DC-1GHz switch from a common toggle switch

Introduction

Professional-grade high-frequency switches are based on coaxial structures with centre conductors that can be physically moved by electromagnets. The coaxial signal paths allow operation up to microwaves. For most ham applications, the electromagnetically actuated switch is not as convenient as a manually actuated one. Products marketed for ham use are usually implemented from custom rotary switches.

Materials and method

The aim of this project is to create a high frequency switch out of a nondescript toggle switch meant for DC and AC 50Hz operation. Although the toggle switch is more commonly available in the DPDT format than SPDT, we chose the latter because we suspect that latter’s fewer contacts are favourable from the RF standpoint. The second (unused) switch in the DPDT may have parasitic capacitances that can degrade the first switch’s isolation at higher frequencies. Since the toggle switch’s internal structure and solder terminals present significant impedance discontinuities, it will be interesting to find out what is the highest frequency the overall assembly can support. The three solder terminals protrude ~4mm from the switch’s enclosure; their ends are directly soldered to the BNC coaxial connectors’ centre conductors without intermediary wires. BNC connectors of different body lengths have to be used so that their centre conductors can physically contact the switch’s terminals. The RF-tight housing is made by soldering PCB rectangles at their edges. The housing’s removable top and bottom sides are screwed to the threaded brass stand-offs at the corners.

 

Assembly details (box size 40x30mm)

Results

The loss is <0.4 from DC to 1GHz. The loss in the B path is ~0.2dB higher than the A path – likely caused by different BNC body lengths. We suspect that the loss is mostly contributed by stray inductances in the signal path; the path length of the solder terminal to the COMMON port’s centre conductor is ~10mm. The ripples in the loss response are probably caused by reflections at the discontinuities.

 Insertion loss test

The isolation is better than -25dB at DC-1GHz. The reducing isolation with frequency is caused by the parasitic capacitance between open switch contacts and also by radiated coupling.

Isolation test

The switch’s power handling is unknown. It has been used since 1998 at 5W TX in the 2m ham band. It might be possible to deduce the power handling from the toggle switch’s rating - 3A 250VAC and 6A 125VAC – but we did not try this.

Performance summary

Ideas for improvement

The high frequency loss can be improved by reducing the length of the discontinuities - e.g. the non-coaxial segment of the signal path.

(C) This work is protected by copyright and may not be reproduced in any form without the expressed consent of the owner.