Page 2 of 2 FirstFirst 12
Results 21 to 24 of 24

Thread: Patch panel?

  1. #21

    Join Date
    May 2020
    Location
    Central Ohio
    Posts
    59

    Default

    Another question, this one about HF antennas that can grow and shrink.

    This is just the rudiments of an idea; if it were reduced to practice, it'd probably look a lot different.

    But imagine a length of bare copper wire, with a single hanging loop in the middle. Not too ambitious; say six feet of wire total, with a loop maybe a foot in diameter or so, which would leave about a foot and a half on each side of the loop at the top.

    So squeeze the top of the loop tightly together, so that the wire makes contact with itself there. You've shorted out the loop. If you were to use the wire as an antenna now, the transmitter would think it was about three feet long. (Yes, it would probably burn you too--don't bother me with details.)

    Now stretch the wire and shrink the loop, still keeping the wire in contact with itself at the top of the loop. Now the loop is only six inches in diameter, with a bit over two feet of wire on each side of it. The loop is still shorted out, but there's less wire in it. Now, if it were an antenna, the transmitter would see four and a half feet, instead of three.

    Or you could stretch the loop and shrink the wire, again keeping the loop shorted out, and wind up with four feet of wire in the loop and only two feet in the antenna.

    Obviously there are a number of engineering problems with this, such as how to keep good contact at the top of the loop without the wire wearing out, and how to drive the mechanism, and how to protect it from the weather, and so on; but is there some fundamental reason why this would be a stupid idea? Something maybe about RF and skin effect? (It's probably valid to assume that the circumference of the loop is small compared to the operating wavelength.)

    Or is this maybe already a tremendously common idea, so common that I've already seen it hundreds of times without recognizing it?

  2. #22

    Default

    Quote Originally Posted by dnwiebe View Post
    So I'm in the process of building a little box that will give me access to my feedline. I've enclosed a schematic.

    In case the attachment doesn't come through, it's a 100:1 voltage divider across the feedline, 10 kilohms in all, that should allow the scope to see 1% of the transmitted power.

    That's pretty simple, from an audio standpoint, but are there gremlins hiding somewhere in the shadows for RF? Should I use a 100K divider instead of 10K?
    Be sure the resistors can handle the rf voltage. Also, some newer digital scopes have a max voltage whereas some old scopes have a max current. Make sure you dont fry the front end of your scope! There should also be a ground connection in there for the scope as some scope inputs are isolated from the power source ground. If there is a potential difference between the coax shield and the scope case, the voltage between the scope probe shield and scope case can also cook the input.

    Most of the time when one wishes to connect an oscilloscope to a feed line carrying power, a capacitive device is used that resembles a coax tee but with the scope connection only capacitively coupling with the rf path between the other 2 ports. Check out https://youtu.be/0Kk_N_TpDeo

  3. #23

    Join Date
    May 2020
    Location
    Central Ohio
    Posts
    59

    Default

    Thanks for the video link; it was fascinating.

    I had an interesting time today. I built a 1:49 balun kit for an 80m EFHW that I want to put up, because my OCFD only transmits down to 40m. Iím using 18 AWG PTFE-insulated wire, and running it from my attic window through some trees in the yard.

    I wanted it centered at 3.8MHz, but I couldnít find any decent references for velocity factor, so I cut it a little long (about 112 feet) and figured Iíd trim it.

    I took my little nanoVNA to the window, connected it directly to the balun, and ran an SWR sweep. I found a surprisingly high-Q parabolic-looking SWR curve centered at about 3.705MHz. I calculated that cutting off a little more than two feet would raise it to 3.8MHz, so I cut off about a foot and a half and measured again, getting a center at 3.796. I figured that was close enough, so I hung the balun outside the window, added a feedline back to the radio, and...wow.

    Running another SWR sweep at the radio end of the feedline showed a high-Q curve centered at 4.2MHz.

    Whaaat?

    So I cogitated a bit, and hereís what I came up with. An EFHW always makes its own counterpoise, if one isnít provided, and when I was testing the antenna at the window, there was no connection to ground, so the only counterpoise it could have made was inside the balun itself. The radio shack, though, is pretty well grounded, so the counterpoise would have been much better. Would that affect the SWR center? It seems plausible, although I canít identify the mechanism by which it would.

    It looks like what I should do now is splice fifteen or twenty feet of wire back onto the antenna and then trim it down again, this time sweeping always from the radio end of the feedline. Yes? Will the length of the feedline affect the SWR center, do you think?

    Oh: and is there a way I can reduce the Q of the antenna? When I measured it at the window it was 1:1.15 at 3.796, but at 3.6 and 4.0 it was already past 1:5.

  4. #24

    Join Date
    May 2020
    Location
    Central Ohio
    Posts
    59

    Default

    Well, I have some more news to report.

    I spliced in about thirty feet of antenna wire, checked the SWR sweep on the VNA connected to the radio end of the feedline this time, and cut out antenna wire until it centered at 3.803MHz, which was good enough for me.

    However, I'm noticing that aside from the annoyingly high Q of the antenna (I read that you can increase Q by shortening the antenna and adding loading coils, but I haven't found a way to _decrease_ Q if you don't have any loading coils to remove), I have another problem: now that the antenna wire is longer, minimum SWR is about 1:1.96 rather than 1:1.15. Given the antenna's high Q, this makes my operating window really narrow.

    So I'm thinking what may have happened is that since EFHWs are known to have high impedance at their ends, increasing the length of the antenna may have raised that impedance, so that a 1:49 transformer (three-turn primary, 21-turn secondary) produces an impedance that's still a factor of two too low. I'm thinking the next thing to do is to re-wind the transformer to 3/30 turns, for a 1:100 ratio, which is about twice the 1:49 ratio.

    Let me know if I'm doing something dumb...

Tags for this Thread

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •