Keyboard Shortcuts
ctrl + shift + ? :
Show all keyboard shortcuts
ctrl + g :
Navigate to a group
ctrl + shift + f :
Find
ctrl + / :
Quick actions
esc to dismiss
Likes
Search
"Projector of the Sharpest Beam of Electric Waves"
|
My copy of QEX came today and I was immediately drawn to the famous paper by Yagi & Uda.
Though not meant in any way as a criticism, I had hoped for more mathematical rigor. I don't know if that was presented in another paper as mention is made that more was forthcoming. Does anyone know? I don't recall ever seeing a Fourier analysis of the Yagi-Uda design. Ronald Bracewell discusses many interesting arrays, but at least to my recollection, only covered slotted RF arrays in the 2nd edition. Seismic style point receivers are well treated, but point receivers are trivially simple. At least to me. But that may simply reflect a career spent on elastic waves instead of electromagnetic waves. It's the same wave equation, but the devil is in the details so I get very nervous when I switch the physical domain. The title of the paper should be mathematically provable. I'm interested, but a bit too busy and too lazy to do the analysis required for a proof. Which completely neglects whether I still have the skills for such an undertaking. Math was never my strong suite and it has been far too many years. If anyone knows of a rigorous proof that the Yagi-Uda is the "sharpest beam" I am *very* interested in reading it. On inspection it seems plausible, but mathematics at that level is rather a black art to mere applied math people like me. Thanks and.. Have Fun! Reg |
||
|
I also read the reproduction of the Yagi-Uda paper in QEX and was also struck by the lack of any sort of mathematical basis for the development of the antenna.
toggle quoted message
Show quoted text
There is also another paper that appears in the PROCEEDINGS OF THE IEEE, VOL. 85, NO. 11, NOVEMBER 1997 by David Pozar entitled "Beam Transmission of Ultra Short Waves: An Introduction to the Classic Paper by H. Yagi" which elaborates somewhat on the Yagi-Uda antenna. As nearly as I can tell, the original development was accomplished in a strictly empirical manner. Looking through my antenna books, I find that there is quite a lot of mathematical infromation in Balanis' "Antenna Theory", 3rd Edition. The information appears to be spread throughout the text. Section 10.3.3, starting on page 577, contains (eventually) a rather involved theoretical analysis of the antenna using the integral equation moment method followed by a description of how to calculate the antenna's far field pattern. There are also apparently? MatLab and FORTRAN programs on the CD that was included in the book. Unfortunately, I no longer have access to FORTRAN compilers or MatLab. Weeks' "Antenna Engineering", Section 4.7.3, page 196 has a simpler description which shows how to calculate the antenna element currents and the resulting radiation pattern. Oddly, I do not see any mention of Yagi-Uda arrays in either of the first two editions of Kraus' "Antennas" but I do find a bit of descriptive information in the third edition, Section 8.6, page 246. Finally Elliot's "Antenna Theory and Design, Revised Edition, IEEE Press, has two sections on Yagi-Uda arrays: Section 8.7, page 368, describes two-element antennas and Section 8.8, page 373, describes antennas with three or more elements. I hope that helps. DaveD On 3/13/2021 8:47 PM, Reginald Beardsley via groups.io wrote:
My copy of QEX came today and I was immediately drawn to the famous paper by Yagi & Uda. --
This email has been checked for viruses by Avast antivirus software. |
||
|
I thought there were free Fortran compilers available last I looked into (two decades back).? For some reason, I first was thinking a free Visual Studio version may allow importing and compiling, though I haven't verified.?
A quick Google search of "" returned far more results than I expected.? Worth reading into if you're interested in.? Nice to see a range of options.? I was also pleasantly surprised to find a MATLAB code to Python compatible code translator appears.? , where the compiler was in a shady transition from not able to convert to C code and requiring the MATLAB Component Runtime (MCR) for the MATLAB Compiled apps.? |
||
|
The OMPC referenced download links at?http://ompc.juricap.com/download, from bitbucket, are broken. Though found a github repository for OMPC: ?
Haven't tested since I don't have a Python 2.5 legacy, circa 2008'ish, machine built at the time being.? Figured share however.? In regards to the Yagi-Uda antenna designs, wondering what the inductive logic is regarding the range of phenomena characteristics, i.e. internal, boundary and external conditions.? Even experimenting to deduce, would be a very interesting project to determine the mathematical observations that characterize the design performance.?? |
||
开云体育For some time I have wanted to start using Gnu Octave. The last time I looked, the authors of that code considered any differences between MatLab and Octave to be a bug in Octave. It appears that Octave now has a wide variety of packages which support various computational methods, similar to MatLab's Toolboxes. I did not see an Octave package for PDEs, though. At one time the MathWorks would sell one a copy of MatLab, Simulink and whatever toolboxes one wished to purchase for a fraction of the commercial cost if one signed an agreement that precluded the commercial use of that copy of MatLab. I did this in 2000; when I went to re-up, The MathWorks informed me that they no longer supported that option. Later on, I believe they expanded the capabilities of the student version of MatLab. I briefly pursued the use of that but did not explore it in depth. You are correct that are are various free FORTAN compilers available online. I have not written FORTRAN code since the early- to mid-eighties. DaveD On 3/14/2021 12:16 AM, jafinch78 .
wrote:
I thought there were free Fortran compilers available last I looked into (two decades back).? For some reason, I first was thinking a free Visual Studio version may allow importing and compiling, though I haven't verified.? |
||
|
The article is a? reprint of the original paper that was published in 1926.??I think the point of the original article was to show discovery of this type of antenna.? ? Of course, the general theory of antennas was not what it would be come yet either in 1926.? ?Today, they would have written the paper, and immediately filed a patent.? ?I didn't see any evidence of a patent from them though.
I found that Kraus does discuss this antenna type on Chapter 11, page 320 as a three element linear array.? He references a later paper by Yagi, but not the original.? ?There is no general theoretical treatment of the antenna type, though he does point to a couple of other references that discuss a single reflector.? He goes into dipoles to a great extent in an earlier chapter,? which is still the driven element for a Yagi.? 73's Jim? N8CAH |
||
|
toggle quoted message
Show quoted text
On Mar 14, 2021, at 11:15, James Amos <jimamos@...> wrote:
|
||
|
The 4th ed of Ballanis has an extensive derivation via the method of moments. However, that is an approximation, so it is not an avenue to a proof. Lawson's ARRL monograph goes into much detail, but again is of an applied mathematics vein and relies on approximations. I feel certain that a proof must be via the Fourier transform. Having met a gentleman who was performing FFTs via a desk calculator and trig tables in the 1950's long before the famous paper by Cooley and Tukey, I have great appreciation for practical insight and superposition. The pictorial dictionary of transform pairs in Bracewell's book are invaluable for acquiring such insight. I should not at all be surprised if Uda had a grasp of the Fourier transform such that he could visualize where the optimum solution lay and then simply used empirical experiment to adjust for fringe effects. The fact that Uda referred to it as a "wave canal" lends some support for this hypothesis. My copy of Bracewell is MIA at the moment which is not surprising as it is my most heavily used mathematics text. Neither Churchill's "Operational Mathematics" nor Papoulis' excellent text on the Fourier transform stray far enough from the pure mathematics to offer any insight. Uda's phrase, "wave canal" suggests that an analysis in terms of a transmission line might yield a rigorous proof of the assertion of "sharpest". A transmission line constructed in the form of uniformly spaced dipoles might give further insight and 2-6 GHz is relatively low frequency today. Light gauge wire and heavy fishline might produce some very interesting results. I only wish I had my lab set up to do that instead of just a pile of equipment on mover's dollies. At the moment my research workstation is sitting on the bench as I try to track down a bit fade problem :-( Have Fun! Reg On Saturday, March 13, 2021, 10:21:39 PM CST, Dave Daniel <kc0wjn@...> wrote: I also read the reproduction of the Yagi-Uda paper in QEX and was also struck by the lack of any sort of mathematical basis for the development of the antenna. There is also another paper that appears in the PROCEEDINGS OF THE IEEE, VOL. 85, NO. 11, NOVEMBER 1997 by David Pozar entitled "Beam Transmission of Ultra Short Waves: An Introduction to the Classic Paper by H. Yagi" which elaborates somewhat on the Yagi-Uda antenna. As nearly as I can tell, the original development was accomplished in a strictly empirical manner. Looking through my antenna books, I find that there is quite a lot of mathematical infromation in Balanis' "Antenna Theory", 3rd Edition. The information appears to be spread throughout the text. Section 10.3.3, starting on page 577, contains (eventually) a rather involved theoretical analysis of the antenna using the integral equation moment method followed by a description of how to calculate the antenna's far field pattern. There are also apparently? MatLab and FORTRAN programs on the CD that was included in the book. Unfortunately, I no longer have access to FORTRAN compilers or MatLab. Weeks' "Antenna Engineering", Section 4.7.3, page 196 has a simpler description which shows how to calculate the antenna element currents and the resulting radiation pattern. Oddly, I do not see any mention of Yagi-Uda arrays in either of the first two editions of Kraus' "Antennas" but I do find a bit of descriptive information in the third edition, Section 8.6, page 246. Finally Elliot's "Antenna Theory and Design, Revised Edition, IEEE Press, has two sections on Yagi-Uda arrays: Section 8.7, page 368, describes two-element antennas and Section 8.8, page 373, describes antennas with three or more elements. I hope that helps. DaveD On 3/13/2021 8:47 PM, Reginald Beardsley via groups.io wrote:
> My copy of QEX came today and I was immediately drawn to the famous paper by Yagi & Uda. > > Though not meant in any way as a criticism, I had hoped for more mathematical rigor.? I don't know if that was presented in another paper as mention is made that more was forthcoming.? Does anyone know? > > I don't recall ever seeing a Fourier analysis of the Yagi-Uda design.? Ronald Bracewell discusses many interesting arrays, but at least to my recollection, only covered slotted RF arrays in the 2nd edition.? Seismic style point receivers are well treated, but point receivers are trivially simple.? At least to me.? But that may simply reflect a career spent on elastic waves instead of electromagnetic? waves.? It's the same wave equation, but the devil is in the details so I get very nervous when I switch the physical domain. > > The title of the paper should be mathematically provable.? I'm interested, but a bit too busy and too lazy to do the analysis required for a proof.? Which completely neglects whether I still have the skills for such an undertaking.? Math was never my strong suite and it has been far too many years. > > If anyone knows of a rigorous proof that the Yagi-Uda is the "sharpest beam" I am *very* interested in reading it.? On inspection it seems plausible, but mathematics at that level is rather a black art to mere applied math people like me. > > Thanks and.. > > Have Fun! > Reg > > > > > -- This email has been checked for viruses by Avast antivirus software. |
||
|
The Gnu development tool suite is of very high quality. It includes FORTRAN 77, 95, C, C++, Objective C and possibly some others for a wide range of machine architectures and operating systems. http://gcc.gnu.org/releases.html Octave is a very good Matlab clone. I know that Octave is supported on Windows, Linux and FreeBSD. Very likely the Gnu development tools are supported on Windows, but I only use Windows for binary only software. https://www.gnu.org/software/octave/index Have Fun! Reg On Sunday, March 14, 2021, 09:28:31 AM CDT, jafinch78 . <jafinch78@...> wrote: I thought there were free Fortran compilers available last I looked into (two decades back).? For some reason, I first was thinking a free Visual Studio version may allow importing and compiling, though I haven't verified.? A quick Google search of "" returned far more results than I expected.? Worth reading into if you're interested in.? Nice to see a range of options.? I was also pleasantly surprised to find a MATLAB code to Python compatible code translator appears.? , where the compiler was in a shady transition from not able to convert to C code and requiring the MATLAB Component Runtime (MCR) for the MATLAB Compiled apps.? |
||
|
toggle quoted message
Show quoted text
On Mar 14, 2021, at 13:10, Sean Turner <[email protected]> wrote:
|
||
|
Indeed...I couldn't think of the co-author's name to save my life either!
toggle quoted message
Show quoted text
Sean On Sun, Mar 14, 2021 at 12:06 PM, Dave Daniel wrote:
|
||
|
Ouch!!!!! $500 and up used! On Sunday, March 14, 2021, 02:10:37 PM CDT, Sean Turner <[email protected]> wrote: Indeed...I couldn't think of the co-author's name to save my life either! Sean On Sun, Mar 14, 2021 at 12:06 PM, Dave Daniel wrote:
|
||
开云体育Thank you, Sean. Yes, that is a very good set of texts about antennas. I haven’t looked in a while, but I believe it is now completely unavailable. Krieger or someone (is Dover still around?) should see if they can get the rights to re-publish it. DaveD On Mar 14, 2021, at 15:10, Sean Turner <[email protected]> wrote:
|
||
开云体育I see one copy available. $485. Wow. ? I bought my set used in the early 80s when I was collecting all of Collin’s work. I think I paid around $100 for the set. DaveD On Mar 14, 2021, at 15:17, Reginald Beardsley via <pulaskite@...> wrote:
|
||
|
On Sun, 14 Mar 2021 at 04:21, Dave Daniel <kc0wjn@...> wrote: There are also apparently? MatLab and FORTRAN programs on the CD that gcc is a free compiler, that compiles Fortran, as well as numerous other languages.? Octave is a free clone of MATLAB, but it does not have the extensive range of toolboxes support that MATLAB has, so if the code depended on a toolbox, you might be out of luck. I think there's a pretty inexpensive home version of MATLAB. DaveD |
||
|
Dr David Kirkby Ph.D C.Eng MIET Email: drkirkby@... Web: Kirkby Microwave Ltd (Tel 01621-680100 / +44 1621-680100) Stokes Hall Lodge, Burnham Rd, Chelmsford, Essex, CM3 6DT. On Sun, 14 Mar 2021 at 14:50, Dave Daniel <kc0wjn@...> wrote:
MATLAB is very economical for home use Although good for numerical work, I believe Mathematica is king when it comes to symbolic maths. There's a home version of that, but it is more than twice the price of MATLAB. However, if you have a lot of patience, then you can use Mathematica for free on a Raspberry Pi. I tried it on an earlier Pi (not sure what version), but not a very early version of the Pi. I found Mathematica to be painfully slow.
|
||
|
Out of print books, ugh. Reminds me of a class I took where the instructor wanted to use Christopher Van Wyk's excellent "Data Structures and C Programming" which is out of print. Van Wyk was enthusiastic about getting it redone for use, but the publisher nixed that. Wouldn't even allow it to be scanned for the class.
toggle quoted message
Show quoted text
Perhaps if enough people are interested they might reissue "Antenna Theory", but these are very niche books whose audience is probably the kind of people who have them already. :o) On Sun, Mar 14, 2021 at 12:17 PM, Reginald Beardsley wrote: Ouch!!!!! $500 and up used! |
||
|
Might be worth trying on a Pi 4. It is _much_ more powerful than any previous Pi.
toggle quoted message
Show quoted text
Sean On Sun, Mar 14, 2021 at 12:47 PM, David Kirkby wrote: However, if you have a lot of patience, then you can use Mathematica for free on a Raspberry Pi. I tried it on an earlier Pi (not sure what version), but not a very early version of the Pi. I found Mathematica to be painfully slow. |
||
|
On Sun, 14 Mar 2021 at 01:47, Reginald Beardsley via <pulaskite=[email protected]> wrote:
I find it extremely unlikely that such proof exists. You would of course need to start by defining what is the "sharpest beam". The diffraction limit puts a limit on the first sidelobe and some Yagi-Uda antennas have the first sidelobe approximately that. I find it worrying that the ARRL no longer sell Lawson's book on Yagi-Uda antennas. It was an ARRL publication.? It was the book that taught me enough to write my own program to analyze the antenna. It is a fairly different concept to NEC. You could argue it give you an analytical expression for the far-field, as the far-field pattern depends on the self impedance of elements and the mutual impedance between them. But the formula for mutual impedance in Lawson's book makes the assumption the elements have zero diameter and are a half-wave long. The
National Bureau of Standards (NBS)? made lots of careful measurements on Yagi-Uda antennas. Lawson shows that one of them must be wrong - probably just a typo. Dave |
jafinch78 .
to navigate to use esc to dismiss