Hartley on Air - 2
The first operation hours have been full of teachings on the 'Hartley' configuration transmitter. The proper feedback tap on the coil and the proper position of the antenna link are essential for a very good tone.
The FEEDBACK TAP determines the portion of the alternated voltage fed back to the oscillating tube. The greater this voltage, the higher the drive level. Beyond a certain drive level there is no power output increase, and lot of power starts to be dissipated on the triode control grid. At low feedback voltages, the oscillations start randomnly and are very much influenced by the coupling level to the antenna.
The ANTENNA LINK should take the most possible power from the tuned coil, but with the lowest deleterious influence on the frequency stability and the tone quality. A link coil should work only with the magnetic field component generated by the primary coil. Any stray capacitance between the main coil and the link deteriorates the oscillator behaviour. They can't be eliminated but can be reduced at a minimum if the link is shaped and positioned so that to stay the most possible close and confined in the vicinity of that portion of the main coil having the lowest voltage towards ground. The feedback tap is that region and so the link shall be placed in its correspondence.
Any change in the operating frequency, for the same given antenna, determines a variation of the load impedance that the oscillator sees through the link coil. The current that flows in the link (the 'secondary'), by the working principle of any electrical transformer, develop a magnetic flux with opposite sign of that developed by the main coil (the 'primary'). Therefore, the load, through the antenna link, tends to chocke the oscillations in the main resonating tank. If the load match is not correct, either the transferred power would be very little, or the oscillations do not start. So, for any ample frequency change, according to the antenna impedance variation law, both the feedback tap and the antenna link should be moved, and in general the number of turns in the link coil should also be changed.
In the picture it is shown where the tap and the link are located, for a reasonable performance between 3 and 4.5 MHz with the antenna in use and without any need of modifications.
The big chocke at bottom left of the picture is the initial plate chocke that has been chosen to decouple the oscillator from the power supply. It turned out it has too many turns and made the oscillator being unstable on a wider frequency range than what is currently. For relatively high coupling levels to the antenna, the main resonating tank looses its 'driving' effectiveness and the circuit becomes a 'TNT' oscillator, where the main resonating tank becomes the plate chocke + plate bypass capacitor to grodund. The 'spurious' oscillating frequency is a few kc/s and at the receiver a crowd of intermodulation products appear around the resonating frequency of the main tank, at intervals equal to the spurious generated frequency. It has been exclued and a much smaller, less critical, chocke is now used instead.
Now, this transmitter allows to make ordinary radio contacts through all its operating range (1.7 - 4.5 MHz), with a clean tone, even though very slightly and still influenced by the antenna wires oscillations in the wind ... T9 reports were achieved on 300 miles QRB radio contacts around 3.8 MHz, in the typical summer statics noise. The power is well beyond 5 W, on all the operating range, so plenty enough. In total, there are 10 components including the triode, and they appear enough to manage CW traffic reliably and with acceptable quality on average paths.
It is a real pleasure ...
Radioman, July 2016
Posted by Radioman | Permanent Link