Here is a summary of some measurements I made this summer while in the process revising my beverage antennas.
1. The pick-up in the vertical feed and termination is only 15db. to 18 db. down and as such dominates the side and back reception of the antenna.
2. There is absolutely no difference between sloping or vertical feeds.
3. Raising the feed to the height of the beverage and using a counterpoise of two ¼ wavelength horizontal wires increases the rear/side rejection so that the antenna pattern agrees with the theoretical. The difference is dramatic … as much as 12-13 db. The matching transformer must have better isolation than most commercial tri-filar designs. The W8JI design on a Multi-Aperture core works well.
4. Raising the feed and using a single ¼ wavelength stub in the direction of the beverage has a good null off the side, but the null is sharp and the pick-up off the back is worse than with the vertical feed.
5. Although the raised feed yields a great pattern, it is very hard to implement, especially in cases of multiple antennas.
6. The vertical feed, as poor as it is, seems to hear as well as the raised feed. The noise picked up by the beverage in an omni-directional noise environment appears dominated by the front lobe.
7. If you want better signal to noise, reduce the beam-width. Going from 15 to 30db off the sides or back, won't do much.