The oscillation of R33

The May 2015 issue of Fortean Times (a periodical I warmly recommend) has a fascinating article by Daniel Wilson about a type of radio interference known as oscillation, which afflicted radio broadcasting in the 1920s and 1930s, about which, I'm ashamed to say, I previously knew nothing at all.1 What's fascinating about oscillation is not the technical aspects, but rather the social ones, because it was a type of interference that listeners could create as well as experience as they were trying to tune in to a particular radio station, interfering not only with their own wireless set but any others nearby trying to listen to the same frequency. This led to oscillators becoming a social pest: they were told off by the press, by the government, and by other members of the public. They were even hunted down by radio detector vans (the start of a great British tradition). While many oscillations were accidental, a consequence of domesticating a technology which wasn't quite ready to be domesticated yet, it seems that others were intentional -- it was done to annoy other listeners, or at least that was the suspicion. (The trolls are always with us.)

The problem arose with a particular type of radio receiver, the regen or regenerative receiver. This had the advantage of high amplification, meaning that a regenerative radio set only needed as little as one valve (or vacuum tube) compared with half a dozen for tuned radio frequency receivers, the other main early alternative. So regenerative sets were cheaper and hence more common. But they were harder to use, because of the non-linear interaction between frequency and gain, particularly when tuning close to a strong signal. I must confess to not fully understanding how oscillation works, so it's probably safest to just quote Wilson here:

Oscillation occurred whenever listeners over-strained their radios to increase volume. There was no 'volume' dial per se, rather there was a 'reaction' dial controlling the relationship between two coil components that interacted electromagnetically. Volume and tuning sensitivity were tied together. Through the use of the reaction dial, an output coil inside the radio fed back its energy into an input could. In this way, the valve's input could 'hear' some of its output, augmenting both the signal volume and the inaudible radio tuning frequency. Beyond a certain threshold, radio feedback tones were produced, energising the aerial and transmitting a radio frequency back from the set.2

Oscillation occurred near, either side of, and not at, the frequency transmitted by a radio station, as it's effectively a beat (or heterodyne). This meant that oscillation was actually useful for quickly finding a station. First the tuning dial would be used to sweep the frequencies until the squelch of the associated oscillation was heard. Then it would be turned back and forth until the oscillation's pitch was lowest, which was at the station's frequency. Finally, the reaction dial would be turned down to allow the station itself to be heard. But in the process, the radio's own aerial became a low-powered transmitter, strong enough to be picked up by other radio sets (of any type) in the neighbourhood -- a particular problem in urban areas where everyone was trying to listen in to a popular programme, and an even worse one when they were tempted to oscillate further to overcome poor reception.

So there was a lot of policing of correct wireless etiquette going on. For example, the Exeter Gazette provided the following list of '13 Don'ts':

2. Don't forget that when you oscillate, you are spoiling the pleasure of hundreds of other people.
3. Don't forgot that when your receiver is oscillating you cannot get the best results from it.
4. Don't sacrifice quality of reproduction for volume of noise.
5. Don't attempt to get on two valves all that your neighbours say they get on three.
6. Don't deliberately oscillate your receiver in order to revenge yourself upon a neighbour who is oscillating; it only makes it worse for others.
7. Don't try to improve your tuning if your results are satisfactory.
8. Don't try to work a loud speaker from a single valve.
9. Don't search for distant stations by their carrier waves.
10. Don't forget that the B.B.C. issues an anti-oscillation pamphlet free on application.
11. Don't forget that when you oscillate, you are disobeying the regulations printed your license.
12. Don't forget that reaction should the last adjustment and not the first.
13. DON'T DO IT!3

The Hull Daily Mail recognised that some oscillation was unintentional:

It is always particularly bad in the autumn, because there are many people who return from their summer holidays and take up their hobby again. Some of them have forgotten how to tune in; others attempt to make use of batteries and accumulators that have been neglected all the summer, and broadcast sundry crackles and whistles for the 'benefit' of their neighbours. Each season, too, there are new listeners, whose knowledge of tuning is very limited, and who seem to take an unconscionable time in learning to 'twiddle the knobs' without emitting howls from their receivers.4

But alongside these September newbies, there was the far more insidious troll: the 'wild man of the ether':

He spends his time, apparently, in experimenting on lines of discovering the best way of creating the maximum amount of interference in the shortest possible time. When there is an item in the programme (which is fairly frequent) that he does not like, or which does not appeal to him, he tightens up his reaction coupling, twists his condenser knobs, and sends out a series of howls comparable only to the lamentations of Scrooge's Ghost.5

The Mail wanted the Postmaster-General, or the British Broadcasting Company (Corporation from 1927), to do more about the oscillation problem. But it was difficult. In December 1924 the BBC asked listeners whose enjoyment of 'the whole reception of the transmission of dance music from America' had been spoiled by what was suspected to be deliberate interference to report their experiences, so that the oscillator could be localised (the story was reported under the fantastic, if unlikely, headline 'ONE MAN ANNOYS BRITISH ISLES').6 The Post Office soon began looking for them directly, using 'Motorcars, fitted up with direction-finding sets [...] in order to locate houses where sets are being badly handled and causing oscillation', with the view of 'bringing test prosecutions against the offenders'.7 By 1932 this had evolved into a system of detector vans which would investigate complaints, issue warnings on a first offence, and cancel licenses on a second (effectively a fine of 10s., since that's how much a new license cost). Since only 'Several holders of wireless licenses -- one in London, one in Birmingham, and several in the North of England' were reported to have been fined in this way, it doesn't seem to have been hugely effective in catching oscillators.8 Technology ultimately solved the problem by the later 1930s, when cheaper tubes allowed the superior superheterodyne sets to become universal, and regens faded from the scene.

So what did oscillation sound like? Wilson, who among other things is a composer of electronic music, has created the following piece from oscillating vacuum tubes, explaining that 'it gives a rather exaggerated flavour of the kind of sounds that polluted the airwaves of the 1920s and 30s':

Observant readers will have noticed the photo of a damaged R33. This was taken after the breakaway incident of 16 April 1925, when the moored R33 had been blown from Pulham out over the North Sea in a gale. The skeleton crew on board managed to prevent a crash, and flew the airship back to Pulham and safety. But this daring feat was imperilled, so it was claimed, by the wild men of the ether:

The Postmaster-General's attention has been called to the fact that during the recent flight of the R33 communication between the airship and the Croydon and Pulham aerodromes was seriously impeded by oscillation from wireless receiving sets. The consequences which might result from interference with communication with aircraft in danger are very grave, and the Postmaster-General warns the owners of wireless receiving sets that the use of reaction to such an extent as to cause interference is an offence which directly contravenes one of the conditions of their wireless licences and renders the licence subject to withdrawal.9

This suggestion that oscillation posed a danger to aircraft safety is fascinating, but it doesn't actually seem to have been a widespread concern. At least, I can find few further references to the problem. But unsurprisingly, the R33 incident was used in yet another attempt to police the behaviour of wireless listeners, this time by the Tunbridge Wells & District Wireless Society:

we hope that all readers noticed the report in the papers concerning the oscillation that was experienced by the authorities who were trying to keep in touch with the airship R33. The report stated that the messages were greatly interfered with owing to amateurs who were endeavouring probably to receive those messages. These constant reports of oscillation are very regretable [sic], inasmuch as they may induce the authorities to prevent the use of reaction by the public at large, which would be a great drawback to the amateurs concerned, who would only have themselves to blame.10

I haven't done much more than partly recapitulate Wilson's article using examples I've found in BNA, so it's definitely worth checking out the source: Daniel Wilson, 'Rogue oscillators', Fortean Times 327 (May 2015), 38-44.

  1. Reading it I was put in mind of an equally fascinating Fortean Times article about something else from my period I knew nothing about, A Victim's bizarre account of persecution by ventriloquist stalkers, Crook Frightfulness (1935). Turns out Wilson wrote that too. 

  2. Daniel Wilson, 'Rogue oscillators', Fortean Times 327 (May 2015), 40. 

  3. Exeter Gazette, 16 June 1926, 4

  4. Hull Daily Mail, 19 November 1926, 4

  5. Ibid. 

  6. Courier (Dundee), 12 December 1924, 5

  7. Tamworth Herald, 23 May 1925, 2

  8. Nottingham Evening Post, 13 February 1932, 1

  9. Aberdeen Press and Journal, 22 April 1925, 8

  10. Kent & Sussex Courier, 15 May 1925, 5

CC BY-NC-ND 4.0 This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License. Permissions beyond the scope of this license may be available at

5 thoughts on “The oscillation of R33

  1. Neil Howlett

    'Crook Frightfulness' has echoes of the marvellous Flann O'Brien's wonderfully developed tales of the WAAMA League's hordes of unemployed ventriloquists.

  2. ajay

    "It was by a continual and personal badgering of the Chief Engineer at 2LO on the question of "Why is Oscillation and How is it Done?" that his lordship incidentally unmasked the great Ploffsky gang of Anarchist conspirators, who were accustomed to converse in code by a methodical system of howls, superimposed (to the great annoyance of listeners in British and European stations) upon the London wave-length and duly relayed by 5XX over a radius of some five or six hundred miles." - From Dorothy Sayers' "The Article in Question", published in "Lord Peter Views the Body", 1928. (I am dubious that this would work.)

  3. Post author


    I've not read Flann O'Brien, though I do seem to know a few of his fans!


    Very interesting; I wonder if Sayers was parodying a specific paranoia or just the tendency of some in the Establishment to see plots everywhere. Wilson makes the point that it wasn't possible to make a melody of any sort with oscillation; but I don't see why you couldn't do Morse (or some other code) simply by varying the length of time of the interference (short, short, short, long...)

  4. ajay

    You could certainly transmit signals by oscillation, but I thought the dubious bit is "duly relayed by 5XX over a radius of some five or six hundred miles". Sayers seemed to think that the way 5XX worked was by listening in to what was happening on the London wavelength (2LO) and then rebroadcasting it at a different frequency and much higher power - in other words, 5XX was simply what we'd call a rebro station. So anything on 2LO's freq got automatically forwarded on by 5XX, whether it was actually news broadcast by 2LO, or howls broadcast by someone else (such as an oscillating anarchist).
    I wasn't sure this was the case. But thinking about it, it might well have been. 5XX did broadcast 2LO programming, and not from London - from Daventry - and so it must have had some link to 2LO - either it used a landline, or it actually did just rebroadcast exactly what it received.

Leave a Reply

Your email address will not be published. Required fields are marked *