Antenna Feed Line Length

[Toll_Free]

Re: Death, Taxes, and Coax Woes

In an ideal system, the length of the coax antenna feedline should not make any difference to either the antenna’s SWR or its efficiency, see ARRL Handbook, 2006 Edition, Sec 21.1. However, as others have pointed, out its length is often critical real-world mobile installations. I suggest (as Hamin' X did) that this can only happen if the coax shield is radiating RF energy. In order for the coax antenna lead to work as an ideal feedline, the current carried on the outer braid must be equal and opposite from the transmitted signal traveling along its core; otherwise, the signal attempting to radiate from the core cannot be cancelled out. The ideal feedline does not radiate any of the transmitted signal, and only the antenna forms the resistive load. However, if the feedline does leak, it must then be considered part of the antenna, and it’s length will certainly effect both efficiency and SWR.

When you think about it, it seems inevitable that you will not be able to achieve an ideal ground in your car or truck. Almost certainly the chassis to which the coax shield is grounded will carry stray RF from whatever electronic control systems are found in modern cars, including the ignition, the CPU/ECU, the fuel injectors, the fuel pump, among others; not only this, but chassis itself is probably segmented, and unable to act as an efficient ground plane. The result will be common-mode currents in the coax shield not be present in the core, which will, in turn, allow the transmitted RF to escape from the core. In short, your feedline in a mobile installation will almost certainly become part of your antenna.

Since the nature of the stray current introduced onto the outer braid of the coax feedline cannot be predicted, it would follow that it is impossible to predict any specific length of feedline necessary to achieve a low SWR. Thus, the proper length of your feedline will have to be determined by experimentation on a case by case basis, but it’s important to understand that it will make a difference, nevertheless.

Perhaps one solution is to coil the feedline so that it acts as an RF choke, as I’ve seen suggested in some articles, but then other writers advise strongly against doing that. Has anyone tried this?

Yes. It works.

However, the length necessary to choke off antenna currents, and prevent the thing from becoming an antenna itself (the feedline) is usually long enough to reach the amplifier itself.

Your 100 percent right, too. You need to choke off the feedpoint. You need to get the feedpoint to 50 ohms, which is next to impossible on a mobile install. Remember, most vertical antennas DON'T present 50ohm-0j. Anywhere.

The choke works. Ferrite cores work even better, just because of the ease of use. I prefer a coaxial choke, however, and have noticed better performance out of them when used on my base station.

They do, however, need to be wound for a single frequency.

Your coax is not balanced, though. The shield is supposed to be "cold" as far as RF is concerned.

--Toll_Free

 

[freecell]

common mode CURRENT radiation from the outer surface of the shield in a coaxial feedline is the result of feeding BALanced antennas with transmitters and feedlines that are designed for UNbalanced operation.

 

[bob85]

And the long of it is, if you use a piece of coax that is not a multiple of a quarter wave, electrically, and you don't measure your VSWR at the FEEDPOINT of the antenna, you don't have a friggin clue as to what your real feedpoint match is.

Don't buy into this. The proper way, is to use an electrical quarter wave to match your system. Then, use as long a piece as necessary, if you want to. I stick with the electrical quarter waves. Works for me, and I KNOW my measurements are accurate, not reading the capacitance values in the coaxial cable itself, and not reading a skewed reading from the feedpoint reactance.

--Toll_Free

thats the last thing i would use,
since only an electrical halfwave, multiple electrical halfwaves or an even number of electrical quarterwaves of coaxial cable will mirror its true terminating impedance regardless of that termination been the same as the characteristic impedance of the coax or not how can you possibly know the true antenna feedpoint impedance if you look at it through single electrical 1/4wave or any odd multiple thereof,

is this a test or are you winding us up??? please explain???

 

[bob85]

Please, I beg you. Explain this? How a good match at your feedpoint can be repeated as a bad match at the 'OTHER' end of the coaxial cable, but it really doesn't mean anything, other than the foldback circuit in the radio will cut your power back, you won't see max power transfer, and you can see hotspots on the coax, the radio, etc.

Not to mention, this causes more stray RF around.

--Toll_Free



--Toll_Free

imho you must either be measuring the antenna with a rf source that has a impedance that is different to that which you used at the feedpoint
eg, you used a 50ohm mfj 259 at the feedpoint then added the feedline and line section then used a radio that does not have a 50ohm output,
or the coaxial or line section/directional coupler you added is not 50ohm characteristic impedance,

somebody show me the error of my ways..

 

[bob85]

would there be anything to be gained by using ferrite sleeves on the coax of a tripple 7" magnetic mount antenna even if the lower than 50ohm antenna is matched to the coaxial with a "dollar match" at the feedpoint??

ps the magmount is also grounded with a 4" braid to the car roof, the roof is double skinned with factory sliding sunroof so no way to direct mount antenna to roof.

 

[Master Chief]

every length of feedline is a matching transformer....Cebik.

Wow. Someone that actually reads, comprehends, and gets it.

Cebik wrote it, freecell gets it. He does have a callsign.

 

[Hamin' X]

Well, that's one way to look at it.

I guess the other way is that coaxial cable length can play into account when your trying to get a flat match, and don't know what the complex impedance is on the other side, huh?

You are correct, I don't have an antenna analyzer, so I have no clue as to the true impedance of the antenna.

Let me get this straight. When you measured as close to the feedpoint as possible, it worked OK, and you got a smaller amount of reflected power.

the Bird43 was connected to the base of the antenna with a 12" jumper. I would have connected it with an adapter, but could not find my N male-PL259 adapter.

But, when you put a piece on that was "convenient", your radio saw a worse match.

The original length of cable was about 14 feet of rg58u non-foam. I don't like to use foam coax for mobile installations, due to center conductor migration when making sharp bends.

And you think that your coaxial cable isn't being used as a radiator? As part of the counterpoise of your antenna system?

I've reread my original posts several time and cannot find where I ever said to the contrary. It seems pretty evident that there are common mode currents on the coax.

Please, I beg you. Explain this? How a good match at your feedpoint can be repeated as a bad match at the 'OTHER' end of the coaxial cable, but it really doesn't mean anything, other than the foldback circuit in the radio will cut your power back, you won't see max power transfer, and you can see hotspots on the coax, the radio, etc.

Not to mention, this causes more stray RF around.

OK, off your knees and stop begging, it's unbecoming of one that only uses electrical-quarter-wave-multiples :shock: .

The fact is that I stumbled onto the correct matching section by accident, when I used the 43 and 12" jumper. So, rather than correct the problem with the antenna (there was a problem, or the length of coax would not have made a whits worth of difference), I added 16 inches of coax (4 inches for the Bird line section and 12 inches for the for the jumper). The only thing that this did was, mask the problem from the radio and make it happy, so that it would produce full power (about 100 Watts). No coax melted and 99% of the power was transferred to the antenna. I'm not saying that the antenna was 99% efficient, only that the power was not lost in the feedline.

Rich

 

[bob85]

hey guys theres a forum somewhere on the net where a bunch of guys are arguing about just this kind of thing,

one claims to be a BBC engineer one HARRIS engineer and a guy claiming to have constructed antennas for NASA's space program, and one guy that claims to write the arrl articles for the ham books plus some other hams,

they are arguing about what a bird meter does and does not do when placed at different points in a feedline,

they also argue over what happens to any reflected power,
some claiming it is re-reflected by the tank circuit and eventually all power is delivered to the load while others say the power is absorbed by the transmitter output matching circuit and lost,

they quote CEBIK and others

i have never read anything by CEBIK ( most likely i would not understand him anyways ) nor do i know who he is,

i read it over several times and probalbly forgot most of what i read :roll:

the one thing i do remember and will never forget is my thoughts :!:

IF THESE GUYS CANNOT AGREE ON WHATS REALLY GOING ON WHERE THAT HELL DOES THAT LEAVE ME :shock: ,

EXACTLY WHERE DO I LOOK FOR FACTUAL INFORMATION IF THE GUYS WRITING ABOUT AND BUILDING THIS STUFF CANNOT AGREE ON SOMETHING LIKE A BIRD METER AND WHAT HAPPENS TO REFLECTED POWER :?:

i talked to dennis ( ccm ) about it and he laughed, he said they are always arguing over something,


YUP

 

[Hamin' X]

I just try to state the facts as I see them. Like changing the length of the coax to my screwdriver antenna. It worked. As far as what happens to the signal, I believe that it is re-reflected to the antenna. Why? Because my field strength meter tells me so. I tune my screwdriver antennas with a FS meter on the dash. Max FS reading=max output from the antenna.

You're right though Bob, it is all theory and a theory is only correct until another one comes all to prove it wrong. I work on the gimmick-method principle. If it works sometimes, but not every time, it is a gimmick. If it works every time, it is a method. Neither one requires that you know the underlying principles as to why it worked.
Rich

 

[freecell]

of course it was re-reflected toward the load. it has no choice since it (the bulk of the reflected power involved) cannot travel beyond the match point. approximately 1/64th of a wavelength at 7.26 mhz. made enough difference in Z = E/I in your case to restore nominal operation.

246 / 7.26 mhz. X .66 = 22.3636 feet or 268.3632"
don't misunderstand here, this is the calculation for an electrical 1/4 wave line (90 degrees) at 7.26 mhz. for RG58 feedline with a VF of approximately .66.

168" would be your 14' RG58 line or 56.3415 degrees.

22.3636 / 16 = 1.3977 feet or 16.7724", give or take 3/4 of an inch. there's your 12" patchcord and the line section in your bird.

add 5.6248 degrees of length. total length of the new line in degrees = 61.9663 degrees, or just a bit longer than 2/3 of a 1/4 wavelength.

that seemingly insignificant change was enough so that the new values of E/I present at the input of the line satisfied the transmitters loading requirements. had the 706m2g been used with an AT the problem would have never appeared.

"every length of feedline is a matching transformer"....Cebik.

 

[bob85]

the re-reflected idea is my current stance on the subject as i posted earlier, not because i pretend to understand it because i dont but their argument made more sense to me, i am glad you agree :!:

 

[Hamin' X]

they quote CEBIK and others

i have never read anything by CEBIK ( most likely i would not understand him anyways ) nor do i know who he is,

You can read L. B. Cebik's stuff here:Cebik.com

Like all info on the proper care and feeding of antennas, run the info thru your BS filter and remember: It's just another tool in the toolbox. If it fits, use it.

Rich

 

[bob85]

:shock: thats some reading right there, i just hope he knows what hes talking about lolol

thanks rich

 

[Lazybones1222]

Man. Just use an antenna tuner.

 

[freecell]

the section of cebiks site pertinent to the discussion here which is well worth reading and studying is:

SWR, Feedlines and Reactance