Thoughts on Vertical Coaxial Sleeve Dipole-Driven Slim-Jim Collinear Antenna

Discussion in 'Antennas, Feedlines, Towers & Rotors' started by KAOSFM, Dec 11, 2019.

ad: L-HROutlet
ad: l-rl
ad: MessiPaoloni-1
ad: Left-3
ad: Subscribe
ad: Left-2
ad: L-MFJ
  1. KAOSFM

    KAOSFM QRZ Member

    Hi All,

    I hope all is well and was hoping to get some feedback from the forum for a new antenna project I am working on. I've been attempting to learn all I can about vertical collinear antennas and the various methods of loading and phasing them. I want to avoid a jpole design due to the extra space wasted by the j-match section, their issues with common-mode and just based on mediocre results obtained by the j-poles I have used, including a copper standard j-pole, a twin-lead j-pole in PVC and finally a twin-lead super j-pole collinear in PVC. I found that my 5/8 wave Comet vertical dominated all of those, including the super-j which was supposed to have higher gain on paper.

    I came across an interesting collinear design in Google patents, as shown below in the first antenna diagrammed on the left. It appears to be a center-fed dipole at the base which is co-phased in series with a coil matching section and an upper half-wave parallel radiator element that closely resembles a slim jim. They claim 6dbi gain in the patent, which is questionable because they appear to add extra mythical slim jim gain simply because of the folded vertical element on top.

    The closest thing I could find in practice was a so-called "Franklin Spider" ground plane antenna. It uses a half wave matching stub instead of a series coil, substitutes radials for the single ground dipole element and lacks the slim jim top. They claim 3.58dbi gain for that one. I live in an HOA and want to avoid radials, opting for a slim low profile vertical antenna mounted to a discreet tall fiberglass rod. My design keeps the center-fed dipole like the original patent at the base, but I elected to use a coaxial sleeve dipole arrangement to keep that clean vertical profile.

    The sleeve portion is 3/4" aluminum tube with thin RG174 coax inside being choked off by the sleeve and tied to the coax ground at the top of the element. I'm also using the same size tubing for the upper + dipole element to enhance broadbandedness. I plan on adding some clamp on ferrites at the base of the sleeve/top of the feedline to fight common mode. I plan on opting for the 1/2 wave tuning stub hairpin like the franklin design due to simplicity to feed the top half wave element. I'd do a coil to increase element spacing, but there's no easy guidance on how to do that correctly. The patent just says it uses a coil less than a 1/4 wavelength in height. I'm debating whether to use a standard 1/2 wave top element or modifying it to the folded slim jim style of the original patent. I also may or may not fold the middle phasing hairpin around the radiating element as I've heard this somewhat cleans up the pattern.

    I'm open to thoughts on the design in general, whether to use the standard or folded top element, etc. I'd be especially grateful if there are any NEC wizards out there willing to model these. I haven't settled on a single design direction yet because I don't want to waste materials or deal with multiple dangerous trips on and off our steep roof. I've seen so many super jpole designs out there which are basically stacked 1/2 wave elements, but with the bottom element being end-fed with a 1/4 wave j-match. I've yet to find any that substitute a center-fed half--wave dipole for the bottom 1/2 wave element. The original patent was intended for 2.4 GHz wi-fi applications. I see no reason why this cannot be scaled down to VHF.

    In my mind, in the coaxial dipole arrangement, you save dealing with the j-match and save a 1/4 wavelength of wasted mast space. I plan on using 75 ohm coax to feed it, which would match the dipole driver section nicely. This also adds a ground element via the lower dipole sleeve section, which would hopefully fight the feedline coupling problem from jpoles. I just don't know if I am looking at this wrong. I'm having trouble visualizing the waveform/phase relationship of all the elements between the end-fed base of a jpole or a center-fed dipole. Your thoughts are appreciated. Once a design option is settled upon, I look forward to building it and sharing the results with the forum. Thanks!

    -Nick

    [​IMG]
     
    W6KCS and NH7RO like this.
  2. K7TRF

    K7TRF Premium Subscriber QRZ Page

    Here's a quick EZNEC model of the folded Design #2 shown on the right.

    The coaxial dipole is modeled as a center fed vertical dipole with ideal feed which is not completely accurate but is what the coaxial sleeve and common mode choke is emulating. The NEC based tools are limited to 2 dimensional modeling and though it's possible to model a bunch of wires built into a mesh sleeve it would take some time. The main thing this model looks at is the use of the phasing line which is modeled literally, not as a transmission line stub and the folded top element.

    I had to lengthen the folded top element by a few inches to tune the overall antenna to the 2 meter band. I also don't see a complete reversal in currents with the horizontal half wave phasing section so don't consider this fully optimized but just a starting point to see if the basic idea works. The lowest point of the primary vertical dipole is at 20 feet above average ground and of course it's a clear field model with no nearby objects or clutter.

    upload_2019-12-11_8-50-41.png

    With the caveats above, this seems to match pretty well into 50 ohm coax though the match as you suspected is a bit better looking into 75 ohm coax. There's definitely some gain improvement with the upper collinear element approaching 6 dBi for very low elevation angles.
     
    W6KCS likes this.
  3. K7JEM

    K7JEM Ham Member QRZ Page

    Does the folded part on the top add to the gain, as compared to just a standard HW section? I don't see how that could help, except maybe with bandwidth.
     
    W6KCS likes this.
  4. K7TRF

    K7TRF Premium Subscriber QRZ Page

    It looks like folding the top section helps a great deal with matching at the cost of some peak pattern gain:

    upload_2019-12-11_9-5-41.png
     
    W6KCS likes this.
  5. K7JEM

    K7JEM Ham Member QRZ Page

    I guess now I am wondering how much gain a simple dipole shows under the same conditions? I wouldn't think a collinear antenna like this could show even 3dB gain over a dipole, which would mean that a dipole should be showing around 5+dBi with the same conditions as above.
     
    W6KCS likes this.
  6. K7TRF

    K7TRF Premium Subscriber QRZ Page

    Yeah, a simple vertical dipole over medium ground gives you most of the pattern gain and the collinear element doesn't add a ton.

    upload_2019-12-11_9-17-29.png
     
    W6KCS likes this.
  7. K7JEM

    K7JEM Ham Member QRZ Page

    It looks like the simple dipole and the first folded collinear model have nearly the same gain. The "unfolded" collinear has around 2dB of gain, which I would expect. Somehow, the folded antenna is cancelling the gain, like the currents are out of phase in that folded section or something. Seems strange anyway.
     
    W6KCS and K7TRF like this.
  8. NH7RO

    NH7RO Ham Member QRZ Page

    To the OP: This probably what I'd do for a collinear if/when I ever get around to building one; scale this from 70cm design down to 2m and use the fattest Heliax I could find instead of regular coax (for better "phasing management" as that is apparently the devil in the details with many co-linear designs, as I understand it). I believe it would take around 9 feet of Heliax at a minimum; double that would be better in the gain dept. IIRC)

    Hope this helps:

    http://www.rason.org/Projects/collant/collant.htm

    While one of these would be fatter than a standard coax model it wouldn't require any protrusions or radials and still could be placed inside a plastic/fiberglass radome of some sort (that is also tricky since it affects the velocity factor in turn which affects the phasing).

    73,

    Jeff
     
    W6KCS likes this.
  9. KAOSFM

    KAOSFM QRZ Member

    Excellent, thanks for clarifying the design parameters with these revealing antenna models. It's unfortunate as each of these prospective designs have advantages, the folded one in terms of a nice broadband match and the non-folded design has a superb flattened pattern that doesn't waste as much RF up in the sky. The fact that any gain advantage compared to a single dipole is lost with the folded design is interesting. I think the assumption that the folded variant is somehow cancelling the gain is on point. It's just odd that you wouldn't see this with a regular single slim-jim antenna. It suggests to me that there is some interaction with the j-match on a single slim jim that adds an advantage that is clearly absent with my folded collinear proposal.

    I noticed that most folded monopole antennas, which the top of the folded collinear design looks like, always has the open end of the monopole terminated to a ground plane. That ground plane is totally absent in the collinear for obvious reasons. Perhaps if a coil was used instead of a phasing stub, adding height/element spacing, these things would improve, but I'm suspect of that. Based on the popular excel super j-pole collinear calculator that's floating online, for the center frequency of 146mhz, they suggest using a coil 20.2 inches in height, using approximately 72.98 inches of #14 enameled wire.

    If there's no way to restore any gain advantage over a single dipole with the folded model, I think it's likely not worth bothering with. I would be better off just raising the single coaxial sleeve dipole as high as possible and by itself if the gain can't be made on par with the other non-folded model. I suppose I am surprised by the poor 3:1 or worse match for the non-folded version.

    Does the modeling data reveal what the actual impedance is at the center frequency? I'm curious if adding radials at the base of the sleeve dipole element, either in downslope, flat or upslope would bring the SWR down to 2:1 or better. My other thought was if using a 4:1 1/2 wave coax balun and giving it a balanced feed would help. The only problem is that it's only practical to install it at the base of the antenna below the sleeve. That would send a balanced signal up via the coax inside the sleeve towards the true center of the dipole.

    The only other solution I can think of is a series 1/4 wave transformer balun (coax), but again, I installed my SO-239 feed connector at the bottom of the sleeve dipole, not at the center feed. Perhaps also, like the folded design, replacing the middle 1/2 wave phasing stub with a coil would clean up the SWR. I've updated my drawings to reflect these suggestions. Any advice to make the design more workable is appreciated. Originally I was pursuing a CoCo (coaxial collinear) design using RG58, but the multitude of problems introduced by the fact that the VF is different between the signal in the phasing sections and the signal on the the shield for the radiating sections. Thanks! -Nick

    [​IMG]
     
  10. K7JEM

    K7JEM Ham Member QRZ Page

    A folded monopole is a QW antenna, hence the GP and the connection to it. The bottom ends of a QW monopole are low impedance, the ends of a folded HW antenna (like you are describing) are high impedance, and can't be attached to ground, therefore your design #2 will not work.
     
    NH7RO likes this.

Share This Page