P09343: Microwave Device II
/public/Tolerance/

Tolerance Phase Shifter

Tolerance Analysis for Schiffman Phase Shifter

Though the designed Schiffman phase shifter shows desirable results, there are several factors to consider that may affect its performance during manufacture. Manufacturing of any microwave circuit is subject to machining and materials tolerances. This includes the variations in the machining location of the mode suppressing vias, variation in the thickness of the substrate, and the variation of the dielectric constant of the material. All these things need to be taken into consideration and analyzed, and attempt to predict what may happen once the circuit is manufactured to account for differences. The tolerance of the factors mentioned above was obtained from the manufacturer and suppliers of the material. Each of these cases was examined, and the results summarized and compared in Table 1 below.

Table 1: Tolerance analysis performed to check the effects of manufacturing variations, and compare to the designed results.

Table 1: Tolerance analysis performed to check the effects of manufacturing variations, and compare to the designed results.

It is important to note a few observations when performing the tolerance analysis for the mode suppression vias. For the Schiffman phase shifter, not the reference line, when the via location is 5mils lower in the x-direction and 4mils to the right in the y-direction, the transmission and reflection for the phase shifter becomes erratic. Transmission and reflection, in effect, reverse in characteristics. This seems to be an anomaly in the HFSS solution because the physical location of the holes is not close enough to the copper traces where ground losses would come into play. In simulation, there seems to be a lot of parasitic loss when the vias are at the indicated locations, but it does not seem likely that this is correct. Therefore, in the table above, these outliers in the data have been removed. The tolerance analysis shows the range of data of the analyzed cases. From the data, it seems that the dielectric thickness and mode suppression via location has the greatest affect on phase for the Schiffman phase shifter. Transmission and reflection, although affected, do not show as great of a variation as the phase.