**3. Results: Phase Shifter Prototype Performance**

The design has been validated through a prototype. Figure 6.a yields some details of the shifter prototype, and the TRL (transmission-reflection-line) kit for the calibration prior to the measuring process. The measurement results are provided in the rest of the figures in Figure 6, in order to validate its proper functioning and performance.

(**a**) **Figure 6.** *Cont*.

11

**Figure 6.** Prototype results of the 18 GHz phase shifter: (**a**) prototype and TRL kit (central image corresponds to a microscope zoom view); (**b**) arg(S21); (**c**) Δarg(S21) (related to min arg(S21) value, 0 v); (**d**) |S21|; (**e**) |S11|.

As it can be noticed, there is a good agreement between the design performance and the prototype measurements. The insertion loss levels of the prototype are higher than the expected ones in the design. There exists a variety of possible causes for that: parasitic components of the varactor, conductive epoxy resistance, and roughness of the shifter printed line limits. For the sake of completeness, Figure 7 compares the simulation phase results with the experimental ones (Figure 7a), and provides an insight of the linearity of the device in terms of phase/voltage relation (Figure 7b).

**Figure 7.** Phase shifting performance: (**a**) comparison between simulation and measurement phase results of the 18 GHz phase shifter; (**b**) phase/voltage relation (the dashed line is the linear reference and the grey zone is the one that provides at least 360◦ variation).

Notice that the simulation results in Figure 7a are in terms of capacitance and the experimental ones are in terms of voltage. It must be noticed that there is a non-negligible difference between pairs of simulation and experimental curves. However, the phase range of variation is preserved and the phase shifter prototype is fully functional, only needing a voltage/phase conversion table. In addition, it must be pointed out that, although the phase/capacitance relation is not linear, it is compensated by additional non-linear behavior of the capacitance/voltage relation. It is worth mentioning that, although the device provides a quasi linear phase/voltage response at the central frequency, this linearity is strongly degraded at the band limits, as it can be seen in Figure 7b. However, since the prototype provides a complete phase shift of around 600◦, part of this possible variation can be sacrificed in order to increase the linearity of the device, as detailed in Figure 7b, where the grey region is the one that provides at least 360◦ phase variation.

For the sake of completeness, the next table (Table 2) provides a comparison of the performance of our device with the previously referenced literature works [10–15].


**Table 2.** Performance comparison of our device with the previously referenced literature works.

#### **4. Conclusions**

This communication presents the design and validation through prototyping of a microstrip phase shifter at 18 GHz for K band applications, with electronically controllable phase shift. The design is based on the use of loaded 3 dB/90◦ couplers. The performance of the reflective loads is improved due to the use of non-sequential impedance transformers. This design is particularly useful for reconfigurable phased arrays such as transmitarrays or reflectarrays, being fully integrable in such designs because of the reduced footprint. Furthermore, the shifter architecture is suitable for miniaturization, increasing the permittivity of the substrate. The dynamic phase range of the phase shifter is higher than 600◦ for the entire shifter bandwidth (>1.5 GHz). This phase range is higher than a complete phase turn of 360◦, which allows the introduction of phase wrapping. The device losses are lower than 8 dB.

**Author Contributions:** M.T.E., A.P.-C., and A.A.-A. designed and simulated the phase shifter. P.E., C.S.-G., and P.P. manufactured and measured the prototype. A.P.-C. and P.P. wrote the document. J.F.V.-V. and P.P. supervised the whole study. All the authors participated in revising the article.

**Funding:** This work has been partially supported by the TIN2016-75097-P, RTI2018-102002-A-I00, and EQC2018- 004988-P projects of the Spanish National Program of Research, Development, and Innovation and project B-TIC-402-UGR18 of Junta de Andalucía.

**Conflicts of Interest:** The authors declare no conflict of interest.

### **References**


© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
