The type of  N connector that i use on the array (with the white coax). It's abt 8mm longer than standard N.... Influence on the bazooka resonance ?

The connectors again and the schematixs of the bazooka

I have changed the coax to a KX13 with standard connectors + bazooka but no plastic protection (test made on 1 yagi only). SWR lowered a bit, but still not 1.1:1. The problem is that i can't use standard N connectors on the Gedelex, and i can't use KX13 because of loss. I have to keep the gedelex with its N ...

So i tried this :

Bending the upper side of the dipole towards the reflector makes the swr decrease on 432. SWR not changing a lot on 432 when changing coax length (1.1:1 to 1.3:1 max). Before (without bending) SWR was reaching 1.6:1 on 432 when changing coax length.

Upper side of dipole bent towards reflector.

Hard to say where i'm really at the best return loss, as value is still changing with the coax length.....

Here are the measures (MFJ 269) i have in shack with single antenna, bent dipole.

 

QRG RL with the phasing coax alone SWR RL with the phasing coax + extra length 1 SWR RL with the phasing coax + extra length 2 SWR
428 18 1.2 12 1.6 17 1.2
429.6 12 1.6 26 1.1 14 1.4
431.7 28 1.0 16 1.3 20 1.2
432.1 22 1.1 18 1.2 23 1.1
433.1 17 1.3 19 1.2 16 1.3
434 19 1.2 13 1.5 14 1.4

I guess i'm a bit too low on qrg...but am i going the right way and what else can be done to get same results with any coax length. Which qrg should i tune for best RL on each antenna ? Do i need to tune a bit higher than 432.1 ? I hope the F/B and pattern are not much affected bending the dipole...