with the capacitor shows the value of R for the match is much worse...compared to a 5/8 wave radiator without the capacitor added. IMO, this high resistance would seem to increase the required transformation for the match. Maybe this accounts for the lack of an increase in gain. DB suggested the pattern is improved with a proper capacitor added, and I agree.

With the capacitor the value for R is higher. When it comes to actual matching circuits for antennas, matching the R variable a given amount is less efficient than matching the X variable the same amount. Half wavelength antennas, and the Imax with its cap will have this problem. This is that other reason I mentioned above, as losses in an antenna system have the effect of widening an antenna systems bandwidth.

The current distribution on the antenna is improved as more of the currents on the antenna with the capacitor were in phase. That doesn't necessarily translate into more gain, however, and doesn't necessarily translate into a lower angle of radiation either for the low angle lobe (for those that want to go their). There is more going on with antenna systems in a real world (and even a fully fleshed out modeling) environment and the other factors have their effects as well... In the case of the Imax. however, it seems to put the feed point impedance in the range of the matching system that they use for their a99 antenna, which would be a benefit for their company to use the same matching system on multiple different antenna designs.

And Marconi, I know you wanted data from one of my models in another thread. I'm sorry for not getting back to you on that, but I lost my storage drive. All of my models and my notes about them are gone, along with almost everything else I collected for decades now. For me to show anything with regard to this or anything else I will need to remake the models from scratch, and I don't even have the software installed for that yet... That being said, the data on my web server should be good as that is a different device entirely, however, that device only had results, not the models themselves.


The DB

 

[W2EKG]

That POS flaps around like a wet noodle in the wind.