Total Copper Content revisited: The Fat Factor

Discussion in 'Antennas, Feedlines, Towers & Rotors' started by KL7AJ, Jun 1, 2011.

Thread Status:
Not open for further replies.
ad: L-HROutlet
ad: l-rl
ad: L-MFJ
ad: Left-3
ad: abrind-2
ad: Subscribe
ad: Left-2
  1. KL7AJ

    KL7AJ XML Subscriber QRZ Page

    The TCC (Total Copper Content) factor of amateur radio station performance has been well documented and seldom disputed. The more copper (or aluminum) you have up in the air, the better your station will work. This is a corollary of the NFL parameter (No Free Lunch).

    Fat dipoles work better than skinny dipoles. I remember hearing this ages ago, but hadn't paid much attention to it until I started working at HIPAS Observatory and HAARP; our findings were also confirmed by the fine work of W4RNL on his second book of Quads. The increase in bandwidth is not the only reason to use fatter antenna elements. It can be consistently demonstrated by measuirement and modelinig, that a fat dipole can have over 1dB of gain over a thin wire dipole.

    Stay tuned for more details

    Eric
     
  2. KD0CAC

    KD0CAC Ham Member QRZ Page

    Eric , could you give some specifics , like gauge of wires or diameter of element , does it take 2-4 times the size to get that 1 db gain .
    Haven't heard is the new book is out yet ?
    Thanks John
     
  3. W0BTU

    W0BTU Ham Member QRZ Page

    That's interesting. Seems like a conductor loss thing (very thin=more loss).
     
  4. WB3BEL

    WB3BEL Ham Member QRZ Page

    There is no simple rule of thumb. The devil is in the details. For simple antennas like verticals and dipoles, you will not get much gain at all. The main impact of larger diameter conductor is in the possibility of increased bandwidth.

    For the case of the quarter wave vertical or similar there is very little gain difference in the diameter. But if you took all that metal and put it into the ground system you could very well have dBs of difference vs the teeny fractional dB in the radiating element diameter.

    The gain issue becomes more apparent for high Q arrays especially those using electrically shortened elements or very tight coupling. Also antennas like small Quads or loops with low radiation resistance are an example.
     
  5. KL7AJ

    KL7AJ XML Subscriber QRZ Page

    Hi Mike:

    Actually it's a CURRENT DISTRUBUTION function, not a resistance function. In a full sized dipole, at lower H.f. frequencies, the resistance of say, 12 ga copper wire is pretty negligible, even considering skin effect. If you were to make a cage dipole with four such wires (simulating a 1" diameter tube, for instance) the reduction in copper losses would be almost unmeasurable. However the effective 1" diameter tube has a non-sinusoidal current distribution...you have more current toward the ends than a thin dipole. So you gain more than the mere reduction in copper loss would suggest. Now the question still comes up....is it worth 4 times the copper to get 1dB of gain? Maybe not. But if these conductors are part of a large array...say a phased array wire antenna...this improvement can add up considerably!

    Eric
     
  6. AD4J

    AD4J Ham Member QRZ Page

    Is the 1 dB advantage coming from a change in radiation pattern or in efficiency?

    I just modeled a 40 meter dipole (67' long) up 65' above average ground. I did it with both #16 wire (0.05" diameter) a 2" diameter wire. This is a difference of 40X in diameter. However, the thicker wire showed only a 0.01 dB improvement in gain (from 7.19 to 7.20 dBi). I'm not sure what we are doing differently in the models.

    73,
    Jim AD4J
     
  7. W0BTU

    W0BTU Ham Member QRZ Page

    That's interesting, Eric. Thanks.

    To get the extra gain, the pattern of the thick antenna must of necessity be more directional, right?
     
  8. KL7AJ

    KL7AJ XML Subscriber QRZ Page

    Hi Mike:

    That is correct. If you want to model something very simple, create a "flat biconical" or "bowtie" antenna. Just build a dipole with TWO wires extending from the feedpoint with about a 10 degree angles from each other....like a fan dipole, except for one frequency. On 40 meters, the wires will be separated by 10 feet or so at the ends. Now run the model. You will have over 3dBi in free space....and lots less wire than a 4 wire cage! This arrangement gives you a lot more capacitive end loading.

    Eric
     
  9. N3OX

    N3OX Ham Member QRZ Page

    Eric, I agree with AD4J and am interested in the specific models or measurements you're looking at.

    In a model with simple thick vs. thin wires, I indeed found the rise in current near the ends that you stated. There's about 70mA more current about 70% out from the feedpoint to the tips in a 6" diameter EZNEC model fed with 1A at the feedpoint than there is in an 18AWG dipole.. but that change isn't enough to increase the gain more than a small fraction of a dB, maybe 0.02dB.

    My first pass at an actual cage dipole with eight wires, shows about 140% efficiency (average gain in EZNEC), so that needs a lot of tweaking to fix it. A simple fan resonant around 13MHz shows 142% efficiency (about 1.5dB).

    I checked a bicone I modeled a while ago and it failed the average gain test by a similar amount.

    Big fat wires, however, show the current increase toward the ends that you suggest, but are fine in the Average Gain Test and then show very similar gain to a dipole, maybe 0.02 or 0.03dB better, consistent with the small percentage current increase out in the ends.

    Here's my plot of the difference between a six inch diameter resonant 14MHz dipole and one with 18AWG wire. The antennas are therefore slightly different lengths and I've plotted the difference vs. segment number. Feedpoint current is 1A in each case:

    http://n3ox.net/files/deltaI_thickdip

    But the gains are nearly identical for the "fat wire" models. The current increase is probably measurable on a real antenna but the increase in current is just too small and the increased regions are too closely spaced together in space to cause a big gain differential. And I just can't seem to get a multiple wire cage to show 0dB average gain for a decent comparison, but it should be the same. Different current distribution, but in a way that's almost immaterial to the gain.

    73
    Dan
     
  10. KL7AJ

    KL7AJ XML Subscriber QRZ Page

    Just to do a double check, I modeled this on both NEC-2 and EZnec, witih very similar results. Try modeling this antenna and see what you get. (use free space model)


    40 meter dipole consisting of two pairs of wires on each end, separated by two feet at the outside extremity. (Two element "fan" dipole, each element self resonant) I use a short straight section between the two "apexes" to insert the current source. I get 3.4 Dbi on NEC-2 and 3.7 dbi on EZnec. It's actually more than I expected. I'm going to run an optimizaton program and see what separation gives the MOST gain.

    Eric
     
Thread Status:
Not open for further replies.

Share This Page

ad: vanity