The water tank is a convex surface (at least on the side you can get to) not a concave surface. However offhand I don't think that you will be able to get far enough away from the surface of the tank to illuminate it properly and the curvature will most likely not be anywhere near optimum for the desired pattern.
I would just mount the antenna on the rail and see what it does. As to a second antenna on the opposite side of the tank there is no need to have the two cables from the power splitter to be the same length. The two antennas are not going to see one another and they are not part of a phased array so cable lengths will be immaterial. There will be some nulls (and peaks) in directions where both antennas happen to be in view but moving a few feet one way or the other by the unit in view will change the phase relationship between the two antennas anyway. Nothing you can do about it. Regards Gary K4FMX _____ From: [email protected] [mailto:[EMAIL PROTECTED] On Behalf Of allan crites Sent: Wednesday, January 31, 2007 12:39 PM To: [email protected] Subject: Re: [Repeater-Builder] Re: Antenna on the side of a water tower I'd suggest you consider the water tank as a reflector of the type called Gregorian after James Gregory of England who devised it about 1660, and the operation of which is described in the book "Antennas for all applications" by John Kraus 3rd edition on pp. 680-684. Illumination of the (water tank as a) reflector (called a concave ellipsoidal surface) with a point source ( a yagi or other such similar directional type of antenna pointed at the reflector ) will yield a wide distribution of RF energy in the desired area of operation (close to 180 degrees) without the multi-lobes and nulls which can and do occur with the use of an omni-directional antenna regardless what ever the spacing from the antenna to the reflector (water tank surface). Since the reflector (water tank) is not a flat sheet, determination of the appropriate spacing from an omni-directional antenna to a spherical reflector is a compromise at best, if one hopes to achieve an optimum radiation pattern in the area of desired operation with-out undesirable nulls. One thing is for sure, you cannot expect more than 180 degrees of operation from the side of the water tank on which the antenna is mounted. Any signal found on the opposite side of the tank is the result of multi-path reflections, and will not and can not be dependable or predictable for use. You may want to mount another directional antenna on the opposite side of the tank to improve the coverage in the opposite area. This obviously would require a splitter ( power divider) to couple the two antennas and /or an additional length of transmission line from the xmtr to the 2nd antenna. Both lines feeding the antennas from the power splitter should be of the same length. And there may be nulls at the 90 degree and 270 degree locations around the tank with two antennas mounted at 0 and 180 degrees. Good Luck! 73, Allan Crites, WA9ZZU "Jim B." <[EMAIL PROTECTED]> wrote: Dave VanHorn wrote: >> I would suggest 1/2 wave away from the surrounding metal as a > minimum, but try and get as far away as you can. >> > > Yes, but what's bugging me is that I'm sure there are BAD distances, > especially up close within 1-2 wavelengths Normally for side-mounting on a normal tower, one wavelength will get you close to an omni pattern, but something as big as a water tower, I don't know. I don't think you'll be able to mount it far enough away, practically speaking. I'd still mount it facing towards the most important area to cover from that site, and vote it. -- Jim Barbour WD8CHL
