One director extra, huge difference. Why?

Discussion in 'Antennas, Feedlines, Towers & Rotors' started by BX2ABT, Feb 13, 2019 at 3:38 AM.

Tags:
ad: L-HROutlet
ad: l-rl
ad: L-MFJ
ad: Subscribe
ad: MessiPaoloni-1
ad: Left-2
ad: Left-3
  1. BX2ABT

    BX2ABT Ham Member QRZ Page

    Still trying to master the black art of VHF/UHF yagi making and before I head back for Hogwarts I want to ask you professors here about the following curious happening.

    I used VK5DJ's yagi calculator and came up with a 6 element folded dipole yagi (see pdf). 1 1/4" boom, with all elements separated by using clamps (see photo). SWR is/was very bad about 2:1, so I tore it down and rebuild it from scratch.

    SWR is a flat 1:1 at 145 MHz when I use 5 elements (R, DE, D1, D2, D3). Yea! Then I mount the last element (D4) and the SWR tanks to 2:1. What is happening here?

    More clarification: happens both in horizontal and vertical positions. The yagi hangs with some 5 feet distance from both ceiling and floor. I also tried it up 8 feet in the garden, but same result. All dimensions are up to the millimeter correct, including the transformer. I studies many yagi designs and most measurements come down to the measurements I use.

    How can one director make such a huge difference in SWR reading? Enlighten me, please, and any suggestions to help me troubleshoot this are welcome (including recipes for magic potions). 73 de Hans (BX2ABT)
     

    Attached Files:

  2. KB4QAA

    KB4QAA XML Subscriber QRZ Page

    Looks like good basic construction!

    I note this comment from the VK5DJ site:
    A boom length of 2 wavelengths (or 10 elements) would be a minimum sized antenna. On the other hand, yagis with as few as 8 elements have used the design and worked very well.

    -You are attempting a 6 element design that is even smaller, and even further from the programs intended design theory. Perhaps you have shrunk things too far...

    -I note a photo on the site that shows one antenna with a Balun. Are using one? Why would one use a Balun, when I would think the looped feed element should provide a 50 Ohm impedance?

    Suggestion: Don't allow the coax to hang parallel to any elements when taking measurements. Rotate the antenna horizontal and let the coax hang perpendicular.
     
    NH7RO and W7UUU like this.
  3. BX2ABT

    BX2ABT Ham Member QRZ Page

    Thanks for the quick reply.

    True, but then it's strange that another yagi I made with three elements works fine, the bigger one with five elements works fine, but adding one messes up big time.

    A folded dipole has, in theory, an impedance of 280~300 ohms. What you say is a balun is a half lambda transformer to match to the 50 ohm coax.

    When I measure the coax is routed to the back of the antenna and connected to the SWR meter/HT behind the reflector.
     
    KB4QAA likes this.
  4. KE7RGP

    KE7RGP Ham Member QRZ Page

    I cannot fully answer your question, but you shouldn't expect Zin to be 280-300 Ohms. That is for an isolated folded dipole. Consider a single center fed half wave dipole. It's impedance is about 73 Ohms. Now add a single parasitic element nearby. The input impedance is now Zin=Vin/Iin=Z11+Z12*I2/Iin. For closely spaced wires the currents are approximately equal in magnitude and opposite in phase. So Zin=Z11-Z12. When these closely spaced elements are nearly equal in length, Z12~Z11 and Zin can be vanishingly small. I think typically, designers sacrifice directivity to get a non-zero input impedance. OR, they use the folded dipole with an isolated Zin of 4*73 Ohms. Gives them more room to avoid the vanishingly small Zin when the parasitic elements are added.

    As a funny side note, I've been able to achieve the same effect with an end-fed Yagi. Haven't seen that one in the literature. Still needs a matching section.
     
  5. KC8VWM

    KC8VWM Moderator Volunteer Moderator QRZ Page

    Hans,

    A few things come to mind as follows...

    The instructions don't specify the type of tubing used to construct the boom. Round boom, square boom, 2" round - 1" x 1" square or 1.5" x 1.5" etc.. etc...

    Secondly, where are the 4:1 balun construction details indicated the instructions? How exactly is it to be wired and connected to the feed point exactly?

    The boom dimensions in the photo appears to look different than what is written in the instructions. The photo appears to have boom overhangs much longer than specified below.


    This may also be the problem...

    This is "assuming" the cable you are using to construct the 4:1 balun has a velocity factor or .66, but does it really?

    So have you actually measured and/or otherwise confirmed this with the cable manufacturing spec. sheets? Measured with an antenna analyzer? etc..

    Also, tell us more about what frequency you taking these SWR measurements / readings.

    What does the SWR look like on other frequencies at each end of the band?
     
    Last edited: Feb 13, 2019 at 6:07 AM
  6. AA5MT

    AA5MT Ham Member QRZ Page

    As important, what does the swr curve look like? Does it shift with the extra element?

    Tom
     
  7. BX2ABT

    BX2ABT Ham Member QRZ Page

    @KC8VWM: you are right, the boom is not specified in the pdf: 1 1/4" square alu, which is the parameter I used when running the yagi calculator. The overhang is more than necessary because it is supposed to become a cross yagi for satellite reception with 1/4 lambda distance between the two yagis.

    I don't have an analyzer (will get one in a couple of weeks), but I've made many baluns and they all seem to work. I use 145 MHz for all measurements (middle of the 2m band here in Taiwan).

    @KE7RGP: I do use an isolated folded dipole. With the balun it gives me a 1:1 SWR at 145 MHz. With 3 directors also, just not with 4 directors.

    @AA5MT: the antenna is in pieces on my work bench, so I can't measure right now, but iirc it shifted up, iow the 1:1 SWR went above 146 MHz.

    @KB4QAA: you're comment on shrinking the design too much from the design theory made me do a little experiment.

    I searched on-line for another yagi calculator program. I found QuickYagi which is an old DOS program, but it runs fine under DosBox on Linux.

    I designed a standard yagi, took the element spacings and applied them to my yagi. D1 needed some fine tuning, but then I got a 1:1 SWR. Adding D2~4 the spacings were spot on and I got an SWR of 1:1, albeit with a much longer yagi (around 1.2 lambda), because the spacings were much bigger than my original design.

    Then I took away D4 and lo-and-behold the SWR tanked again from 1:1 to 2:1. So the number of directors with their spacings do have a profound effect of the impedance of the driven element.

    My antenna books don't have much theory in them and on-line it's difficult to find yagi theory, so any links to that are appreciated so I can understand more. Also, any tips on good yagi calculators for shorter beams would be appreciated, too. I like the VK5DJ one, because it compensates for boom width and calculates the length of the folded dipole for you.

    What I will probably do now is make my yagi into a 5 element one and live with a little less gain. Once I get some new alu-booms I can attempt to make a longer one.
     
    KB4QAA likes this.
  8. KB4QAA

    KB4QAA XML Subscriber QRZ Page

    A trick I have used in the past to tweak the SWR once an antenna is built is to adjust the reflector position. You would have to run a model to see what that does to the pattern.
     
  9. N8CMQ

    N8CMQ XML Subscriber QRZ Page

    Input impedance changes with the added elements and the matching network needs adjustment when the impedance changes.

    Unless there isn't a matching network, then the SWR goes up without a way to adjust for it...
     
    Last edited: Feb 14, 2019 at 1:09 PM
  10. KC8VWM

    KC8VWM Moderator Volunteer Moderator QRZ Page

    In this case, I suspect antenna SWR matching is achieved by adjusting the length of the RG-58 used to construct the 4:1 Balun.

    Design parameters are specifying a 1/2 wavelength section of RG-58A with a velocity factor of .66

    Everything works but the Balun in this case also functions as a matching device and therefore it's critically important to know exactly what the velocity factor and other electrical measurements of everything is.

    This is not a component where getting things close to the ballpark of things is good enough, because even small variations and changes to the length of this cable can result in profound effects on the antenna SWR.
     
    Last edited: Feb 14, 2019 at 2:18 PM

Share This Page

ad: Schulman-1