[ Reply | Next | Previous | Up ]
From: Alan G3NOQ
Date: 27 Nov 2002
Time: 08:14:56
Remote Name: 20.138.254.2
David, Marc - What I said could be clarified. When I said the "radiation field" I do not mean the far field, I mean the total EM field round the antenna and further away, that transports the power away from the antenna. If you imagine a notional surface like a ballon enclosing an isolated antenna, all the radiated power, as it leaves the antenna, passes through the rubber surface of the balloon from the inside to the outside. In a lossless medium like free space, or with microwave antennas that are clear of the ground, it does not matter what the shape or size of the balloon is, as long as it completely encloses all the conducting surfaces of the antenna, and then all the radiated power passes through the balloon regardless of its shape and size (this was all linked to the Maxwell theory by J.H.Poynting, a professor at Birmingham University, in his paper of 1884). Therefore the efficiency is easy to define as the power passing through the enclosing surface (balloon), divided by the power accepted at the terminals. . . . Now with HF antennas it is more complicated because the medium (ground etc.) is not lossess. In this case the power passing through the "balloon" reduces as its size increases, or, to put it another way, the fields decrease faster than 1/r. Strictly speaking and to be pedantic, the definition of gain and efficiency is now meaningless, but it has been adapted to cope with this situation because gain and efficiency are concepts which are too valuable to throw out. . . . . I do not see a conceptual problem for small HF antennas with two terminals, like dipoles and loops, it just means that the experiment has to be thought out in such a way that the result is based on power through the "balloon" and not involving the lossy medium, so the balloon has to be a small one enclosing only the antenna itself. There are various ways of doing this, but I should stop to see if you are with me so far (I know you understand surface integrals but the balloon idea is to help casual readers who might not wish to get involved with that kind of thing). . . . .On the question about NEC, lossy ground was coded into NEC3 and NEC4 and it seems to work well if used properly. NEC4 is preferable especially for small antennas because the basis functions were redefined to improve numerical stability. With NEC2 and NEC3 the basis functions were constant, sine and cosine, so that when the segments were short the constant term would fight the cosine. With NEC4 the basis functions are constant, sine and 1-cosine, which is better because sinx approximates x and 1-cos approximates x^2, so there is less conflict. . . . . . Kind regards, Alan