A QRPP VXO Transmitter for  5 MHz

Contents:

1.   Features Summary
2.   Introduction
3.   Overview
4.   Variable Crystal Oscillator (VXO)
5.   Pin-out Diagrams
6.   Introducing 'Ugly' Construction
7.   Initial Testing
8.   Feedback from other constructors
 

Features Summary

This transmitter has been designed for use on 5 MHz and has the following features:
- 12 v operation
- VXO transmitter, with 2 kHz tuning range;
- key-click filter;
- 40 milliwatt output power;
- built-in antenna changeover.

Introduction

This transmitter is a resurrected project that was started in 2005 when I was living in Wales.  In April 2023, Phil, G4XVP mentioned to me his early experiments with a one transistor transmitter and this inspired me to re-visit this project.

Overview

This prototype transmitter uses a 2N2222 transistor as a Colpitts VXO (variable frequency crystal oscillator) to generate 40 mW on 5 MHz. 

Variable Crystal Oscillator (VXO)

Two switched crystals are used to provide access to both the UK QRP centre of activity (5.262 MHz) and part of the WARC allocation (5.355 MHz).  A 100 pF tuning capacitor provides a tuning range of approximately 2 kHz.   This little TX may be expected to draw about 9 mA at 12 volts.  The core of the tuned circuit should be adjusted to provide maximum RF out, consistent with quick oscillator start-up, and minimum 'chirp'.

Further to the interesting feedback from Kevin MW0KXN (below), I recently re-visited the values of the two potential divider capacitors in the oscillator.  I had originally specified 530 pF (base-to-emitter) and 200 pF (emitter-to-ground).  As in Kevin's observations, I also found the circuit operation to be insensitive to the values of these capacitors.  Nevertheless, the circuit diagram below now shows the values used in my latest version of the transmitter (August 2024) in which two 300 pF capacitors are used.

The crystals were obtained from the G-QRP Club, and the KANK 3334 coils are available on eBay.

Pin-out Diagrams

Introducing 'Ugly' Construction

Those with lots of experience of building homemade rigs will have no trouble building the transmitter into a smart box.  Those with less experience would, perhaps, benefit by starting with a simpler approach. This prototype can be constructed quickly on copper-clad board using 'ugly' construction techniques (see picture below).  It doesn't look very pretty, but that really doesn't matter - the important thing was that it is possible to use the prototype to test the design, and have those first QSOs!

Note the construction of this ugly construction workbench.  A large piece of copper-clad board has been screwed to a spare piece of shelving (chipboard, laminated with a decorative coating).  The front and rear runners use more copper-clad board to support the switches and sockets.  At the end of each project, the components that have been soldered to the main board (and each other!) are removed so that the workbench is ready for your next magical creation.

For more information about ugly construction techniques, click here.
 

Initial testing

My first listener report was from Rich, G0GGA in Hinckley, Leicestershire (118 km, my signals 329 to 459).  A good start!

Later, I worked Jon, GW0FJT in Llandeilo (100 km, my signals 529); and Rog, GW3UEP in Cardigan (150 km, my signals 539).  Neither QSO was scheduled, with both stations responding to my CQ calls. 

On 8th June, I worked two 5 watt QRP stations: Ken, GW4JGW in Swansea (100 km, my signals 559); and Colin, G3VTT in Dovercourt, near Harwich for my best DX so far (250 km, my signals 239).   Meanwhile, Norman, GM4KGK on the Isle of Lewis (770 km) was recording my 40 mW CQ calls that, at times, were just above his noise level.  You can hear one of Norman's four recordings here.

Then, on the morning of 9th June, something even more magical happens.  You'll have to understand that, by this time, I've gained the impression that a 5 watt QRP station has to be S8 or above for them to hear my 40 mW.  At 07:40 UTC, I hear G4EXQ calling CQ on 5.262.  He's not strong at 559.  Anyway, I give him a call and, after a slight delay, he comes back with my callsign!  A single call does it, and we're in contact!  We manage a full-format QSO and I learn that Andy is in Bude (100 km) and running 5 watts to a G5RV.  Andy receives my name, QTH, power, antenna and the fact that my TX is home made.  We are both astounded by this contact!  

But there's more!

Later that day, on the evening of the 9th June 2023, I put out more CQ calls on 5.262 Mc/s at 40 mW.  During this time, Norman GM4KGK, who had previously sent me a recording of my 40 mW signals, is in the shack working on something, but with the receiver still tuned to 5.262 MHz.  Norman hears my CQ calls, and reckons that a QSO might be possible.  So Norman responds to my CQ call!  There's a lot of QSB, but we make a proper QSO -- and I have a new 'best QRPP DX' in the log! The distance is in the region of 770 km (to be confirmed).

Happily, Norman made some more recordings of my signals, recorded during our QSO.  This one is of me sending Norman's report:
https://www.4alg.uk/radio_g/qrp/20230609_GM4KGK_QSX_RPRT.wav

And this one is of me sending my power:
https://www.4alg.uk/radio_g/qrp/20230609_GM4KGK_QSX_PWR.wav

Of course, all credit goes to the operating skill and the fine receiving set-up of my QSO partners.

Feedback from other constructors

From Kevin, MW0KXN

Kevin writes: 
"Good evening Steve.  Thanks for a fantastic QSO.  It was difficult copy for me with your 3 W, so I am very grateful to you for persevering with my 60 mW.    I am delighted to say that your QRPP VXO Transmitter for 5 MHz has worked a treat.

I didn't have a 120 ohm resistor to hand, so I used 100 and 50 ohm resistors on the emitter circuit in place of the 120 and 50 ohm resistors in your design; and I used 400 pF and 200 pF in the voltage divider.

I have tuned the coil by eye to the cleanest sine wave, and was getting 78 mW by my calculations. The 2nd harmonic was about -30dB, and the 3rd harmonic about -40dB.  I have added a low pass filter and the 2nd harmonic is down to about -47dB at the expense of a 1dB insertion loss and am now getting about 62 mW RF output.

The output appears to be pretty insensitive to the capacitor values in the voltage divider.  I have tried adding a 200pF capacitor to make the voltage divider 600 pf and 200 pF, and then tried 400 pF and 400 pF with no noticeable difference to the output power or start-up."

Kevin has kindly provided the following photos, including details of his low pass filter.