Home |
Search |
Today's Posts |
|
#1
|
|||
|
|||
Coaxial Antenna question
I'm trying to build a decent performing 2M coaxial skirted antenna and have a question about its design for maximum efficiency. I based the "hub" on an SO-239 connector. I soldered my RG8X cable center conductor to the solder pin center conductor and brought the shield braid out in two places. I cut a 19" piece of half inch copper pipe, cut 4 half inch slots on one end, fanned the slotted end out slightly to fit nicely against the SO-239, and slid this pipe over my coax and up to the SO-239. I brought the two braid lengths previously prepared out through two of the rather fat slots, soldered the pipe to the SO-239 and the braids were soldered to the pipe where they came protruded out the slots. I soldered an 18 1/2" brass welding rod to a PL-259 center conductor and screwed that the the SO-239 for my radiator. The copper pipe assy and coax slide nicely down into a length of 3/4" PVC . This makes a very nice break-down package for transportation and portable work. Back to my question: I am measuring about 125 watts forward and 4-5 watts reflected at my desired frequency ( 146.000 ). While this may be acceptable to some, I would like to get the match down to 1:1 SWR. Are there any design deficiencies in my proto-type? Can anyone suggest something I might try to improve the match? Is there a "rule of thumb" regarding the construction of such antennas?... length of coaxial skirt vs. antenna element? I appreciate any feedback on this. Thanks. I suppose I could provide a picture if anyone requires it.... Ed K7AAT |
#2
|
|||
|
|||
Coaxial Antenna question
"Ed" wrote in message . 192.196... I'm trying to build a decent performing 2M coaxial skirted antenna and have a question about its design for maximum efficiency. I based the "hub" on an SO-239 connector. I soldered my RG8X cable center conductor to the solder pin center conductor and brought the shield braid out in two places. I cut a 19" piece of half inch copper pipe, cut 4 half inch slots on one end, fanned the slotted end out slightly to fit nicely against the SO-239, and slid this pipe over my coax and up to the SO-239. I brought the two braid lengths previously prepared out through two of the rather fat slots, soldered the pipe to the SO-239 and the braids were soldered to the pipe where they came protruded out the slots. I soldered an 18 1/2" brass welding rod to a PL-259 center conductor and screwed that the the SO-239 for my radiator. The copper pipe assy and coax slide nicely down into a length of 3/4" PVC . This makes a very nice break-down package for transportation and portable work. Back to my question: I am measuring about 125 watts forward and 4-5 watts reflected at my desired frequency ( 146.000 ). While this may be acceptable to some, I would like to get the match down to 1:1 SWR. Are there any design deficiencies in my proto-type? Can anyone suggest something I might try to improve the match? Is there a "rule of thumb" regarding the construction of such antennas?... length of coaxial skirt vs. antenna element? I appreciate any feedback on this. Thanks. I suppose I could provide a picture if anyone requires it.... Ed K7AAT remember, too low an swr can kill you! To get exactly 50 ohms you may need a fatter copper pipe, or some other change to the geometry. if ti works well as it is i would leave it alone. |
#3
|
|||
|
|||
Coaxial Antenna question
On 02 Nov 2008 00:12:29 GMT, Ed
wrote: I'm trying to build a decent performing 2M coaxial skirted antenna and have a question about its design for maximum efficiency. (...) Back to my question: I am measuring about 125 watts forward and 4-5 watts reflected at my desired frequency ( 146.000 ). While this may be acceptable to some, I would like to get the match down to 1:1 SWR. 125 watts forward with 5 watts back is a VSWR = 1.5:1 The characteristic impedance of a coaxial antenna is about 75 ohms. The best you can do with your present arrangement is therefore about 1.5:1 which is what you're getting. To do any better, you'll need some way to match the 75 ohm antenna to your 50 ohm system. That's usually an odd multiple of 1/4 wave coax section, with an impedance of 61 ohms. Something like: http://www.repeater-builder.com/rbtip/matchingstubs.html I suppose I could provide a picture if anyone requires it.... A JPG is worth 1000 guesses. Also, you might want to do your testing with something less than 125 watts, as accidents and miscalculations might become expensive. -- Jeff Liebermann 150 Felker St #D http://www.LearnByDestroying.com Santa Cruz CA 95060 http://802.11junk.com Skype: JeffLiebermann AE6KS 831-336-2558 |
#4
|
|||
|
|||
Coaxial Antenna question
Ed wrote:
. . . Back to my question: I am measuring about 125 watts forward and 4-5 watts reflected at my desired frequency ( 146.000 ). While this may be acceptable to some, I would like to get the match down to 1:1 SWR. . . . Actually, this is acceptable to just about everyone. And those for whom it's not don't usually have a rational reason for it. Here are some ways to do it: 1. Lower the impedance at resonance by making the top of the antenna into an inverted cone shape, by using fanned out multiple conductors. You'll have to adjust the length of the conductors and possibly the sleeve to maintain the same resonant frequency. 2. Build and use a simple matching network, preferably placed as close to the antenna as possible. Then you can adjust it for a perfect match. 3. Use some other form of impedance transformation such as a stub matching network or transmission line transformer. If you do things right, it'll work the same when you finish as it did when you start. But you'll feel better seeing zero "reflected" power on your meter, and the placebo effect is not to be sneezed at. Roy Lewallen, W7EL |
#5
|
|||
|
|||
Coaxial Antenna question
"Roy Lewallen" wrote:
But you'll feel better seeing zero "reflected" power on your meter, ... _____________ The meter indication will be essentially zero only if the directivity of the device measuring reflected energy is extremely good (as in 60+ dB). The leakage of forward energy into the reflected sample that is present in typical commercial products will lead to (mis)adjusting the load Z to whatever produces zero on the reflected meter. However some amount of reflected energy will be needed to cancel the leakage of the reflected coupler, in order to obtain that "zero" reading. And then the reverse energy is not really zero, and the true load SWR is not really 1:1. RF |
#6
|
|||
|
|||
Coaxial Antenna question
Ed, You might have expected the feedpoint impedance to be around 70 ohms. It will depend on the feedline configuration, because you haven't taken much is the way of measures to decouple the feedline. Your measured fwd and ref indicates VSWR~=1.5 which is consistent with 70 ohms, but you haven't measured 70 ohms. Assuming though that such an antenna should be close to 70+j0 at resonance... If you did want to incorporate an impedance matching system that doesn't compromise the portability you have described, you could try a twelfth wave transformer with 29.3° of 50 ohm coax from the feedpoint, then 29.3° of 75 ohm coax then any length of 50 ohm coax to the transmitter. For example, for 146MHz, that could be 137mm of Belden 9258 (RG8/X) then 139mm of Belden 1189A (RG6/U) then any length of 50 ohm coax to the tx. Owen |
#7
|
|||
|
|||
Coaxial Antenna question
Owen Duffy wrote in
: Ed, You might have expected the feedpoint impedance to be around 70 ohms. It will depend on the feedline configuration, because you haven't taken much is the way of measures to decouple the feedline. Your measured fwd and ref indicates VSWR~=1.5 which is consistent with 70 ohms, but you haven't measured 70 ohms. Assuming though that such an antenna should be close to 70+j0 at resonance... If you did want to incorporate an impedance matching system that doesn't compromise the portability you have described, you could try a twelfth wave transformer with 29.3° of 50 ohm coax from the feedpoint, then 29.3° of 75 ohm coax then any length of 50 ohm coax to the transmitter. For example, for 146MHz, that could be 137mm of Belden 9258 (RG8/X) then 139mm of Belden 1189A (RG6/U) then any length of 50 ohm coax to the tx. Owen Very nice, Owen. Saved me a lot of difficult math.... since I have those materials on hand I may see what I can throw together tomorrow. Ed |
#8
|
|||
|
|||
Coaxial Antenna question
Ed wrote in
36.82: Owen Duffy wrote in : Ed, You might have expected the feedpoint impedance to be around 70 ohms. It will depend on the feedline configuration, because you haven't taken much is the way of measures to decouple the feedline. Your measured fwd and ref indicates VSWR~=1.5 which is consistent with 70 ohms, but you haven't measured 70 ohms. Assuming though that such an antenna should be close to 70+j0 at resonance... If you did want to incorporate an impedance matching system that doesn't compromise the portability you have described, you could try a twelfth wave transformer with 29.3° of 50 ohm coax from the feedpoint, then 29.3° of 75 ohm coax then any length of 50 ohm coax to the transmitter. For example, for 146MHz, that could be 137mm of Belden 9258 (RG8/X) then 139mm of Belden 1189A (RG6/U) then any length of 50 ohm coax to the tx. Owen Very nice, Owen. Saved me a lot of difficult math.... since I have those materials on hand I may see what I can throw together tomorrow. I didn't do the math, I punched the numbers into TLLC (http://www.vk1od.net/tl/tllc.php). Of course, the reason I was so specific is that translation from the 29.3° depends on the velocity factor... so use the velocity factor for the cables you have at hand. (For example, if you use RG59, it has a very different velocity factor to te 1189A, and you need to adjust accordingly.) There is a little on the twelfth wave transformer, including a graph of the lengths for different transformation ratios at http://www.vk1od.net/RG6/index.htm . Have fun. Owen |
#9
|
|||
|
|||
Coaxial Antenna question
Owen Duffy wrote in news:Xns9B4AAC084EA4nonenowhere@
61.9.191.5: .... There is a little on the twelfth wave transformer, including a graph of the lengths for different transformation ratios at http://www.vk1od.net/RG6/index.htm . If you read the article, you might try evaluating the systemusing TLLC with say 10m of feedline configured with: - the twelfth wave transformer near the antenna and 10m of RG8/X, and - the twelfth wave transformer near the tx and 10m of RG6. Intesting, the cheapest option (mostly RG6) is the one with the least loss! Owen |
#10
|
|||
|
|||
Coaxial Antenna question
"Owen Duffy" wrote in message ... Ed wrote in 36.82: Owen Duffy wrote in : Ed, You might have expected the feedpoint impedance to be around 70 ohms. It will depend on the feedline configuration, because you haven't taken much is the way of measures to decouple the feedline. Your measured fwd and ref indicates VSWR~=1.5 which is consistent with 70 ohms, but you haven't measured 70 ohms. Assuming though that such an antenna should be close to 70+j0 at resonance... If you did want to incorporate an impedance matching system that doesn't compromise the portability you have described, you could try a twelfth wave transformer with 29.3° of 50 ohm coax from the feedpoint, then 29.3° of 75 ohm coax then any length of 50 ohm coax to the transmitter. For example, for 146MHz, that could be 137mm of Belden 9258 (RG8/X) then 139mm of Belden 1189A (RG6/U) then any length of 50 ohm coax to the tx. Owen Very nice, Owen. Saved me a lot of difficult math.... since I have those materials on hand I may see what I can throw together tomorrow. I didn't do the math, I punched the numbers into TLLC (http://www.vk1od.net/tl/tllc.php). Of course, the reason I was so specific is that translation from the 29.3° depends on the velocity factor... so use the velocity factor for the cables you have at hand. (For example, if you use RG59, it has a very different velocity factor to te 1189A, and you need to adjust accordingly.) There is a little on the twelfth wave transformer, including a graph of the lengths for different transformation ratios at http://www.vk1od.net/RG6/index.htm . Have fun. Owen Hi Owen Another way of avoiding the math is to use both a Smith Chart and an overlay of a Z Theta Chart. The problem of choosing line lengths and their Zo them becomes intuitive. But any "perfect match" does depend heavily on knowing impedance rather than VSWR, as you know. The load impedance ploted on the Smith Chart can be assummed to translate to any impedance on the circle of constant VSWR for any load impedance. The impedance moves along the line of constant "Theta" on the Z Theta Chart for a change of Chart Z. With the overlay of the two charts, it is fairly easy to see what lengths and Zo will produce the best match. Jerry KD6JDJ |
Reply |
|
Thread Tools | Search this Thread |
Display Modes | |
|
|
Similar Threads | ||||
Thread | Forum | |||
16 el coaxial colinear (2.4ghz) question | Antenna | |||
antenna coaxial switch | Homebrew | |||
Coaxial Cable Question | Scanner | |||
Coaxial sleeve antenna? | Antenna | |||
FS: Coaxial skirted VHF antenna | Swap |