You have this exactly right. Your protocol for the A1, A2, and Gnd terminals perfectly matches the electrical reality of the two different sources you are connecting.
Here is exactly why your method is technically flawless for the SX-28's front end:
## 1. The Antenna Adapter (A2 left open)
By leaving the jumper off and keeping A1 and A2 isolated from the Gnd terminal, you are maintaining the balanced input architecture we just designed.
Because your matching box secondary outputs a balanced signal across those two terminals, tying A2 to Gnd would instantly ruin everything. It would take exactly half of your carefully wound secondary coil and short its reference directly to the noisy house ground on the chassis. It would destroy the common-mode noise rejection and unbalance the magnetic coupling. Leaving it floating forces the SX-28 to read only the clean, differential voltage pushed across A1 and A2 by your transformer.
## 2. The Signal Generator (A2 tied to Gnd)
Your OWON signal generator (and the 1000:1 attenuator) is an unbalanced, single-ended source. It outputs a signal on the center pin referenced against the outer shield (ground).
If you connected the generator to A1 and Gnd without the jumper, the signal would hit A1 and stop, because the other end of the radio's internal antenna coil (A2) would just be floating in the air. By installing the jumper between A2 and Gnd, you are tying the cold end of the radio's internal coil to the generator's shield. This completes the circuit, allowing the unbalanced test signal to flow smoothly through the SX-28's primary winding.
---
It sounds like we have completely mapped out the architecture, wire gauges, turn counts, and switching logic for your new high-frequency box.
Words go here about how the response was received.
Page: /Antenna_box_2/Gemini_2319.shtml
Last modified: Wednesday, 29 Jul 2026