I once went to an antenna tuning/design/factory for my startup's product and I told the guy there, I want an antenna to support X frequencies, this that dBs, good VSWR, hopefully achieve X kilometers range, etc etc.
Two days later he had 3 designs of antennas ready to be made into a flexible PCB and two days after that we got the FPCB samples.
I was amazed so I asked to see how he works. He took copper tape, and with a boxcutter he carves the antenna, and adds solder blobs to tune with the network analyzer. Then, once he was satisfied, he shoved them into the anechoic chamber and boom, done. Black magic stuff.
Antenna intuition is really hard to attain, it takes years playing with the right equipment in the right environment..
I remember this from a professor in college as well. I spent weeks modeling an antenna element and then once I had it made, he brought out the copper tape and a knife and tuned it to closer match what we wanted. It is humbling to see this kind of craftsmanship in technology.
I think with good simulation tools, optimization processes, maybe even deep learning, we could make it a lot less of black magic. Make it boring and approachable instead.
It kind of reminds me of chemistry. You have physical laws that are fairly simple but the interactions are hard to describe without a bunch of computation.
That's a good point, but given the search space is a set of voxels, you can probably brute force a yuge amount of possibilities and fit some sort of paramterized curve to them.
The analysis part of engineering is, thanks to modern computers, practically "done". However, we are still not able to do synthesis well. We also have exact mathematical models for that, but it usually amounts to NP-hard problems (think integer optimization).
I don't know about this specific case, but often the exact models are so complex that you can't optimize them analytically, so you resort to a global optimization best effort, where the results are not guaranteed to be any good - so you might as well try something DL-based too, to see how it would perform.
I saw a NASA Tech Brief from 25 years ago where they did that for a small multi-band satellite antenna. The result looked like a pin cushion someone sat on.
There is a surprising lack of easily digestable antenna/ham radio related material on the internet. I know because it took me 3 weeks to learn the basics of antennas when I expected to finish in 2 nights. Some of the best information I read was from old Royal Canadian Airforce videos, atleast several decades old[0]
I still haven't been able to find a general equation for a have wavelength dipole antenna explained in simple English. I do have one based on empirical evidence, though[1]. I've even bought a copy of the ARRL Handbook, but I find that it goes from 0 to OMG-language-is-this too quickly.
>I still haven't been able to find a general equation for a have wavelength dipole antenna explained in simple English.
Equation for what? Length at resonance? Feed point impedance? Other interesting parameters?
Antennas are one of those things where it takes a long time to develop intuition, and there are no simple formulas for anything, just useful models that get you close, and simulation programs that work well enough to give an answer within your manufacturing tolerances. (All models are wrong, some are useful.)
So start with a couple of fundamental ideas: 1) Accelerate an electron, get a photon. 2) An antenna is a transformer that couples the end of your feed line to free space.
The reason the general family of dipole antennas is efficient is that the Ohmic resistance is usually around an Ohm or less, and the "radiation resistance" can be raised to around 70 to 80 Ohms. So 80/(80+1) is the ratio of energy coupled to space versus total energy input. Pretty good efficiency.
In a center fed dipole, the driving voltage creates an electrostatic force that attempts to slosh electrons in the conductor one way or the other. At resonance, a small amount of energy input creates lots of sloshing, because the driving voltage just needs to give a boost to the resonant sloshing. Off resonance, effectiveness is much lower. Actual length at resonance depends on the length:diameter ratio of the conductor, the dielectric constant of the surrounding medium, the height above ground, and the dielectric constant and conductivity of the ground. If you know all of those, the dipole can be modeled as a just barely tractable boundary value problem.
The empirical formulas that you see usually assume a practical conductor diameter and practical height. You might enjoy playing with one of the NEC2-based antenna modeling programs out there. NEC2-family solvers do "method of moments", where each wire is chopped up into segments, and then for an N-segment model, an NxN matrix of mutual inductances models the coupling among wire segments.
The ARRL Antenna Book takes more time to explain fundamentals than the Handbook. The ARRL also publishes an antenna physics book that I haven't read. I notice that PhD committee chair for the author of the Antenna Theory web site we are yakking about was Balanis, who wrote a pretty good book called "Antenna Theory" -- but the book assumes you are an EE graduate student with at least a semester of multi-dimmensional DiffEQ beyond the elementary DiffEQ course.
That's the problem I suppose. Most of the material is either geared for EEs, or assumes I need a refresher. Not someone completely new to the hobby.
>Equation for what? Length at resonance? Feed point impedance? Other interesting parameters?
I should have clarified: The arm length of a half wavelength dipole and all the variables that go into it. I assumed it was 1/4 wavelength, but while building mine, I discovered calculators that gave calculations different from mine.
>So start with a couple of fundamental ideas: 1) Accelerate an electron, get a photon. 2) An antenna is a transformer that couples the end of your feed line to free space.
I don't mean to sound thick, but you've already assumed too much. Before I signed up for a membership with my local radio society, I didn't even know Ohm's law.
I pride myself in being technical. If I can teach myself to code and, in a few years, craft tested API's and decoupled front-ends that are tested through CI pipelines and deploy through CD, I can surely teach myself enough physics to build an antenna — no.
I still struggle to understand basic concepts like:
+ Baluns
+ Why does the height of an antenna effect its effectiveness?
+ Gain
+ Circuit design
+ Transistors
+ Honestly, I still think radio waves are magic sometimes, even though I think I've seen the effects of electrically generated magnetic fields on coils
>You might enjoy playing with one of the NEC2-based antenna modeling programs out there.
Tried playing with CocoaNEC 2.0, but the lack of documentation left me feeling like an air head.
I'm hoping a little more exposure to electrical systems will help.
To a first-order approximation, the length is 1/4 wavelength per leg. However, that assumes an ideal conductor in free space. We rarely have the luxury to suspend an antenna a few wavelengths from the ground, water, buildings, and anything else remotely conductive. The way the antenna interacts with those things in the near-field affects its impedance and makes it behave electrically longer.
The common formula you'll see is: Total Length (in feet) = 468 / f (in MHz). If we suppose exactly half wavelength and do the unit conversions, we would expect L (ft) = 492 / f (MHz). So why do the calculators use the shorter length? It's an empirical compromise. Taking 5-10% off total length is generally what's needed to account for things in the near-field. The 468 number has been repeated enough that it's stuck. In practice, I almost always cut dipoles for a full half-wavelength, hook them up to an antenna analyzer and then trim them down. With as cheap as hookup wire is, I'd rather not take the risk of being too short and having to field solder a splice (not fun on ARRL Field Day).
There's not really a general formula for finding length. The physics is nothing more than Maxwell's equations, but many of the deviations from an ideal dipole come from interactions with the environment. It's difficult to measure and/or predict how the environment will behave, so you're often better off building the antenna and then adjusting it in place. And so we end up with empirical rules of thumb like L = 468 / f.
Antennas are one of the harder topics for amateurs, for sure. The theory is well-understood, sitting somewhere at the intersection of EE and physics. I'm lucky enough to have a strong background in both fields, but there's a clear lack of curriculum for amateurs without that background. This website at least seems useful for building intuition, so hopefully it helps you some.
You're trying for a huge breadth of material, stuff that is covered in multiple specialities in electrical engineering. Multivariable calculus is really the entry point for engineering level antenna design.
Balanis' book is a great reference but I wouldn't recommend it for learning. Honestly the ARRL antenna book and antenna handbook are the best practical books on putting an antenna together without getting bogged down in the details. The real hard part is getting some radio equipment so that you can experiment and learn since Spectrum and Network analyzers are waaaay out of most people's hobby budgets.
True. I already own a Baofeng, and an SDR, but for the rest I'm honestly thinking of using my Arduino in conjunction with some diy circuitry (if I can wrap my head around enough of it) to build an oscilloscope, and I found an article that used a noise generator in conjunction with an SDR as an SWR meter [0]
It's definitely not high end, but my wife would kill me if I she found out how much I'd have had spent a halfway decent SWR meter, if I went that route.
Not a bad idea but a few things to keep in mind.
- The sample rate on an Arduino is pretty low, depending on which model you have it may only be a few MHz and the absolute best you can theoretically possibly achieve with your home o-scope bandwidth is half the Arduino clock frequency. You'll also have to cast off the Arduino language and start moving to C/assembly to get the full performance out of it.
-The dynamic range on the Arduino ADCs is pretty low. It's going to be limited by the number of bits in the Arduino ADC.
-The input on an Arduino ADC is going to be either very low or very high impedance. This is also common on lower frequency scopes (low frequency in the case being 1GHz or less bandwidth) but if you try driving your input directly the impedance mismatch from the RF is going to cause poor signal and lots of noise on your measurement.
I don't know what SDR you have but it might be suitable for the signal capture and you can probably gin up some filters in post processing to approximate a Spectrum analyzer. If you have a source and 2 directional bridge couplers you could use it to make a poor mans Scalar Network Analyzer. A noise generator or a swept tone (Chirp) can be used in both these cases. The article you linked basically did a 1-port Network analyzer using the bridge coupler and used to to measure VSWR.
A lot of modern oscilloscopes, even hobby ones, now include the ability to capture scope traces to a PC and an FFT math function. You're going to be limited in frequency without a downconverter but it's a good way to go on a budget when working on a bench. The Rigol's are a lot better than they were when they first came out and make a good hobby scope for a reasonable price. If you can be spendy, the Keysight hobby level scopes are a joy to use. Best bet on a budget is to troll the internet for an old Tektronix, Lecroy, or HP/Agilent/Keysight but you'll probably not get something with fast trace capture or a built in FFT.
All of that said, making test equipment out of your Arduino is a great hobby project that will teach you tons of useful engineering. I highly recommend it!
The main reason antennas are hard to understand is that their dimensions are comparable to the wavelength they are tuned for. Unless you are dealing with microwaves, the systems used to generate and detect RF are modeled using "lumped circuit elements". For example, resistors, capacitors, inductors, and transistors. And, at first, big sparks. These elements are small compared to the wavelength.
The design of those elements and the circuits using them was historically pretty independent from formal electromagnetic theory, as developed by Maxwell. The intersection was Oliver Heaviside, at the end of the 19th century.
Before Heaviside, RF electronic design was a largely a matter of groping though the practicalities of employing those circuit elements to create oscillations and couple them to resonant wires supported as high as possible.
As the peer posts explained, antennas are resonant structures in which electrons are caused to slosh back and forth. As they slosh, they accelerate, and as they accelerate, they radiate electromagnetic energy. But, it is best to stick to the rules of thumb at the level of the ARRL handbooks. To understand the EM theory related to radiation and antennas, you really need to work through to the final chapters of Griffiths, "Introduction to Electrodynamics".
But that is not necessary to get an intuitive understanding of antennas, to construct them, or to run the modeling software.
Resonant antennas are only one class of antennas, with the other being traveling wave antennas. A log periodic looks similar to a Yagi, but it’s not. Same with a biconical and dipole.
It’s an easy day when I have dimensions on the order of a wavelength. Usually it’s 1/10 or less, and shoved up against metal.
How do I develop my intuition regarding the low resistive impedance of such antennas? I understand the high capacitive reactance, but haven't got a grip on why the restive part is so low.
What is your general approach to matching these very short radiators?
When you bring an antenna close to a conductor, say a dipole next to a plate, the radiation resistance decreases. This is due to the currents induced in the plate, creating an image and reinforcing the currents in the dipole.
It’s really the ratio of radiation resistance to conductor resistance. You can shrink an antenna to infinitesimal size, made of perfect conductor, but as the radiation resistance decreases, it’s more difficult to impedance match. An infinitesimal antenna would have zero bandwidth. Sort of like Bode Fano criteria limiting bandwidth versus impedance.
There is a Chu theoretical limit which limits antenna efficiency and bandwidth given volume, hence 3D fractals and other stuff. Ain’t no free lunch. A lot of antenna research is who can get closest to the Chu limit. Sort of like coding and the Shannon capacity.
Thanks, I had not heard of that. Also, that link lead me to the WP article Electrically Small Antenna. It's remarkable how much they knew 70 years ago.
> I still struggle to understand basic concepts like:
+ Baluns + Why does the height of an antenna effect its effectiveness? + Gain + Circuit design + Transistors
That's a lot of topics. People spend a lot of time understanding each one. Keep working at it, you will traction eventually.
I'll try to make some helpful comments.
> Baluns
Well, a lot of the confusion comes from the fact that two fundamentally different widgets are called "balun". A form of transformer, and a form of choke. Bottom line: to feed a balanced antenna with an unbalanced feed line (coax) you want to keep common-mode currents off of the shield. The electric field should be entirely contained between the inside of the shield and the center conductor. The choke style balun (stack of ferrite beads) creates a high impedance on the outside of the shield, so the current flows inside. The transformer style accomplishes the same goal by different means. Usually a transmission line wound on a toroid.
> Gain
A finite amount of power is going into the antenna. Nothing you do in the antenna can increase the power, but you can direct it. It almost always works out to optimizing the phase difference among different radiating elements such that you get constructive interference in the desired direction, and destructive interference in undesired directions. To visualize, gin up some code that plots two sine waves of the same frequency and their sum. Alter the phase of the two source sine waves and observe the result. A 3 element Yagi-Uda works on this principal: The "reflector" is a bit longer than 1/2 wave, so the energy that it absorbs in the near field is reradiated with a phase lead w.r.t. the driven element. The "director" is a bit short, and reradiates with a phase lag. The radiated energy from the elements sums constructively going forward.
> height above ground
Some energy from the near field impinges on the ground, and is reflected with a phase reversal. It will sum constructively or destructively at various angles above the horizon as a simple trigonometric function of height. (N6BV's HFTA - HF Terrain Analysis program -- uses GTD - General Theory of Diffraction -- to model antennas above ground terrain. Good fun for optimizing antenna height.)
> Circuit design
Start with circuit analysis. Once you understand the building blocks, you will be able to synthesize something with them. This should feel a lot like programming eventually. Start by learning how to read simple schematics. The best advice in this regard that I ever got was from my first semester circuit analysis prof: "Keep redrawing the circuit until it makes sense." Which means: redraw simpler circuits representing each regime of operation. Start with DC: caps are open circuits, inductors are shorts. Draw that. Now you understand the DC bias, or can figure out the bias voltages pretty easily with Ohm's law. Then redraw at the operating frequency with freq-dependent impedances.
> Transistors
A bipolar transistor is a current-controlled current source, and FET is a voltage-controlled current source. So to analyze a transistor, you have a bunch of things to draw: the input side at DC to find the operating point (bias) the input side at AC to understand the controlling signal, the output side at DC to understand the operating point and output impedance, and the output at AC. Also identify the interstage coupling elements. Not sure if my comments on transistors are helpful yet until you are more confident at circuit analysis. Keep at it! It isn't so different from programming, just a different grammar.
> I'm hoping a little more exposure to electrical systems will help.
I'll give you a nugget that I think is correct.
The radiation efficiency is determined by the cross product of the electrostatic and magnetic fields the electrons are exposed to. No cross product no radiation. AKA why most circuits don't radiate very well.
A dipole is a resonator that generates a large electrical and magnetic cross product. Because it resonates the energy stored/flowing through it is many times the energy being pumped into it.
Lots of things effect the resonance peak of a physical resonator. That's the difference between theoretical antenna's and physical one.
"I am a practicing antenna engineer, with a PhD in antennas and I have worked for many years in defense, university and the consumer electronics field as an antenna engineer."
Nice that some of the "old web" soldiers on. Firsthand info from actual experts.
This is a great summary. I too have struggled to get decent information on the web about antenna theory and design. As it turned out, I was searching wrong :-) The keyword is 'electrodynamics' and the canonical text is "Classic Electrodynamics" by Jackson. I am told that if you can understand the contents of this book, antennas are pretty straight forward. I started in on it, got whacked upside the brain a number of times, then backed off to "Introduction to Electrodynamics" by Griffth which is the undergraduate version and does a bit more math review, which was essential in my case. My plan is that once I am through that I'll go back and re-start Jackson.
The fun bit here is that if you look for computer code to simulate this stuff you will run into a lot of Fortan code. So if you ever wanted to learn Fortran this will give you some code to puzzle over.
By which you probably refer to the NEC2 code base. NEC2 is public domain, so that is what most hackers use. The native UI is column-senstive punch cards. Blessedly, there are people that have put more modern front-ends on the NEC2 back-end.
NEC2 is OK-ish, as long as you avoid the well-known bugs. NEC4 fixes some of the bugs, but falls under ITAR, so requires a license and can't be exported (last I knew, anyway). There are also multi-kilo-dollar-per-seat antenna modeling packages available commercially.
NEC2 is pretty old, and an interesting story I heard about the validation of the model was that the DOD, having helicopters handy, stuffed a helicopter full of instruments and flew it around an antenna range to capture ground-truth data for antennas that had been built from models. Last week I was talking with my friend N6BT, who has been in the antenna business for decades. For 3 or so years he has had a quad-rotor that he flies around with a signal generator, and uses the GPS time from the quad-rotor to correlate GPS time-stamped data from his ground-based spectrum analyzer to collect actuals. He is finding MANY discrepancies (primarily at low angles) between NEC4 and actual, due to the sketchy ground models.
That has been my experience as well, at work we have a multi-kilobuck simulation package but we still put the antenna on our range to test it. The range consists of basically a robot arm that can hold the antenna under test (AUT) in any orientation, a transceiver/spectrum analyzer that can move forward or back to get into the near, Fresnel, and far ranges, and a transceiver/spectrum analyzer that is connected to the AUT.
Comparing simulations to actual always yields some interesting nuggets of information.
Jackson is really good. There are actually a lot of other good books in that general category depending on exactly what you are looking for.
Cheng's "Field and Wave ELectromagnetics" is an excellent undergrad book to start to learn then you have
Harrington's "Time Harmonic Electromagnetic Fields" and Stratton's "Electromagnetic Theory" which are both really fundamental and thorough
and more high end you have Collins' "Foundation of Microwave Engineering" and "Field Theory of guided waves"
A bit more practical if you like smith charts you've and matching networks you've got Pozar's "Microwave Engineering"
and for Antenna theory you've got
Stutzman and Thiele's "Antenna Theory and Design" or Kraus' "Anetnnas"
Balanis' book is good but a lot of the material in in comes from these books + Jackson and I think it makes a much better reference than a learning text.
For the purposes of learning antenna theory at the level of Balanis's 'Antenna Theory', I think Jackson is severely overkill. It's also written for physicists rather than engineers, so the emphasis is different. On the other hand, Jackson's problems are quite excellent. There's all kinds of interesting stuff in there that is hard to find in most other places (plasma physics and MHD come to mind). Something like Griffiths or Cheng will be just right. I studied Computer Engineering in undergrad, and worked through several books on electrodynamics at the time (and did several courses in EM), including Jackson. I didn't study RF, but my classmates in RF were all using Balanis for their RF and Antenna design courses, which clearly comes before the level of EM where most people are working (suffering?) through Jackson, or Balanis's 'Advanced Engineering Electromagnetics'.
As a physicist rather than an EE (although I'm neither now), antennna always led to confusion. In particular antenna effective area and reciprocity. Trying to imagine a 1d dipole antenna funneling some 2D part of the incoming wavefront into its output waveguide just felt like magic. As did the trying to intuitively see that an antenna's gain is the same in transmission and reception when the wavefronts seem totally unequivalent. Would have liked to have really studied it more but antenna don't get covered in a typical physics course.
Think of it like this: The wavefront around a dipole is comprised of near and far fields, with the near field being spherical and the far field planar.
That near field has a reactive component (stored energy) that does not propagate, but falls off at 1/r^3 or faster.
So an incoming plane wave induces charge motion, which builds up the reactive near field over many cycles, generating that spherical wavefront.
So that seems to indicate reciprocity is only valid for a steady state, but it’s still valid. If your transmit antenna were fed a monocycle (i.e. not time for the near-field to build up), the receiving antenna wouldn’t have enough time either.
Evanescent fields exist sure but Far field is really just a useful mathematical construct. It is typified by a wave-front where you can approximate no phase difference wherever it lands on a planar antenna.
It's like how you can approximate the Earth as flat when making a platting because it is very large and you are very small. If you look at the Farfield approximation calculation for a large antenna or phased array, you'll see that the equation is a function of distance, wavelength, and aperture size.
Edit: I should point out that Evanescent waves do not carry power (no net energy flow) so the power transfer is always reciprocal between 2 antennas.
As an EE with sense of physics I ran into the same confusion. Your instincts regarding reciprocity, effective area and all that were in the right direction and with a few more steps you would have had your satisfying 'Eureka' moment.
Most of these comes from the Pathloss Equation which it turns out is a stitch up to make things simpler and easy (but wrong).
It's best explained with respect to parabolic antennas but applies to all antennas. The key point to picking this apart is to consider reciprocity which states that the that an antenna is "the same" as either a transmitter or a receiver. In particular the gain is is the same.
So gain is the increased power with respect to an isotropic radiator. With a parabolic antenna the focus (ie beam width) of the transmitting beam does depend on frequency due to geometric concerns and as such, in the path loss equation the the antenna gain appears as frequency dependant as it ought to.
However, reciprocity requires that the receiver also have an identical, frequency dependant gain. The gain of the receiver though depends only on its physical (or effective) area and not on its frequency.
In the pathloss equation you can see that the loss goes as the reciprocal of the square of the distance, which it should, but also goes the reciprocal of the square of the frequency. This frequency term (which causes the equation to violate conservation of energy, normally a bad thing) is there to cancel out the bogus frequency term incorporated into the gain of the receiving antenna due to the also bogus reciprocity law.
So to simplify, in an electromagnetic link between two antennas, the gain of the transmitting antenna depends upon the frequency of the transmitting carrier, because the focus of the beam varies with frequency. The signal then drops off as 1/r^2 in the normal way (with no frequency component) and the gain of the receiver depends only on its size. A bigger receiver antenna captures more energy from the receiver. That's it, simple and sensible.
Effective area is a separate topic for long wavelength transmissions but also sensible in the end.
I have an etymological question. The chap who supervised my ham tests insists that bugs have antennae and radios have aerials. I wonder if it is a US/UK thing? He is British.
Primarily, yes. Though Wiktionary has the following usage note:
> Some make a distinction between an antenna and an aerial, with the former used to indicate a rigid structure, and the latter consisting of a wire strung in the air. For those who do not make a distinction, antenna is more commonly used in the United States and aerial is more commonly used in the United Kingdom.
Well even funnier than that he said bugs and then corrected himself (although I am pretty sure the two words have a strict definition that means they are not interchangeable). He also used the word antenna lots of times too! Most ham textbooks in the UK use the term antenna, but common usage is definitely still aerial
We'll definitely always use aerial for the TV or radio receiving thing on the roof, or the thing you pull out the back of the set. Seeing the sibling comment with a distinction, and yeah I have heard it occasionally for transmission capable. A 30 foot pole in the back yard might indeed get called antenna.
We don't seem to use bugs as a catch all much any more, though it's coming back, and we always kept it for bed bugs. Not especially consistent - but few of these transatlantic complaints are!
Am I overlooking something or is the "cantenna" not considered a fundamental antenna type? It does not seem to be listed in the page about different antennta types.
It seems to be quite easy to build once you find a suitable can (easier than a Yagi-Uda antenna) and it seems it can easily keep pace with a Yagi-Uda antenna of similar size.
It's slightly ironic that the source page cites the Einstein quote about simplicity and yet, one of the central equations of antenna link design is the "Path Loss Equation" which is in fact, simpler than possible. :)
Two days later he had 3 designs of antennas ready to be made into a flexible PCB and two days after that we got the FPCB samples.
I was amazed so I asked to see how he works. He took copper tape, and with a boxcutter he carves the antenna, and adds solder blobs to tune with the network analyzer. Then, once he was satisfied, he shoved them into the anechoic chamber and boom, done. Black magic stuff.
Antenna intuition is really hard to attain, it takes years playing with the right equipment in the right environment..