POTY and return loss plot

  • Hi Group

    I've build a POTY from Hans - I'm pretty sure, that I've done it right ;-) please no comments on my soldering skills on this type of hardware.

    But the return loss plot is a little bit strange.

    Anyone has a hint, what I've done wrong - the plate distance is 3mm ...

  • Looks OK to me - the spurious resonance around 1.9 GHz is the reflector resonating and that varies depending on its size. The other resonance looks to be in the right place. It could be a little better match but there is the expected double dip. A little light bending will probably help improve the VSWR if it is a problem for your PA. The two dips are a little high - the aim is to have them evenly either side of the operating frequency.

  • Thanks OMs for ultra fast reply!

    The analyser is directly coupled to the feed with an SMA M-M - when I use a 1m Hyperflex the 1.8GHz dip is still there.

    Fine tuning will be done by bending - I'll try it later, when I've installed it at the dish.

    But when I check the origin, I also have this double-dip but far more deep >30dB compared to my week 11dB - but maybe as proposed fine-tuning will help.

    Thanks again!

  • But when I check the origin, I also have this double-dip but far more deep >30dB compared to my week 11dB - but maybe as proposed fine-tuning will help.

    I am not sure exactly what you mean but the dip ought to be 16-20 dB, no more as if it is it means both resonances are the same frequency and that's not s good for circular - (still pretty good though). The main thing is your amplifier is happy and you are a good signal into the satellite.

  • Like G0MJW said and having built 23 pieces myself: The plot looks good.

    The sharp dip at 1880 MHz is due to the resonance of the reflector.

    Perhaps you may tweak a little to have both 2.4 GHz dips such that 2400 MHz is centered (and thus not having the least RL) for truely circular polarization.

    16 - 20 dB RL is good enough.

  • It is not clear to me in which ways the patch can be "tweaked" ?.

    Mine is waiting to be soldered soon.

    73 Ed

  • Tweaking is possible by pressing sides/corners of the patch towards the reflector (so decreasing the distance somewhat) when the max RL frequency is too high, or pulling the sides/corners of the patch from the reflector (so increasing the distance between patch and reflector somewhat) when the resonant frequency is too low.

  • What is very important is that the distance between the plates is 3.0mm (measure it e.g. with a 3.0mm drill) and that the screws to mount the flange are not protruding. Personally I also remove some teflon from the flange. Success!

    PA1EW Regarding your question concerning the 'double dip' of a helix: "No". A helix can be considered as a 'traveling wave' antenna.

    The patch can be considered as two antennas: one somewhat too short (high frequency, therefore capacitive), the second somewhat too long (low frequency, therefore inductive). We learned LEICIE, and the trick now is to have each antenna lag or lead by 45° , so that the total becomes 45 -(-45) = 90° ( or -45 -(+45) = -90° ) and therefore producing a cork screw, aka circular polarization (CP).

  • Finished soldering my patch and RL measurements. After the first results (spot-on frequency) I tried a few times to resolder the patch radiator to obtain the "magic double-dip", but this I could not get in any way. When re-checking the information I realised that it may not appear at all (?). The main thing seems to be that the RL dip is not very deep, but instead a broad, shallow one. This was achieved with abt. 19dB so it should produce proper LHCP.

    Hope that my Kuhne blackbox returns soon after modification so I can continue with on-air tests.....

  • G0MJW Yes, it is silvered indeed. Just a very thin layer applied with rubb-on chemical process from a bottle with (toxic!) liquid containing silver. I used this also on my helix copper wires. It is more for the "looks" and perhaps does a good job regarding skin-effects.

    73 Ed

  • DD0KP Thanks Heiner, I am very curious how my patch will perform. Once I have the Kuhne up-converter back I can start testing ( and getting back on-air!).

    I think when your helix was properly built it should work as good as the patch. Anyway, whishing you good luck with the new patch-feed!. I now can see some pro's and con's and some depending on preference as well.

    73 Ed

  • PA1EW Concerning the (theoretical) double dips, I read a thread somewhere on the BATC forum that some equipment are not able to resolve this double dip.

    The trick then is to tune/tweak the response so that a shallow (broad) dip of ca. 20 dB occurs at 2400 MHz. If you've sharper dips then the two resonances coincide and and affect the axial ratio (read, it'll produce CP elsewhere).

    Your response looks fine to me, so I am curious how it'll perform compared to your previous feeds.

  • While doing some shopping this afternoon I philosophized a little why these dips do not appear on some equipment (see BATC link in posting above).

    Could it be that 'scalar equipment' is not able to visualize the dips, while 'vectored equipment' is?

    In other words, while the 'balloon' circles around Z = 1 (normalized) in the Smith Chart, scalar equipment cannot derive the sign of 'j', while a real VNA can?