OTR goodie: Old copy of The Radio Amateur's Handbook online

I enjoyed reading this old edition of the Radio Amater's Handbook tremendously. It was published at a time when spark gaps and vacuum tubes were used simultaneously. Because the technology is so old, most people don't know how the spark-gap transmitter works. Here is an expanation:

In an electrical arc, as the current increases, the ionization of the gaseous substance in the vicinity of the arc increases, decreasing the resistance of the arc, so that the voltage across the arc decreases. The E-I curve of the arc goes in the negative direction, producing a negative AC resistance. The negative resistance of the arc counteracts the positive resistance of the tuned circuit, causing the arc, with its tuned circuit, to oscillate. The oscillation ocurs at the point in the E-I curve where the resistance is zero. It happens that any vacuum tube or FET oscillator can also be considered to be a negative resistance oscillator.

Ermi,

Your post was most interesting to me since I never realized that the spark gaps worked as negative resistance devices. My take was that they generated a wide spectrum of RF junk which was filtered by the tank (sort of like a class C amp).

I do not take issue with what you wrote but I would like to ask if you know of any work with glow tubes (also negative resistance devices) which might have been tried?

Neil

I've seen glow tubes used in RC relaxation oscillators at audio frequencies. I don't know how high in frequency they can go. I have not heard of anyone driving an LC circuit with a glow tube. I suspect that the response time of glow tubes is not fast enough to operate at higher than the lower radio frequencies, but I don't know that for sure.

I should point out that there are different kinds of spark gaps. There is the original spark gap used by Marconi, which produces wideband damped oscillations. Then, there is the far more stable spark gap of the arc lamp, which was used by Frederick Collins, who is the author of the edition of the ARRL Handbook discussed in this thread. In 1899, Collins, as he describes in the Handbook he wrote, used a microphone to modulate the spark of an arc lamp. Collins's transmitter produced waves that were nearly sinusoidal, and had low enough noise to allow AM to be transmitted.

In the arc lamp transmitter, there is a delay in the change in ionization due to a change in the current through the arc. So, the arc lamp AM transmitter could be used only at the lower radio frequencies. This was fine during the early days of radio, because groundwave was used for long-distance communications, and groundwave transmission is best at the lower frequencies.

Radio amateurs were at one time restricted to frequencies above 1500 kHz, which were considered to be unsuitable for long-distance radio communications. Thanks to the pioneering efforts of these radio amateurs, methods of achieving reliable long distance communications using skywave propagation, were developed.

Thanks for that great expose' on the topic. In my youth, I experimented with relaxation oscillators made with NE-2 type bulbs. They maxed out at about 10 kHz. Though they were negative resistance devices, the operation was on/off for the bulbs so perhaps they are not truly negative resistance oscillators in the sense of a tunnel or Gunn diode.

I asked since there were many very clever things devised in the early era of radio and thought perhaps someone invented a glow tube which would work at higher frequencies.

Neil

Glow tubes function on a similar principle as spark gaps. In both cases, conduction is caused by ionized gas. In the spark gap, the ionized gas is air. In glow tubes, the ionized gas is a noble gas, such as neon, at low pressure. Ignition is caused in both types of devices by supplying enough voltage across the electrodes. As the current increases, the ionization increases, causing the voltage across the electrodes to decrease, resulting in the AC resistance across the electrodes being negative. The negative resistance near the ignitin voltage is very high, and becomes less negative with increased current. As current increases, the negative resistance approaches zero. There is a delay between the current increase and the ionization increase, and this causes the maximum frequency of an oscillator using the negative resistance of a spark gap or a glow tube to be limited.

Because of the similarity between the spark gap and the glow tube, it should be possible to use a glow tube as an LC oscillator that produces sinusoidal waves. Yet, all glow tube oscillators are RC relaxation oscillators that produce triangular waves, and not LC oscillators producing sinusoidal waves.

I tried to find an E-I curve for an NE-2 bulb, but I had no luck. So, I measured the curve myself. I connected an NE-2 to a 120 kohm current-limiting resistor, and connected the bulb and the resistor, joined in series, to a variable, high voltage, power supply. The ignition voltage is 91 volts, and the voltage reduced to 66 volts when the current was 180 uA. Since the rated current is 500 uA, I did not use more than 630 uA in my test. Between 180 uA and 630 uA, the voltage across the bulb remained at 66 volts, making the AC resistance of the bulb practically zero in this region. Between just greater than zero, and 180 uA, the AC resistance is negative.

I did not have enough resolution in the adjustment of my high-voltage power supply to adjust for a stable current less than 180 uA. Perhaps some sophisticated, fast responding, current source can keep the bulb in the negative-resistance region. This would allow making a glow tube oscillator with an LC circuit.