TRANSMISSION
performance can often be LINES ... improved a hundredfold! In this article we'll give you some tips on how to select the best transmission line for your YOUR PIPELINE station and how to hook it up. Balanced feedlines TO THE Years ago all amateurs used balanced open-wire feedlines made from no. 12 or 14 (1 .5- OUTSIDE 2.0mm) copper wire, separated every two or three feet (1 m) by spacers as shown in Fig. 1A. In the very early days most \NORLD amateurs made their own feedline spacers by boiling wooden dowels in beeswax; later on, after ceramics were perfected, several firms made low-cost ceramic spacers How to choose which were very popular. the right feedline Following the war, plastics with good rf insulating properties for your station became available and they and how to get quickly replaced the more the most out of it expensive ceramic spacers. Plastic spacers also lended themselves to the manufacture of commercial "ladder line" - whereas wood and ceramic spacers had to be tied in place, plastic spacers had a low BY JIM FISK, W1HR melting point so they could be attached by simply heating the wire and pressing it into the Receiver, transmitter, building blocks in your station plastic. transmission line, antenna - - like a chain which is only as Twinlead or ribbon line, Fig. each is equally important to the strong as its weakest link, you 1 B, was the next step in the operation (and success) of your can't neglect one of the links development of balanced amateur station. No matter how without affecting overall station feedlines. Originally designed good your antenna is, or how performance. Your for use with television powerful your transmitter, if transmission line is one of receivers, it consists of two your receiver is insensitive or these links. parallel wires imbedded in a unable to separate strong, After spending a good chunk web of polyethylene plastic. nearby signals, you won't be of your hard-earned money on a Although television twinlead able to hear all the stations transmitter and receiver, and was an inexpensive answer to that can hear you. Likewise, if putting up the best antenna the feedline problem, the your receiver is super, but your you can afford (or your landlord conductors were pretty small transmitter isn't doing the job will allow), the transmission for transmitting purposes, and it should, you will be able to line is often an afterthought. losses increased dramatically hear weak, faraway stations Little attention is paid to its whenever it rained. The that your transmitter just can't selection or installation, and manufacturers solved the reach. So it is with each of the yet, for only a few dollars, its conductor problem by
32 m May1977 introducing heavy-duty twinlead conduits, taped to tower legs, with heavier conductors which buried in the ground, or even was marketed specifically for run underwater. Although transmitting purposes. The balanced feedlines predate effects of rain were minimized coaxial cable so far as amateur to a certain extent by the use use is concerned, the basic of the tubular construction coaxial structure received shown in Fig. 1C which considerable attention from provides a longer path between telephone engineers in the late the two conductors. 1800s who were trying to There was a time, 45 or 50 Fig. 2. Construction of modern , flexible improve telephone coaxial cable. Center insulator may be years ago, when practically polyethylene, Teflon, polyfoam , or other transmission, particularly in every transmitting station in low-loss plastic. Some types may have a transatlantic submarine cables. the world used open-wire d oub l e bra i d coveri ng for e xtra In the early 1930s Bell feeders - there was simply shielding, or silvered conductors for low Telephone engineers started nothing else available. loss. experimenting with coaxial Although balanced feeders had lines for the transmission of very low loss, a big plus, they high-frequency radio waves, also had their problems: They by everyone, the transmitters of and found that they worked had to be kept well away from the time were designed to very well. The rigid, air-spaced, metal objects; they had to be match them. That isn't true solid copper lines used in installed high above ground; anymore, but some of the these studies weren't practical bends and turns in the line had popular amateur antenna for amateur stations, but to be very gradual, and they handbooks still persist in commercial broadcasters often twisted in the wind, pushing the use of twinlead quickly adopted them as shorting out the two and open-wire feeders. This is standard equipment. Most of conductors. Since balanced, not to say that balanced these early coaxial lines were two-wire feedlines were used feedlines are all bad - in some built with ceramic disks or applications they are a very beads spaced at intervals along good choice - but modern the line to support the inner amateur equipment is designed conductor, a technique that is for use with coaxial cable and still used today (Fig. 3). 0 BALANCED OPEN·WtRE LINE is not directly compatible with Flexible coaxial cable first balanced feedlines. In fact, if appeared on the scene in the you were to make a survey of late 1930s, but it was =JC: --=------_J active amateurs, you'd probably 0 T WI NLEAD find that 99 out of 100 are using a coaxial transmission line. 3~- -- - j Coaxial cable e TUBULAR TWINLEAO The coaxial transmission line Fig . 1. Three types of balanced feedlines is inherently unbalanced and which are sometimes used by radio consists of a center conductor amateurs. Many years ago the open-wire which is surrounded by a feeder in A was used by practically every grounded shield or outer radio station in the world; today it is seldom seen. Tw inlead B was originally conductor as shown in Fig. 2. developed for home television receivers, Since the rf field is contained Fig. 3. Co nstruction of coaxial air altho ugh heavy-duty versions are completely within the shield, insulated line which is used by high avai lable fo r transmitting use. The this type of transmission line is power, commercial stations. This ar tubular twinlead in C has somewhat rangement was first devised more than lower loss than conventional twinlead, not sensitive to nearby objects 40 years ago, and except for the use of but is available only in low power ver - unlike balanced feeders, it modern plastic spacers (usually Teflon), sions. can be run through metal it has changed very little.
May1977 m 33 expensive, and amateur trouble. However, commercial transmitters of the day were television was on its way, and designed for balanced the impact of television feedlines, so few amateurs interference quickly put an end were interested in giving it a to unshielded radio try. Although this cable used transmitters - and their woven copper braid for the balanced feedlines. Although outer conductor (and a natural balanced feeders don't radiate Fig. 4. Every transmission line is made rubber weather shield), the under ideal conditions, few up of distributed capacitance and in· resemblance to modern flexible amateur stations were ideal, ductance, as shown here. When rf coax stopped there. with the result that the feedline energy is applied to t he line, these com ponents set up a fixed relationship be· Polyethylene was discovered in often radiated unwanted rf into tween voltage a nd current, t hereby 1937, but wasn't available in nearby television sets. Coaxial establ ishing t he characteristic im commercial quantities until transmission line didn't solve pedance of the line. much later, so this first flexible all the TVI problems but it did coax used ceramic spacers like keep the rf bottled up until it Any transmission line, its more expensive cousins. reached the antenna where it balanced or unbalanced, Some manufacturers tried belonged. twinlead or coaxial line, has a using natural rubber as a certain amount of capacitance flexible dielectric, but the Feedllne characteristics and inductance per unit length losses were far too great for If you've read the operating as shown in Fig. 4. These use at radio frequencies. manual for your transmitter (or components are distributed Most of the military transceiver), you know that the equally along the length of the equipment built during the war rig likes to " look into" a 50- line (called distributed was designed for use with ohm load, with a standing-wave components) and don't behave coaxial transmission lines, and ratio (swr) not greater than 2:1. exactly like individual the armed services bought You have probably asked capacitors and inductors thousands and thousands of yourself, " Why a 50-ohm load? (which are called lumped feet of flexible coaxial cable Why not 25, 75, or 100 ohms, or components). When rt energy is which used polyethylene as the some other value?" This is a appl ied to the transmission center insulator. After the war good question - with an line, the distributed ended, much of this cable answer that has an interesting components establish a fixed showed up on the surplus story behind it. But before we relationship between voltage market, but many amateurs get to the story, let's talk a and current which, from Ohm's were still using transmitters little about one feedline law, results in the from the 1930s and didn't think parameter that's always listed characteristic impedance of the conversion from balanced in the catalogs: Characteristic the line. feedlines was worth the impedance. Since the distributed capacitance and inductance along the line are a direct result of conductor diameter and spacing, and the insulating material used between them, it follows that the characteristic impedance of the line can also be calculated from the same dimensions. For coaxial transmission line the characteristic impedance is determined by the ratio of the size of the outer conductor, D, to that of the inner conductor, d , as shown in Fig. 5. Assuming the outer diameter is held constant, say at 1 inch (25mm), the smaller the center conductor, the higher the characteristic impedance. Unfortunately, there is no one ratio of conductor diameters which is optimum for Braider machines in Amphenol's Chicago plant; machines cover thousands of feet of all important transmission coaxial cable daily with copper braiding (photo courtesy Amphenol). parameters. For example, a
34 ~ May1977 coaxial line which has high-frequency amateur bands 2 .0 minimum loss (called because most of the commonly \ attenuation) is not optimized used 50-ohm coaxial cables can J. 8 for maximum power, and a line easily handle the maximum \ which is designed for amateur power limit of 1000 1.6 \ maximum breakdown voltage watts. ... does not have minimum loss. "'u \.-' The attenuation minimum ~ 1.4 ' ~ "'!. This is shown graphically in which occurs at 77 ohms (Fig. 0: ~ f( \~ Fig. 6. As can be seen, power 6) explains why the CATV ~ 1.2 handling peaks at a companies use 75-ohm coax in n. characteristic impedance of 30 their cable distribution systems ~ !...... ~ 1..- ~ 1. 0 ,,...... " ohms, breakdown voltage ..J r-:- .... - they're looking for minimum 4J"' 0: ,. -- -- peaks at 60 ohms, and signal loss over a cable which i , '· - 0.8 ,..,o minimum attenuation occurs at may be thousands of feet long. I ' 'i--~~~" -~~ about 77 ohms. In amateur stations the I ~ .,.. ":
May1977 m 35 feet (30m) of transmission line, it's very easy to translate this figure to the actual length of 10 _ r"'-. . : -:~---, ~--~-.-.....: line which you are using. This 60 - - J """" I +---._J__ - -\--_..__-I--- is another benefit of using dB - "- ____j_ I I 50 values - they are directly "- I -'-+----+--.,...-t-·...J -~I proportional to line length. If I 40:-- :~·+__j-= j _:_I _ · -,~~I - L~ - - I• I 100 feet (30m) of cable has 4 1 : - ~ . 1 ...... ·-""==- ~-,--+---+-+--t--i dB loss, for example, half that length or 50 (15m) feet will ~ ~ ~ -, 1+ I ', ~- : --- exhibit half the loss or 2 dB; 40 feet (12m) w i 11 have 40 per cent 1- 1- iii 20 -=P-:_J ~_J -11-: --· I f-· ~"' _:: =- of the loss or 1.6 dB; 30 feet : ~~- - -- ~ - 1-t--t-~ (9m) will have 30 per cent of •5 :_ ' _ :__j t J -"'l~,._-,:::=::::=:t=1 the loss or 1.2 dB, and so on. : : : j ! i ...___.._. t--t---i Conversely, 150 feet (45m) of - I . . L l____l_ -+----+--+-+----+--+-~----< 10=.LL1 - I i --- - ~K line will have 1.5 times the loss 0 .:; ~ 6 1 10 of 100 feet (30m) or 6 dB, 200 TRANSMISSION LINE LOSS (dB ) feet (60m) will have twice the Fig. 7. Decibel or dB values expressed in term s of power ratio. loss, etc. Since cable loss in dB is directly proportional to cable length, the use of dB simplifies loss calculations, as discussed in the text. Choosing coaxial cable Table 1 lists the important operating parameters for the various types of coaxial cable has 100 watts output. Because lines commonly used by which are commonly used by of loss through the relay, only amateurs. RG-8/U, for example, amateurs. At last count there 89 per cent of the power at the has 0.66 dB loss per 100 feet were something like 300 input is available at the output, (30m) at 14 MHz. With 100 different types of coaxial cable or 89 watts. The loss through watts into the transmitter end on the market, but many of the filter reduces this still of the transmission line, this these are specialized types further, so 93 per cent of 89 means that 86 watts will make which are not designed for use watts (about 83 watts) is it to the antenna to be radiated. as rt transmission lines; available at the input connector However, this figure assumes RG-62/U , for example, a 93-ohm to the transmission line. Since that the antenna is matched to line which is used for pulse only 50 per cent of the power the feedline - that the input work, or RG -125/U, a 150-ohm makes it through the impedance to the antenna is line with extremely low transmission line, the power the same as the characteristic capacitance per unit length. into the antenna is about 41 .5 impedance of the coaxial cable. There are coaxial cables with watts or 41 .5 per cent of the If the line is not matched to the lead sheaths for underwater total power delivered by the antenna, the loss will be installations, cables with transmitter. This is exactly the slightly higher, as will be braided steel armor, and even same as 3.8 dB loss, as can be discussed later. some with very high seen from Fig. 7, but the dB Note from Table 1 that the attenuation for use in test values are obviously a lot higher the frequency, the equipment. However, most of easier to use because they can higher the loss per 100 feet. the types used by amateurs are be simply added together. This is a characteristic of all listed in Table 1. If you're not familiar with types of feedlines and explains Perhaps the most popular using dB, here are some why cable selection at vhf and type of coax is RG-8/U, a guidelines which may be uhf is even more important medium-size cable with a maxi helpful: First of all, 1 dB is than it is for the high-frequency mum power rating of 2000 watts about the minimum change bands. Note also that the at 28 MHz, and somewhat more which you can detect with your smaller diameter cables have ANTENNA ear; - 3 dB represents 50 per much higher losses. RG-58/U, ~ 1.5W cent loss; -6 dB corresponds for example, has 3.5 dB loss to 75 per cent loss; and -10 per 100 feet (30m) at 50 MHz - dB is the same as 90 per cent with 100 watts into the TIR LOWPASS RELAY FILTER T loss. With these values in transmission line, only 45 watts get to the antenna - the other ·0., d8 · O.J dB mind, it's not too difficult to I0.891 10.931 interpolate for values in 55 watts are used to heat up Fig. 8. Example of a loss calculation us between. the cable! ing dB. The total loss of 3.8 dB cor Given in Table 1 are the loss Although dB loss figures are responds to power transmission of figures for coaxial transmission almost always given for 100 41 .5%.
36 ~ May1977 Table 1. Important operating characteristics of 50- and 75-ohm coaxial cables used by amateurs. Types with non contaminating jackets (designated NV} should be selected for long life. Power rating is given for 30 MHz and is somewhat higher on the lower frequencies; this is power into the transmission line, not transmitter input power.
Loss In dB/100 Feet (30m) Power Rating Cable Zo at 30 MHz Type Jacket* (Ohms) 3.5MHz 7.0MHz 14.0 MHz 21 MHz 28MHz (watts) RG-8/U v 52.0 0.29 0.43 0.67 0.81 0.98 1900 RG-8A/U NV 52.0 0.29 0.43 0.67 0.81 0.98 1900 RG-11/U v 75.0 0.34 0.55 0.81 1.0 1.2 1375 RG-11A/U NV 75.0 0.34 0.55 0.81 1.0 1.2 1375 RG-17/U v 52.0 0.13 0.22 0.35 0.42 0.52 6400 RG-17A/U NV 52.0 0.13 0.22 0.35 0.42 0.52 6400 RG -58/U v 53.5 0.54 0.82 1.25 1.6 1.9 575 RG-58A/U v 52.0 0.69 1.05 1.6 2.1 2.5 500 RG-588/U NV 53.5 0.54 0.82 1.25 1.6 1.9 575 RG-58C/U NV 50.0 0.69 1.05 1.0 2.1 2.5 500 RG-59/U v 73.0 0.55 0.80 1.15 1.4 1.7 700 RG-59A/U NV 75.0 0.55 0.80 1.15 1.4 1.7 700 RG-598/U NV 75.0 0.55 0.80 1.15 1.4 1.7 700 RG-8 Poly foam v 50.0 0.28 0.41 0.60 0.75 0.90 2000 RG·11 Polyfoam v 75.0 0.32 0.51 0.73 0.93 1.10 1425 RG-58 Polyfoam v 50.0 0.80 1.18 1.70 2.1 2.5 500 RG-59 Polyfoam v 75.0 0.40 0.59 0.88 1.1 1.3 750 •vindicates a vinyl jacket; NV Indicates a non-contaminating vinyl jacket. than that on the lower high loss and lower power limits a small diameter cable frequency bands. Attenuation than RG-8/U. It also costs less, recommended for short cable ranges from about 0.3 dB per which may be a consideration, runs between equipment in 100 feet (30m) on 3.5 MHz, up but since the difference is your station; RG-11 /U is to nearly 1 dB per 100 feet something less than $10 per medium sized cable (30m) on 10 meters. RG-8A/U 100-foot roll, in the long run I recommended for long runs to (also marketed as RG-213/U) think RG-8/U is well worth the your antenna. Except for their has identical electrical char slight extra cost. However, for impedance difference there is acteristics but has a so-called connections between various little practical difference "non-contaminating" outer equipment in your station between RG-8/U and RG-11/U, jacket which means that it has where short lengths of cable or between RG-58/U and considerably longer life. The are used, RG-58/U is an RG-59/U for that matter. black plastic covering on excellent choice. RG-8/U contains chemicals At the other end of the size Standing-wave ratio (called plasticizers) which, spectrum Is RG-17/U , a heavy, As was mentioned earlier, when exposed to sun and large diameter cable which has the feedline losses given in weather, leach into the center less than half the loss of Table 1 assume the input insulator and increase cable RG-8/U. It is also much more impedance to the antenna is loss. In practical terms this expensive so it's seldom used the same as the characteristic means that it should be by amateurs unless the impedance of the transmission replaced about every five years. transmission line is several line. However, if the antenna The newer RG-8A/U (or hundred feet long. "Non feedpoint impedance is not the RG-213/U), however, doesn't contaminating" versions of same as that of the line, a have this problem and has a RG-17/U are RG -17A/U (also mismatch results, and a portion useful life of 20 years or more. designated as RG-218/U), and of the rf energy is reflected Since the cost difference is RG-178/U, which has a silver back down the line toward the only a few cents per foot, plated double braid and is transmitter. This is most easily RG-8A/U or RG-213/U is a better designed for uhf use (also measured In terms of standing choice. designated as RG-177/U). wave ratio or swr. Although the (or Its later "non-contami Although I haven't mentioned complete story of reflected nating" version, RG-588/U and the 75-ohm cables listed in power and standing-wave ratios RG-58C/U), but note that It has Table 1, the same sort of is beyond the scope of this significantly greater arguments apply. RG-59/U is article, Just let me say that,
May1977 m 37 90
80
- !dB 70 ' . 60
-2 dB ~ 00 2 0 Vi 100 90 :ii"' 2 80 "'
70 60 21 MHz - 2dB ,_ , 6 0 0 0
6 10 12 6 8 10 12 SWR " SWR Fig. 9. The effect of swr on transmission line loss. So long as line loss with a matched load is low, as it is on the high· frequency amateur bands, swr has little effect. contrary to popular belief, higher loss cables such as can be reduced only at the load reflected power is not lost. RG-58/U , high swr can end of the line, so you may be However, a mismatched load drastically increase power loss. able to get a better match to does have the effect of This is another good reason to the line by pruning your increasing power loss because choose lower loss lines such antenna - a better match the portion of the rf power as RG-8/U or RG-11 /U. If high means lower swr. On the other which is reflected makes two swr doesn't cause significant hand, you may be using an complete trips through the power loss on the high antenna with a feedpoint cable - once toward the frequency bands, why, you may impedance that presents more antenna, the other back toward ask, do so many operators than 2:1 mismatch to your the transmitter. Since the worry about getting the swr transmission line - such as a reflected power makes two down to 2:1 or less? Because very short vertical, or a dipole trips through the line, each the tank circuits in modern that is operated at a frequency with attenuation, it receives transmitters simply aren't well away from resonance. This twice as much loss as that designed to cope with an swr often happens on 80 meters portion of the power which greater than that. They could when you want to use a single makes only one trip down the be, but it would increase their antenna to work CW on the low line to the antenna. selling price. Also, high swr end near 3.5 MHz, and 75-meter However, this slight seriously degrades the power sideband near the upper band additional loss is not serious handling ability of the line so if edge. If you cut the dipole for on the high-frequency bands you run high power and have the center of the band, at 3750 where cable attenuation is low, high swr you may find that you kHz, you'll find that the swr at as shown in the four graphs in are operating above the power 3.5 and 4.0 MHz will be close to Fig. 9 for 100 feet (30m) of limits of the line. If this 5:1. The solution here is to use RG-8/U on 80, 40, 20 and 15 happens, the line may actually an antenna tuner or meters. Even at 21 MHz, where get hot enough to burst. Transmatch between your cable attenuation is relatively On the other hand, you're not transmitter and the feedline as high at 0.81 dB per 100 feet, an going to gain much by shown in Fig. 10. When the swr of 5:1 - high compared to twiddling with your high Transmatch is tuned properly, normal standards - increases frequency antenna to get a it will present a 50-ohm load to total line loss by less that 1 dB; perfect match. So long as the your transmitter. So long as the the difference would be barely swr is less than 2:1, your swr on the line is within perceptible in terms of received transmitter doesn't care, and reasonable limits - say 10:1 signal strength. On 3.5 MHz the the station at the other end with RG-8/U on 80 meters - effect of swr is perhaps even won't be able to tell the the stations you are working more surprising: An swr of 10:1 difference. won't know the difference. increases line loss by the If the swr on your same, barely perceptible 1 dB. transmission line is greater Installing your If cable attenuation is high, than 2:1, you probably won't be transmission line as would be the case at vhf or able to tune your transmitter to If you're using coaxial cable, it's if you were using long runs of its maximum rated input. Swr pretty hard to get into trouble,
38 m May1977 but here are some tips that will small cables like RG-58/U and help get the most out of it. RG-59/U . 1. Use coaxial connectors 6. Don 't install coaxial cables which are designed for the TRANSMtnER LOWPASS ANT(NHA near hot-air ducts, furnace FILTER TUNER cable you are using, and make pipes, or steam radiators. The sure they are properly installed. Fig. 10. Using an antenna tuner to pro· thermoplastics used in cable Complete assembly directions vide a good match to your transmitter. manufacturing get soft with are usually provided when you Swr on the line between the tuner and rising temperatures and may the antenna is of little importance if line swell ou.t through the braid. buy the connectors, but if loss is low. The lowpass filter behaves they're not, any recent edition as it should only when it's terminated in The center conductor can also of the ARRL Antenna a matched load, so it should be placed sink, from the force of gravity, Handbook will show you how. between the transmitter and the antenna through the temperature-soft tuner. ened center in sulator and short 2. If you use a coaxial out the cable. When either of connector where it's exposed these things happen, the cable to the weather, weather-proof it you do find evidence of is permanently damaged and with several coverings of moisture you probably won't must be replaced. plastic electrical tape. If have to replace the entire moisture seeps into the cable length of cable - unless it has 7. Use cable hangers or through an unprotected been exposed to moisture for a supports for long cable runs. connector, at the fairly long time. If you cut it The "Pop-Top" lids from drink very least it will increase loss back a foot (30cm) at a time, cans fit nicely over RG-8/U and - at the very worst it will you'll probably find that several make excellent cable hangers short-circuit the line. feet (1-2m) back from the - just drive a small nail damaged end the line is as through the tab into the nearest 3. If you suspect that good as new. support. If you run a coaxial moisture has gotten into the cable to the top of your tower, line, you can often tell by 4. If you have to run your coaxial line under a window to tape it to one of the tower legs removing the connector. The every four or five feet (1-2m). copper braid should be bright get to your antenna, don't push and shiny, and the center the sash down so hard it 8. Inspect the cable every six insulator should look the same mashes the cable. If the cable months or so to make sure it as it did when you originally is pinc hed, it changes its hasn't been knicked, cut, or put the connector on. If you're characteristic impedance and otherwise damaged. If you in doubt on this point, compare inc reases the swr on the line, discover the problem soon it with a piece of new cable. If no matter how well your enough, you may be able to antenna is matched. If you repair it before moisture want to run your cable under a destroys the usefulness of the window, make a wood spacer complete cable. If you find a that runs the full width of the damaged spot in the center of window and close the sash on a long run, cut the cable in two, it. Then drill a hole through the remove the damaged portion, spacer slightly larger than the and install connectors on each outside diameter of the cable end. Then buy a straight (or large enough to pass the adapter (such as a PL-258) and connector if it is already connect the two cables installed). together. If the junction is 5. Don't install coaxial cable outside, seal it well with plastic with tight bends; this changes electrical tape. the conductor spacing which If you carefully choose your affects the impedance of the feedline, and follow these line. If you have a very tight simple guidelines, your bend in the cable, the center transmission line should last conductor may eventually work for a good many years. In fact, its way through the insulator if you buy coaxial cable with a until it touches the outer braid, non-contaminating jacket you shorting out the line. What do I may never have to replace it; I mean by a tight bend? A bend know of one length of RG-8A/U radius less than about ten in an underground run to a times the diameter of the cable tower that doesn't have any Some of the many types of coaxial cable which are available on the market. - about 4 inches (10cm) for more loss now than it did when Several of the cables shown here have medium-sized cable such as it was buried fifteen years ago! double braids for additional shielding . RG -8/U , or 2 inches (5cm) for HRH
May 1977 ~ 39