Topband: Phasing 2 inverted L question

[Eduardo Araujo]

Hi everyone, I have an inverted L on 160 with 15 mts of vertical side over a good ground. It measures Rin 15 ohms.
I am planning to install another one a 1/4 wave away and phase them with 90 degrees difference looking for gain in broad angle rather than good f/B ratio.
I have no previous experience on phasing arrays. Reading ARRL antenna book and ON4UN book, I am a little confuse about how to proceed in using coaxial as phasing lines.
I wonder what could it be better:

1 - Make the elements longer and use a capacitor to adapt them to 50 ohms 
individually with the other element disconnected, then measure the combined element resultant Z with both elements coupled and start the phasing line length calculation from that point (ON4UN SW), or

2 - Make the elements longer and use a capacitor to adapt them to 50 ohms 
individually with the other element disconnected, and use 1/4 wave vertical line length examples in ON4UN book of 90 - 70 - 90 or 84 - 71 - 84, or

3 - Measure the combined resultant Z without any Z transformation at the 
antenna bases, and start the phasing line length calculation from that point.

As I am planning to use 9913, I think that without any Z transformation at bases, I might have a higher SWR on phasing line with a little increase in losses and even more phase shift. 
Finally, as I will not have more than 20 mts of coaxial to the connection point of the 2 verticals.

4 - Would it be better to do the final match to 50 at the phasing line input point or at the shack end? 

Any suggestion, practical recommendation or consideration will be highly 
appreciated for this first attempt.

Many thanks in advance '73...... Eddy, LU2DKT