I agree with 'packrat', a fan dipole does work. It tends to work as well as the amount of effort put into it's construction and tuning. Do both of those thingys close to 'right', and it performs very well. It's a lot like stacking monoband beams on a tower, with a few "but's" thrown in, sort of.
The placement of the 'legs' of each dipole works 'best' if they are as far apart as possible. Meaning that two dipoles would be at 90 degrees from each other. If more than just two dipoles, then as far apart as possible, sort of 'fan' them out (and where the name comes from). That's all well and good if you've got that much open area and supports for each leg of that fan-dipole. One 'catch' with that is that people typically don't have that much room. So, you place all those 'legs' closer together. A separation of a couple'a few inches works. The closer those legs ar e to each other, the 'pickier' the tuning though, since they tend to interact more.
The closer two (or more) antennas are in relation to frequency, or in harmonic relations, the more they interact. Some antennas for one band can be used on another band without much problems. Those tend to act 'oddly' when used in a fan-dipole. The ones that don't work well on other bands tend to work very well together in a fan-dipole. Is that a biggy? Not really, just be aware of it as a possible reason for 'things' not working as expected sometimes.
And as with any antenna, get the thing as high as possible for the situation. I wouldn't expect it to be very directional, most people can't typically get that kind'a height.
All the other things about dipoles apply too. maybe the most important one of those 'things' is to make the thing out of pink insulated wire. It tends to radiate better. An electron sees where it came out of and gets as far as possible from it as quickly as possible to keep from being embarrassed! Makes sense to me.
- 'Doc
(There may be some 'rocket science' involved in antennas, but it's more a 'bottle-rocket' thingy than an 'Atlas Booster', sort of.)