I am a EE and an Extra Class Ham and you did a nice job explaining this subject, I think you hit the nail on the head regarding the coax length of the shorter antenna, as I am sure you know as one moves along a quarter length of coax you form a perfect circle around the impendence of the load on a smith chart, one can also use a quarter length of coax of a different impedance to match the source to the load, lastly at these frequencies conductors and pcb traces also have significant capacitance and inductance that must be taken into account. Like you said, lots of back magic going on. Thanks for sharing.
Thanks, Bruce. Watching with interest this minute. I hope to do a followup test and I'm sure this will improve my methodology. Or perhaps show it's not even needed.
I have been a ham radio operator for over 40 years. In that time I have built hundreds of antennas. I was totally confused by Josh's demonstration. For the reasons you mentioned and more it was not an apple to Apple comparison. You did a great job of explaining how things work!
My dad had a huge understanding as far as antennae and the importance of lenght and the effects. He used to match short wave and CB antennas for truck drivers back in the 70's. We always had a big rig sitting in our yard while my dad set up the truckers radios. He was really good at it too. We had a 60 ft antenna mounted outside our house. At night,when the sky waa clear,I used to talk to guys in Hawaii on the CB radio. His radio was so powerful and well matched,he could talk to anyone as long as the signal could reach his antenna. Some guys were in total disbelief that they were actually talking to a kid from Michigan. I know this is different,but its not. My dad understood all about lenght of antenna, the lenght of the actual cable and the power needed to work the whole thing. Many times guys living close by would complain my dads radios would bleed through several channels if they lived too close to us. He was well within the Fcc regulations and trust me,they showed up several times over complaints. He could probably set up an RC radio set up to fly a drone around the world if he were still alive. I know once the FAA showed up at the door because he was running an illegal linear booster on the antenna and it was interfering with the radio equipment at the airport 25 miles away from us. You could light up a florescent tube up when he turned on the booster. I remember them asking me questions about it but at that age,I didnt know much,so I didnt tell them anything. I do know that night he disassembled the booster and buried it in the back yard. That was the end of his " long range" radio. Too bad,I would have loved flying FPV around the world. For the record,my dad passed away years ago and all his equipment sold. I have no idea how he built that booster,so dont ask.
Perhaps his antenna was more like 36 feet long and he used SSB frequenties?. You could with low power like 500mw get a signal around the globe under perfect conditions bouncing up and down to the atmosphere and back to earth and skip over huge parts of the world. We spoke to all sorts of countries back in the 80's - 90's (from the Netherlands) until everything got regulated. I did some 90 cm experimenting and 2 meter + 11 meter and build some radio transmitters and finally one tv transmitter. Bad filtering on my part made the people nearby knock at my parents house to complain they heard me talking through the tv sets :) Filtering became important :) But one night I got caught and police confiscated all sort's of electronic equipment I have made and everything that was connected to the antenna (some rule!) So i kinda quit... I was about +14-ish at the time. Now I do FPV :) I liked your story, Brought back some nice history.
I talked on 10 & 11 meters. I still have my President HR - 2510. Standing wave ratios, Peaks and troughs in the waveform. I remember building antennas out of 102" of Romex & PVC pipe, and then fiberglassing it all when you're done. Good times!
I found a somewhat glaring error in your whiteboard graphics, trying to discus the "sweet spot". Before continuing, allow me to say that I'm a retired Sr. Electronic Technician, and an Extra Class Amateur Radio operator (for over 40 years). When you shortened the antenna in your diagram, you indicated that it lowered the resonant frequency. You also indicated that when the antenna was too long, it raised the resonant frequency of the antenna. This is completely opposite of reality. The shorter the antenna, the higher the frequency it will resonate at, while the longer the antenna, the lower the frequency it will resonate at. This is why, in amateur radio parlance, an 80 meter (3.5 mhz) antenna has a 1/4 wavelength of almost 67 feet. A 10 meter (28.5mhz) antenna has a 1/4 wavelength of approximately 8 1/4 feet (recall from "CB" days - 27mhz - that many were putting 9 ft. quarter wavelength whips on their vehicles). Another possible point of contention was in your showing that one of the receivers apparently had no allowance for the length of the resonant part of the antenna. Perhaps you didn't notice that the ground traces had "bumps" in them, making them come closer to the antenna feed trace. It is common practice for RF engineers to make up either inductance or capacitance via "tricks" with the traces on the PCB. At SHF frequencies (such as in our RC gear at 2.4ghz), it takes very little variance in the trace to induce either inductance or capacitance. (designing SHF radio PCBs is an EXTREMELY unforgiving endeavor). I still enjoyed your discussion of resonant antennas. In amateur radio, 1/4 wavelength antennas are often used because they exhibit a characteristic impedance "near" 50 to 75 ohms, which makes it easy to feed them with 50 or 75 ohm coax. A 3/4 wavelength antenna (and most odd numbered 1/4 wavelengths, ie., 5/4, 7/4, etc.) will come somewhat close to the 50 - 75 ohm impedance. This is why radio amateurs have used 40 meter antennas (for 7mhz operation) also as a 15 meter antenna (where it is 3/4 wavelength long). I hope you find some of this of interest. While my 40 years of amateur radio antenna experience is in mostly the LF, HF, and VHF frequency ranges, the same theories of operation apply to UHF and SHF antennas. Cheers.
Hello, like your return comment on the video, so you may be able to give me an answer I thought I was right on my thinking from many years ago when I was in RC (35-40) told to me by more experienced flyers back then, that an antenna needed to be 1/4 wavelength and 1/4 wavelength could still be established by having the wire in increments of so many inches IE 2,4,6", etc or 4,8,12" long and still maintain the 1/4 wavelength right or wrong? I'm just mostly trying to establish a more scale look for warbird with long antennas from the cockpit to the tails, but I'm thinking would come in handy for other applications as well if I was right? If you don't mind, What is your thinking/experience on this? Most appreciated, Charlie
@@crotaflyingsolo4759 Hello Charlie. You can use ODD multiples of the 1/4 wavelength & still remain pretty close to resonance. As I mentioned above, ham radio (amateur radio) operators have used 40 meter antennas (approx 33 feet long 1/4 wavelength) to also operate on 15 meters (approx. 11 feet long 1/4 wavelength). The opposite is NOT true .. you cannot resonate an 11 foot long 1/4 wavelength 15 meter antenna on 40 meters without some sort of turning circuitry. So, stick with ODD multiples of the 1/4 wavelength antenna: x3, x5, x7 .. etc. In you warbird models, to make an antenna look more realistic, perhaps you can maintain the 1/4 wavelength & simulate longer lengths by adding on thread/string to the end of the wire to make up the additional length. Good luck.
@@billsomrak First thanks for the reply 2nd what's ODD I may know but forgot? Third, I thought of the string thing but I've got a fine-scale miniature twisted cable meant for the look but thought if I connected the two would it give ant better reception with an Apoximatly 2' long cut to the proper wavelength times 2?? Rec has 2 leads plus the Sat. rec.. I thought it was as you said, had to ask someone of knowledge I asked one of the Tech LOL people at a distributor, and he said ABSOLUTELY NOT! I said to myself yea right and moved on, glad you chimed in the vid. Thanks
@@crotaflyingsolo4759 Odd ... the opposite of Even. I assume you know the difference between odd & even. Just be sure your total length is an odd multiple. Example: if you use 32mm as 1/4 wavelength, then 96mm would be the 3rd multiple ... 160mm would be the 5th multiple ... etc.
@@billsomrak ☹️ okay now that I feel really stupid it's not all my fault when you capitalize the ODD instead of odd, my brain registered it as an abbreviation for a technical term honestly really really! 😁 But thanks for the clarification, think of me what you will I appreciate it. Hopefully I'm not the only one reading this that feels a little stupid right now but pleased to have the information verified okay sir thank you very much I appreciate it! I'll experiment with my extension antenna's. But one more question I think you didn't answer was does / will extending the antennas give a better reception or just longer antennas? Thanks again for your help much appreciated!
Regarding the extra length on the X4R antennas, there are significant reactive components in the feedline (PCB trace, connector and coax) that can be either inductive or capacitive. For this reason, antennas of exactly 1/4 wavelength are rarely very well tuned. These components can be modelled reasonably well, but there's always a bit of variability, especially if you're using low-cost components. If you're installing or replacing a stripped coax antenna, it's usually easiest to start with a radiating element that's a bit too long and trim it back until you hit minimum SWR.
Mr. Bruce. No one seems to ever take into account the length of the shielded cable that terminates in an antenna, the transmission line. I built antennas in my youth. Even a full wave antenna will perform poorly if your transmission line isn't cut to the proper length. The trick is in the diameter of the line with respect to its length. The impedance value. Glad you mentioned the tuning of the transmission line to reduce SWR. A line of an odd length, greater than or less than 3/4 wavelength messes with the 1/4 wave antennas tuning. I have built tuneable antenna loading coils to fine tune the signal phasing to compensate for line odd length. I used a simple signal strength meter placed a short distance from the antenna to peak out the signal output. I also used a SWR meter occasionally. The meter would slightly alter signal strength as compared to an uninterrupted transmission line's actual output. ;-) You impart valuable knowledge to the uneducated hobbyists. Keep up the good work!
Yes, I measured the length of the transmission lines for good reason - which I'll go into in part 2. I have a feeling that those people who have "lengthened" the active element of the 26mm antennas by simply cutting away more of the screen may have introduced additional problems by changing the feeder length ;-)
Absolutely correct! If the antenna's feeder coax was tuned at the factory, any shortening of the shielding will effectively make the antenna slightly longer, slightly out of faze. As long as neither coax or antenna were shortened there should be only a small phasing shift. Slight drop in radiated power. At such low wastage even a slight decrease could make a measurable difference in DB output. ;-)
It thought the velocity factor came in too, so the final length is a product of the frequency, PCB layout, transmission line length, impedance, distance to earth, velocity factor and voodoo. I have to say for some reason antennas have always fascinated me (sad) I like the idea that someone could invent a new 3d design tomorrow that will become a standard for tens of years yet make it at home with some wire, snips and solder. I suppose also where an optional connector is possible by pads on the PCB that will change the dimensions required for a tuned direct solder.
Also the circumference of the transmission wire, actual impedance per measured length, distance between core wire and shielding, and more voodoo, to achieve maximum radiated signal off of the tip of the antenna. YIKES! ;-)
Apparently hobbyist's have performed experiments that has shown antenna lengths make no difference in RF reception. Bruce here on his channel RCmodelreviews has chimed in but has failed to show these experiments to be wrong. Even halfway has to agree. The crazy thing about that is it is proving my contention that there is no 'wavelength' because light has no speed. Unless you call instantaneous a speed. We've been lied to about light-speed.
Bruce I enjoy your videos. I have been designing wireless gadgets for most of my career and have some comments: 1) I agree on your observation about JB’s test up. When we performed radiated tests that are very controlled and try to eliminate as many variables as possible including human interaction. 2) Antenna length is reversely proportional to frequency. So longer means smaller frequency and shorter mean higher frequency. 3) in general a matching circuit is inserted between the RFPA output pins and the antenna. The match is usually tweaked to cancel out the affects of the pcb and connector and 50 ohms is assumed. So typically any 1/4 antenna should be 50ohms but connectors and coax and tolerances can provide variability. 4) I decided myself to just change all of my R-XSR antennas to 150mm with the 32mm ish radiator. It works great for me but I have to admit you and JB bring up good points as to why FrSky is doing this. In the end, FrSky should just come out and say why they are shipping different antennas with their different receivers.
Excellent explanation, Bruce. As for inductors between signal and ground: As you said, an inductor acts like a wire for low frequency, and a resistor for high frequency. In this case, it is shorting the lower frequencies to ground. While it may not be tuning the antenna, it may be there to help avoid other transmission problems, such as noise from motors, escs, etc from affecting the reception of the modulated signal.
One more comment: if the answer is that FrSky is loading the circuit, then it would mean that antennas are not interchangeable between receivers and spare antennas must exactly match the receiver. If that were true, then spare antennas would need to be sold per-receiver. Since this isn't done, my gut feeling is that the receivers aren't being loaded.
Good job Bruce, you've got a good start on the explanation. I questioned Joshua's set-up for testing tx on the pvc. I compared it to truckers dual antennas which change the radiating signal pattern into front to back not omnidirectional. Also when his Taranis tx board failed, the probable cause was the length of the feeder cable between tx board and antenna, which numerous people said didn't matter, but I know that it does, being you can change the swr ratio by lengthening or shortening the feeder line as you've showed. Also ideally the best length for most frequencies for antenna length is a 5/8 wave, anything more or less is not as efficient. Looking forward to part 2
I really enjoy the whiteboard days and educational presentation. Even if some information shakes out as not so correct, kudos for putting it out there. Comments usually clean up the mistakes.
Yey, Whiteboard stuff! 😁👍 It's some kind of crazy, transmitters, receivers and antennas are the absolute center of our hobby - and only a handful of people know how it really works 🤔 What about a educational video series, where you explain things step by step deeper in details to get more antenna professionals out here?
Boy Bruce, I’m a just digital design engineer, but loved fields and waves and RF material because it’s so cool. You and Josh sound like professionals. I have a feeling my remarks, but mostly questions below get right at the heart of the issues involved here. Simplifications in a video like this are essential, but can leave out areas that may explain a lot of issues. First, I’m sure FrSky have some excellent RF/Antenna designers. So the length difference being an accident is nearly impossible. It’s on purpose. But could still be wrong. Between static protection (presumably) L’s to ground, funky shaped circuit board traces, and a transmission line a multiple of the wavelength, and antenna length different from the frequencies resonant length, FrSky had something in mind regarding antenna tuning and impedance matching. I always thought the reactance of an ideal antenna was purely resistive, so to the transmitter it looks as though the power is being completely dissipated into this “resistor”, with none being reflected back. [Remove the antenna, replace with resistive load, transmitter can’t tell.] In reality, no energy is being dissipated in this “R”; it’s really all being transmitted into free space. COOL TRICK! And I thought that the characteristic impedance of an ideal transmission line, which is modeled with L’s and C’s, is also purely resistive, but lossless. An infinite length of this line would appear to be this characteristic impedance, and to a transmitter, the current flow would be accordingly. Then if the characteristic impedance of the line, matches the impedance of the antenna (both 50 ohms, say), the length of the transmission line is irrelevant. But FrSky has the transmission line length a function of the wavelenth, with the antenna not being the 1/4 wavelength “resonant length”, plus some “tuning like” stuff is being done at the front end. My limited and rusty intro knowledge doesn’t withstand this. But I know non-optimal antennas can be tuned at the transmitter end. Question: Will this tuning result in maximal ideal antenna power output, or is it matching non-ideal impedences, to minimize the VSWR, and the efficiency of the whole combination (i.e. maximize the power output, which is lower than the ideal 1/4 wave transmitter length)? Which might fall short of not having to tune in the first space. Or not. Also, it does seem like people are complaining of reduced RC range, being measured by actual failsafes vs telemetry data, and have found that modding to 1/4 wavelength improves the range dramatically. True?? I guess Josh was trying to quantify this. Question: Has anyone looked at the telemetry range before and after these antenna mods?? BIG QUESTION: HAS ANYONE ASKED FRSKY ABOUT IT? Are specific implementations so filled with RF “black magic” that this information would tend to be considered highly proprietary? If so, apparent RC range problems, with their buyers hacking on their lovingly designed, precise length antennas, would tend to make them more forthcoming. I would one would think! I speak all Chinese dialects fluently, so I’m calling, and talking to one of their RF guys. LOL
18:30 It's very likely that the purpose of the inductors is simply to bias or match the output of the IC to the antenna. Not all chips are matched to 50 ohm by default, often the manufacturers will instruct you to add external inductors or capacitors. Also, even though a 1/4-wavelength monopole should be purely resistive (no inductance or capacitance), that still doesn't mean that it will be matched to 50 ohm. A perfect monopole (with infinite ground plane) has an impedance of about 37 ohm. A perfect dipole, or monopole with sleeve balun (pretty much the same thing) has an impedance of about 73 ohm. A simple wire antenna (monopole without ground plane or balun) will be very unpredictable. It's perfectly possible that adding inductors and/or capacitors improves the matching. It's also possible that incorrect antenna lengths will compensate for poor matching at the IC side. In fact we don't even know for sure whether the coax is 50 ohm. It could be 75 ohm or something completely random. Long story short, if you have several different designers doing matching by trial and error, using different antenna lengths, different coax lengths, different ICs and different PCB layouts, it's very likely that they will end up with different solutions that may work equally well.
@@renzevenir4853 Haha WTF this might be the most relevant comment possible especially considering it's coming from a person who actually designs antenna's.
i know that you are an amazing antenna creator,,marten i would like to ask if you could make a pair of 2,4ghz receiver antennas for a long range build hexacopter.. could you ?
Great video, theese ”dumbed down” videos are great. They create a basic understanding of how it works. Most of us aren’t gonna engineer a reciver any time soon anyway.
Wow. I love you for making this video! I am an antenna engineer and i've been wondering about this for a while but havent had the time to look into it myself. One observation about the xsr antenna: if you understand smith charts ( and im assuming you do) then you know that a 3/4 wavelenth of coax will make the capacitance of the short antenna look like an equal amount of inductance at the output of the coax. However that still doesnt explain the extra inductor in parallel on the board. I look forward to part 2!
@RCModelReviews Joshua's test was fine. We know about all those details, he was just doing what we call a 'sanity check', where we are just looking for serious outliers. As expected, he found very little difference. That's great, that means we don't need to worry too much about antennas. All the antennas from reliable brands work well. I also like your video though. Thumbs up, very interesting.
This video is great, and much more informative than Joshua’s video, I look forward to part 2. I know that your telemetry with FrSky was lost a while ago and your RSSI with them is low, but maybe reaching out to them and seeing if they can explain their seemingly arbitrary antenna choices would be really useful 🙂
Yeah, my rather lengthy comment above says FrSky did this on purpose. RC transmission range appears to be improved largely by lengthening antenna to 1/4 wavelength. RSSI return measurement should reflect this. FrSky cut to less than 1/4 wavelength on purpose. Has anyone noticed what the mod did to the telemetry range? That’s the receiver being the transmitter Bruce is talking about. LOL. IDEALLY, I think you’d want your RC range to be as long as possible, but with the telemetry range being LONGER, so the telemetry data always reflects the true RSSI. If the telemetry range is SHORTER, as I’d suspect it would be, then you’d lose telemetry information, before the receiver actually lost RC signal. Question: Does “Loss of signal” mean telemetry transmitted a zero value for RSSI, or that the TELEMETRY signal itself was lost? BIG DIFFERENCE! Since everyone is confused, and it sure looks like FrSky has a problem, I think FrSky would answer the question about “WTF is going on?”
Very informative. It's good to see you and Joshua Bardwell working on the same problem at the same time. I will be watching both with interest as I'm just in the process of switching from Spektrum to Frsky. I'm looking to find out if there is any big difference between receivers and antenna lengths. Thanks Bruce !! Great work !!
Best way to test an antenna is to use an RF cavity... The basic idea is that it both guides the RF signal and acts as a faraday-cage against any background interference / reflections. One side has the transmitting antenna poking into the center of the cavity and angled to provide the strongest possible output. Along with this a ground is attached from the transmitter to the cavity's skin to bleed off any RF coming into contact with the cavity walls. Then a RF step attenuator is used to step up / down the power to simulate required RF intensity / range desired. The other end has the receiving antenna at an exact distance from the transmitting antenna. This is usually a harmonic multiple of the frequency wavelength being tested. This antenna is insulated from the cavity and positioned such that has the greatest possible reception of the incoming RF signal. The received RF is then shunted to a Spectrum-Analyzer to look at both the signal strength and harmonics present. Ideally, from one test to the next the antenna to be tested is placed in the exact position and orientation as the antenna proceeding it.
Awesome video and your diagrams helped me understand the function of the antenna on an RC CAR. Most people say online that you cannot change the length of an RC antenna due to frequency but you just proved how depending where it falls as capacitive or inductive. I have seen nitro RC Cars with super long antennas out above the car body so now it shows why! Thanks for I have learned something to day I did not know before. You should write a book by the way.
Bruce, thank you very much for your contribution of knowledge to this community! As always, you've done an incredible job explaining this complicated topic. I can't wait to see what you'll come up with for field testing in part 2.
I have been researching this for some time now as i was puzzled after measuring different antenna length and designs. A lot of people didnt come up with the right answers. I think "Nominal Velocity of Propagation" is the keyword. The Antenna is 1/4 Wavelength x Propagation constant. The "Propagation constant" of the specific dialectic is specified in the datasheet of the antenna cable beeing used. To be exact its one half of a 1/2 wavelength design dipole antenna that was built concidering Nominal Velocity of Propagation. To test along you should build one antenna without dialectic that is 32mm or longer and cut it down in succession.
BTW if I could just add in my defense, the claim I wanted to investigate was from people saying they were seeing drastically reduced range with 26mm antennas, like 50% range or something. So I felt that even though there were lots of flaws in my test methodology, I would be able to at least show whether there was any merit to that. Without an anechoic chamber or a network analyzer, concrete answers are impossible to come by.
@Milan Karakas channel - Light- travel is instantaneous (if you can call that a speed), but 'travel' is not quite correct. Photo-electricity exists in a matrix. One photon 'ignites' charging it's neighboring photon to ignite and so on 'down the line' till out of human perception instantly. Like a row of dominoes butted up against each other. No matter how long your domino rally, the instant the first domino goes over is the same instant the last domino goes over. Photons are the "ether" of space/existence. A "Holographic Universe". My interest in antennas and reception is with toy drones. We get allot of vids going into FPV reception/transmission distances and mili-amp power ratings in 5.8ghz but when it comes to the technical details of 2.4 ghz transmission we get zilch and even appears youtube channels are trying to get viewers to sabotage their transmission and reception with their claimed 'mods'. - Extending the drone antennas to longer half wave or even full wavelengths that in fact don't work. Modding the transmitters that already have 6 batteries/9volts to lipo 7 volts as if that makes no difference while they try and sell you RF boosters that will not boost if your transmitter is already sending out signal at that amperage. Telling people to throw out a dipole antenna for a clover-leaf further degrading ones transmission range. Where is the specs on the transmitter controller amp output on 6 volt, on 9 volt? These answers they hide from us while they direct our attention at FPV specs and allow charlatans to have us sabotage our drones and their controller transmitters. First things I want to know is the transmission amperage output of these toy transmitters in comparison to their operating voltages rather than being misdirected over wavelength bullcrap.
@Joshua Bardwell - But we do have 'at home' instrumentation to check the efficacy of 'wavelength' MATHEMATICAL theory. In this video th-cam.com/video/pPgd16kFmOA/w-d-xo.html Andrew Mcneil makes a log antenna to specific wire lengths based on hertz divided into lightspeed mathematics 'wavelength' theory. And he has some instrumentation that tests how well it works. It tested great. Now we have a SIMPLE "at home" experiment to test the efficacy of wavelength. Simply have Andrew make another log antenna except NOT to wavelength specks and use that same instrumentation to test this other log antenna against the first. If this second antenna test out low then it proves wavelength is true. But if it test out equal to the first then THAT debunks wavelength and lightspeed along with it. Accusing me of 'nonsense' out of hand is not the scientific method. This experiment is the scientific method. Videos EXPLAINING a misnomered oven is not the scientific method. This EXPERIMENT I'm proposing is the scientific method. Do You guys have balls enough to take the challenge or you gonna coward out and run away screaming 'nonsense nonsense' cause you are too inferiority complexed to ever be proven wrong about anything. ? YOU started this Joshua Bardwell. Now lets see you finish it.
You had a great test, well good enough for the purposes of flying quads. Even with a chamber and vna, it would still be difficult to measure the true properties of the received 2.4g signal at the input pin of the receiver IC. you would need a pcb made that accurately captures the various impedances as they sit on the board with the layout of the passives, specific board (probbaly fr4?), etc. I think the test performed in your yard was more than sufficient.
Good video. Allowing for the disclaimer at the beginning, you appear to be pointing out that, within reason, the actual length of the active segment of the antennas may be irrelevant if they are 'tuned' correctly..... *sits back and watches the arguments*
I just stumbled across your video, so I'm obviously very late to the party. I don't have time to dig through all the comments to see if this has been addressed, but thought I'd chime in anyway. The spacing between pins and/or board traces introduce capacitance. I believe the inductors are used in parallel with the inherent capacitance to form an LC tank filter. At resonance, there would be very little signal passed to ground, but at freqs on either side of Q, stray signals would be attenuated. IOW, the inductors plus natural capacitance combine to form a cheap band pass filter between chip and antenna.
Hello. First of all, you are doing excellent job. Now, there is some things I wont to point out, circuit paths on board have their own capacitance and inductance that can not be neglected, at this frequencies. At about 17:45 on your video, it can be clearly seen that ground paths have some round like shape along side. Those are been used for tuning antenna impedance. As you said, antennas are like voodoo magic. We have to assume that manufacturer was done their best job to cut antenna at "perfect" size for particular circuit for particular type of receiver to tune them well. What you was not said is what is the consequence on radiation path of shortened and elongated antenna, assuming they are both perfectly tuned. Have a nice day and safe flights.
Apparently hobbyist's have performed experiments that has shown antenna lengths make no difference in RF reception. Bruce here on his channel RCmodelreviews has chimed in but has failed to show these experiments to be wrong. Even halfway has to agree. The crazy thing about that is it is proving my contention that there is no 'wavelength' because light has no speed. Unless you call instantaneous a speed. We've been lied to about light-speed.
The round shapes look more like filled vias to me. Without vias, the ground tracks wouldn't be connected to the internal ground plane, which causes all kinds of issues.
Termi, I’m embarrassed for you. The speed of light was measured centuries ago. It’s around 186,000 mi/s in a vacumn. You’re flashlight observations have made you think it’s instantaneous. LOL The speed gets a little slower outside a vacumn, like an electrical signal moving down a transmission line. And c = f*l where c is the speed of light, in the media it’s moving through, and l is the wavelength. I = c/f although I doubt you remember algebra. The speed of light is not instantaneous, but is invariant between different observers moving relative to each other (Thank you Dr. Einstein!), has partical and wave properties (Photons, thank you Einstein, which hit at the heart of quantum mechanics), and via the simple formula, certainly has a wavelenth, as any traveling wave with a velocity will have. Your comment reeks of flat earth people. Your assignment: Write a white paper explaining how the wave properties of light, and corresponding things like interference patterns, are compatable with light being “quantized” into partical like packets called “photons”. That will keep you busy, and that question is deep enough, so as simple as it sounds, your white paper, with the right words, could be Nobel Prize material. That would redeem you for your STUPID ass comment.
the Rx antenna is the single most important and sensitive element in the entire chain, when talking about the final performance of the system. The design of the Tx antenna determines efficiency and reliability of the xmitter, mostly. Those are my 2c as a physicist and as an aside, i do not mind *at all* watering down the details to get a message across!
Nice job on a tricky to explain subject. You have a talent for stripping the fluff off and getting to the stuff that matters. I've noticed many of the camera/vtx combos use the inductor which may reinforce the static shunt theory for the inductor. (old school HF antennas often include a spark gap and very large resistor to take care of wind generated static, which can be shockingly (haha) large. A look at the RF chip specsheet will probably add some insight to the antenna connection LC. My 1st guess on why they use off length antennas on some models is to accommodate the chip and/or PCB inherent inductance or capacitance, One minor point (don't you love it when people pick on your presentation) shorter antenna raise the resonant f, but it doesn't effect your excellent whiteboard presentation one bit. Ignore the nit-pickers and keep digging into the mystery, inquiring RCers wan to know! Looking forward to part 2
Hi folks. Very, very good explanation. This, quite literally, is a 'complex' subject. Reactance cannot be measured with a simple 'real' number. It has a real part, which represents the resistance, and what mathematicians call an 'imaginary' part which represents the combined inductance and capacitance. Together they make up a complex number. Resonance occurs when the complex bit is equal to zero. Enough of my show off pommie gooblygook! Hopefully though it helps to appreciate what an excellent job has been done here explaining this challenging subject. I've subscribed. BobUK.
Shortening the antenna gives a similar graph to shifting the frequency lower on the same length antenna. He's not saying short antennas are for lower frequency.
Yes Ming, you're right. Although I must admit that I had to think twice myself. I hate getting old and after 5 days trying to make this look simple I have to say that I risked confusing myself ;-)
Reminds me of what I got to do early on in RF class with designing an RF front end using lumped element L matching networks for narrow bands. Great video otherwise!
Hey Bruce! That chip the antennas are wired to is a Skyworks SE2431L-r, which is a 2.4ghz RF front end who's job is (basically) to handle antenna switching in diversity applications and and translate it into data the CC2500 Rf tranciever can understand, or vice versa. The datasheet for the front end can be found here: www.skyworksinc.com/uploads/documents/SE2431L_202410K.pdf but what you really want to look at is probably this one: www.ti.com/lit/an/swra439/swra439.pdf At section 4 in that datasheet, they describe some circuit design considerations for power decoupling (and other buzzwords), but particularly around figure 7 where they show the antenna output circuitry, with capacitors in parallel, and the inductors (L8 and L10) in series with them. Note that the difference in these antenna circuits is meant to describe antenna 2 being a PCB antenna, and antenna 1 going to an SMA connector. (section 4.4) More light reading: www.ti.com/lit/an/swra161b/swra161b.pdf
Bruce, thanks for the explanation of the xmttr/rcvr concept...! I can sympathize with your reported confusion, since it has been quite a few years since I studied RF in school...! This is a nice (and concise) refresher!!!
stingerswarm and steele did a video a while back (rotor riot?) about this, they did both long and short leads in range test mode and the shorter one failsafed way fast (in an open field). then they trimmed the short ones shielding back to match the long and the range more than doubled
Some thoughts: 1.The new batch of XSR comes with 32mm antenna 2. Maybe the maximum power transfer theorem/Stanting wave ratio can help to understand the concept and overheating of transmitters. By the way, good video as always Mr. Simpson!
More likely those inductors on that Rx are there to protect the frontend, when the antenna get lopped off by a prop, as happens fairly often on race/freestyle quads. BTW, right up front, it's incorrect to state that only the active element beyond the shield ground is relevant to the antenna tuning. Unless there's a balun between the ground and active element, the entire shield ground resonates and *is* part of the antenna. Because the ground is a different length than the active element, it means the antenna is not a 1/4 wave whip, nor a dipole. A true 1/4 wave whip (such as you'd find on many handheld transmitters) starts right at the antenna connector, and has no ground element at all. It is possible that FrSky did some VSWR tuning and discovered that looking at the antenna as a whole, they got better resonance if the active element is a bit shorter or longer than 1/4 wave. BTW, if you look at UHF LRS Rx antennas which are usually one of 1. "Sander style" 1/4 wave whip (no ground element), 2. a simple coax extension with 1/4 wave active element, or 3. a 1/2 wave dipole (even just the extension with 1/4 wave ground element soldered onto the side below the active element), the dipole blows the other two away. Usually we're talking easily double the range in virtually any orientation. Sander style whips sometimes work, but I think it's because the Rx board itself is used as the ground plane. The coax with 1/4 wave active element is generally quite terrible. I think 2.4ghz Rxs use em mostly due to convenience and ultra low cost, and virtually any other option would be superior. Note that FrSky does make proper dipoles as well.
Given that there's no practical commercial advantage in having slightly different lengths for these antennas you have to conclude that it's either due to poor manufacturing quality control (very unlikely) or simply the result of testing and tuning for the particular design characteristics of each receiver.
Thanks for the video, very cool! It will (might) be interesting to see the "physical" tests because I know I get less range on some receivers vs others, even though they are described as "full" range. I "think" it may be related to antennas, but not sure. I remember when I first started FPV and you came out with a "FPV made easy (part 1) video". That was the video that got me started watching your channel. So, I waited, and waited , and waited... but no part 2. My point is, I hope you continue this subject as I think it will be relevant (at least as of now) and interesting.
I've wondered about this topic in the past and really appreciate your explanation. I saw your previous video a year or two ago about calculating the antenna length, and noticed at the time that my FrSky antennas were actually a little longer than you calculated at 33mm. I considered shortening them at the time but wasn't sure if FrSky was doing something a little differently, and so stuck with the original length. I look forward to part 2 as it sounds like you might test it. I have a potential idea for a new topic for a video: How dangerous is it to be shocked by a typical high-performance, 4 cell Lipo battery? When I have the occasional exceptionally bad crash that really smashes up my quad, I like to examine it before I pick it up to determine if I've damaged the battery wires. What might happen if I accidentally touch the exposed wires with one hand when picking it up? And in a worse case scenario, as unlikely as it might be, what if I grabbed the + wire with one hand and the - with the other, so the current would flow through my chest, (which I've heard is a particularly dangerous scenario with AC power)? Does the high amperage of these batteries make them more dangerous?
Bruce, at 2.4GHz frequencies, will the traces on the circuit board not add inductance and/or capacitance to the circuits, depending upon their length and proximity to the ground traces. I think a lot is made of that in the design of digital circuitry, like motherboards and the internal circuitry of CPU's.
A bit tangential to this vid but I've just seen Josh' Bardwells video on the Trannis antenna mod, and he's pointed out potential issues with over-torquing and in some cases the *seriously* limited max insertion count of SMA connectors. It might be worth mentioning in your next vid that you should only remove antennas if *absolutely nessercery* and avoid over-tightening them.
In the ye olde days of steam radio we were able to tune out the reactance in a transmission line / antenna with the use of a Pi or T network in front of the valve power amplifier. Today’s solid state RF devices don’t have the same range of operation as a Valve/Tube/Bottle and so feedline impedance and antenna length are more critical to ensure efficiency on the intended frequency of operation and safeguard the RF device from destroying itself.
"I'm Bruce Campbell, and you're going to learn something today!" Lol! Great lesson on my favorite brand of receivers! ...you should have played Joshua's creepy piano music at the end!
The question to ask is 'Why?'. Surely cutting to the correct length is significantly more cost effective than implimenting work around circuit design and manufacture?
CyclopsFPV only guys who are doing it knows. Like, what if circuit traces have some impedance or if speed of light in antenna is not the same as in vacuum or ... God knows what.
like bruce said this is voodoo black magic stuff really. RF circuitry has to take into account parasitic capacitances which arise from all sorts of places (like just the traces on the PCB). You may have noticed that the traces on the PCB for the ground connection to the antenna weren't completely straight, and that itself may be to reduce capacitive coupling between the antenna line and the ground plane. There are tons of circuit considerations you have to take into account when designing an RF circuit, even as an Electronics Engineer I don't understand half it. There are industry professionals who have done it for decades who don't completely understand why some things work, just that they do. There are a very handful of people around the world who actually know what they're doing when it comes to RF circuitry.
It's like tuning a xylophone. You calculate what the length and thickness of the bells should be, cut them to length, install them and find they're not quite on pitch; so sand or notch or weight them until they are spot on. When tuning antenna circuits, you do so in an RF screen room, and make the final adjustments based on measurements from a spectrum analyzer or vector network analyser. Not in this case, but when you're dealing with interchangeable antennas or rigid antennas, you can't change the antenna or standard 50Z termination, and it's too late to trim the board, so you want to leave yourself an component that you can quickly swap out in the screen room, and change on the BOM for the determined value when you release to production.
Everyone seems to be assuming you tune the circuit to the antenna, but it seems to me it's probably the other way around as changing a circuit is much harder than cutting an antenna
Some time ago I replaced anntennas on my XSR. I acually soldered uFL connectors to XSR and used some cheap Chinees antennas which have 1/4 wave length antenna. I have done tests on our model airfield and in both tests I reached almost 900m. My feeling from the test is that with modded antennas I had slightly better signal. But this was completly unscientific test so I am looking forward to yours.
Broken Tarannis antenna.. Hi Bruce... Great video keep it up please..just coming to the basics for many of us is a trauma you make it a little bit understandable. My Tarranis antenna just broke and I puzzled as to why it was 'sodered' to the board...why not just put mini sma connector on the RF board so as to just unclip and replace ....or just put a regular sma to unscrew & replace. Now I see it all changes the effecting of the whole set up.... Hmmmm will the type of 'sodder'/ size of blob of said 'sodder' make a difference... Have flown/ range tested and all seems ok... Ps UK is about to legislate re: drones...first reading seems reasonable... Thanks again
One of the factors that you are not considering is the parasitic Capacitance of other metals (including other antennas) that will shift the center of the resonance for the antenna, especially for a simple monopole.In addition, there is no counterpoise (ground plane), which means that the image plane is back against the shield of the coax.If you compare this to a 5/8 wavelength with a counterpoise, then it has a better radiation pattern than these cheap simple wire antennas. It would also be better to create a simple dipole with an associated counterpoise than these antennas.On other factor that you are not considering is polarization; with simple wire monopoles they are vertical or horizontal. For FPV the better antennas have a Circular polarization - Right Hand (RHCP) or Left Hand (LHCP). Again these are most effective with a reference plane.
AT 12:00 , when you shorten the antenna you DON'T lower the operating frequency , you increase it , wavelenght is opositly proportional to the frequancy .
Hi Bruce, I really like the explanation. So thanks. When you tackle the part 2, can you also maybe explain/comment if the length of the coax between the Rx (or Tx) and the active element matters. I think I understand that there is resistance in the wire, but does the coax length also have something to do with tuning?
I appreciate this explanation. However, I don't feel the need to slam other TH-cam "personalities" in the process of commending you, lol. While it might be highly simplified, works good enough to get the principle across to folks like me. So I think it is safe to say this video is a success! Thanks.
Hi Bruce, I have been looking for ages for a decent explanation of antenna theory. All you posts on the subject have been extremely helpful. May I ask you to clarify something else on a related subject? What is the required length of RG316 coax for the Taranis antenna mod, when used with a 4dBi Omni-directional antenna? I watched the post by IBCrazy on the Taranis antenna mod, but there was a couple of instances where I could not follow his logic. As I understand it, we need to align the coax length to the wavelength for the required frequency to negate the effects of impedance and thereby avoid internal reflections \potential signal attenuation. In his video, Alex states the coax cable length is to the base the SMA adaptor not to the entire length (SMA housing + coax) . I would have thought the SMA housing would also influence the wavelength calculations. Would you include the housing or not in your calculation? Secondly , his calculation for determining the length of cable is based on: (∁*VF)/Hz , where ∁ = speed of light (m/s), Vf= Velocity factor of coax material and Hz = frequency (cycles/s) So for a 2.4Ghz signal this should equate to 0.087m for a full wavelength, when using RG316 cable with a VF of 69.5%. Again Alex calculations are different. Obviously he knows far more than me, but I would very much appreciate your thoughts on the subject. To that end, please can you give us your definitive guide to the Taranis antenna MoD? Keep up the brilliant work! Cheers Nick
Simple answer... There are parasitic (unintentional) capacitances and inductances on the receiver PCB, solder joints, U.FL connector and coax cable also that they are trying to compensate for by altering antenna length slightly from design to design (and perhaps from unit to unit to a lesser degree depending on their production test capabilities).
If f increments along the X axis from left to right, then I believe the too short and too long antennas were drawn on the wrong side of the resonant frequency. Outcome is the same, though.
The critical frequency has units of Hz, but wavelength has units of m. How did you go from Hz to m? Did I miss a step? I enjoy and appreciate your videos. Please keep it up.
The circuit temperature would be the best indicator of reflected power. Adjust antenna length for lowest temperature. FrSky probably knows this. At home, adjust antenna length for best range/performance. That's what you want anyway.
18:07 Why are there dots etched on the Antenna lines from the Chip to the antenna connectors on that PCB ?? 18:20 Why would they put extra parts (inductors) on this PCB when all they had to do is make the antennas a tiny bit longer ?? Will be waiting for the outdoor range test results.
I think the dots are just blind vias (vias are connecting holes to other PCB layer; blind vias are to internal layers so you don't see the holes), to make a low impedance path to ground on the other side of the board to make what's called a stripline (impedance matched transmission line etched on a PCB) or waveguide. If you look at RF PCBs you usually see this "stiching" along transmission line to antennas or antenna connectors. Example: iot-bits.com/wp-content/uploads/2017/10/PCB-trace-antenna-layout-via-stitching-technique-1024x885.png
Bruce, despite saying that you are not sponsored by any company, it does seem like you are sending subtle messages. You’re wearing a lovely crisp lumenier t-shirt that makes you look like a lumenier radio engineer whilst at the same time wiping your board/arse with the hobby king t-shirt :)
Bruce, can you specifically do some tests with the XM+? I've found it gets the best range of all FrSky RX other than the L9R and the XM+ has had antenna changes through out its revisions. Strangely, the first revision of the XM+ seems to be the best performing, with the latest with the ultra short antennas being the worst.
Hi, I have an E-Foil question , Does it matter how long the sheilded /braided part of a 2.4 antenna is before the 31mm bare section ? I have an efoild surf board , the reciever is on the ESC at the rear ,I need to extend the antenna towards the front of the board because when starting off with the rear of the board submerged . also the hatch at the rear is carbon fibre wheras the board is not . ( am I correct in thinking this could block the signal to a degree? My plan was to use a 500 mm extention cable with micro UFL on the reciver end and remove the shielding for the last 31mm the other end. Will this work and not damage the reciever please?
As mentioned previously in the comments, I wonder if velocity factor is part of the issue. Are they both the same coax and is the velocity factor the same. Are the capacitance of the 2 antennas the same.
19:40 inductors are not used to implement static discharge protection, it's not rational because it will complicate RF front end matching. In industry common practice is to use very low parasitic capacitance TVS diodes (TVS - transient voltage suppressor). You took basic examples of a simple antenna matching with series inductor/capacitor and made a mystery why inductor is used in parallel. Looking from a Smith chart you can use different approaches to match impedance.
Once again, another excellent technical video with some humor mixed in. SIR: could you maybe look into the possible harmful effects from a 1 watt rf transmitter at 5.8 ghz? I am not qualified to address this but I think you may be. Perhaps we should be telling folks to keep active microwave radiators away from their eyes. As power gets higher and higher, we are starting to make little microwave ovens! It is time someone addressed this and everyone, including me, trusts your opinion.
I am a EE and an Extra Class Ham and you did a nice job explaining this subject, I think you hit the nail on the head regarding the coax length of the shorter antenna, as I am sure you know as one moves along a quarter length of coax you form a perfect circle around the impendence of the load on a smith chart, one can also use a quarter length of coax of a different impedance to match the source to the load, lastly at these frequencies conductors and pcb traces also have significant capacitance and inductance that must be taken into account. Like you said, lots of back magic going on. Thanks for sharing.
Thanks, Bruce. Watching with interest this minute. I hope to do a followup test and I'm sure this will improve my methodology. Or perhaps show it's not even needed.
FrSky are still selling dodgy R9 antennas in the EU. Thank you
Paweł Spychalski channel for providing a workaround.
I you haven't, watch Andrew McNeil. He has some great vids on this subject
I have been a ham radio operator for over 40 years. In that time I have built hundreds of antennas. I was totally confused by Josh's demonstration. For the reasons you mentioned and more it was not an apple to Apple comparison. You did a great job of explaining how things work!
Josh makes me cringe at times. Dosn't he work with RF professionally?
At a black box level. His job is with wireless networking.
My dad had a huge understanding as far as antennae and the importance of lenght and the effects. He used to match short wave and CB antennas for truck drivers back in the 70's. We always had a big rig sitting in our yard while my dad set up the truckers radios. He was really good at it too. We had a 60 ft antenna mounted outside our house. At night,when the sky waa clear,I used to talk to guys in Hawaii on the CB radio. His radio was so powerful and well matched,he could talk to anyone as long as the signal could reach his antenna. Some guys were in total disbelief that they were actually talking to a kid from Michigan. I know this is different,but its not. My dad understood all about lenght of antenna, the lenght of the actual cable and the power needed to work the whole thing. Many times guys living close by would complain my dads radios would bleed through several channels if they lived too close to us. He was well within the Fcc regulations and trust me,they showed up several times over complaints. He could probably set up an RC radio set up to fly a drone around the world if he were still alive. I know once the FAA showed up at the door because he was running an illegal linear booster on the antenna and it was interfering with the radio equipment at the airport 25 miles away from us. You could light up a florescent tube up when he turned on the booster. I remember them asking me questions about it but at that age,I didnt know much,so I didnt tell them anything. I do know that night he disassembled the booster and buried it in the back yard. That was the end of his " long range" radio. Too bad,I would have loved flying FPV around the world. For the record,my dad passed away years ago and all his equipment sold. I have no idea how he built that booster,so dont ask.
Bruce could it be that the uneven lengths counter each others capacitance/induction by themselves thus not needing helping circuits?
Perhaps his antenna was more like 36 feet long and he used SSB frequenties?. You could with low power like 500mw get a signal around the globe under perfect conditions bouncing up and down to the atmosphere and back to earth and skip over huge parts of the world. We spoke to all sorts of countries back in the 80's - 90's (from the Netherlands) until everything got regulated. I did some 90 cm experimenting and 2 meter + 11 meter and build some radio transmitters and finally one tv transmitter. Bad filtering on my part made the people nearby knock at my parents house to complain they heard me talking through the tv sets :) Filtering became important :) But one night I got caught and police confiscated all sort's of electronic equipment I have made and everything that was connected to the antenna (some rule!) So i kinda quit... I was about +14-ish at the time. Now I do FPV :) I liked your story, Brought back some nice history.
I talked on 10 & 11 meters. I still have my President HR - 2510.
Standing wave ratios, Peaks and troughs in the waveform. I remember building antennas out of 102" of Romex & PVC pipe, and then fiberglassing it all when you're done. Good times!
I found a somewhat glaring error in your whiteboard graphics, trying to discus the "sweet spot". Before continuing, allow me to say that I'm a retired Sr. Electronic Technician, and an Extra Class Amateur Radio operator (for over 40 years). When you shortened the antenna in your diagram, you indicated that it lowered the resonant frequency. You also indicated that when the antenna was too long, it raised the resonant frequency of the antenna. This is completely opposite of reality. The shorter the antenna, the higher the frequency it will resonate at, while the longer the antenna, the lower the frequency it will resonate at. This is why, in amateur radio parlance, an 80 meter (3.5 mhz) antenna has a 1/4 wavelength of almost 67 feet. A 10 meter (28.5mhz) antenna has a 1/4 wavelength of approximately 8 1/4 feet (recall from "CB" days - 27mhz - that many were putting 9 ft. quarter wavelength whips on their vehicles). Another possible point of contention was in your showing that one of the receivers apparently had no allowance for the length of the resonant part of the antenna. Perhaps you didn't notice that the ground traces had "bumps" in them, making them come closer to the antenna feed trace. It is common practice for RF engineers to make up either inductance or capacitance via "tricks" with the traces on the PCB. At SHF frequencies (such as in our RC gear at 2.4ghz), it takes very little variance in the trace to induce either inductance or capacitance. (designing SHF radio PCBs is an EXTREMELY unforgiving endeavor). I still enjoyed your discussion of resonant antennas. In amateur radio, 1/4 wavelength antennas are often used because they exhibit a characteristic impedance "near" 50 to 75 ohms, which makes it easy to feed them with 50 or 75 ohm coax. A 3/4 wavelength antenna (and most odd numbered 1/4 wavelengths, ie., 5/4, 7/4, etc.) will come somewhat close to the 50 - 75 ohm impedance. This is why radio amateurs have used 40 meter antennas (for 7mhz operation) also as a 15 meter antenna (where it is 3/4 wavelength long). I hope you find some of this of interest. While my 40 years of amateur radio antenna experience is in mostly the LF, HF, and VHF frequency ranges, the same theories of operation apply to UHF and SHF antennas. Cheers.
Hello, like your return comment on the video, so you may be able to give me an answer I thought I was right on my thinking from many years ago when I was in RC (35-40) told to me by more experienced flyers back then, that an antenna needed to be 1/4 wavelength and 1/4 wavelength could still be established by having the wire in increments of so many inches IE 2,4,6", etc or 4,8,12" long and still maintain the 1/4 wavelength right or wrong?
I'm just mostly trying to establish a more scale look for warbird with long antennas from the cockpit to the tails, but I'm thinking would come in handy for other applications as well if I was right?
If you don't mind, What is your thinking/experience on this?
Most appreciated, Charlie
@@crotaflyingsolo4759 Hello Charlie. You can use ODD multiples of the 1/4 wavelength & still remain pretty close to resonance. As I mentioned above, ham radio (amateur radio) operators have used 40 meter antennas (approx 33 feet long 1/4 wavelength) to also operate on 15 meters (approx. 11 feet long 1/4 wavelength). The opposite is NOT true .. you cannot resonate an 11 foot long 1/4 wavelength 15 meter antenna on 40 meters without some sort of turning circuitry. So, stick with ODD multiples of the 1/4 wavelength antenna: x3, x5, x7 .. etc. In you warbird models, to make an antenna look more realistic, perhaps you can maintain the 1/4 wavelength & simulate longer lengths by adding on thread/string to the end of the wire to make up the additional length. Good luck.
@@billsomrak First thanks for the reply 2nd what's ODD I may know but forgot?
Third, I thought of the string thing but I've got a fine-scale miniature twisted cable meant for the look but thought if I connected the two would it give ant better reception with an Apoximatly 2' long cut to the proper wavelength times 2?? Rec has 2 leads plus the Sat. rec..
I thought it was as you said, had to ask someone of knowledge I asked one of the Tech LOL people at a distributor, and he said ABSOLUTELY NOT! I said to myself yea right and moved on, glad you chimed in the vid.
Thanks
@@crotaflyingsolo4759 Odd ... the opposite of Even. I assume you know the difference between odd & even. Just be sure your total length is an odd multiple. Example: if you use 32mm as 1/4 wavelength, then 96mm would be the 3rd multiple ... 160mm would be the 5th multiple ... etc.
@@billsomrak ☹️ okay now that I feel really stupid it's not all my fault when you capitalize the ODD instead of odd, my brain registered it as an abbreviation for a technical term honestly really really! 😁
But thanks for the clarification, think of me what you will I appreciate it. Hopefully I'm not the only one reading this that feels a little stupid right now but pleased to have the information verified okay sir thank you very much I appreciate it! I'll experiment with my extension antenna's.
But one more question I think you didn't answer was does / will extending the antennas give a better reception or just longer antennas?
Thanks again for your help much appreciated!
Regarding the extra length on the X4R antennas, there are significant reactive components in the feedline (PCB trace, connector and coax) that can be either inductive or capacitive. For this reason, antennas of exactly 1/4 wavelength are rarely very well tuned. These components can be modelled reasonably well, but there's always a bit of variability, especially if you're using low-cost components. If you're installing or replacing a stripped coax antenna, it's usually easiest to start with a radiating element that's a bit too long and trim it back until you hit minimum SWR.
SWR?
Mr. Bruce. No one seems to ever take into account the length of the shielded cable that terminates in an antenna, the transmission line. I built antennas in my youth. Even a full wave antenna will perform poorly if your transmission line isn't cut to the proper length. The trick is in the diameter of the line with respect to its length. The impedance value. Glad you mentioned the tuning of the transmission line to reduce SWR. A line of an odd length, greater than or less than 3/4 wavelength messes with the 1/4 wave antennas tuning. I have built tuneable antenna loading coils to fine tune the signal phasing to compensate for line odd length. I used a simple signal strength meter placed a short distance from the antenna to peak out the signal output. I also used a SWR meter occasionally. The meter would slightly alter signal strength as compared to an uninterrupted transmission line's actual output. ;-) You impart valuable knowledge to the uneducated hobbyists. Keep up the good work!
Yes, I measured the length of the transmission lines for good reason - which I'll go into in part 2. I have a feeling that those people who have "lengthened" the active element of the 26mm antennas by simply cutting away more of the screen may have introduced additional problems by changing the feeder length ;-)
Absolutely correct! If the antenna's feeder coax was tuned at the factory, any shortening of the shielding will effectively make the antenna slightly longer, slightly out of faze. As long as neither coax or antenna were shortened there should be only a small phasing shift. Slight drop in radiated power. At such low wastage even a slight decrease could make a measurable difference in DB output. ;-)
It thought the velocity factor came in too, so the final length is a product of the frequency, PCB layout, transmission line length, impedance, distance to earth, velocity factor and voodoo. I have to say for some reason antennas have always fascinated me (sad) I like the idea that someone could invent a new 3d design tomorrow that will become a standard for tens of years yet make it at home with some wire, snips and solder. I suppose also where an optional connector is possible by pads on the PCB that will change the dimensions required for a tuned direct solder.
Also the circumference of the transmission wire, actual impedance per measured length, distance between core wire and shielding, and more voodoo, to achieve maximum radiated signal off of the tip of the antenna. YIKES! ;-)
Apparently hobbyist's have performed experiments that has shown antenna
lengths make no difference in RF reception. Bruce here on his channel
RCmodelreviews has chimed in but has failed to show these experiments to
be wrong. Even halfway has to agree. The crazy thing about that is it
is proving my contention that there is no 'wavelength' because light has
no speed. Unless you call instantaneous a speed. We've been lied to about light-speed.
Bruce I enjoy your videos. I have been designing wireless gadgets for most of my career and have some comments:
1) I agree on your observation about JB’s test up. When we performed radiated tests that are very controlled and try to eliminate as many variables as possible including human interaction.
2) Antenna length is reversely proportional to frequency. So longer means smaller frequency and shorter mean higher frequency.
3) in general a matching circuit is inserted between the RFPA output pins and the antenna. The match is usually tweaked to cancel out the affects of the pcb and connector and 50 ohms is assumed. So typically any 1/4 antenna should be 50ohms but connectors and coax and tolerances can provide variability.
4) I decided myself to just change all of my R-XSR antennas to 150mm with the 32mm ish radiator. It works great for me but I have to admit you and JB bring up good points as to why FrSky is doing this. In the end, FrSky should just come out and say why they are shipping different antennas with their different receivers.
My Gratitude to you sir...Your contributions have made you immortal on this planet.
Excellent explanation, Bruce. As for inductors between signal and ground:
As you said, an inductor acts like a wire for low frequency, and a resistor for high frequency. In this case, it is shorting the lower frequencies to ground. While it may not be tuning the antenna, it may be there to help avoid other transmission problems, such as noise from motors, escs, etc from affecting the reception of the modulated signal.
One more comment: if the answer is that FrSky is loading the circuit, then it would mean that antennas are not interchangeable between receivers and spare antennas must exactly match the receiver. If that were true, then spare antennas would need to be sold per-receiver. Since this isn't done, my gut feeling is that the receivers aren't being loaded.
Good job Bruce, you've got a good start on the explanation. I questioned Joshua's set-up for testing tx on the pvc. I compared it to truckers dual antennas which change the radiating signal pattern into front to back not omnidirectional. Also when his Taranis tx board failed, the probable cause was the length of the feeder cable between tx board and antenna, which numerous people said didn't matter, but I know that it does, being you can change the swr ratio by lengthening or shortening the feeder line as you've showed. Also ideally the best length for most frequencies for antenna length is a 5/8 wave, anything more or less is not as efficient. Looking forward to part 2
I really enjoy the whiteboard days and educational presentation. Even if some information shakes out as not so correct, kudos for putting it out there. Comments usually clean up the mistakes.
Yey, Whiteboard stuff! 😁👍
It's some kind of crazy, transmitters, receivers and antennas are the absolute center of our hobby - and only a handful of people know how it really works 🤔
What about a educational video series, where you explain things step by step deeper in details to get more antenna professionals out here?
RCModelReviews, You are an asset to the FPV community and a legend mate !
Boy Bruce, I’m a just digital design engineer, but loved fields and waves and RF material because it’s so cool. You and Josh sound like professionals. I have a feeling my remarks, but mostly questions below get right at the heart of the issues involved here.
Simplifications in a video like this are essential, but can leave out areas that may explain a lot of issues. First, I’m sure FrSky have some excellent RF/Antenna designers. So the length difference being an accident is nearly impossible. It’s on purpose. But could still be wrong. Between static protection (presumably) L’s to ground, funky shaped circuit board traces, and a transmission line a multiple of the wavelength, and antenna length different from the frequencies resonant length, FrSky had something in mind regarding antenna tuning and impedance matching.
I always thought the reactance of an ideal antenna was purely resistive, so to the transmitter it looks as though the power is being completely dissipated into this “resistor”, with none being reflected back. [Remove the antenna, replace with resistive load, transmitter can’t tell.] In reality, no energy is being dissipated in this “R”; it’s really all being transmitted into free space. COOL TRICK! And I thought that the characteristic impedance of an ideal transmission line, which is modeled with L’s and C’s, is also purely resistive, but lossless. An infinite length of this line would appear to be this characteristic impedance, and to a transmitter, the current flow would be accordingly. Then if the characteristic impedance of the line, matches the impedance of the antenna (both 50 ohms, say), the length of the transmission line is irrelevant. But FrSky has the transmission line length a function of the wavelenth, with the antenna not being the 1/4 wavelength “resonant length”, plus some “tuning like” stuff is being done at the front end.
My limited and rusty intro knowledge doesn’t withstand this. But I know non-optimal antennas can be tuned at the transmitter end. Question: Will this tuning result in maximal ideal antenna power output, or is it matching non-ideal impedences, to minimize the VSWR, and the efficiency of the whole combination (i.e. maximize the power output, which is lower than the ideal 1/4 wave transmitter length)? Which might fall short of not having to tune in the first space. Or not.
Also, it does seem like people are complaining of reduced RC range, being measured by actual failsafes vs telemetry data, and have found that modding to 1/4 wavelength improves the range dramatically. True?? I guess Josh was trying to quantify this. Question: Has anyone looked at the telemetry range before and after these antenna mods?? BIG QUESTION: HAS ANYONE ASKED FRSKY ABOUT IT? Are specific implementations so filled with RF “black magic” that this information would tend to be considered highly proprietary? If so, apparent RC range problems, with their buyers hacking on their lovingly designed, precise length antennas, would tend to make them more forthcoming. I would one would think! I speak all Chinese dialects fluently, so I’m calling, and talking to one of their RF guys. LOL
18:30 It's very likely that the purpose of the inductors is simply to bias or match the output of the IC to the antenna. Not all chips are matched to 50 ohm by default, often the manufacturers will instruct you to add external inductors or capacitors.
Also, even though a 1/4-wavelength monopole should be purely resistive (no inductance or capacitance), that still doesn't mean that it will be matched to 50 ohm. A perfect monopole (with infinite ground plane) has an impedance of about 37 ohm. A perfect dipole, or monopole with sleeve balun (pretty much the same thing) has an impedance of about 73 ohm. A simple wire antenna (monopole without ground plane or balun) will be very unpredictable. It's perfectly possible that adding inductors and/or capacitors improves the matching. It's also possible that incorrect antenna lengths will compensate for poor matching at the IC side. In fact we don't even know for sure whether the coax is 50 ohm. It could be 75 ohm or something completely random.
Long story short, if you have several different designers doing matching by trial and error, using different antenna lengths, different coax lengths, different ICs and different PCB layouts, it's very likely that they will end up with different solutions that may work equally well.
@@renzevenir4853 Haha WTF this might be the most relevant comment possible especially considering it's coming from a person who actually designs antenna's.
i know that you are an amazing antenna creator,,marten i would like to ask if you could make a pair of 2,4ghz receiver antennas for a long range build hexacopter.. could you ?
Great video, theese ”dumbed down” videos are great. They create a basic understanding of how it works. Most of us aren’t gonna engineer a reciver any time soon anyway.
Wow. I love you for making this video! I am an antenna engineer and i've been wondering about this for a while but havent had the time to look into it myself. One observation about the xsr antenna: if you understand smith charts ( and im assuming you do) then you know that a 3/4 wavelenth of coax will make the capacitance of the short antenna look like an equal amount of inductance at the output of the coax. However that still doesnt explain the extra inductor in parallel on the board. I look forward to part 2!
@RCModelReviews
Joshua's test was fine. We know about all those details, he was just doing what we call a 'sanity check', where we are just looking for serious outliers. As expected, he found very little difference. That's great, that means we don't need to worry too much about antennas. All the antennas from reliable brands work well.
I also like your video though. Thumbs up, very interesting.
This video is great, and much more informative than Joshua’s video, I look forward to part 2. I know that your telemetry with FrSky was lost a while ago and your RSSI with them is low, but maybe reaching out to them and seeing if they can explain their seemingly arbitrary antenna choices would be really useful 🙂
Well joshua's video was also greate becouse it simple. Mesurmenting method was practical.
Yeah, my rather lengthy comment above says FrSky did this on purpose. RC transmission range appears to be improved largely by lengthening antenna to 1/4 wavelength. RSSI return measurement should reflect this. FrSky cut to less than 1/4 wavelength on purpose. Has anyone noticed what the mod did to the telemetry range? That’s the receiver being the transmitter Bruce is talking about. LOL. IDEALLY, I think you’d want your RC range to be as long as possible, but with the telemetry range being LONGER, so the telemetry data always reflects the true RSSI. If the telemetry range is SHORTER, as I’d suspect it would be, then you’d lose telemetry information, before the receiver actually lost RC signal. Question: Does “Loss of signal” mean telemetry transmitted a zero value for RSSI, or that the TELEMETRY signal itself was lost? BIG DIFFERENCE! Since everyone is confused, and it sure looks like FrSky has a problem, I think FrSky would answer the question about “WTF is going on?”
Very informative. It's good to see you and Joshua Bardwell working on the same problem at the same time. I will be watching both with interest as I'm just in the process of switching from Spektrum to Frsky. I'm looking to find out if there is any big difference between receivers and antenna lengths. Thanks Bruce !! Great work !!
Best way to test an antenna is to use an RF cavity...
The basic idea is that it both guides the RF signal and acts as a faraday-cage against any background interference / reflections. One side has the transmitting antenna poking into the center of the cavity and angled to provide the strongest possible output. Along with this a ground is attached from the transmitter to the cavity's skin to bleed off any RF coming into contact with the cavity walls. Then a RF step attenuator is used to step up / down the power to simulate required RF intensity / range desired.
The other end has the receiving antenna at an exact distance from the transmitting antenna. This is usually a harmonic multiple of the frequency wavelength being tested. This antenna is insulated from the cavity and positioned such that has the greatest possible reception of the incoming RF signal. The received RF is then shunted to a Spectrum-Analyzer to look at both the signal strength and harmonics present. Ideally, from one test to the next the antenna to be tested is placed in the exact position and orientation as the antenna proceeding it.
Looking forward to your tests, also looking forward to the long range Tx/Rx systems tests.
Awesome video and your diagrams helped me understand the function of the antenna on an RC CAR. Most people say online that you cannot change the length of an RC antenna due to frequency but you just proved how depending where it falls as capacitive or inductive. I have seen nitro RC Cars with super long antennas out above the car body so now it shows why! Thanks for I have learned something to day I did not know before. You should write a book by the way.
Bruce, thank you very much for your contribution of knowledge to this community! As always, you've done an incredible job explaining this complicated topic. I can't wait to see what you'll come up with for field testing in part 2.
I have been researching this for some time now as i was puzzled after measuring different antenna length and designs. A lot of people didnt come up with the right answers. I think "Nominal Velocity of Propagation" is the keyword. The Antenna is 1/4 Wavelength x Propagation constant. The "Propagation constant" of the specific dialectic is specified in the datasheet of the antenna cable beeing used. To be exact its one half of a 1/2 wavelength design dipole antenna that was built concidering Nominal Velocity of Propagation. To test along you should build one antenna without dialectic that is 32mm or longer and cut it down in succession.
Your explanation was a bit simple but enjoyable none the less. Thanks Bruce!
BTW if I could just add in my defense, the claim I wanted to investigate was from people saying they were seeing drastically reduced range with 26mm antennas, like 50% range or something. So I felt that even though there were lots of flaws in my test methodology, I would be able to at least show whether there was any merit to that. Without an anechoic chamber or a network analyzer, concrete answers are impossible to come by.
well indeed sir, chopped the antenna to the 31mm after watching your video and gained a lot of range, no more failsafes at the end of the track
No one is critiquing your work JB, between you and Bruce we’re destined to get the best answers for RC in both hemispheres.
@Milan Karakas channel - Light- travel is instantaneous (if you can call that a speed), but
'travel' is not quite correct. Photo-electricity
exists in a matrix. One photon 'ignites' charging it's neighboring
photon to ignite and so on 'down the line' till out of human perception
instantly. Like a row of dominoes butted up against each other. No
matter how long your domino rally, the instant the first domino goes
over is the same instant the last domino goes over. Photons are the
"ether" of space/existence. A "Holographic Universe".
My interest in antennas and reception is with toy drones. We get allot
of vids going into FPV reception/transmission distances and mili-amp
power ratings in 5.8ghz but when it comes to the technical details of
2.4 ghz transmission we get zilch and even appears youtube channels are
trying to get viewers to sabotage their transmission and reception with
their claimed 'mods'. - Extending the drone antennas to longer half
wave or even full wavelengths that in fact don't work. Modding the
transmitters that already have 6 batteries/9volts to lipo 7 volts as if
that makes no difference while they try and sell you RF boosters that
will not boost if your transmitter is already sending out signal
at that amperage. Telling people to throw out a dipole antenna for
a clover-leaf further degrading ones transmission range. Where is the
specs on the transmitter controller amp output on 6 volt, on 9 volt?
These answers they hide from us while they direct our attention at FPV
specs and allow charlatans to have us sabotage our drones and their
controller transmitters. First things I want to know is the transmission
amperage output of these toy transmitters in comparison to their
operating voltages rather than being misdirected over wavelength
bullcrap.
@Joshua Bardwell - But we do have 'at home' instrumentation to check the efficacy of 'wavelength' MATHEMATICAL theory. In this video th-cam.com/video/pPgd16kFmOA/w-d-xo.html
Andrew Mcneil makes a log antenna to specific wire lengths based on
hertz divided into lightspeed mathematics 'wavelength' theory. And he
has some instrumentation that tests how well it works. It tested great.
Now we have a SIMPLE "at home" experiment to test the efficacy of
wavelength. Simply have Andrew make another log antenna except NOT to
wavelength specks and use that same instrumentation to test this other
log antenna against the first. If this second antenna test out low then
it proves wavelength is true. But if it test out equal to the first then
THAT debunks wavelength and lightspeed along with it. Accusing me of
'nonsense' out of hand is not the scientific method. This experiment is
the scientific method. Videos EXPLAINING a misnomered oven is not the
scientific method. This EXPERIMENT I'm proposing is the scientific
method. Do You guys have balls enough to take the challenge or you gonna
coward out and run away screaming 'nonsense nonsense' cause you are too
inferiority complexed to ever be proven wrong about anything. ? YOU
started this Joshua Bardwell. Now lets see you finish it.
You had a great test, well good enough for the purposes of flying quads. Even with a chamber and vna, it would still be difficult to measure the true properties of the received 2.4g signal at the input pin of the receiver IC. you would need a pcb made that accurately captures the various impedances as they sit on the board with the layout of the passives, specific board (probbaly fr4?), etc. I think the test performed in your yard was more than sufficient.
Good video. Allowing for the disclaimer at the beginning, you appear to be pointing out that, within reason, the actual length of the active segment of the antennas may be irrelevant if they are 'tuned' correctly.....
*sits back and watches the arguments*
Nicely done , a qualitative understanding is always a good idea before quantitative . Because only then do the numbers make sense .
Stop apologizing for other people’s lack of ability to understand. I like your videos.
Oddly enough this is nearly like tuning a speaker enclosure crossover network circuit . . . sort of kind of . Brilliant stuff . Thanks man !
I just stumbled across your video, so I'm obviously very late to the party. I don't have time to dig through all the comments to see if this has been addressed, but thought I'd chime in anyway. The spacing between pins and/or board traces introduce capacitance. I believe the inductors are used in parallel with the inherent capacitance to form an LC tank filter. At resonance, there would be very little signal passed to ground, but at freqs on either side of Q, stray signals would be attenuated. IOW, the inductors plus natural capacitance combine to form a cheap band pass filter between chip and antenna.
Hello. First of all, you are doing excellent job. Now, there is some things I wont to point out, circuit paths on board have their own capacitance and inductance that can not be neglected, at this frequencies.
At about 17:45 on your video, it can be clearly seen that ground paths have some round like shape along side. Those are been used for tuning antenna impedance. As you said, antennas are like voodoo magic. We have to assume that manufacturer was done their best job to cut antenna at "perfect" size for particular circuit for particular type of receiver to tune them well. What you was not said is what is the consequence on radiation path of shortened and elongated antenna, assuming they are both perfectly tuned. Have a nice day and safe flights.
Apparently hobbyist's have performed experiments that has shown antenna
lengths make no difference in RF reception. Bruce here on his channel
RCmodelreviews has chimed in but has failed to show these experiments to
be wrong. Even halfway has to agree. The crazy thing about that is it
is proving my contention that there is no 'wavelength' because light has
no speed. Unless you call instantaneous a speed. We've been lied to about light-speed.
The round shapes look more like filled vias to me. Without vias, the ground tracks wouldn't be connected to the internal ground plane, which causes all kinds of issues.
also: earth is a cylinder
Termi, I’m embarrassed for you. The speed of light was measured centuries ago. It’s around 186,000 mi/s in a vacumn. You’re flashlight observations have made you think it’s instantaneous. LOL The speed gets a little slower outside a vacumn, like an electrical signal moving down a transmission line. And c = f*l where c is the speed of light, in the media it’s moving through, and l is the wavelength. I = c/f although I doubt you remember algebra. The speed of light is not instantaneous, but is invariant between different observers moving relative to each other (Thank you Dr. Einstein!), has partical and wave properties (Photons, thank you Einstein, which hit at the heart of quantum mechanics), and via the simple formula, certainly has a wavelenth, as any traveling wave with a velocity will have. Your comment reeks of flat earth people. Your assignment: Write a white paper explaining how the wave properties of light, and corresponding things like interference patterns, are compatable with light being “quantized” into partical like packets called “photons”. That will keep you busy, and that question is deep enough, so as simple as it sounds, your white paper, with the right words, could be Nobel Prize material. That would redeem you for your STUPID ass comment.
Thanks, Bruce. I really enjoy these whiteboard talks!
Don't worry about the haters. I love the way you explain things. Keep it simple... thanks.
the Rx antenna is the single most important and sensitive element in the entire chain, when talking about the final performance of the system. The design of the Tx antenna determines efficiency and reliability of the xmitter, mostly. Those are my 2c as a physicist and as an aside, i do not mind *at all* watering down the details to get a message across!
Nice job on a tricky to explain subject. You have a talent for stripping the fluff off and getting to the stuff that matters.
I've noticed many of the camera/vtx combos use the inductor which may reinforce the static shunt theory for the inductor. (old school HF antennas often include a spark gap and very large resistor to take care of wind generated static, which can be shockingly (haha) large.
A look at the RF chip specsheet will probably add some insight to the antenna connection LC. My 1st guess on why they use off length antennas on some models is to accommodate the chip and/or PCB inherent inductance or capacitance,
One minor point (don't you love it when people pick on your presentation) shorter antenna raise the resonant f, but it doesn't effect your excellent whiteboard presentation one bit.
Ignore the nit-pickers and keep digging into the mystery, inquiring RCers wan to know!
Looking forward to part 2
Hi folks. Very, very good explanation. This, quite literally, is a 'complex' subject. Reactance cannot be measured with a simple 'real' number. It has a real part, which represents the resistance, and what mathematicians call an 'imaginary' part which represents the combined inductance and capacitance. Together they make up a complex number. Resonance occurs when the complex bit is equal to zero. Enough of my show off pommie gooblygook! Hopefully though it helps to appreciate what an excellent job has been done here explaining this challenging subject. I've subscribed. BobUK.
Bruce. Isn't shortening your antenna actually moving the resonant frequency up not down? Reverse of what you said at 12:35.
westernradio that caught my eye also.
Hey, I'm old and easily confused! :-)
Shortening the antenna gives a similar graph to shifting the frequency lower on the same length antenna. He's not saying short antennas are for lower frequency.
Yes Ming, you're right. Although I must admit that I had to think twice myself. I hate getting old and after 5 days trying to make this look simple I have to say that I risked confusing myself ;-)
Reminds me of what I got to do early on in RF class with designing an RF front end using lumped element L matching networks for narrow bands. Great video otherwise!
Hey Bruce! That chip the antennas are wired to is a Skyworks SE2431L-r, which is a 2.4ghz RF front end who's job is (basically) to handle antenna switching in diversity applications and and translate it into data the CC2500 Rf tranciever can understand, or vice versa. The datasheet for the front end can be found here: www.skyworksinc.com/uploads/documents/SE2431L_202410K.pdf
but what you really want to look at is probably this one:
www.ti.com/lit/an/swra439/swra439.pdf
At section 4 in that datasheet, they describe some circuit design considerations for power decoupling (and other buzzwords), but particularly around figure 7 where they show the antenna output circuitry, with capacitors in parallel, and the inductors (L8 and L10) in series with them.
Note that the difference in these antenna circuits is meant to describe antenna 2 being a PCB antenna, and antenna 1 going to an SMA connector. (section 4.4)
More light reading:
www.ti.com/lit/an/swra161b/swra161b.pdf
I love "Mr. Wizard" (watched when we where kids) now he's back!
Thats mr wizard hes aged well lol
Your videos are gold. Thank you for sharing your experience and spending time to explain such things.
Bruce, thanks for the explanation of the xmttr/rcvr concept...! I can sympathize with your reported confusion, since it has been quite a few years since I studied RF in school...! This is a nice (and concise) refresher!!!
stingerswarm and steele did a video a while back (rotor riot?) about this, they did both long and short leads in range test mode and the shorter one failsafed way fast (in an open field). then they trimmed the short ones shielding back to match the long and the range more than doubled
Some thoughts:
1.The new batch of XSR comes with 32mm antenna
2. Maybe the maximum power transfer theorem/Stanting wave ratio can help to understand the concept and overheating of transmitters.
By the way, good video as always Mr. Simpson!
More likely those inductors on that Rx are there to protect the frontend, when the antenna get lopped off by a prop, as happens fairly often on race/freestyle quads.
BTW, right up front, it's incorrect to state that only the active element beyond the shield ground is relevant to the antenna tuning. Unless there's a balun between the ground and active element, the entire shield ground resonates and *is* part of the antenna. Because the ground is a different length than the active element, it means the antenna is not a 1/4 wave whip, nor a dipole.
A true 1/4 wave whip (such as you'd find on many handheld transmitters) starts right at the antenna connector, and has no ground element at all.
It is possible that FrSky did some VSWR tuning and discovered that looking at the antenna as a whole, they got better resonance if the active element is a bit shorter or longer than 1/4 wave.
BTW, if you look at UHF LRS Rx antennas which are usually one of
1. "Sander style" 1/4 wave whip (no ground element),
2. a simple coax extension with 1/4 wave active element, or
3. a 1/2 wave dipole (even just the extension with 1/4 wave ground element soldered onto the side below the active element), the dipole blows the other two away. Usually we're talking easily double the range in virtually any orientation. Sander style whips sometimes work, but I think it's because the Rx board itself is used as the ground plane. The coax with 1/4 wave active element is generally quite terrible. I think 2.4ghz Rxs use em mostly due to convenience and ultra low cost, and virtually any other option would be superior. Note that FrSky does make proper dipoles as well.
I've been watching Joshua Bardwell's video too and in my humble opinion, he is the witch doctor and you are the neurosurgeon.
This video is great! But.... it just makes the question we're all asking make more sense. We need part 2!! Please tell us the answer!!! :D
No mention of characteristic impedance of antenna, coax and RF chip Bruce.A good match of all three is important for maximum power transmission.
That is some gnarly looking solder.
Given that there's no practical commercial advantage in having slightly different lengths for these antennas you have to conclude that it's either due to poor manufacturing quality control (very unlikely) or simply the result of testing and tuning for the particular design characteristics of each receiver.
Thanks for the video, very cool! It will (might) be interesting to see the "physical" tests because I know I get less range on some receivers vs others, even though they are described as "full" range. I "think" it may be related to antennas, but not sure.
I remember when I first started FPV and you came out with a "FPV made easy (part 1) video". That was the video that got me started watching your channel. So, I waited, and waited , and waited... but no part 2. My point is, I hope you continue this subject as I think it will be relevant (at least as of now) and interesting.
I've wondered about this topic in the past and really appreciate your explanation. I saw your previous video a year or two ago about calculating the antenna length, and noticed at the time that my FrSky antennas were actually a little longer than you calculated at 33mm. I considered shortening them at the time but wasn't sure if FrSky was doing something a little differently, and so stuck with the original length. I look forward to part 2 as it sounds like you might test it.
I have a potential idea for a new topic for a video: How dangerous is it to be shocked by a typical high-performance, 4 cell Lipo battery? When I have the occasional exceptionally bad crash that really smashes up my quad, I like to examine it before I pick it up to determine if I've damaged the battery wires.
What might happen if I accidentally touch the exposed wires with one hand when picking it up? And in a worse case scenario, as unlikely as it might be, what if I grabbed the + wire with one hand and the - with the other, so the current would flow through my chest, (which I've heard is a particularly dangerous scenario with AC power)? Does the high amperage of these batteries make them more dangerous?
Bruce, at 2.4GHz frequencies, will the traces on the circuit board not add inductance and/or capacitance to the circuits, depending upon their length and proximity to the ground traces. I think a lot is made of that in the design of digital circuitry, like motherboards and the internal circuitry of CPU's.
Extremely informative and well presented video - great work Bruce
Thanks for putting in the hard work
Looking forward to Part 2
A bit tangential to this vid but I've just seen Josh' Bardwells video on the Trannis antenna mod, and he's pointed out potential issues with over-torquing and in some cases the *seriously* limited max insertion count of SMA connectors.
It might be worth mentioning in your next vid that you should only remove antennas if *absolutely nessercery* and avoid over-tightening them.
In the ye olde days of steam radio we were able to tune out the reactance in a transmission line / antenna with the use of a Pi or T network in front of the valve power amplifier. Today’s solid state RF devices don’t have the same range of operation as a Valve/Tube/Bottle and so feedline impedance and antenna length are more critical to ensure efficiency on the intended frequency of operation and safeguard the RF device from destroying itself.
Thanks for sharing your curiosities Bruce. I wonder if you might re-visit the R9 long-range system in your investigation?
Thank you Bruce for the masterclass, I learned manything today.
"I'm Bruce Campbell, and you're going to learn something today!" Lol! Great lesson on my favorite brand of receivers! ...you should have played Joshua's creepy piano music at the end!
I am a cable tv technician and i learned something new today :)
Really looking forward to the follow-up video
This is why bruce is and stays my RC mentor. JB would only read the manual to you and might sell you one of his crappy FCs or two
Ok Alex....time to chime in on this :)
layman's terms, keep the lessons coming.
The question to ask is 'Why?'. Surely cutting to the correct length is significantly more cost effective than implimenting work around circuit design and manufacture?
CyclopsFPV only guys who are doing it knows. Like, what if circuit traces have some impedance or if speed of light in antenna is not the same as in vacuum or ... God knows what.
like bruce said this is voodoo black magic stuff really. RF circuitry has to take into account parasitic capacitances which arise from all sorts of places (like just the traces on the PCB).
You may have noticed that the traces on the PCB for the ground connection to the antenna weren't completely straight, and that itself may be to reduce capacitive coupling between the antenna line and the ground plane. There are tons of circuit considerations you have to take into account when designing an RF circuit, even as an Electronics Engineer I don't understand half it. There are industry professionals who have done it for decades who don't completely understand why some things work, just that they do. There are a very handful of people around the world who actually know what they're doing when it comes to RF circuitry.
It's like tuning a xylophone. You calculate what the length and thickness of the bells should be, cut them to length, install them and find they're not quite on pitch; so sand or notch or weight them until they are spot on. When tuning antenna circuits, you do so in an RF screen room, and make the final adjustments based on measurements from a spectrum analyzer or vector network analyser. Not in this case, but when you're dealing with interchangeable antennas or rigid antennas, you can't change the antenna or standard 50Z termination, and it's too late to trim the board, so you want to leave yourself an component that you can quickly swap out in the screen room, and change on the BOM for the determined value when you release to production.
Everyone seems to be assuming you tune the circuit to the antenna, but it seems to me it's probably the other way around as changing a circuit is much harder than cutting an antenna
4 YEARS ON, IS PART 2 READY YET PLEASE? :)
Some time ago I replaced anntennas on my XSR. I acually soldered uFL connectors to XSR and used some cheap Chinees antennas which have 1/4 wave length antenna. I have done tests on our model airfield and in both tests I reached almost 900m. My feeling from the test is that with modded antennas I had slightly better signal. But this was completly unscientific test so I am looking forward to yours.
Well said lots to take in but you make it as easy as can be for your average Joe thanks as always
great informative video! Did you do the part 2 of it ? I cant find part 2
Broken Tarannis antenna.. Hi Bruce... Great video keep it up please..just coming to the basics for many of us is a trauma you make it a little bit understandable. My Tarranis antenna just broke and I puzzled as to why it was 'sodered' to the board...why not just put mini sma connector on the RF board so as to just unclip and replace ....or just put a regular sma to unscrew & replace. Now I see it all changes the effecting of the whole set up.... Hmmmm will the type of 'sodder'/ size of blob of said 'sodder' make a difference... Have flown/ range tested and all seems ok...
Ps UK is about to legislate re: drones...first reading seems reasonable... Thanks again
One of the factors that you are not considering is the parasitic Capacitance of other metals (including other antennas) that will shift the center of the resonance for the antenna, especially for a simple monopole.In addition, there is no counterpoise (ground plane), which means that the image plane is back against the shield of the coax.If you compare this to a 5/8 wavelength with a counterpoise, then it has a better radiation pattern than these cheap simple wire antennas. It would also be better to create a simple dipole with an associated counterpoise than these antennas.On other factor that you are not considering is polarization; with simple wire monopoles they are vertical or horizontal. For FPV the better antennas have a Circular polarization - Right Hand (RHCP) or Left Hand (LHCP). Again these are most effective with a reference plane.
Also, both coax and PCBs have capacitance and inductance that will affect the impedance of the circuit.
AT 12:00 , when you shorten the antenna you DON'T lower the operating frequency , you increase it , wavelenght is opositly proportional to the frequancy .
More antenna theory please...very informative and well explained :)
Hi Bruce, I really like the explanation. So thanks.
When you tackle the part 2, can you also maybe explain/comment if the length of the coax between the Rx (or Tx) and the active element matters. I think I understand that there is resistance in the wire, but does the coax length also have something to do with tuning?
That was a very good video. Waiting for the second part!
I appreciate this explanation. However, I don't feel the need to slam other TH-cam "personalities" in the process of commending you, lol. While it might be highly simplified, works good enough to get the principle across to folks like me. So I think it is safe to say this video is a success! Thanks.
looking for part 2!!!! great job
Hi Bruce,
I have been looking for ages for a decent explanation of antenna theory. All you posts on the subject have been extremely helpful.
May I ask you to clarify something else on a related subject? What is the required length of RG316 coax for the Taranis antenna mod, when used with a 4dBi Omni-directional antenna?
I watched the post by IBCrazy on the Taranis antenna mod, but there was a couple of instances where I could not follow his logic.
As I understand it, we need to align the coax length to the wavelength for the required frequency to negate the effects of impedance and thereby avoid internal reflections \potential signal attenuation. In his video, Alex states the coax cable length is to the base the SMA adaptor not to the entire length (SMA housing + coax) . I would have thought the SMA housing would also influence the wavelength calculations. Would you include the housing or not in your calculation?
Secondly , his calculation for determining the length of cable is based on: (∁*VF)/Hz , where ∁ = speed of light (m/s), Vf= Velocity factor of coax material and Hz = frequency (cycles/s)
So for a 2.4Ghz signal this should equate to 0.087m for a full wavelength, when using RG316 cable with a VF of 69.5%. Again Alex calculations are different. Obviously he knows far more than me, but I would very much appreciate your thoughts on the subject.
To that end, please can you give us your definitive guide to the Taranis antenna MoD?
Keep up the brilliant work!
Cheers
Nick
May we have Part 2, Bruce? Please? Merry Christmas!
Simple answer... There are parasitic (unintentional) capacitances and inductances on the receiver PCB, solder joints, U.FL connector and coax cable also that they are trying to compensate for by altering antenna length slightly from design to design (and perhaps from unit to unit to a lesser degree depending on their production test capabilities).
Are we allowed to say this is terrible anyway? Great video Bruce.
If f increments along the X axis from left to right, then I believe the too short and too long antennas were drawn on the wrong side of the resonant frequency.
Outcome is the same, though.
The critical frequency has units of Hz, but wavelength has units of m. How did you go from Hz to m? Did I miss a step? I enjoy and appreciate your videos. Please keep it up.
I love and understand your Videos. You learned me the function of a P.I.D. System. Thank you so much!
The circuit temperature would be the best indicator of reflected power. Adjust antenna length for lowest temperature. FrSky probably knows this. At home, adjust antenna length for best range/performance. That's what you want anyway.
18:07 Why are there dots etched on the Antenna lines from the Chip to the antenna connectors on that PCB ??
18:20 Why would they put extra parts (inductors) on this PCB when all they had to do is make the antennas a tiny bit longer ??
Will be waiting for the outdoor range test results.
I think the dots are just blind vias (vias are connecting holes to other PCB layer; blind vias are to internal layers so you don't see the holes), to make a low impedance path to ground on the other side of the board to make what's called a stripline (impedance matched transmission line etched on a PCB) or waveguide. If you look at RF PCBs you usually see this "stiching" along transmission line to antennas or antenna connectors. Example: iot-bits.com/wp-content/uploads/2017/10/PCB-trace-antenna-layout-via-stitching-technique-1024x885.png
Bruce, despite saying that you are not sponsored by any company, it does seem like you are sending subtle messages. You’re wearing a lovely crisp lumenier t-shirt that makes you look like a lumenier radio engineer whilst at the same time wiping your board/arse with the hobby king t-shirt :)
Hahaha "subtle messages."
Bruce, can you specifically do some tests with the XM+? I've found it gets the best range of all FrSky RX other than the L9R and the XM+ has had antenna changes through out its revisions. Strangely, the first revision of the XM+ seems to be the best performing, with the latest with the ultra short antennas being the worst.
Adam-g1 I cant figure this out, stock lengths are different so how do you known if replacement ufl is worse, ugh
Hi, I have an E-Foil question , Does it matter how long the sheilded /braided part of a 2.4 antenna is before the 31mm bare section ? I have an efoild surf board , the reciever is on the ESC at the rear ,I need to extend the antenna towards the front of the board because when starting off with the rear of the board submerged . also the hatch at the rear is carbon fibre wheras the board is not . ( am I correct in thinking this could block the signal to a degree?
My plan was to use a 500 mm extention cable with micro UFL on the reciver end and remove the shielding for the last 31mm the other end. Will this work and not damage the reciever please?
Great video! Very interesting. Thanks for putting your time into this subject.
Hey Bruce! Can you do a video on putting a dummy load on a transmitter, both Radio and FPV?
Strictly speaking the signal is not lost, but rather shifted into the infra-red region of the em spectrum.
As mentioned previously in the comments, I wonder if velocity factor is part of the issue. Are they both the same coax and is the velocity factor the same. Are the capacitance of the 2 antennas the same.
Thank you! Looking forward to part 2!
19:40 inductors are not used to implement static discharge protection, it's not rational because it will complicate RF front end matching. In industry common practice is to use very low parasitic capacitance TVS diodes (TVS - transient voltage suppressor). You took basic examples of a simple antenna matching with series inductor/capacitor and made a mystery why inductor is used in parallel. Looking from a Smith chart you can use different approaches to match impedance.
Once again, another excellent technical video with some humor mixed in. SIR: could you maybe look into the possible harmful effects from a 1 watt rf transmitter at 5.8 ghz? I am not qualified to address this but I think you may be. Perhaps we should be telling folks to keep active microwave radiators away from their eyes. As power gets higher and higher, we are starting to make little microwave ovens!
It is time someone addressed this and everyone, including me, trusts your opinion.