Ham Radio - Do you really need the center conductor in a coax magloop?
Vložit
- čas přidán 7. 09. 2024
- Testing if the center conductor in the outer loop is needed to improve efficiency in a coax magloop antenna.
Magloop build video:
• Ham Radio Building the...
Web link to discussion on currents in coax transmission line:
gm3sek.com/202...
If you like my videos, please subscribe here:
www.youtube.com...
If you want me to keep making more videos like this, consider becoming my patron at Patreon: / kb9rlw
Discuss this and other videos on Facebook:
/ kb9rlw
One of the guys I work with has been refining a loop over the past 2 years. He is using a chunk of LDF wave guide in a 1 meter loop. He built an augur slightly larger than the inner conductor to remove it. He then fashioned the gap end of the loop into a trombone capacitor insulated with teflon to prevent arc over. He uses a stepper motor and dynema chord move the trombone in and out. The loop tunes from 13 Mhz to 31 Mhz with very low SWR. The LDF allows him to run 100 Watts to the loop. The teflon keeps it from arcing over as the voltages are huge. de KC0KP, Craig
Kevin at his best, in his natural element. Similar material is why I subscribed years ago. Good show, friend...🇺🇸 👍☕
Glad to see you are back. You look and sound much better than your last video.
Nice to see you back in action producing great content Kevin! One comment on this one tho'; there's one configuration you didn't test, and that's using /only/ the centre conductor. I spoke to Paul OM0ET about his mag-loops, and he tells me he's using /only/ the centre conductor on his...
I don't think it would matter either way. Using only the center, the braid would be capacitively and magnetically coupled and act kind of like a parasitic element, the same as the center does when you just use the braid.
I believe OM0ET splits the shield at the midpoint opposite the tuning capacitor. In other words, the shield is not a continuous conductor.
@@DaDitDa om0et loop does not work well!! keep your hands away!
A very informative discussion, which I found of interest as I have been playing around with coax loops for some time. Thanks for sharing.
👍Thanks for video Kevin. An interesting test. I wondered if the bandwidth increased with the inner connected simply because there are more paths for the RF current. Though, the wider bandwidth may not be desirable on a STL, as it tends to indicate lower Q. Certainly was interesting to see the effect of the inner ends being connected.
There were too many variables. I had disassembled and reassembled the loop between tests. Also the SWR was slightly higher so we were measuring a shallower portion of the dip. That's why I said it was within a margin of error
Thank you for the video Kevin.
Thanks for this as I had wondered, but it just made sense to connect the inner as well.
G4GHB.
I have a coaxial mag loop that uses the center conductor to make it a two turn loop for 40 and down. I wonder if you could experiment with this idea, and show as you have here what the effects are and how you see it working.
The same loop I have has no separate transmitting loop, but instead a toroidal coupling loop inside the capacitor tuning box. It’s an interesting little loop!
WV1T
Another interesting experiment would be to completely remove the inner conductor. It should give the same results as a simple copper pipe. Shorting both ends of the coax makes the loop an inductor in parallel to itself, reducing the overall inductance. On the other hand, using the center conductor with one end open should act as a capacitor in parallel to the tuning capacitor, increasing the capacitance. This could help make the tuning capacitor smaller… Just a thought. I am not an RF engineer.
I did try connecting the center at only one end, and it did not shift the tuning of the loop more than a few hundred Hertz. So it was not acting like a capacitor in parallel with the tuning capacitor.
Great video, many thanks!! Robert K5TPC
Thank you for this interesting video Kevin. I have the Chameleon F loop 3.0 and I believe that they have the center conductor connected to the shield. I'll have to check that. The wider bandwidth that you found with center conductor connected will make tuning the peak slightly easier. That is a plus for my 75 year old hands.
Well as I said, I had disassembled and reassembled the loop between those tests, so that wider bandwidth might have been within a margin of error. I know with the shield the SWR was slightly higher so we were measuring a shallower portion of the dip. So I really don't think there was much of a difference in reality.
very interesting bought a loop off aliexpress .. for shortwave and saw the center of the coax loop was not connected and wonder why your video was the answer great video
"part of the antenna" is an ill defined phrase. In addition to capacitive coupling of the center conductor, there is also magnetic coupling. When the center conductor is not connected to the shield, you can look at the coax as an air core transformer. The source of signal, transmit or receive, is applied to the shield, and the center conductor becomes the secondary winding. An open secondary has very low current (driving only the center to shield capacitance. But when you short the secondary, significant current (as scaled to the signal power) flows, resistive loss in the center conductor becomes significant, and the system is significantly detuned. (This is related to why shielded receiving loops must have a gap in the shield.)
As to whether it is better to include or exclude the center conductor in parallel with the shield, consider the following: While the parallel combination will have lower DC resistance, we don't care about DC. Skin effect causes part of the conductor to be excluded from current flow, that is, the cross sectional area through which the RF current flows is decreased, so the resistance experienced is increased. Resistive loss is a significant factor in a magnetic loop because the circulating currents are high. Since RF does not flow in the center bulk of the center conductor, but only to roughly the same skin depth as the inner and outer surfaces of the shield, its RF resistance will be higher by the ratio of the shield diameter to the center conductor diameter. (it's not actually linear, but a good approximation.) While I would normally suppose that e RF current would divide according to the resistances of the available paths, making the addition of the center a slight reduction in total resistance, and therefore a reduction in losses, your measurement of narrower bandwidth, thus higher Q, and lower loss, without the center conductor, leads me to speculate that perhaps magnetic coupling is leading to higher than suspected RF current in the center conductor.
Note, too, for a transmitting loop, that heat generated by the resistive losses is much harder to remove from the center conductor, with it's small radiating surface all wrapped up nice and cozy in layers of insulation and other conductors heating back at it. I therefore claim that connecting the center conductor significantly reduces the power needed to melt the COAX.
I have used a length of coax for the tuning capacitor for a loop. It will handle a full 100 Watts of power. That was for the length of use which was a weekend camping trip. Mine was fed with the usual coupling loop. It is necessary to carry with you several extra lengths of coax in order to QSY. This was used on 40M. Had excellent results. I cannot tell you if the center conductor was connected or floating
I thought about using a length of coax plus a smaller variable capacitor with larger plate spacing as a fine tuning. The coax would be course tuning, with different lengths for different bands.
I've seen mentioned, although I don't remember where, that if the centre conductor is used as the loop and the screen is grounded at one end only, the loop will pick up less random electrical noise. Might be worth a try.
I doubt that would work. If you grounded the screen, you would kill the resonance of the loop. It would be like hooking up a capacitor to it and grounding one end of the capacitor, you provide an RF path to ground. We already know the center conductors capacitively coupled to the braid.
There is a similar split braid shielded loop for receiving only. That might be what you're thinking of. Definitely won't work for transmitting, since that power goes right to ground.
Great video, Kevin. I love these practical experiments. 73, K7KS
Learn something new everyday. Thanks for that.
Very interesting! I've seen designs where they connect the coax center conductor into series with the outer conductor making a concentric double loop, it would be interesting to test that config!
Glad you are back.
One thing is that the center conductor adds surface area to the loop. This may be related to the improved bandwidth with the center conductor connected. I think that the bandwidth improvement is significant. In that connecting the center conductor is simple to do I would opt for that. IMHO
73
Except that I have not seen consistent improvement in bandwidth in testing.
The bandwidth difference that I saw here was partially due to the SWR change as well. When the SWR came up slightly, we were measuring over a shallower portion of the dip in measuring from the 2:1 points. So part of that difference in bandwidth was an illusion.
Duplicate the test yourself, you might be surprised.
Excellent work. Thank you.
Thanks for this interesting video. One guy OMOET who manufactures a loop claims to only use the inner for the loop. He says the shield helps lesson the capacitive effect of the body whilst adjusting the tuning.
forget om0et
I doubt it. Since the braid is capacitively end magnetically coupled to the center, it becomes something like a parasitic element. The same thing that we saw with the center. So anything that affects the braid would also affect the center.
The most effective way to reduce hand capacitance affecting tuning, is to use a ganged capacitor and wire your loop to the static plates as we did in this loop build. With the rotor plates acting like two capacitors in series, hand capacitance hardly affects tuning at all.
Kevin. Nice. Now, think of the braid as a single turn inductor. The center conductor is also a one turn inductor, tightly coupled to the shield turn. So you have a transformer with primary shield and secondary center conductor. Short the secondary. It cancels the primary. Ooops. Dead loop.
My ears pricked up (around thirteen minutes?) when you shorted the centres together- of course all you actually did was short out the tuning capacitor via a capacitance and so the tank circuit could no longer resonate at any useable QRG, but you knew that.
I have never connected the centre conductor (I used Heliax for three loops.) For one loop, I knocked up a junker example where I arranged for a double ganged on/off/on switch for the secondary loop feed AT ONE END ONLY, and using toroidal coupling simultaneously changed the number of turns on the toroid. This was proof of the series capacitance concept. It worked allowing the loop to "tune" closer to its self resonant frequency, but it's a rubbish way to build a loop. It's fine on receive though! I was prompted to try that out after seeing a (no longer available) uncharacteristically inexpensive loop from the USA which had such a switch. I puzzled over what the switch could be doing inside the box to alter the tuned frequency- I reckon I guessed correctly.
Current can only flow inside the coax if it has something to push against in the form of a second conductor. The coaxial nature of the loop facilitates this but being shorted at each end, how much current actually flows? Answers on a postcard! it simply cannot radiate anyway since it is "shielded" (ha!) The current won't couple into the centre conductor and back again to radiate off the outside- that's just silly. It's a pointless exercise just like trying to screen an earth lead.
What can we take from all this? Just as you said, it is pointless to connect the centre conductor with the possible exception of switching it to one end to benefit from overall lower series tuning capacitance, but using ANY type of switch to accomplish that is ridiculous when you really really should be trying to reduce resistance to the considerable circulating current flow in the secondary loop. Better to design the loop to tune higher by cutting it shorter and lose coverage at the lower end where the bandwidth becomes a lot narrower, and then build another loop!
Good video Kevin!
The shorted center conductor brings the resonance to a different frequency because the center core in this case changes the capacitance of the C. See the VNA. (love your videos friend)
Nope, scanned the entire range with the VNA there was no point of resonance. I demonstrated that in the video by rotating the capacitor over its entire range to show there was no longer a resonant point. It's more like a shorted secondary in a transformer. The load of the shorted turn killed the resonance of the tank circuit.
@@loughkb That makes sense. Thanks for the reply buddy
Hi Kevin. It might be interesting to connect the center of one to the shield of the other (on both sides) so that the center opposes the flow on the inside of the shield. Good discussion re: skin effect. Glad to see you're doing better. 73 OM
I think that would just cancel itself out.
@@loughkb Yes, it should cancel out the fields between the inner shield and the center conductor, but place all the RF on the outside surface of the shield where it can radiate. I see it as a trifilar winding, or a "buck boost" configuration. I also haven't had my coffee yet, so I'm probably completely wrong. I figured since you already have the clip leads attached, it would be easy to check.
Kevin ! Glad to see you coming OUT of the covid FOG.
Interesting results. Can you do this test on HIGHER frequencies? I 'm just curious, and don't have a lab anymore.🤔🤥🤔
I'd have to build a loop that could handle higher frequencies. This test was done in the 10 m band, we'd have to go up to 50 megahertz with a much smaller loop.
*Great information Kevin. Thanks !*
Thanks for the testing
Hi Kevin,
The proof is in the pudding. Nice testing procedure. Always enjoy your videos. Has your long Covid been improving? Stay safe. 73 WJ3U
Very very slowly getting better. I still only measure progress month to month..
The proof is not in the pudding. The proof of the pudding is in the eating.
The center is not connected in commercially made loops but you can use it to make the loop 2 turns to get lower frequencies like 80 Meters.
It is shorted in the chameleon loop I own. Some of them do.
HOWEVER, the best qualities of a loop are when constructing a "shielded" loop where as the shield is split in the center and cross connected at the break with the center conductor. This eliminates noise and causes the loop to purely respond only to the magnetic properties if the E & H fields. Just my humble opinion, 73 KV4WM
If that were true, everybody would be building them that way. Transmitting loops that is.
Interesting info Kevin. I also wonder how much a shielded cable that has a aluminum or copper rap under the mesh will do in your tests. I have a mag loop I purchased about 4 years ago that adds the lower bands by switching in the center wire in series with the outer shield. 80 & 60 meters switched in and 40 to 12 in the switched out position. It also uses a toroid around lead to the shield to inject the signal into the antenna. Seems to work reasonably well. David K6DCH
The coax is parallel conductors so it's naturally a capacitor. When you hook the shield to the center, you get current flowing (yes, on the surface(s)) of both conductors and so you have current flowing through the coax capacitor, which is now in series (i.e. less total capacitance) with your tuning capacitor. When you disconnect the center it's still capacitively coupled, but as an open circuit no current will flow and the effect of that coupling is minimal. When you short the center to itself you've created a path for current to flow but only to be dissipated as heat because of the resistance of that inner loop. You've effectively capacitively coupled your loop (which is only the shield) to a dummy load that kills all the RF you received.
Yes, exactly the point I was making. In response to a comment on the build video indicating that the center is isolated and not part of the antenna.
Yes the centre conductor is capacitively coupled to the braid, BUT .... only to the inside surface of the braid. I am guessing that it may be acting as a loosely coupled shorted turn, dampening the loop, thus increasing the bandwidth?
No, more like a shorted secondary winding in a transformer. It killed the "ringing" of the resonant tank circuit.
How about connecting only the center conductor. Then compare it with only the outer conductor configuration. I believe Paul, OM0ET, builds his coax loops using only the center conductor; and they appear to function well. (Edit: After reviewing several of Paul's CZcams videos, I realized that he uses a center-split outer shield design for his center conductor connected QRP loops. Theoretically, this design improves SNR on receive, but doubtful that it's as efficient as an outer shield connected loop radiator -- which appears to be confirmed by the relatively wide SWR for his loops.)
Hello Kevin. I've done the same test as you, and I don't connect the center conductor on my loops either. 73 ZS2VR
Very nice analysis Kevin, but you didn't do one test I would have liked to see, relative transmit field intensity readings
Very interesting.
If you add capacitance in series with your tuning capacitor you will raise the frequency of the loop to another band.
Or just use a shorter coax for a smaller outer loop.
It already goes up into 10 meters.. To get to the next amateur band, 6 meters, we'd have to go quite a ways up in frequency. No way to get that far with a 2ft. diameter loop.
@@loughkb I see what you are getting at, and I was not thinking in context to your loop size. I recently built one of 3/4 copper pipe that mathematically, should have done up to 10, but tunes from 40-15, it is a 2.6-foot loop with a large air variable out of an old full power tuner. I think I need to put more capacitance in to make it up to 10meters. So, I was transposing my thoughts onto your projects. I have only been a HAM in my retirement from the electronics field, so I look at guys like you as my Elmers. Thank you for putting your work out there for us all to see and try for our selves. 73's🎙KD9OAM🎧
I want results for a 3/4" copper water pipe magloop, empty, and then lined with RG8/U run inside, with all conductors shorted at the terminations. My thoughts are the coax braid may not be as impervious as a solid tube, in your experiment today. Thanks for the information provided so far.
Sounds like a good experiment please let us know what your results are when you do it.
@@loughkb Wish I had the space, I am so busy renovating British "Universal AVO meter" Multimeters, the first in the world to include "Amps Volts and Ohms" measurements in a single case.
Is narrower bandwidth better ? I find having to retune loops on QSY quite annoying. Maybe it helps reduce broadband noise fed into the receiver.
The narrower bandwidth indicates a higher Q factor and a higher efficiency. With magloops, the narrower the bandwidth, the better they radiate.
@@loughkb Aha ! That sounds like an advantage indeed. Thanks for replying and best 73 !
The effect of connecting the center conductor likely also depends on the frequency as the skin effect only dominates at higher frequencies.
Sure, if by "high frequency" you mean anything above 10 kHz. Skin effect at 1 MHz is 65 micrometers. That's 2 thousandths of an inch in freedom units.
Thanks!
Dear Kevin,
It is a pleasure to see you again after a period of time. very interesting experiment . I had a thought that I wanted to share. I have been contemplating the possibility of reducing the Q factor of the antenna by shorting both the inner and outer conductor of the loop. This would result in a slightly broader bandwidth and slightly lower efficiency. On the other hand, by utilizing only the braid, we could achieve slightly less bandwidth and consequently higher Q factor and efficiency. is it correct?
Thank you and regards,
That was the comparison I did in this video.
@@loughkb 👍 perfect
How about connecting the inner to the out at just one end? 😁
interesting!! so i wonder what happens if one place a capacitor between the centre core ends?
Try it out and let us know. :-)
Thanks for your videos! A couple of questions about your results. 1) Did you measure the radiation resistance versus the DC resistance with the center shorted to the shield? Since the radiation resistance of a small loop antenna is very low, I would assume that any reduction in the DC resistance would improve the antenna performance. 2) When you grounded the center conductor, you added a distributed 200pF capacitor to your tuned circuit which would probably seriously affect the tuning of the antenna. What is the capacitance range of your tuning capacitor? Thanks.
Number one, I made the same assumptions, but this test shows there really isn't much of a difference.
As to number two, I don't recall off the top of my head the range of capacitor, but it was discussed in both the design and build videos. The build video is linked in the video description. When I shorted the center, it didn't just affect the tuning, it killed the resonance of the tuned circuit the loop and capacitor make. Off camera I had swept it with my VNA and there was no point of resonance at all when the center was shorted.
I have read that increasing the diameter of an inductor reduces its inductance. In the case of an antenna element, it would also reduce its Q, if one does not change its length, according to Stan Gibilisco. So, when the center conductor is shorted to the outer conductor, the diameter of the inductor is increased, and not only so, the two inductors are in parallel, reducing the inductance. After all these words, I am simply asking, can you also measure the inductance for both cases?
Hi kev you missed one thing if you split the braid and use inner core only you get a higher Q factor narrow bandwidth ang electrically quieter less QRN good to see you back
Doubt it would be any different. The isolated braid would just couple to the center magnetically and capacitvely, becoming something like a parasitic element, just as the center does in our test.
But if you have done this test, I would be interested in the data and images of your build.
@@loughkb will get back to you on this
You can't manage what you can't measure. Or, as we say here in The Netherlands: "meten is weten" ("measuring is knowing"). We should always measure and verify our assumptions, and, like my grandpa always said: using good gear is half the work.
What happens if you only use the inner wire? Does the skin effect go trough the capacitor in effect?
Science in doing whats never done before can sometimes be interesting.
Why didnt you retune the cap when you attached the wires? I would also think a large bandwidth would be the goal.
Because I missed the detail until editing.
It is a characteristic of a magnetic loop that the bandwidth becomes narrower the higher the efficiency of the antenna. That's just the way they work.
Dummy loads have very wide bandwiths. They make terrible antennas. It is the narrow bandwith that makes this antenna efficient.
If you would place an insolated wire inside of the tubing of a copper tube loop, would that change the rating of the variable cap needed to tune the loop, or would you be able to tune that loop to another band with the variable cap you are using? After you shorted the center conductor did you check to see it the loop would not tune to a different band?
I swept it with a VNA with the center shorted. There was no point of resonance. It was like a shorted secondary in a transformer, it killed the resonance of the antenna.
Please excuse my poor english. My question is, does the center conductor would act as the feeding loop instead of the usual small loop (Faraday coupling)? I'm testing the MLA of the Wonder Loop type with 1:100 UnUn using 43 material toroid to get 100 times larger inductance for the secondary side of UnUn than that of main loop. This is the way to concentrate RF current to the main loop. If the center conductor would act as the feeding loop, it would be worth to challenge it instead of the Wonder Loop type for me.
That is an interesting idea. Sounds like a good experiment. I might do that.
@@loughkb Thank you for your comment. I think, the center conducter of the coax shows the same impedance of the outer sheath due to the working function as an UnUn.
Usually the length of feeding loop is about 1/5 of the main loop, but the closely linked part is half or less than it of the small feeding loop. Let we assume that the half of the feeding loop is working as a closely coupled area, the ratio becomes the 10:1 (1:1/(5*2)) in length. Thus, it become the 100:1 UnUn. This ratio is in good agreement with my Wonder Loop Experiment. For the case that the whole length of center conducter of the main loop is used as a feeding loop, the ratio of the closely linked length becomes the 1:1, and the resulted impedance of the center conducter becomes the same value of the resonating outer sheath.
This is my speculations. I would appreciate your comment.
@@user-sv4ef3mu5u I tried it, it did not work at all. It completely killed the resonance of the loop
@@loughkb Your result is totally different to may experiment. Outer sheath and center conducter is closely linked and forms the 1:1 UnUn. Thus, the impedance of center conducter become high as same as the LC resonance circuit of the outer sheath and capacitance. So, I connected the center conducter to the 1:100 UnUn and it worked fine. If the main loop length is in the 1/4 to 1/8 wave lengh, 1:49 UnUn is better than 1:100 UnUn. I have to comment that I`m always using the UnUn as isolation trance type (ballanced connection).
Aren’t you severely changing the capacitance when you short the two center conductor ends?
I wouldn't think so. That does not change the gap or the amount of shared surface between the two conductors. But it is like shorting the secondary of a transformer. It kills the antenna. As we saw.
There's no provision on your videos to send "Thanks" with a few dollars. I do that on the other CZcams channels that I watch. I don't do Patreon anymore here but do like to contribute to channels that I find interesting using the "Thanks" feature.
I have PayPal that is registered to my public email which is good on QRZ.com. I didn't know youtube had another angle, I'll have to look for the "thanks" option.
I've looked into the "Super thanks" option on CZcams, I'll consider it, but there are technical issues. I did create a PayPal donation page where viewers can donate a tip if they wish. I'll add a link to my standard description text. Here's the link if you want to have a look. Thanks much.
www.paypal.com/donate/?hosted_button_id=TJDA65P3JDVHJ
Why do you argue that a narrower bandwidth is better? Seems to me the wider BW would be more desireable if the other factors are same (namely SWR)
Narrower bandwidth in a magnetic loop is an indication of a higher Q factor which is an indication of a higher efficiency.
And in a magnetic loop the other factors are not the same if the bandwidth is broader.
Hello OM TU fer video w0asb
Welcome back. Dana VE6KBI