Yeah, just remember that -20dB is a factor of 100, the equivalent is the difference between 0.5W and 50W. It's not an insignificant null. However through our band it's really more like -16dB, which is about a factor of 40. Feeding 50W the repeater see the same as if you're using about 1.25W with a unity gain. IOW, like trying to use a HT on low power. It's unfortunate it's at such an critical take-off angle. :-/
My memory is not good but I think this is exactly why I added the second 70cm. I wanted the Flexi-whip fer sure and use 70cm quite a lot. More now that I'm getting into digital, all the DMR is being done on 70cm here and in most places since the bulk of radios in the commercial world operate in UHF. There are a couple of VHF models but the UHF outnumber them probably 25:1 or more.
Anyway, a repeater I routinely talk through up on Squaw Peak (WB5YOE) couldn't hear me at all with 50W through this antenna. I usually get full quieting at 5W until I get north of Ft. Collins or happen to be in a hole. Most of the time at 50W I can hit the repeater even up into Cheyenne. On RX with the 1/4 wave the S-meter is fully pegged but it bounces around S-6/7 with 3/4 wavelength.
I'm just using NEC-2 to analyze, nothing special. All NEC-2 is free and technically NEC-4/4.2 is free, although being export controlled it's a pain to get. You also have to give Lawrence-Livermore money to obtain it, you know, to administer the software.
Yes, NEC is anything but point-n-click intuitive. But if you're familiar with SPICE or assembly & embedded code it's not tough to figure out. It's just 80-character fields with two-letter line headers. The tough part is accurately describing the system to produce what you think it should, but Dr. Johnk did a fantastic job pounding EM wave into my mush-for-brains noodle.
The actual software I used for those plots is cocoaNEC 2.0, written by Kok Chen, W7AY.
'91 Pickup - Imelda
'08 Tacoma TRD - Donna
"Hey Macarena!" -- Thomas Jefferson