Binomial impedance transformer

[kellogs]

On this paper, a couple of questions:

1. On page 2 just under Fig. 2:

Actually, the load impedance margin in a broadband antenna usually keeps a ratio larger than 3:1. For that reason
the uniform multi-section transformer is discarded while the binomial or Chebyshev types are considered

Is there a knowable relationship between the number of sections in the transformer (or other parameters) and the achievable impedance ratio ?

2. On page 3

III. RESULTS AND DISCUSSION

[...]

Fig. 4 shows the measured balanced dipole impedance without the balun in a band from 200–800 MHz. From that figure it can be seen that the
mean value of the balanced antenna impedance is around 150 ohm (150 +/- 20 ohm)

Fig.4 on the right shows a rather scattered Re an Im on the Smith chart. It seems to me that only |Z| is taken into account when designing the proposed transformer. Does the phase angle of Z have any effect on the transformer ?

3. Besides less bandwidth, what other drawbacks are there when employing a 1-section transformer instead of a 2-section transformer ?

Thank you