From owner-qrp-l@Lehigh.EDU Sun May 23 14:32:20 1999 Received: from fidoii.CC.lehigh.EDU (fidoii.CC.lehigh.EDU [128.180.1.4]) by oucsace.cs.ohiou.edu (8.8.5/8.8.5) with ESMTP id OAA07851 for ; Sun, 23 May 1999 14:32:19 -0400 (EDT) Received: from localhost ([127.0.0.1]:3066 "HELO Lehigh.EDU" ident: "IDENT-NOT-QUERIED") by fidoii.cc.Lehigh.EDU with SMTP id <53950-41060>; Sun, 23 May 1999 14:29:23 -0400 Received: from nss4.CC.Lehigh.EDU ([128.180.1.13]:2598 "EHLO nss4.cc.Lehigh.EDU" ident: "IDENT-NOT-QUERIED") by fidoii.cc.Lehigh.EDU with ESMTP id <53746-40552>; Sun, 23 May 1999 14:28:32 -0400 Received: from mailgw.flash.net (deimos.flash.net [209.30.0.43]) by nss4.cc.Lehigh.EDU (8.9.1a/8.9.1) with ESMTP id OAA52334 for ; Sun, 23 May 1999 14:28:31 -0400 Received: from [209.30.50.46] (p46.utc1.dialup.abq1.flash.net [209.30.50.46]) by mailgw.flash.net (8.9.1/8.9.1) with ESMTP id NAA27155; Sun, 23 May 1999 13:28:16 -0500 (CDT) Message-Id: <199905231828.NAA27155@mailgw.flash.net> Date: Sun, 23 May 1999 12:28:41 -0700 Reply-To: jamesd1@flash.net Sender: owner-qrp-l@Lehigh.EDU Precedence: bulk From: "James R. Duffey" To: "Low Power Amateur Radio Discussion" Subject: Dr. Megacycle's Feedline Primer (long) Mime-version: 1.0 Content-type: text/plain; charset="US-ASCII" Content-transfer-encoding: 7bit Content-Transfer-Encoding: 7bit X-To: kc5tja@dolphin.openprojects.net X-Cc: "qrp-l" X-Mailer: Microsoft Outlook Express for Macintosh - 4.01 (295) X-Priority: 3 X-Orcpt: rfc822;qrp-l@fidoii.CC.Lehigh.EDU X-Listprocessor-Version: 8.1 beta -- ListProcessor(tm) by CREN Status: RO Sam (KC5JA) - You asked several pertinent questions about feedlines. Let me take a stab at answering them; "What's so special about a parallel feedline versus a coaxial system? What are the advantages to each, and why?" "I keep seeing plenty of references to things like, "300-ohm twinlead is much better than coax," or "450-ohm ladder line is...", etc. Yet, I have failed to spot a satisfactory explanation as to WHY it's better." "Also, slightly unrelated, if a parallel feedline system is known to be better than a coaxial system, why are the vast majority of radios built around a 50-ohm, unbalanced feedline system? (My initial guess is cost -- with the popularity of computer networks, RG-58U has become pretty darn cheap in quantity)." You asked for a website on transmission lines. I don't know of any that are explicitly devoted to transmission lines, but L. B.'s Web Page is always a good place to start with when you have an antenna question. He has some good discussions on trasnmission lines. The URL is: http://funnelweb.utcc.utk.edu/~cebik/radio.html Put it in your bookmark file. Lets first look at why coax came to dominate the ham radio feedline world. In doing that I'll answer the last queation first. I have a historical bent and like to approach problems in that fashion. Prior to World War II, coaxial cable was not used widely in the Ham world. It was expensive and hard to get. Almost all ham radio antenna installations were based on balanced feeders such as center fed Zepps, or on single wire feeders such as the Windom, or on end fed antennas with balanced feed lines (the end fed Zepp) or on "no feeder antennas" like the end fed wire. After World War II vast quantities of surplus coaxial cable was dumped on the market: hams got addicted to coax, and as a result operation of antennas with balanced feeders faded out. In the 60s, after ferrites became widely available, baluns became easy to make and balanced feed lines started to come back into use. Today they are as widely used as at any time since before World War II. Cheap coax was not the only undoing of balanced feeders. Television became widespread after World War II, and TVI was a major problem to Hams. At the time most transmitters coupled the antenna directly from the final tank circuits. This was easy to do with balanced feeders, since two taps equally spaced on either side of the center of the final tank inductor did the job. Antenna matching could be done by varying the position of the taps. (Keeping one hand in your pocket of course.) Unfortunately, the class C finals of the day were rich in harmonics and the harmonics fell in the TV bands. Antennas fed in this way radiated the harmonics almost as well as the fundamental. Ham radio manufacturers dealt with this problem in part by using a pi network between the final and the antenna. An unbalanced pi network is easier to build as it only requires one inductor and single variable capacitors rather than dual section capacitors and a second inductor. This led to transmitters (and later transceivers) with an unbalanced 50 Ohm output. The most popular coax in the US was 50 Ohms and the output ended up at that impedance. In fact, the pi network output functioned very well as an antenna tuner and these old rigs were usually capble of loading up antennas that were mismatched with an SWR up to 4:1 or 5:1. Operation of a balanced antenna required an external antenna tuner such as the Johnson Matchbox or baluns. The baluns of the day were airspaced coils and usually efficient only on 1 band. Most Hams of the day just went with coax and decided to forego the added expense and complexity of an external tuner or balun. The introduction of the Novice license in the early 50s added to the stampede to coax as most novices were just learning radio and to a neophyte coax requires less knowledge and understanding to use than balanced feeders. I like to simplify things down to basics. Balanced feedline is lower loss than coax and most of its advantages lie in that fact. Coax is easier to deal with mechanically and most of its advantage lies in that. Look at the table of feedline loss in either the ARRL Handbook or The ARRL Antenna Handbook. If you don't have either of these references you should buy one or the other or both and read the chapter on Transmission Lines. It will explain things in more detail and more accurately than I can here. If we look at 7 MHz, RG-58 shows a loss of 1.0 dB per hundred feet, RG-8 coax shows 0.6 dB loss per hundred feet, 300 Ohm twin lead a loss of 0.25 dB per 100 feet, window line (either 450 Ohm or 300 Ohm) a loss of less than 0.1 dB per hundred feet and homebrew open wire line constructed of 12 ga wire a loss of less than 0.1 dB. At 30 MHz, similar numbers are 1.3 dB for RG-8, 2.4 dB for RG-58, 0.6 dB for 300 Ohm twinlead, 0.15 for window line, and 0.1 dB for the home brew line. As you can see loss rises with frequency. At 7 MHz, the feedline loss with either coax or balanced feeder is acceptable, particularly when one takes into account the potential losses in a tuner and/or balun generally required for the balanced feeders. At 30 MHz the losses in RG-58 have become excessive and it probably should not be used for long runs. Now, the above figures are for feedlines operating when they are terminated in their characteristic impedance. This means that the antenna impedence is equal to the transmission line impedence, and that the transmission line is driven with a driver (transmitter) that has a characteristic impedence of the coaxial line. A half wave dipole has a characteristic impedance of 50 Ohms at 0.15 wavelengths above the ground. Lets stay at 7 MHz. When fed with RG-8 coax the loss in the transmission line loss will be the loss quoted in the table, 0.6 dB. When fed with 300 Ohm twin lead the SWR will be 6:1 and the loss in the transmission line will be increased. The handbook has a table that lets us determine this loss, for the 300 Ohm twin lead the additional loss will be about 0.5 dB for a total loss in the 300 Ohm twinlead of 0.75 dB, or slightly greater than the coax fed antenna!!!! So balanced feeders are not always lower loss than coax feeders. The loss depends on the SWR and the SWR depends on the application. But look at the same antenna fed with window line or home brew open wire line; the loss SWR will be 9:1, from the table the additional loss will be less than 0.4 db for a total loss of less than 0.6 dB which is slightly better than the Coax case. Balanced feeders really come into their own when feeding the same antenna on multiple bands. Lets look at a 100 ft dipole erected at 50 feet. This is pretty close to the popular G5RV antenna and the Handbook elucidates its properties so that I don't have to do the calculations. :^)=. We want to use this antenna on all bands from 40 M up and perhaps for an occasionally excursion to 80 M. Frequency Feedpoint Impedance SWR (50 Ohms) Loss Loss (MHz) (Ohms) for RG-8 for 450 window line (Resistance + jReactance) 1.8 MHz 4.5 - j1670 1800:1 26 dB 12 dB 3.8 MHz 4.0 - j360 63:1 5.7 dB 0.9 dB 7.1 MHz 481 + j960 50:1 5.8 dB 0.3 dB 10.1 MHz 2584 - j3300 134:1 10.4 dB 0.9 dB 14.1 MHz 85 - j120 6:1 1.9 dB 0.5 dB 18.1 MHz 2100 + j1600 65:1 9.0 dB 0.6 dB 21.1 MHz 350 - j1100 73:1 10.0 dB 0.8 dB 24.9 MHz 200 - j370 18:1 5.2 dB 0.4 dB 28.4 MHz 2500 - j1400 65:1 10.1 dB 0.7 dB Well the advantage of balanced feeders should be obvious. The loss is smaller with the balanced feeders in every case. Note that 160 M is unsatisfactory with both feedlines. Other alternates should be pursued on this band, such as tieing the feeders together and feeding it against ground as a vertical. This does not mean you can't feed multiband antennas with coax, you just have to take a different approach. Parallel dipoles or traps are two approaches used very often, but they are beyond the scope of this primer. Another time. One note with coax. You should use a balun or choke balun when feeding a balanced antenna such as a dipole with coax. This will prevent current from flowing on the outside of the coax which can lead to all sorts of unpleasanteries. I would also suggest using a choke balun when feeding verticals for the same reason. The handbook contains easily constructed choke baluns made from winding coax in a coil at the antenna feedpoint. These were designed by W7EL who no longer receives credit for this in the handbook. Balanced feeders are often accused of having feedline radiation, and as a result of this having all sorts of problems including TVI. This is overblown. On a balanced feedline the radiated field will fall away rapidly to zero within a few inches. Unbalanced currents will cause some feedline radiation, but this is a problem with coax as well. Be sure to make balanced antennas as symetrical as possible when fed with balanced feedlines. The balanced feeder should be kept spaced several feeder spacings away from objects. For one inch spaced window line, the line should be kept 3 or more inches away form other objects, particularly metallic ones. When unbalanced antennas such as verticals are fed with balanced line a balun should also be used. Coax is easier to handle than balanced feedline, particularly when bringing the feedline into the shack. Balanced feedline is cheaper, particularly when loss is taken into account. Balanced feedline can be constructed fairly easily, although I find the process some what tedious. Coax is difficult to homebrew. It can be more of a hassle to use balanced feedline. Tuners are usually required and that requires additional adjustments. Getting the balanced feedline into the shack can be a hassle. Some sort of a balanced to unbalanced termination must be made to the trasnmitter. This can be handled by the tuner in some cases. Recommended tuners for balanced feeders include the Z-Match which can be homebrewwed or purchased from EMTECH (kit) commercially, the Johnson Matchbox no longer manufactured but very nice, W6RCA's switched stub tuners, and homebrew link tuners from the handbook. Balanced feeders can be conversation starters. Everybody has seen coax, few people have seen window line. There is no answer to which feedline is "better". It depends on antenna to be fed, what use is made of that antenna, the preferences of the user, and the amount of fuss one is willing to put with in a feedline. I use both feedlines. For a resonant single band or multiple band antenna with impedance near 50 Ohm, I use coax which is the natural choice. For a single antenna to be used on multiple bands with widely varying impedances like the G5RV, I use a balanced feedline. RG-58 is useful below 14 MHz and in short runs (less than 50 feet) at higher frequencies up to 30 MHz or so (but is marginal at 30 MHz). RG-8 is useful at all the HF bands and on short runs at VHF. I prefer the 450 Ohm window line for convenience and low loss. The 300 Ohm TV twin lead is useful in the low loss form, but can be inflexible and difficult to handle. The 300 Ohm window line is a very useful and under utilized line, homebrew line is cheap and can be very low loss, but be prepared to spend an afternoon building it. A good rule of thumb is to try and keep feedline losses below 1 dB. This is easy to do on HF, and in most cases feedline losses can be kept at 0.5 dB or so with only a little more effort at HF. Feedline is a not a good place to economize. It is generally worth spending a little more to get good cable with low losses and high quality construciton. I hope that this helps and is worth the bandwidth. Forgive me Doug? Thanks for letting me get on my soapbox. - Dr. Megacycle KK6MC/5