The late W4RNL has some extensive work on folded dipoles on his website here:http://www.cebik.com/content/a10/wire/wbmw.html
Yes, you have to register (free) for access, but it is well worth it for the wealth of
information he has available.
That said, a folded dipole might not be the best antenna for your needs unless you
are planning to use resistive loading for a wide bandwidth. The conventional folded
dipole doesn't work very well on even harmonics of the design frequency. It presents
a 600 ohm impedance only on the design frequency (where it is about the same length
as an ordinary half wave dipole). Without the terminating resistors the SWR will jump
all over the place as you change frequency. Depending on your ALE system, your
tuner may or may not be able to keep the antenna tuned properly each time the system
The resistive-loaded antennas (there are several types) are popular for this service
because there is no antenna tuning required when you change frequencies. The downside
is the loss in the resistor: for the common 90' folded dipole this can be 10dB on 80m and
20dB on 160m. The Oregon Office of Emergency Services changed to a diamond-shaped
loop (also with terminating resistor) instead of the folded dipole because they said it had
better performance on the frequencies of interest. Unfortunately, critical information such
as antenna efficiency vs. frequency are not always provided by the manufacturers.
Even if you can make your antenna tuner work fast enough, you still have to consider the
radiation pattern of the antenna vs. the desired coverage area. It is difficult to find a
single wire antenna that will give good NVIS radiation from 160m through 40m. For example,
an OCFD designed for 80m will also work on 40, 20 and 10m with relatively low SWR, but
the 40m radiation pattern will have an overhead null. The same thing happens when you
design the antenna for 160m and use it on 80. A straight center-fed doublet will have a
useful NVIS pattern on 160m and 80m, but if it is 1/2 wave on 160m then the pattern on 40m
will have an overhead null. You may be able to shorten it a bit, trading efficiency on 160m
for NVIS coverage on 40m, to some extent. There are some solutions to this problem with
a bit of creativity: I've found a couple designs that should give reasonable NVIS patterns
from 160m to 40m (you don't need them any higher than that) but they require some number
of bends and/or multiple wires.
If the system is just intended to cover the ham bands, then starting with several half wave
dipoles on a common coax feedpoint is difficult to beat in terms of performance, especially for
the lower bands. That gives you good NVIS radiation without needing a tuner. That isn't always
a good solution for commercial/military ALE systems, however, because they have a wider range
of frequencies to chose among.
While Dan mentioned wideband antennas such as the discone, and how much room they take, there
are some reasonable options that might work for you to give low SWR over a wide range of
frequencies. One of the simplest is a true fan dipole: for 40m through 10m, cut 10 wires each 40'
long. Connect 5 wires to each side of a 4 : 1 balun and spread the wires on each side out into
a fan with perhaps 20 to 25 degrees of angle. (This is from memory, as I don't have my notes
handy.) If you have limited end supports, you can support the antenna from the middle wire in
each fan and let the outer wires angle downwards. This should give an SWR less than 2 : 1 or
so over the entire range from 7 to 30 MHz. Obviously it could be made larger to include 80 and
160m, but you start to run into the problems with radiation patterns at some point. If you are
only using ham frequencies, one of these in addition to inverted vee dipoles for 80 and 160m
might give you the coverage you need.