Diagnosis of a 9:1 transformer from NanoVNA plot – part 2

Diagnosis of a 9:1 transformer from NanoVNA plot discussed an example measurement of a 9:1 transformer on a binocular ferrite core. These are often recommended for use with Beverage antennas on 160 and 80m bands, and this was the maker's application. In that article, I hinted that the core might not be #73 as the maker thought, or wished.

This article reports measurements of a 9:1 transformer wound on a Fair-rite 2873000202 (#73) binocular core. The pic above shows the test fixture.

The measurements will be used to calibrate a Simsmith model.

Above, a sweep with open circuit (OC) termination reveals a self resonance at 4.1MHz. This will be modelled as an equivalent shunt capacitance of 66pF on the primary side.

Above, a sweep with short circuit (SC) termination gives an approximate total leakage inductance referred to the primary of 46nH.

Above is a sweep with 450Ω termination from 1.8 to 10MHz, InsertionVSWR is inside the VVSWR=1.2 circle. (The termination pictured is two 1% resistors of 820 and 1000Ω in parallel, nominally 450.5Ω.)

Above is the calibrated Simsmith model including a comparison to measured impedance with a 450Ω load. T1 and T2 are of zero length, so no effect on the model.

The model predicts slightly higher InsertionVSWR (magenta curve) to the measured (blue curve). That is not surprising, ferrite tolerances are quite wide.

Predicted loss is shown as the dotted blue curve, it should be quite acceptable for the receive application where this design is recommended (eg from W8JI). The power in the load for 1W available is shown as the dashed blue line.

It is important to keep the application in mind, and the relevance of performance wrt some nominal impedance when the connected load may depart from that value significantly. The relevance of the core loss figure also depends on the application, if there is system gain to spare, the loss may be accommodated with little impact on performance.