On Point: The Yagi Antenna

If you happened to look up during a drive down a suburban street in the US anytime during the 60s or 70s, you’ll no doubt have noticed a forest of TV antennas. When over-the-air TV was the only option, people went to great lengths to haul in signals, with antennas of sometimes massive proportions flying over rooftops.

Outdoor antennas all but disappeared over the last third of the 20th century as cable providers became dominant, cast to the curb as unsightly relics of a sad and bygone era of limited choices and poor reception. But now cheapskates cable-cutters like yours truly are starting to regrow that once-thick forest, this time lofting antennas to receive digital programming over the air. Many of the new antennas make outrageous claims about performance or tout that they’re designed specifically for HDTV. It’s all marketing nonsense, of course, because then as now, almost every TV antenna is just some form of the classic Yagi design. The physics of this antenna are fascinating, as is the story of how the antenna was invented.

“Uda Who?”

Shintaro Uda. Source: IEEE Cincinnati Section

What would come to be known as the Yagi antenna got its start in the early 1920s in the lab of Professor Shintaro Uda in the Tohoku Imperial University in Sendai Japan. Dr. Uda was working in the VHF band and was looking for ways to make antennas more directional. While experimenting with a resonant loop antenna, he discovered that placing a static loop near the antenna tended to shape the signal away from an omnidirectional pattern, almost as if the loop was acting as a reflector.

Together with his colleague Hidetsugu Yagi, Uda experimented with different configurations. They eventually replaced the loop antenna with a simple dipole, and added additional elements, which they called directors, on a long boom to further shape the beam. Using eight directors on a 15-meter wooden boom mounted to the roof of their laboratory, Uda and Yagi were able to communicate over a distance of 135 km at 68 MHz, no mean feat at the time.

Hidetsugu Yagi and “his” antenna. Source: Physics World.com

Having dubbed their invention the “wave projector directional antenna,” it was inevitable that the antenna would be named after someone. How it came to be credited solely to Dr. Yagi is a tale of some treachery on Yagi’s part with a dash of naiveté on Uda’s. Dr. Uda published the first Japanese language papers on the antenna, but for reasons unknown, Dr. Yagi applied for both Japanese and American patents for the antenna with no mention of Uda. The Japanese patent was assigned to the Marconi Company in England, while the American patent went to RCA. With no mention of Uda, and with Dr. Yagi touring the English-speaking world to discuss “his” antenna at various radio engineering conferences, the antenna gradually became simply the “Yagi antenna” or the “Yagi array.”

Ironically, thanks to inter-service rivalries and a silo mentality in Imperial Japan, it was only the capture of a British radar set during the Battle of Singapore in 1942 that introduced the homegrown invention to the Japanese military. The Japanese intelligence officers didn’t even consider “Yagi” to be a Japanese name — they supposed it was just a code word made up by the British.


The chief characteristics of the Yagi-Uda antenna are high directionality and high gain. Given the fact that the length of each element needs to be close to some fraction of the wavelength of the signal, it’s most practical for the higher frequencies, mostly above 30 MHz. That’s not to say that it can’t be used for the longer wavelengths, though — plenty of hams work the 20 m and 40 m bands through a big Yagi.

A Yagi for the ham bands. Note the driven element with feedline, seven directors, and a single reflector. Source: Antenna-Theory.com

As in Dr. Uda’s original design, a Yagi consists of a single driven element parallel to and coplanar with at least two parasitic elements. A minimal design is a single reflector element located “behind” the driven element (relative to the direction of the radio signal) and a single director element in front of the driven element. A practical antenna is likely to have multiple directors, the more of which there are the tighter the directionality and the higher the gain, at least up to a point.

This gives Yagis their characteristic appearance – a horizontal boom with multiple elements arranged perpendicularly. There are some variations, of course — some Yagis have multiple reflectors, or have a corner reflector arrangement. And some antennas, particularly TV antennas, have the parasitic elements swept back at an angle rather than perpendicular to the boom. Additionally, the elements can be arranged horizontally or vertically, depending on the polarization desired.


To understand the Yagi’s design, recall that a plain old dipole antenna in free space has a radiation pattern that is the strongest broadside to the antenna. That results in two big lobes off the front and the back of the antenna, with little signal off the ends. The driven element of a Yagi is just a half-wave dipole, or sometimes a folded dipole to increase the impedance. The parasitic elements shape and direct the beam using constructive and destructive interference.

As Dr. Uda discovered, the parasitic elements can either be inductively or capacitively coupled to the driven element. Inductive elements are slightly longer than half-wave, while capacitive elements are slightly shorter. The directors are all shorter than half-wave and are therefore capacitively coupled, while the reflector is longer and inductively coupled. The difference from the ideal half-wave is small — usually only 10% to 15%.

Constructive and destructive interference in a Yagi antenna. The green wave represents the sum of the red and blue waves. Source: RadarTutorial.eu

Both the reflector and the directors work by reradiating power from the driven element. The spacing of the parasitic elements relative to the driven element determines the phase of the reradiated signal. The reflector, being inductively reactive, reradiates power 180° out of phase with the driven element. The spacing is set so that this causes destructive interference off the back of the antenna, while at the same time being nearly in-phase with the driven signal off the front of the antenna. This results in constructive interference, boosting the power off the front. Similarly, the capacitively coupled directors are spaced so that they reradiate power more-or-less in-phase in the forward direction, while radiating out-of-phase to the rear.

The result is greatly amplified signal toward the directors, and almost none behind the reflector. And recall that antenna theory states that any antenna that transmits can also receive, and with the same characteristics. It doesn’t matter whether the driven element in a Yagi is driven by a 100-watt power transmitter connected to the feedline, or by a few microwatts picked up from a distant TV tower. The directionality and gain will be the same. And Yagis can have remarkable gain – up to 20 dBm when correctly designed.

As useful as the Yagi antenna is, it’s far from perfect. Because of the critical size and spacing of the parasitic elements, Yagis have a relatively narrow bandwidth. Also, the directionality of the antenna can be an inconvenience, requiring that the antenna be rotated to point more or less exactly at the transmitter or receiver.

But if you need to pull in a single distant signal, that directionality is just what you need. The Yagi is a workhorse antenna, and given the impact it has had it’s probably right and good that many have taken to referring to it as the Yagi-Uda antenna.


102 thoughts on “On Point: The Yagi Antenna

      1. In mathematics you can describe any periodic waveform as a series of any other periodic waveform. Most people choose to say that a square wave IS made up of a series of sine waves. Strictly speaking it is not true, it can be described as such.

        It is equally true to say that a sine wave is made up of a series of triangular waves or that a triangular wave is made up of a series of square waves.

        With a bit of “inventive” number crunching it can also be applied to non-periodic waveforms.

        If you google “basis functions” you should get an explanation though I have never done it so can’t be sure about that.

      2. OK, this is getting bizarre now.

        You cannot receive anything at all with any antenna without something to process the signal it receives be it analogue or digital recovered content.

    1. The antenna doesn’t care about the modulation, so long as the Q is enough to permit the bandwidth necessary for the modulation baseband – and few modulation schemes are available that have a baseband so wide that it couldn’t be accommodated with a Yagi designed for that frequency; spread-spectrum is the only one that comes to mind and that’s a problem for entirely different reasons. ALL RF signals are analog – they just carry information through modulation and digital is just modulation consisting of high and low logic values in the time domain(and additionally, usually just information carried on another modulation scheme such as QAM, FM, or PSK. Morse is a “digital” modulation. As is PSKxx, the various JT modes, RTTY, etc etc. Yagis with perfectly fine with all of these modes, and even high-bandwidth image modes.

      “Square tubing”… really? You should really stop embarrassing yourself. There is no such thing as a “digital” radio wave. It’s all sine being shaped into something else.

      1. can confirm, EM fields at single frequencies are sine waves. the digital modulation scheme is basically modulating the amplitude or phase of a single carrier sine wave, which can be demodulated on the receiving end to recreate the original digital signal.

    2. G1! IMHO you’re somewhere in the top 15% of those here. Definitely above my own level. Little pernicious making all that smoke come outta their ears! Hat’s off to ya!

    3. Annie

      You have no idea what you are talking about. Please educate yourself before saying weird things like that. Radio waves are just Radiofrequency. it doesnt matter if the modulation is digital or analog.

    4. Annie, please educate yourself, before making statements like you just did. You were way off track, and demonstrated NO knowledge of what you are talking about with that statement and so are leading others down the wrong path. An Antennas performance has NOTHING to do with the mode the radio is operating on. It is a simple matter of being resonate on the frequency that you are operating.

  1. In the U.S. most TV antennas are log-periodic arrays, not Yagis. Log periodic antennas have a much wider bandwidth, which is desirable for receiving TV broadcasts, since the channels are spread out over a wide frequency range. Yagis are commonly used by amateur radio operators since the bandwidth requirements are not as large.

    1. That’s a fair assessment, though I think it’s mostly in the past since DTV generally eliminated the lower tv channels. So a lot of antennas now are UHF bowties of some sort.

      But, there was always a need to get that one station that wasn’t local. So you could get channel specific tv antennas, and those were yagis. There were also good antennas for the FM broadcast band, and those were yagis, only needing to cover 20MHz.

      I had a 3element 6 metre antenna, that’s 50MHz, for a while, and that always seemed “big enough”, so it’s hard to imagine 20M yagis, though I know they exist. But I’ve only seen them from distance.


      1. There’s a common design called a “triband Yagi” or just “tribander”, which is a three-element Yagi with coils on each element to make it resonate around 14 MHz (20 m), 21 MHz (15 m), and 28 MHz (10 m). They’re less efficient than single-band optimized antennas, but they’re much easier to put up, and easier to rotate, than a stack of single-band Yagis.

  2. While Yagi antennas have been used for TV, most of the TV antennas from the ’60s and ’70s are actually LPDA (Log-periodic dipole arrays), which are a somewhat different design.

  3. Excellent piece on the yagi-uda, a masterpiece from the days of sliderules. My memory is that in a well designed antenna the director at one frequency would become a reflector at another.

  4. I think that most multi-channel TV antennas (other than the ones that are stacked dipoles in front of a single reflector) are actually log-periodic to get around the bandwidth problem. Similar principal, though – directors/active elements/reflector(s).

  5. Specifying an antenna’s gain in dBm makes no sense. dBm is not a gain specification, it is an absolute amount of power. “20 dBm” is exactly the same as “100 mW”.

      1. “dBi” is gain relative to an isotropic antenna, which radiates equally in all directions. It’s a theoretical construct — no physical antenna can be perfectly isotropic. “dBd”, on the other hand, is gain relative to a simple half-wavelength dipole antenna. There’s no such thing as an “isotropic dipole”.

        1. Sorry for the misnomers. My ‘o so clever’ droid keyboard thinks no word can have a capital letter in its middle. Please replace Dbi with dBi and Dbd with dBd in my previous post.

    1. Wow! I’m gonna have to build me one of these things! 100mW of power for any signal! Who needs a power supply for their radio? I’m gonna power my radio off the antenna and sell the excess electricity! /s

      I found this ^^^ comment when scrolling down to post my own comment about dBm vs dBi. :)

  6. It’s too bad that the did away with 2 to 6 because that also includes the FM band. Those “digital” antennae are useless for FM radio. You can tell by their narrower width.

    1. They renumbered what frequency is on what channel – they’re still stations labeled as channels 2 through 11 in the US, but they’re broadcasting on different (and UHF, IIRC) frequencies now that they’ve gone digital.

  7. I am still using a 30 year old antenna on the roof. I can pull in several stations and over 30 digital channels total without changing the orientation of the antenna. The furthest TV station is about 45 miles away and the nearest one is 20 miles. Almost all are straight west of my house.

    I wanted to get rid of the roof mounted antenna and use something smaller. 3 different outdoor “50-60 miles guaranteed” antenna couldn’t pick up a single channel at all. It didn’t matter what direction, angle, whether it’s outside or inside or even held up by a clown wearing grass skirt while juggling coconuts. It wouldn’t do one measly channel.

    Dunno where I went wrong but to me it seems a modern $100 antenna performs worse than a 30 year old antenna. Maybe digital antenna is also sensitive to digital noise? Lots of wifi signals around my house, and there’s a few spots in the front yard that my Sprint or my brother’s AT&T gets no bar and won’t work.

    1. Same here. I am still using an antenna I mounted in my attic after one of the neighbors tossed it when they got cable in the mid 90’s. It was probably made and originally installed in the late 1970’s. I have it mounted without a rotor pointed at the antenna farm in Chalmette about 45 miles away. It still works fine after the DTV conversion, and much better than any of the modern bowties and miniature “digital antenas” have for friends and neighbors. Every once in awhile I do lose signal for a few minutes, but I’ve determined that’s probably because of large boats passing through the signal beam as it grazes the surface of Lake Pontchartrain. It’s not worth it to me to put it on a pole above the roof to prevent that.

    2. “clown wearing grass skirt while juggling coconuts” I would suggest either red grass skirt or melons instead of coconuts, they do seem to improve the gain quite a bit. For higher frequencies, peanuts instead of coconuts is also a viable improvement.

      The “modern” antennas are designed to be inconspicuous, so that the unsightly antennas of the 80’s are replaced with a “flat material design” that is just glued on (the inside! on) the window. My point is, antennas (antennae?) that work under 1 GHz haven’t improved so much in this century, we still rely on the old proven theory.
      TL;DR: size matters.

  8. Anyone who claims an “analog” antenna cannot be used for “digital’ should be ignored as that is a fundamentally ignorant statement.

    In regards to antenna selection for OTA HDTV signals, I have been using a Radio Shack Attic Format Antenna since ATSC starting being broadcasting in my area. I have had zero issues with reception and signal strength. I have 4 receivers tied through a splitter and everything works fantastic.

  9. So a somewhat related but less technical question. Almost everyone bundles Internet with TV these days so I can'( seem to find a less expensive way to cut the cord. If I drop TV the increased Internet plus netflix/hulu/amazon comes out to be more $. Anyone else finding this issue?
    Also does anyone’s local stations have anything decent to watch?

    1. 20 years go I dropped cable to reallocate the money towards internet access. Every so often there might be something on cable that I would have liked to see, but since I don’t have cable, there’s nothing I can do, so it doesn’t matter.

      Local tv is local tv, for a long time the only tv there was. Though pretty much none of it is local, other than news most programming comes from the network.

      But DTV is great. I used to watch PBS over the air, and it never got better than grainy. I couldn’t get CBS and NBC because they were on channels adjacent to local channels, so the local stations would wipe them out. But with DTV, I get 28 channels, because there are subchannels. I did lose ABC, which moved to a different channel and lower power, so everyone here has problems. But there i usually something to watch. Old tv on one subchannel, old movies on another, more recent tv shows in syndication, shows and movies about black people, lots of PBS and of course Canadian and US network tv, except for ABC.


    2. SiliconDust OTA tuners, Kodi.

      If Windows then recorder is Windows Media Center on an old laptop.

      If Linux then recorder is TVendhead in an old windows box.

      Other recording options are available.

      22 local digital channels give 528 hours of broadcast available and 4 tuners can record 96 hours of that each day.

      Cable completely cut.

    1. Good Man! Didn’t think anybody would come up with that. It’s all over the internet, many construction articles.

      Reliable, non-critical construction plus easy setup, set-top, indoor, attic, or out on a mast, directional and good image rejection, good for local to mid ranges, 300 ohm feed so go find a balun for your 75 ohm coax. Tons of examples “How to Build” all over the internet. Have seen them made out of cardboard and tin foil with elmer’s glue! Much easier and less critical than a yagi to construct. Suggest tin cans and coat hangars or better. Gang 2 or 4 together to increase range… if still need more THEN go to a yagi.

      You can find them in antique shops… and you likely remember one sitting atop grandma’s tv. Radio shack used to sell them for $12. You threw yours away years ago! I know… I found it in the trash and got it to one of the grandmas near me along with a converter box. I’m still getting homemade pies, and they still got free tv!

      1. Actually I’m referring to the Gray-Hoverman, not the original Hoverman. The GH massively outperforms the original, as well as almost any other TV antenna. I am one of the “fathers” of the GH design and am qutie justifiably proud of it. Click on my user name to see the introductory page from 2007, which leads to a ton of material at the home of the GH antenna.

      1. The design is far more forgiving than you seem to expect, and we’re comparing to a yagi which IS of critical construction, often painfully so.

        Been there, done that, mine works very well, don’t mind if my opinion is discounted.

  10. minor nit: is it appropriate to say
    “And Yagis can have remarkable gain – up to 20 dBm when correctly designed.”
    Shouldn’t it just be ‘dB’? I.e. not ‘dBm’, which I thought was an absolute power measurement (relative to a milliwatt). Sorry, it’s been quite a while since school; sanity check for my life-addled brain….

    1. You are correct. dBm is a power measurement relative to a milliwatt.

      For antenna gain, dBi (directional gain relative to isotropic (ideal sphere) ) or dBd (directional gain relative to dipole) are the common representations.

  11. with all of the arguing about antennas – can we just agree that a vintage 60’s TV antenna still receives the digital TV signals just as well as it used to? our lake house has never had cable – the same antenna has been on the roof since it was built, and we still get TV just fine up there – and arguably better than some of the people with the newer “digital” antennas

    1. Unarguably YES!

      There hasn’t been much room to improve antenna design. Some fractal work has been new and only recently developed cause it needs computers to calculate, but not a lot else new. The math and physics have not changed, they are Laws.

      The old stuff was most oft designed by Hams possessing in-depth knowledge grown from a mix of passion and curiosity supported by a wondrous organization called the ARRL and an array of magazines giving comprehensive and detailed in-depth information of all the latest advances, essentially a paper version of the internet but with editors having an eye to technical competence lest the hams drop subscriptions and they go out of business.

      Today’s is oft the work of a college engineering grad that kept the books and likely lacked the passion yet got the job for the right $$$ and worked under specs and pressure from a marketing department with crazy ideas like “make it look like the wings on a jet” or some other such tripe to create a “selling feature” cause there’s money to be made out there and we can make them want the newer despite it not possibly having but a slim ghost of a chance of being a single bit better. Plus marketing would also take an idea from a teenager that looked really kewl and could be talked up.

      It’s people that make the difference. Hams with passion in the old days. Marketing in the quest for money now.

      You can thank Annie for the answer. ROFL… she’s forced me to spend the last two days with it on my mind!

      1. +1
        Personal Opinon: I have a lot of respect but because this place has so many newbs decry that instead of participating simply baits folks to come one step deeper into the pool to swim or drown themselves for personal amusement, throws no life preservers out to help teach or provoke upward movement. Obviously has an education, likely in physics or chemistry such that needed electronics too. Knows math. Great potential to be a very positive force but leaves it to waste. Independent. Worthy of respect.

        1. To Biomed

          That was my point. There are blogs for jokes. This is supposed to be a place where one can get reliable tech info. In my opinion Jokers that mistake newbs should not be allowed to post idiotic posts here.

          1. You win. I concur. Don’t want any newbs drowned. Needs of the many outweighs the needs of the one. I apologize. Shields up. Photon torpedos ready.

        2. Nah. Put that stick down.
          Just my own opinion here. You did seem to contrive a tiger trap and got a few with it, and admit was great fun for a time, but the kittens got spooked, scattered, confused, and it’s not appropriate behavior in any science-based technical type of forum. Only reason it worked is because folks here trust and totally anticipate reasonable collaboration for lots of mistakes to be made and corrected here, but were met with deliberate contrived confusion. Doubt anyone will forget, or trust. You get what you put in. And sister, you did put in!

  12. Many of those television antennas were not Yagi-Uda designs. They were a different type of multiple element antenna, the log-periodic. Log-periodic antennas have less gain for a given size but much wider bandwidth, making them well suited to the television bands. That bandwidth isn’t as important in the US now that the television band has been whittled down to a smaller portion of spectrum, so future UHF TV antennas are likely to shift back to Yagi designs.

    The original UHF television band went all the way from 470 to 890 MHz, a frequency range of nearly 2:1. First channels 14-18 got taken for public service radio. Next channels 70-80 were reallocated for cell phones and public service in 1982. Channels 52-69 went away after the implementation of digital television in 2009. Channels 35-51 will be taken away in the near future; the reverse spectrum auction (where TV stations volunteered to offer their spectrum up for sale) has been completed but the realignment of broadcasters is yet to come. After all the dust settles, the UHF TV band will only range from 500 to 596 MHz.

    There is also still television broadcasting in the US on the two VHF television bands: channels 2-6 (54-88 MHz) and channels 7-13 (174-216 MHz), though not as much as before the digital transition because some stations chose to remain on their transition period UHF channels rather than return to the VHF channels that were used for their analog broadcasts. Additional broadcasters will be moving to VHF after the resolution of the reverse spectrum auction; others will go dark or multiplex with other channels. The move back to VHF will be problematic for some users because most of the antennas they have put up for HDTV reception are UHF-only, and they may not have room for the larger VHF antennas or wish to incur the disapproval from neighbors that those larger antennas will draw. (Here in Boston, for example, there are currently no VHF stations on the air, but WGBH will return to VHF.)

    1. Behind the Iron Curtain most of the antennas were Yagi-Uda, They allowed pulling in distant stations that deviated from the usual propaganda. Once satellite equipment became affordable, they were mostly relegated to “cold-war decorations”. Then cable TV took over and replaced the state propaganda with something more subtle.
      Yagi-Uda antennas allowed the narrow-bandwidth and directionality that was needed to tune in to those distant (100-500km) stations.

      But I guess we are discussing US here, so sorry about interfering.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s