Small antenna - progress, of sorts (UPDATED)

Had a quick attempt to make an antenna for 28MHz within a length restriction of 50cm this afternoon.

I used online calculators to guesstimate the correct loading for a top coil, which came out as 23 turns - I actually needed only 20 - on a 50mm diameter PVC tube, placed 37cm up from the feedpoint.  I used a random wire radial, which turned out to be roughly 2m long.

 

Ugly as hell.  38cm of wire and a 20-turn coil on top, plus radial.

Placed at about 30cm from the ground, the initial match was quite encouraging - 1.7:1 - and the coil was resonating things nicely, if extremely narrowly, at the bottom end of 28MHz - perfect for FT8, but pretty useless across the rest of the band.  That's the price to pay for a very small antenna, of course!  I later managed to get it down to 1.5:1, by sloping the radial up from the feedpoint. 

On adding a second radial, and adjusting the position of the base relative to ground slightly (the height on any given day will change with ground moisture, rain, etc for such a narrow banded antenna), I managed to significantly improve the matching.  This is where I'm at now, ready for some testing tomorrow (note the vertical scale is now zoomed in):

[UPDATE]

A mostly receive-only test for ~3 hours this morning, from a truly rubbish domestic testing environment, shows good promise with FT8, remembering also that this is less than 1/20th wave tall!  Strongest TX from here was -9dB into UR3QL, and farthest so far R3KDV at 2925km, -14dB.  1W at the beach will be boosted to about 10-15W, so this 15W garden test might be very approximately what I might see there (although I expect the clear environment to mean it will be much better in practice).  Indonesia (YE9CDL) came in at a very healthy -8dB.