Announcement

Collapse
No announcement yet.

best antenna

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • @trigger:
    A crude way to check insulation of coax is by it's physical appearance, as follows:

    The PE insulation is transparent/translucent (like egg's white before whipping) and the core wire is visible through it (like pot's bottom is visible through unwhipped egg's white).

    FPE is milky in color and is opaque (liked egg's whipped white) and core wire is not visible through it (like the the pot's bottom is not visible through egg's whipped white).

    Comment


    • The core wire is not visible through the insulation. Thanks for the egg analogy
      T-EGUB1

      Comment


      • Originally posted by trigger View Post
        The core wire is not visible through the insulation. Thanks for the egg analogy
        Let me clarify, PE is not transparent, it is translucent. The central wire is visible through it, but hazy.

        FPE is opaque and the central wire is not at all visible through it
        Last edited by abcd567; 2014-11-27, 17:09.

        Comment


        • Originally posted by trigger View Post
          Hi ab cd,
          I don't know what the coax is as it only has sat 100 written on it so possibly RG6? The OD of the cable is 6mm, the OD of the core is 1mm and the OD of the insulation is 4mm. I know the insulation is polyethylene as I had it checked by one of my chemist friends.


          Thanks

          This is an impossible one to answer.... what you have is bottom of the pile satellite coax cable, specs vary VASTLY - some is so bad, SKY (UK satellite company) forbid its use in installations !!!!!!

          The only way to get its specifications is to use an analyser.
          If you MUST apply complex maths to the project, buy some new cable of known spec !

          ...or just "suck it and see" and wonder what all the fuss was about ;-) lol

          Comment


          • @Rooster:
            @trigger:
            I suggest Dave collapse the Twin-Lead to zero length (by pushing feeder & bottom element towards each other) and then see how his new coco performs. This will make it clear whether or not the twin-lead improves the performance.

            Comment


            • Originally posted by Rooster View Post
              This is an impossible one to answer.... what you have is bottom of the pile satellite coax cable, specs vary VASTLY - some is so bad, SKY (UK satellite company) forbid its use in installations !!!!!!

              The only way to get its specifications is to use an analyser.
              If you MUST apply complex maths to the project, buy some new cable of known spec !

              ...or just "suck it and see" and wonder what all the fuss was about ;-) lol
              I agree with Rooster that SAT 100 specs vary VASTLY. I searched the web, but could not find any specs for it!
              Dave's SAT100 (1 mm dia conductor/FPE??PE?? 4 mm dia)
              is something in between
              RG6 (1 mm dia conductor/FPE 4.8 mm dia)
              RG59 (0.64 mm dia conductor/FPE 3.7 mm dia)
              RG58 (0.81 mm dia conductor/PE 2.9 mm dia).

              I also agree with Rooster that "If you MUST apply complex maths to the project, buy some new cable of known spec !"
              Last edited by abcd567; 2014-11-27, 07:42.

              Comment


              • Now this is what I call "Dark Art of CoCo Making".
                In spite of "assuming" or "guessing" the insulation material & VF, Dave hit a jackpot:
                a coco with performance near equal to his accurately built, impedance-matched Franklin.

                Was this due to his assumptions turning out to be accurate, or because of "twin-lead"? We have yet to see.
                When Dave collapses his twin-lead to zero (or near zero) length and compares again his coco with his Franklin, we will know.

                Comment


                • Originally posted by abcd567 View Post
                  Now this is what I call "Dark Art of CoCo Making".
                  In spite of "assuming" or "guessing" the insulation material & VF, Dave hit a jackpot:
                  a coco with performance near equal to his accurately built, impedance-matched Franklin.

                  Was this due to his assumptions turning out to be accurate, or because of "twin-lead"? We have yet to see.
                  When Dave collapses his twin-lead to zero (or near zero) length and compares again his coco with his Franklin, we will know.
                  There was some trial and error (no assuming or guessing ) as well as a bit of science The science was getting my chemist friend to confirm the insulation was Polyethylene. I wanted to eliminate the possibility that the insulation was PTFE.

                  I made about six 4 element CoCos with different element lengths and gradually homed in to what appeared to be the best. I then made an 8 element one and it performed really well so I went the next step and introduced the twin-lead. I didn't seen any difference in performance by introducing the twin-lead, hence my question in post #1870.

                  The SAT 100 cable's core and screen is non-magnetic (so pure copper?) I do have a short length of RG6 U which is magnetic so must have some iron in it? The SAT 100 insulation is not opaque like RG6 but it is not translucent either! My experiments suggest the SAT 100 has a velocity factor of 0.66 (PE rather than FPE)
                  T-EGUB1

                  Comment


                  • Originally posted by trigger View Post
                    There was some trial and error (no assuming or guessing ) as well as a bit of science The science was getting my chemist friend to confirm the insulation was Polyethylene. I wanted to eliminate the possibility that the insulation was PTFE.

                    I made about six 4 element CoCos with different element lengths and gradually homed in to what appeared to be the best. I then made an 8 element one and it performed really well so I went the next step and introduced the twin-lead. I didn't seen any difference in performance by introducing the twin-lead, hence my question in post #1870.

                    The SAT 100 cable's core and screen is non-magnetic (so pure copper?) I do have a short length of RG6 U which is magnetic so must have some iron in it? The SAT 100 insulation is not opaque like RG6 but it is not translucent either! My experiments suggest the SAT 100 has a velocity factor of 0.66 (PE rather than FPE)
                    Good to know that the VF you used is not a guess, but determined by trial & error method, which also a scientific way.

                    Now why Twin-Lead did not make a difference? I could think of following possible reasons:

                    (1) Twin-Lead is a failed concept/experiment.

                    (2) I have made some mistake in my calculations or simulations.

                    (2) Scince your coco is near perfect, twin-lead does not make any noticeable difference.

                    (3) Twin-Lead is not adjusted to correct length, as your original question suggested.


                    I will check the model for any bugs, & re-simulte 8-element coco to find out twin-lead's correct length, using VF=0.66 conductor dia=1mm, insulation dia=4mm, element length=91mm.

                    I will also re-check my Smith Chart's calculation to see any mistakes or bugs.

                    Comment


                    • A quick check showed using twin lead (or equivalent ladder line) for impedance matching is in common practice with amatuer radio. So it is NOT a failed concept/experiment.

                      Another basic check: my 4 element coco has impedance of 180-j180 ohms. The twin lead center to center spacing=thickness of insulation+wire dia=1.9+1=2.9mm. Twin lead impedance=276xlog (2x2.9/1)=211 ohms. Now twin lead resistance is higher than antenna resistance. The twin lead can provide impedance match only when its resistance is less than antenna resistance!!

                      That was a quick check. Why simulation shows successful impedance match/improved swr with twin lead of above dimension is not clear. Needs further study/head scraching.
                      Last edited by abcd567; 2014-11-28, 06:00.

                      Comment


                      • Dave's Twin-Lead:
                        Dia of wires=1mm, dia of insulation=4mm, hence center of wire to center of wire=4/2+1/2=2.5mm.

                        Impedance = 276xlog(2x2.5/1)+0xj = 193+0xj ohms.

                        Now I have to run simulation of 8 element coco with SAT100 coax parameters (PE, 0.66, 91mm) to get coco's impedance. If coco resisance is lower than twin lead resistance, then twin lead will not help in impedance matching.

                        Comment


                        • When I simulated simple wire antennas on 4nec2 software, it gave accurate results. I then started simulating surfaces, cylindrical to be exact, as coco and cantenna have cylindrical surfaces. The initial cocos were air insulated. The software gave accurate results. I then added insulation to coco, results did not seem wrong.

                          Last I added impedance matching by twin lead formed by coax wires, the software gives excellent SWR. But when tried by Dave, and now cross checked by me by fundamental and simple calculations, the twin lead formed by coax wires should not improve SWR.


                          Why? Is this due to limitation of software? By moving from simple wire antenna to insulated coco with impedance matching stub, did I exceed software's capability? For wire antennas like Franklin, the simulation took about say 30 seconds, but insulated coco with twin lead takes about 4 minutes**!

                          I am still scratching my head (not yet reached the stage of banging my head to the wall)

                          **Later Addition: I checked the time for simulation of 8-element coco WITHOUT twin lead: It takes 7 minutes ! With twin lead, it should take more time.
                          Last edited by abcd567; 2014-11-28, 19:00.

                          Comment


                          • Ithink I have reached a stage where I need to purchse an Antenna Analyzer or VNA (vector network analyzer), so that I can verify the results of simulaltion & theory/maths by measurements.

                            Any suggestions for reasonably priced analyzer?

                            Comment


                            • I have modeled Dave's 8 element coco using SAT100 coax (PE, 0.66 VF, element length 91mm, wire dia 1mm, insulation dia 4mm).
                              I am now posting two simulation screenshots for 8-element made from SAT100 & from RG6.

                              There is a marked difference in radiation pattern of two. It seems longer element of RG6 gives better radiation pattern and somewhat higher gain.

                              Since the impedance of SAT100 Coco is markedly different from impedance of RG6 Coco, the Twin-Lead length for SAT100 Coco should be markedly different from what is required for RG6 Coco.

                              Now I have to run another simulation of SAT100 CoCo with Twin Lead.
                              I have also to calculate the length of Twin Lead by Smith Chart using Twin Lead impedance calculated by formula Z=276 x log (2S/d) + j 0, where S= center to center spacing of twin lead wires, and d is dia of twin lead wires.
                              I will try to do it during this weekend.


                              Coax SAT100 - 8 element Coco
                              Daves 8 Element.PNG

                              Coax RG6 - 8 element Coco
                              My 8 Element.PNG
                              Last edited by abcd567; 2014-11-28, 08:08.

                              Comment


                              • I am still scratching my head (not yet reached the stage of banging my head to the wall)

                                Your head is too valuable to bang against a wall. Thanks for all the work you are doing. I'll pop out this weekend and get some coax of known VF.
                                T-EGUB1

                                Comment

                                Working...
                                X