...any suggestions how basically how to make a 1/4 wave ground mounted vertical effective with only 2 radials?
I ran a number of models. My program works best when the wires are elevated, so
I placed them about 8cm above ground. I'm not sure just how much difference there
will be between that and laying on the ground: putting them up about 30cm high
improved the signal by nearly 1dB. Soil conditions were "good".
I tried a center loading coil, and two off-center coils, equal or not, as well as a 2-wire
capacity hat, all of which resulted in a high feedpoint impedance. About the best I was
able to get was a 1dB improvement in radiated power at a 15 degree vertical angle, and
I was being perhaps a bit optimistic about the coil losses to do that. It's not clear that
you would actually get much improvement over the quarter wave wire once you include
possible losses in the matching network: certainly the quarter wave wire is easier to
feed.
My recommendation is that you start with a quarter wave wire and see how it works with
the best radial system you can manage (which might mean elevating them along a fence
if you can.) Then you can experiment with other loading options to see how much
difference it actually makes: find a nearby ham to take measurements, or build a simple
Field Strength Meter and mount it in a fixed location, then check to see how much it
changes when you try different antennas.
If you want to try the different designs, I got about the best results using a 35uH loading
coil placed 3/4 of the way up the wire. Using 20uH there and another 20uH at the 1/4 point
gave similar results. These all have a high input impedance, so will need a tuner at the
feedpoint. I'd suggest about 6uH and 80pf as an "L" network, adjusted as necessary
for best SWR.