Antenna surprize

Thread Starter

Janis59

Joined Aug 21, 2017
1,834
This is question demanding an answer.

Many occasions had seen the antenna construction drawings into internet what forces me to think about author competence. Would it be happen I have stepped in the same mistake (to believe ant written) again?

So, I find the site telling very interesting vertical quarter wave dipole. Standard in such situation is coaxial with lambda quarter multiplied with shortening factor about 0.95...0.98 with taken off braiding and santech copper-pipe down with length of other lambda quarter with cable genuine shortening factor. Cable down must contain at least a coil of some 5-7 turns or ferrite core over cable surplus coil wound of cable as it is. But this website told, it is possible to feed the lower part straight from below, thus the copper pipe is unnecessary. Thus, lower part is fed straight from output of coaxial coil, whilst upper part is fed in the middle, where upper and lower parts comes near. Made it, switched to VNA and see the hedgehog`s coat - needles over needles (below the hundred by count) up to horizon down and up by frequency.

So, may it be that antenna construction belongs those class of ""non-working" or the problem is cased that antenna ends in the PL252 whilst VNA with SMA. Thus I had to have SMA to PL252 converter, then PL259 to PL259, then PL252 to antenna. Thus the plug to plug reshaping parts are probably too much and that what I see probably are reflections between badly made Chinese plugs?? Just I obtained those and this was first trial with those plug converters.

For to provide the standard test of plug-to-plug re-shapers quality I need a 50 Ohm ultra-good reference load into PL252 shape, whilst all mine are in shape of SMA. Thus I cannot to test it. So, what are thoughts do the antenna is faulty or testing cabling is faulty, or both?

P.S. Today the post office made a surprise, arrived one year ago paid parcel from China with right plug in one-piece. Result on the VNA is the same. Pictures attached.

PPS: find one similar antenna construction what is written in manner making easy to believe: https://www.vic.wicen.org.au/?page_id=1353 and found many similar as well, thus seems it isnt the an-alphabetical construction claim. But then what else?

PPPS: My target is 165 +/- 9 MHz range; thus BW=18 MHz (quite much).

PPPPS: Okay, You may suspect me blamed on calcs? My cable is LMR-400, v(f)=0.85, solid teflon, inner d 2.8 mm, outer 10.3 mm. Thus the lambda=1.82 and upper part (physical) length is 455 mm. Lower part (electrical) is 455*0.85=386 mm. Correction for Large Unknown 5% makes both parts 477 mm and 406 mm. Coil contains 5 turns on 30 mm PVC canalization white-pipe. That is all. Okay, may apply the nude wire V(f)=0.97 and 0.95 for plastic pipe covering the upper radiator - what gives out the 440 mm and 406 mm. Checked em too.

PPPPPS: Read about coaxial coil inter-turn gap has a strong impact - if gap-less turn to turn then VSWR is lower but sharper, whilst if coil is gapped, VSWR is not so low but much wider. Read my 5 turns on 30 mm makes coil resonant at 136 MHz what is more than demanded 5-6% under the operating frequency. No, may not blame the coil.

PPPPPPS: at last decomposed the antenna with idea that probably cable bending place where it becomes a coil was too sharp and probably double braiding one layer had been broken. Made a hari-kiri to cable. No, it wasn`t, however there was about 1 mm deep compression crater in the teflon insulator, thus some minor reflection may take a place.

PPPPPPPS: had an idea, this antenna demands a rather long cable about 20 meters, thus I rolled it to the coil on the shoulder. Now I de-coiled that maximum straight and picture wasnt changed not for a yotta. Still in my BW=18 MHz stays 3 waves with R=5 Ohm to 20 Ohm; X=-125 Ohm at all frequencies and VSWR 6...60. Catastrophic.
 

Attachments

Last edited:

Thread Starter

Janis59

Joined Aug 21, 2017
1,834
Hi!
Yaakov; Crutschow, do You have nothing to add here helpful?

External view with upper plastic tubulus taken off - see in the attachment. VNA measurer attached in the far end.
 

Attachments

Last edited:

Ya’akov

Joined Jan 27, 2019
9,071
Hi!
Yaakov; Crutschow, do You have nothing to add here helpful?

External view with upper plastic tubulus taken off - see in the attachment. VNA measurer attached in the far end.
I am having a bit of a hard time following your description, but is that made of PVC pipe?
PVC has a permittivity of something like 4 and can have noticeable effects on antenna tuning.
 

Thread Starter

Janis59

Joined Aug 21, 2017
1,834
Hmmm, the lower part indeed is PVC, but the 80% of antenna length I took it away to be able to see and touch every part. Thus, probability is small. Especiqally, because PP and PE have rather similar epsilon values.

Last experiment today, I straightened a coil, thus the X was reverted to stay between -120 and +120 Ohms via the zero in some 5 points in span of that 155-175 MHz, up and down, up and down while R=0.2 to 5 Ohms. So, seems the demolizing all around capacitance comes from the coil, BUT still stays the question what creates the multiple resonances distant about 3 MHz one from each another.
For better understand the construction, let read the first reference, that to vic.wicen.org
And anycase, thanks a lot for at last trying to immerse in my problem.
 

Ya’akov

Joined Jan 27, 2019
9,071
Hmmm, the lower part indeed is PVC, but the 80% of antenna length I took it away to be able to see and touch every part. Thus, probability is small. Especiqally, because PP and PE have rather similar epsilon values.

Last experiment today, I straightened a coil, thus the X was reverted to stay between -120 and +120 Ohms via the zero in some 5 points in span of that 155-175 MHz, up and down, up and down while R=0.2 to 5 Ohms. So, seems the demolizing all around capacitance comes from the coil, BUT still stays the question what creates the multiple resonances distant about 3 MHz one from each another.
For better understand the construction, let read the first reference, that to vic.wicen.org
And anycase, thanks a lot for at last trying to immerse in my problem.
I am trying to follow, when I get the chance I will review all this. I need to be less distracted.
 

Ylli

Joined Nov 13, 2015
1,086
Looks like what you are trying to do is often called a "Flower Pot Antenna". Going to be tough to get 18 MHz of bandwidth out of it. A J-pole might do it for you - For maximum bandwidth, use 1/2" or even 3/4" copper pipe and pick a design that has the greatest spacing between the main section and matching section.
 

sagor

Joined Mar 10, 2019
903
With a VNA, you have to calibrate with the cable you use, at the antenna feed point - first. What you may be seeing is the reaction of the feed coax feeding the antenna. It is not clear how you are doing this, so that is just a suggestion.
 

Thread Starter

Janis59

Joined Aug 21, 2017
1,834
For sure it isnt the flower pot, as the flower pot have part punching outward with precise diameter and length, whilst here are just two lambda fourth radiators and coil for symmetrisation between assymetric cable and symmetrical radiators. I recognize with flower pot some construction like this http://vk2zoi.com/assets/flowerpot-4.png however thank to You I see some authors surprizingly name in the same label even any construction with "bubble" in the middle (what for my imho isnt much correct). So, thank to You I have broader chances to read about it and probably find sth about. Thanks for that.

If You speak about copper pipe down and feed point for both radiators in the middle of construction, it may stand the last save point for me, however copper pipe means I shall not able to use a planned plastic pipe, it must be larger and more ugly and I must grinding a special adapter for it, so... its the last what I plan to do.

Rather question here stands WHAT is giving that damn large capacitance in the system minus 125 reactive Ohms if I have an inductance with plus signed amount of reactive Ohms? And second and no less intriguing question,m what gives a comb-like reflections there, staying in physical distance 300/165-300/168=32 mm? Simply I dont recognize any reflection-able elements be so near in all the construction.
 
Last edited:

Thread Starter

Janis59

Joined Aug 21, 2017
1,834
[QUOTE="sagor: With a VNA, you have to calibrate with the cable you use, at the antenna feed point - first. What you may be seeing is the reaction of the feed coax feeding the antenna.
[/QUOTE]
Think this may happen to be very near to "warm". Indeed this cable is bit new for me and damn long, so it may hide a well bunch of surprises. Or not. Shall check it asap.
 

Thread Starter

Janis59

Joined Aug 21, 2017
1,834
RE: Ylli: didnt knew that pipe diameter have impact to BW. Thanks, very useful knowledge, however in concrete construction any copper pipe is strongly unwilled, yet not completely unimpossible.
 
Last edited:

Thread Starter

Janis59

Joined Aug 21, 2017
1,834
Some more questions went around with silence in hundreds of webpages about it:
1) Does it really for upper radiator must not be applied the cable velocity factor, if source of velocity loss inherrently is insulation. Or indeed is right those publications saying there ought be applied naked metal factor near to 1.0?
2) Does the outer piping effect ir indeed 0.95 if already there is inner "piping" aka central lead insulator, thus the effect must be far less notable, isnt?
3) I found one reference stating that symmetrizing coil must be calculated with extreme accuracy, if it is bit too large or to small, its catastrophic. I previously understood it just must be larger by 6% of frequency and thats all. So, what to believe on?
 

Thread Starter

Janis59

Joined Aug 21, 2017
1,834
Just received answer of local antenna authority: ""For measurements with VNA, the absolute reliability is only when connected directly at the radiator or, in the worst case, after a 1/2L coax. Through a longer cable, if the 50 ohm active (non-reactance) load cannot be handled, you can see all kinds of miracles ...""
 

Ylli

Joined Nov 13, 2015
1,086
Just received answer of local antenna authority: ""For measurements with VNA, the absolute reliability is only when connected directly at the radiator or, in the worst case, after a 1/2L coax. Through a longer cable, if the 50 ohm active (non-reactance) load cannot be handled, you can see all kinds of miracles ...""
If you calibrate at the far end of the cable, the VNA can (within reason) compensate for the cable. If you count on using a half wave cable, then you can make a measurement only at one frequency as the cable is a half wavelength only at one frequency.

Coaxial dipole, sometimes known as a 'Flower Pot' antenna:
Screenshot 2021-04-29 095822.jpghttps://g6ohm.webs.com/antennadesigns.htm
 

Thread Starter

Janis59

Joined Aug 21, 2017
1,834
Serie No 2
Made an AIS antenna according this picture below. In the source (https://currents.bluewatercruising....ensive-easy-high-performance-vhf-ais-antenna/) was said that both dipole shoulders must be 41 cm and 40 cm for braiding side. Made it. While in no inside the rope measured resonance - VSWR about 5, R-18 Ohm, resonance farr off. After shortened lengthes to 36 and 37 cm (V(f)=0.59!!?? for RG58??) got prefect resonance and VSWR=1.12. Then fitted into rope and VSWR again 4 and far off resonance. After reversing length back to the 40 and 41 cm got perfect minimum for resonance and VSWR=1.15 with R-45 Ohms. Tried to put one 9 turn coil with no impact at all, and ferrite ring having no impact as well. Cabling is not hand-sensitive, why - dont understand as the radiators are symmetric whilst cable assymetric, but okay. Then half of day tried to pull it tied in the masthead between crossbeam and vant. Measured - VSWR is 1.46 at R=45 Ohms, but still near the perfect resonance. So, driving back from shore to home meditated, probably I mixed construction with braiding up and central wire downward? Next weekend shall try to change up-down. Or problem is caused by mast and vants draining energy inspite them are well in the 45 degree angle but so near as one meter for 2 meter wavelength? So, what may went wrong at installation?
1622471741596.png
 
Top