Recording Technologies and Strategies for British Radio Transmission Before the 2nd World War

Prof. Seán Street

The Bournemouth Media School

Bournemouth University

url: http://www.ukc.ac.uk/sdfva/sound-journal/street002.html

E-mail: sstreet@bournemouth.ac.uk

For .txt version click here.


The development of recording for radio in Britain between 1930 and 1939 was based on a series of responses to specific broadcasting requirements in the areas both of Public Service and competing Commercial interests. The three principle technologies - metal tape, coated film and disc - may be seen as complementary in that they answered differing needs at various times through the decade. The strategies which informed these systems were based on issues of quality, durability, editing capability and portability. These issues, and the control over broadcast material they imply, led ultimately to changes in the way radio was made, preparing the way for the journalistic needs of the Second World War, and the possibilities afforded by increased mobility and quality in the post-war era.




The evolution of recording for radio in Britain before the Second World War was a response to particular circumstances quite distinct from the production of gramophone records for public consumption. It embraced a varied range of technologies in an era when Germany was already developing the tape system (ultimately to become universal after the war), all based on the following criteria which individually or collectively affected both the BBC and its rivals, the commercial stations transmitting populist programming from the Continent. These were:

Quality of sound reproduction


The requirements of programming (duration, editability etc)

Issues of portability

Control of material

The purpose of this paper is to examine the technologies for radio recording between 1930 -39 and to explore the practical philosophies upon which such technical strategies were based. I shall also seek to examine how recording changed these philosophies. In order to do this I shall utilise oral history material by some of the pioneers of British broadcasting and recording, held in the National Sound Archive. I shall also draw on material from interviews and correspondence with the Conservation Manager at the British Library National Sound Archive, Peter Copeland, the BBC Producer and Recording Consultant Antony Askew, and Dr Michael Biel, Professor of Radio and Television at Morehead University, Kentucky. The paper will investigate the sometimes parallel development of the technologies vying for use as a broadcast medium.


Of these there were three:

1. Magnetic

2. Disc

3. Film

These three technologies found their practical realities in British broadcasting in the following forms:

1. The Blattnerphone/Marconi-Stille system

2. The Cellulose Nitrate Disc (The 'Watts' Disc)

3. The Philips-Miller Film System


I shall examine these methods of recording in turn, at the same time seeking to place them in the context of both BBC and Commercial Radio usage during the years from 1930 up to the outbreak of the Second World War.



The principle of magnetic recording was arrived at as early as 1878, when Edison patented a magneto-mechanical system using a sheet of steel. Seven years later Charles Sumner Tainter also developed a magnetic recording device, and in 1887 Wilhelm Hedic of The Netherlands created a system using a tape which contained magnetic particles.1 The following year, Oberlin Smith of Cincinnati suggested a magnetic recording machine which would use cotton or silk thread impregnated with steel dust.2

The real breakthrough however was made by a Dane, Valdemar Poulsen (1869-1942) in 1898, when he built his Telegraphone, consisting of a length of steel piano wire wound around a drum. The device was intended for the recording of telegraphy transmissions; an electro-magnet tracked along the wire to record and reproduce audio signals. Poulsen demonstrated his invention at the Paris Universal Exhibition of 1900. A number of these machines were made in the United States, but were not marketed; it would be nearly a quarter of a century before the emerging medium of radio made the development of such a discovery relevant.


A German engineer, Dr. Kurt Stille (1873-1957), created an improved version of Poulsen's invention in 1924, marketed as a dictating machine by the Vox Gramophone Company. Initially Stille - like Poulsen - used wire, later moving on to develop a metal tape. In 1925 Harold (later Sir Harold) Bishop and L.W. Hayes of the BBC's Engineering Department visited the Vox works in Berlin for a demonstration of the system, but found the fidelity to be considerably below the standard required for transmission.3 The idea of ' bottled programmes' - that is to say material which could be recorded and broadcast at a future date - was further debated in 1927 when the BBC became a Corporation.


The reason for its initially slow development - particularly within world organisations such as the BBC - was only partially a technical and financial issue, as Asa Briggs has written:

'Live' broadcasting was greatly preferred, almost on moral grounds, to recorded broadcasting: it suggested to the listener, 'this is it'. Suggestions were made also that at the other side of the microphone if artists knew they were being recorded and retakes would be made, they would give mediocre performances. The Talks Department, which insisted on scripts that it could vet, went further and made an art form out of the scripted Talk (with a self-conscious capital T) delivered live. The art lay in properly relating writing to performance: what was natural had first to become artificial before it would sound natural again.4

While the BBC could afford to take such a moral stance on the issue of recording, the competing commercial stations, recording sponsored programming in Britain for transmission from European broadcast sites, had an interest in developing recording techniques to the highest and most efficient standards of quality and durability. It is a further argument of this paper that this competitive element in the use of the technology of the time actually accelerated development. That said, when the first exploratory moves were made, they were by the BBC, and the subject under tentative investigation was a variant of the Stille system.



The Blattnerphone/Marconi-Stille System


It was Louis Blattner, a German settled in England since the turn of the century, who was to take up Stille's idea of magnetic recording on steel and develop it to broadcast standards. Blattner was for a time the Manager of the Gaiety Cinema in Manchester, and had the reputation of something of a showman, using early versions of his recording machine as an attraction, giving the public the opportunity of paying to hear their own voice. It is believed that the title ' Blattnerphone' originated with these demonstrations.5

In the late 1920's The British Blattnerphone (Stille System) Company Ltd. was formed and studios were set up at Elstree, although its first machines were manufactured in Germany. In 1929, the BBC's Controller of Programmes, Admiral Sir Charles Carpendale and Noel Ashbridge, recently appointed as Director of BBC Technical Services witnessed a demonstration of the machine, and were sufficiently impressed to agree to a years trial installation at Avenue House, Clapham. This was the headquarters of the BBC Research Department, formed originally from the former Marconi Company research group based at Writtle.


Subsequently the machine was re-housed in Room 66, Savoy Hill at a royalty of £500. The BBC Year Book for 1932 ( which covered the programme year from November 1 1930 to October 31 1931) claimed that:

In some ways the most important event of the year has been the adoption by the BBC of the Blattnerphone recording apparatus…6




The tape used in the original Blattnerphone system was steel, 6 mm wide and 0.08mm thick. A spool contained somewhat more than a mile of such tape, weighing 21 pounds. Travelling vertically through the heads at 5 feet per second this gave a playing time of 20 minutes. The device was cumbersome, driven by a dc motor, and its speed had to be regulated by observing a stroboscope and adjusting a sliding rheostat.7 Uneven tape speed was a problem which plagued the early Blattnerphones, and continued to be a problem when a second generation of ac powered machines appeared in late 1932. These however had the advantage for programme makers in that they were capable of recording 32 minutes on a reduced width tape (3mm), thus enabling half hour programmes to be recorded on a single tape.



The machine was found initially to be unsuited to music recording, although it was useful for rehearsal purposes, due to the advantage of immediate playback. In March 1932, while broadcast equipment was being installed in the BBC's new headquarters in Portland Place, the Blattnerphone - together with a second machine to enable continuous recording - was re-sited on the seventh floor of Broadcasting House. On 23 May - just eight days after the change of premises - it was set an unexpected challenge. Amelia Earhart came to Broadcasting House soon after completing her transatlantic flight, and at extremely short notice there came a request that she should be recorded. The machine had not been fully set up, and was untested in its new site. Nevertheless the recording was a success, and went a long way towards proving the Blattnerphone - and recording - as an adjunct to the technology of wireless broadcasting.



The motive for the development of recording for broadcast became more insistent in December 1932, with the inauguration of the BBC's Empire Service; the requirement for 'bottling' programmes for repeated transmissions across various time zones meant that demand for the machines was heavy, and maintenance became a problem. Inspite of this - and the fact that tape editing was extremely laborious, the metal having to be either welded or soldered in the process - there is evidence that the use of recording had begun to change the way programmes were conceived. In 1932 Pieces of Tape, a composite programme made up of items recorded during the year was broadcast. The following year came Stars in their Courses, an edited programme featuring Irene Vanbrugh, Fay Compton, Sir Frank Benson and Matheson Lang.8



In March 1933 the Marconi Company bought the rights to Blattnerphone. This was to see the start of a fruitful period of development, with a partnership between Marconi's for the mechanical design and the BBC Research Department for the electronics.9The BBC Year Book for 1934 contains an article entitled "The Application of Sound Recording to Broadcasting", in which is discussed the Blattner system alongside developments in disc recording. This will be examined later in the paper. It is clear however that at this time the BBC was thoroughly committed to the steel tape principle of recording:

It is probable…that whatever other [recording] system is used the Blattnerphone will serve as a supplement to it. 10


There is an interesting aside on blattnerphone recording in the 1934 novel Death at Broadcasting House by BBC producer Val Geilgud and Holt Marvell.11 A murdered actor's last radio performance is captured on blattnerphone tape, but even the police have to wait to listen to the evidence, "as the blattnerphone machines were in particular demand for the Empire Service". 12 The book also gives a sense almost of superstition with which the recording process could be regarded:

For a few seconds there was only the hiss of the running of the steel tape. Then a whining cockney voice, vibrant with passion, echoed weirdly through the darkened room. Almost furtively Caird looked round at the faces. Which of them, he wondered, shared his own feelings of horror - almost of incredulity - as they listened to the voice of a dead man…And although he had suggested the replay of this blattnerphone record of Parsons's murder, now that he actually heard it, he experienced both fear and disgust, combined with an overwhelming conviction that the use of such a method must be unlucky and might well be something worse…13



The BBC set Marconi's the task of creating a new generation of machines- now named the Marconi-Stille process in time for King George V's Jubilee in May 1935. By the end of the year there was a recording suite of six machines installed at Maida Vale, having outgrown the original premises at Broadcasting House. The Marconi-Stilles built in consultation with the BBC were extremely reliable; Edward Pawley, in BBC Engineering,1922 - 1972 states that in a year's running, in which 1,957 hours of recordings were made, the percentage of breakdown time was 0.33%, with none of this due to the machines themselves, mainly being attributed to tapes breaking at the joints.14




Direct Disc Recording


As has been stated, during the years 1930 -39 broadcast recording technologies tended to overlap, and while Blattnerphone recordings were used within the BBC, other systems were being developed both there and within the Commercial Radio sector. Indeed there is no evidence to suggest that the Blattnerphone/Marconi-Stille systems were used in British broadcasting other than by the BBC. This was not true of the disc and film systems. Peter Copeland, Conservation Manager at the British Library National Sound Archive suggests that these technologies should not be seen as competing but as complementary.15



In the early part of the decade the BBC had formed a partnership with The Gramophone Company to enable the recording on disc of historic events. One of the first to be so recorded was on 21 January 1930, when King George V opened the London Naval Conference in the House of Lords. There were two major problems as far as the broadcasters were concerned. The first was financial: the system used was the same as that for making commercial recordings; the pressing process cost £50.00 per hour. The second issue was one of time: it was a lengthy process, taking in excess of 12 hours to accomplish. 16This will be discussed later.

A further partnership was developed between the BBC and The British Homophone Company Ltd, enabling the cutting of soft wax discs. At a given moment a number of discs were set in motion together to record a programme or news item. The BBC Year Book for 1934 described the relative merits and problems attendant to this method:

A fine cutting thread is used in the tracking of the recorder giving approximately 150 revolutions to an inch of track, as compared with 84 revolutions in the standard record. This, together with slow-speed turn-tables running at 60 r.p.m., enables a 12-inch record to play for approximately 9 minutes. For immediate play-backs the Blattner system requires a little time (about half that taken for the recording) to re-wind the tape before running off again. The wax system is quicker than this, as the cutting head on a recording machine is changed for a specially designed pick-up which is tracked by the cutting mechanism and so can follow the grooves in the soft wax. 17



There were variations available within the system:


Event Duration


Grooves Per Inch

Up to 5 minutes

78 r.p.m.


Up to 7 minutes

78 r.p.m.


Up to 9 minutes

60 r.p.m.



* Some depreciation of quality likely in the last minute of recording.18



The benefit of this system was that speculative recordings could be made at small cost, enabling broadcast decisions to be made and implemented quickly. This was particularly useful in respect of running commentaries, where highlights could be identified and less interesting sections rejected. The major problem was that one playback destroyed the soft wax recording. Thus it was necessary to make several waxes simultaneously, at least one being retained to make a pressing where the event was considered worth preserving. In 1934 - the last year the BBC operated this system in association with British Homophone and The Gramophone Company - more than 600 waxes were cut, and approximately 300 of these were processed to permanent discs. 19



In the meantime, Cecil Watts, a musician and entrepreneur, had begun the development of a metal-based lacquer-coated disc,which was to revolutionise broadcast recording techniques. In an taped interview in the National Sound Archives oral history section, Watts explained that the original impetus for the invention came in 1930, and combined two family enthusiasms - music and recording:

My father was an enthusiast and expert on the Edison Bell phonograph. I'd been apprenticed as a piano tuner, so I was keen on tone values, sound, that sort of thing. The disc recording system came about because I became a musician and had various bands, and I longed for a play-back recording machine that would give an immediate play-back to rehearsing musicians.20



It is important to understand the nature of Watts' invention, and to compare it with conventional disc recording, in order to explain its attractiveness to programme-makers. In commercial recording, the initial wax disc was electroplated to create a 'master' . A 'mother' was then made from the 'master' by a second process of electroplating, and from this a number of working matrices or 'stampers' from which the final commercial disc was pressed. The process was time-consuming, although the BBC did use it in certain cases where a permanent record of an historic event was required, or for some other reason it was necessary to retain a recording for posterity.

Watts method - an aluminium-based cellulose-nitrate laquer-coated disc - enabled playback directly after recording; hence the term direct disc recording. The advantage of Watts' disc was that while it was soft enough to enable the recorder to cut, it was durable enough to withstand repeated play-backs of up to twenty times. The first Watts discs were cut from the centre outwards, which enabled swarf from the recording head's cutting of the lacquer to collect at the centre of the disc, out of the way of the head itself.



Watts' wife Agnes - later his biographer - was heavily involved in the firm and was also interviewed by the NSA, recalling how the business grew according to industry demand:

He developed the first - very crude - machine over a period of months in a flat over 107 Shaftesbury Avenue. The first discs were hand-sprayed, but later they were dipped. Initially he had a little factory spread between two premises in Old Compton Street and Charing Cross Road. Later, when demand grew heavy, we moved to Cable House, Kew Green, where we had a combined house and office.21


Watts' company was named the Marguerite Sound Studios, later to become famous for its initials alone - MSS. The name 'Marguerite' was a family one, and was used by Watts and his wife for their new venture, because it was believed to be lucky.



For a number of years before the BBC showed interest in the system, the firm manufactured Acetate discs for Commercial radio companies such as the IBC and later Radio Publicity Ltd, the British agents for Radio Luxembourg, as Agnes Watts recalled:

We made records for advertising agencies making programmes to be broadcast from Fecamp and Luxembourg; this is what aroused the BBC's interest. 22


This is borne out by Edward Pawley who states:

By the early 1930s Marguerite Sound Studios were taking on all kinds of recording commissions, such as advertisements for Radio Luxembourg. [This is imprecise, as Radio Luxembourg did not begin broadcasting until 1933. It is likely that Pawley is referring to initial use by the I.B.C., although certainly Luxembourg used Watts discs from the outset of broadcasting] The first approach from the BBC came in the autumn of 1933.23


The BBC obtained a trial machine on loan in April 1934, and after successful experimentation with the system a pair of Watts recorders were ordered for a channel at Maida Vale in June 1935, with a further four at a cost of £200 each in September. It is worth noting that in December 1934 the BBC Board agreed an allocation of £7,600 to develop recording generally. The discs came in two sizes, 12 inch (4.5 minutes recording/playback time, cost two shillings) and 13 inch ( 5 minutes recording/playback time, cost three shillings). The increase in recording by BBC programme makers, using the Watts system may be judged by the following expenditure on discs:


 1.  1935  £900
 2.  1936  £2,500
 3.  1937  £3,000



1939 saw the BBC using the Watts system at the rate of 25,000 discs a year.24 With the disc technology came a new playback system; initially called 'the Watts Desk'. In August 1935 a Recorded Programmes Mixer was installed in Broadcasting House, consisting of three desks, giving access to up to six discs at a time. The degree of accuracy possible in such channels as these was extraordinary. In the mixing room a dial informed the operator of the number of grooves cut by the head as it moved over the disc, enabling the user to locate a specific sentence or even word by groove number.


Pawley gives a graphic example of the precision thus available, at the recording of the Proclamation of the Accession of King George Vl:

The Garter King at Arms stammered over the phrase 'our rightful liege Lord' but when the disc was reproduced it was possible to cut out the stammer and to reproduce the speech as it had been intended. Asa Briggs says that this story cannot be traced in the archives, but the present author remembers the recording very clearly.25



As is sometimes the case with inventions, a number of interests seem to have been developing the principle of the direct disc at about the same time. Professor Michael Biel recently discovered that the Speak-O-Phone company was demonstrating an uncoated aluminium system similar to early Watts experiments, in a St Louis department store in November 1928.26 In America, the equivalent of Watts' research was being carried out by the Presto Recording Corporation in Brooklyn, N.Y.

Their product - "The Presto Recording Disc", was devised in 1933 and advertised with what may be its first official announcement in Radio Craft magazine, dated October 1934:

The Presto Disc

For instantaneous recording. After years of research, we have developed a heavily coated disc that for quality of tone, fidelity and reproduction, and low surface noise, equals the commercial wax record. It is black in color, non-breakable, and non-inflammable…27



The arrival of direct disc recording enhanced the development of mobile broadcasting. The BBC invested £1200 in its first mobile unit, a 30-cwt Morris van, its equipment powered by 24-V accumulators charged by a generator driven by the engine. This initial unit, M53, was first used to record part of Laurence Gilliam's feature, Gale Warning at Battersea Power Station in March 1935. The magazine Wireless World christened the unit 'the BBC's Flying Squad'.28

From this point the BBC continued to develop mobile recording, and the 1937 BBC Annual refers to the fact that " a small fleet of recording vans has been designed and constructed for collecting programme material from all parts of the country for subsequent inclusion in the programmes." 29

Elsewhere in the world, major networks were converted to a new kind of programme making. In the United States for example, where the large nationals had not followed the lead of the smaller private stations in recorded programming, it was Herbert Morrison's dramatic commentary on the Hindenburg disaster - while testing the Presto Direct Disc system in the field on May 6 1937 - that began to break down resistance to recorded output at NBC.

This event is chronicled in the US National Archives as part of an Inspection Report on the recordings:

To test the practicability of recording an event on the scene of action and rushing the transcription back by plane for broadcast a few hours later, WLS, Prairie Farmer Station, Chicago, sent announcer Herbert Morrison and Engineer Charles Nehlsen to cover the first landing in 1937 of the "Hindenburg" at Lakehurst, N.J., on May 6th. As Morrison was describing its approach, the ship burst into flames and exploded. The men stayed on the job nearly three hours and recorded for radio the first eye-witness account of the great disaster. At the end of their recording, they caught a plane to Chicago in order that the account could be broadcast early on May 7th 1937. The National Broadcasting Company waived its rule against the broadcasting of records and sent this recording over 144 stations from coast to coast in the US.30



In UK Commercial radio too the emphasis moved towards the use of the outside broadcast recording, using the Watts disc system. Roy Plomley, later to become famous for his BBC programme, Desert Island Discs, began his career in radio with the IBC. In 1936 he was put in charge of outside broadcast recording. 31 The first programme made for IBC stations by mobile unit was called Radio Parade, recorded under Plomley's direction on Sunday afternoons in a cinema in Kingston-upon-Thames. It quickly became popular and was subsequently sponsored by Stork Margarine. Initially the I.B.C. had two mobile units, but this was expanded to three very soon, developing the touring show, Radio Normandy Calling. During 1937 Plomley was supervising an OB in Blackpool which was to underline the potential poignancy of the recording process:

Between the time we recorded the show and the time it was due on air, the theatre burned down and all the scenery and costumes were destroyed, an event which made front page news all over the country. But there, on Radio Normandy, were Bertini and his Band, and the Five Sherry Brothers and Tessie O'Shea, and all the rest of the big cast, still performing on a stage that no longer existed.32


The use of recorded material from outside sources as opposed to live transmissions, gave commercial companies such as the IBC an advantage over the BBC in terms of flexibility. Many of the variety shows recorded by the IBC, were also broadcast 'live' by BBC radio. Plomley was charged with the task of planning a schedule so as to pre-empt the competition, by changing transmission dates:

On several occasions we were able to put a show on air just a few days before it was broadcast by the BBC. 33



In British radio the requirements of mobile recording with the onset of the war were to enhance the reputation of the Watts discs within the BBC. It is interesting to reflect on the later adherence to disc recording in the face of wartime and post-war developments in magnetic tape technology. (See below) Peter Copeland remembers being one of the last Technical operators in the BBC to be trained in use of the system as late as 1960.34



The Philips-Miller System


After a number of experiments by such as Hymmen, Berthon and Nublat into sound on film, in 1931 Dr. J.A. Miller of Flushing, N.Y., developed his Millerfilm system which became the basis of the Philips-Miller recording process, with tape and equipment manufactured by Philips in Eindhoven, Holland. This was a sound-only system, and its use was as a radio recording medium. The requirements of quality and instant play-back were answered at a stroke in this superior technical advance which, like the Watts discs, was picked up by both BBC and Commercial companies alike, with the most widespread use initially among the latter.



The idea of film that required processing before broadcast clearly prevented the instant use of recorded work in play-back form. This was circumvented by having a groove or pattern cut in a cellulose base of film, coated with gelatine, on which was placed a skin of black mercuric oxide some three microns thick. A v-shaped cutter recorded sound signals by tracing a pattern in the oxide, leaving a transparent track down the centre of the film. This optical pattern could be 'read' by a photo-electric cell. The result was the highest quality of recorded sound known before World War Two, even given advances then taking place in Germany. 35

The benefits and disadvantages of the Philips-Miller system - marketed as Philimil - may be summed up as these:


High quality instant playback of up to 15 minutes per reel.


P-M film was not erasable and therefore not reusable for recording.

Inspite of improvements in efficiency, it remained a costly system, partly because of (1)

Editing of the Philips-Miller film was somewhat problematic in that care was needed to prevent edits being audible. The techniques for avoiding this involved a time-consuming process of cutting the tape then painting over the splice with a black solution comparable to the original oxide, which attenuated the light comparatively slowly as it went past the photo-electric cell.


Peter Copeland's explanation clarifies this further:

The technique - known as 'blooping' - was rather like fading a volume control very quickly up and down, creating a 20 millisecond or so gap in the programme as the splice went through, which - if you chose the point of edit sensibly, that is to say between words or bars, you would possibly not notice it.36


It was argued that while its prime use would be for 'bottling' programmes on the Empire Service, the increased quality would be negated and made irrelevant by the poor reception of Short Wave broadcasts.37

Nevertheless, the use of the system for classical music recording and other situations where high quality reproduction was required, was unsurpassed in the late 'thirties, as Pawley, writing in 1972, remarks:

But for subsequent improvements in magnetic recording, the Philips-Miller system might well have remained the best recording system to this day.38


It is interesting to compare the approaches of the BBC and its Commercial rivals in relation to this technology. Unlike the Blattnerphone, Marconi-Stille and the Cellulose Nitrate disc systems, Philips-Miller was not to the same extent developed in partnership between the BBC and its originator:

The BBC had nothing to do with its development, the first machines being delivered in their fully-developed state in 1937 by their Dutch manufacturers, Messrs. Philips Lamps, Ltd. 39


In fact the BBC had begun experimental recordings with the film system in 1936. 40 However, Howard Thomas, then working for the London Press Exchange, and later to be a key figure in the development of Commercial Television, is correct in his assertion that the process was adopted more quickly by the Continental stations and their agents:

The BBC engineers had been experimenting with this revolutionary advance but they had failed to enthuse the programmers about its merits.41


The BBC Annual 1937 also implies that the Commercial interests were leading the way in this field, mentioning as it does all three technologies then current:

One employs cellulose-coated metal disks (sic.), used chiefly for recording short items or components of a programme. Secondly there is the steel tape recording apparatus, which has recently been very much improved, [this is a reference to the Marconi-Stille machines which replaced the original Blattnerphones] used chiefly for complete recorded programmes for the Empire Service. The third system, at present only in an experimental stage (so far as the Corporation is concerned) [my italics] consists of a mechanical method of recording on celluloid strip, similar to that used for film. 42



It is difficult to ascertain the extent of eventual BBC usage of the Philips-Miller system, although at its peak it would appear to have been considerable. Pawley states that by April 1945, when the BBC was phasing out its use of the system, "at least one reproducing channel would have to be retained indefinitely to cope with the stock of Philips-Miller recordings (some 10,000)." 43 A programme of copying was instigated to transfer recordings considered of lasting value to disc, and later to tape. In 1963, the only Philimil films left in the archives were transferred to 7.5 inch ips tape by Rowena Taylor at Maida Vale, and the originals destroyed as a fire hazard.



The I.B.C. stations, and later Radio Luxembourg, had taken an early interest in the use of film as a recording/playback medium, for reasons of durability as well as quality. The transporting of recorded programmes on discs from London studios to Continental broadcast sites brought with it obvious hazards. The progression from disc to Philips-Miller here came via the use of standard film sound recording. The well known broadcaster Bob Danvers-Walker was in charge of setting up many of the I.B.C. stations, including Radio Normandy, and remembered the installation of play-back projectors at the station's studios in Fecamp:

We had two Western Electric 35mm projectors as you would find in the projection room of a cinema. Programmes would be recorded using just the soundtrack: The Horlicks Hour, The Ballito Stockings Hour, The Ovaltine Programme: the major sponsors would have their programmes put onto this film. These were sent over from England, and we were not only the announcers, we were the engineers as well - we did everything! 44


A similar system was installed at Radio Luxembourg - slightly modified from the Radio Normandy prototype - in the Spring of 1934, supervised by a subsidiary of the Gaumont British Group. The late Stephen Williams, interviewed shortly before his death in November 1994, was chief announcer on Luxembourg and remembered the system operating:

They didn't like the Blattnerphone system, as it was too heavy, too big, too cumbersome…so we went in for ordinary sound film…What we did was install two systems by British Acoustics Ltd…which operated on 35mm film cut in half lengthways so that it fed through the sound-heads on the single remaining line of sprocket-holes. That gave you a complete half-hour without any change at all and was easily transportable…The first programme to be recorded in this way featured Carroll Gibbons, Olive Groves and Paul England.45

Williams remembered the system as being superceded by the Philips-Miller process in "about 1936". Certainly the system was well established in Commercial radio by 1937 when the J. Walter Thompson Organisation opened a new studio, claiming it to be "the only one of its kind in Europe"46, at Bush House, Aldwych, London.

This system …has the advantage that the record may be played back instantly, and at the opening demonstration Foster Richmond, the well-known singer, recorded an item which was reproduced through the loud speakers in the studio almost before he had time to sit down.47

Apart from his enthusiastic examination of the P-M system, the reporter notes that 'in addition to a monitoring room there is a 'dubbing' or editing room in which are a number of special disc recorders as well as sound-on-film recorders'. Equally he is clearly impressed by the studio itself:

The walls of the studio are panelled in such a way that the panels may be reversed, and one side is 'quilted' to reduce echo, whilst the other side is panelled to provide echo, and thus any desired studio acoustics may be obtained in an instant.48


Situated above a swimming pool which had at one time been used by Bush House employees, the studio floor was "rubber-sprung…built upon layers of rubber alternating with thick layers of cork down to a depth of 2 feet." Howard Thomas, at that time working for the rival London Press Exchange, recalls the Thompson Organisation's investment in this studio in his book, With An Independent Air, although he incorrectly refers to the Philips-Miller system as using metal tape.

Furthermore, the commercial radio commitment to pre-recording of programmes eventually influenced the BBC to yield in their devotion to live broadcasting and go over to recording.49


Thomas goes on to observe that with the coming of war, the BBC was to be the beneficiary of Commercial radio investment.

When the BBC took over Bush House in wartime for overseas broadcasting the JWT studios and tape-recorders [sic] became an immediate asset for the propaganda drive.50


This is borne out by Peter Copeland:

Before World War 11 there was only one Philips-Miller recording channel at the BBC. Then with the coming of war, suddenly realising with censorship of broadcasts amd other phenomena coming in the political sense, the BBC understood that it was going to need a great deal more recording equipment than it had ever done before…So round about September 1939 the BBC became aware that the commercial broadcasting organisations in London had got some Philips-Miller machines and they got their hands on them - I don't know if it was compulsory purchase or lease-lend - but they got their hands on them together with the staff who knew how to operate them, and those machines, which were redundant from the point of view of broadcasting from the Continent of course, became BBC-used machines.51


Pawley mentions "three…machines taken over at the beginning of the war" 52 which are likely to be from the Commercial Studios. Martin Pulling, the BBC's Deputy director of Engineering at the time remembered Philips-Miller channels at Maida Vale, London, as well as in Bristol and Manchester.53



Later Developments


The commercial companies in the last years of the decade began a move back to disc recording, as did the BBC itself, although the Corporation's policy of detailed testing and customising of technology meant that this happened at a slower rate. In 1939 the I.B.C. issued a marketing publication, This is the I.B.C. Here there is a good description of how commercial programmes were made, including a comparison of the two technologies as they stood at the time. This section was written by the I.B.C.'s Chief Engineer, Norman Angier. Angier had formerly been an employee in the Research Laboratories of The Gramophone Company Ltd (E.M.I) and subsequently Technical Recording Manager for Decca. He joined the I.B.C. in 1938. His remarks give the clear sense of a moving away from the Philips-Miller process:

At the time of introduction it appeared to offer many advantages over some methods of disc recording then in use. Improvements in disc recording have been so considerable since that this is no longer true…A thorough examination of the two systems has convinced us that since the only purpose of transcriptions is to record advertising programmes, the advertiser is best served is best served by using disc which permits his programme to be recorded at the lowest cost consistent with the necessary standard of quality. Nevertheless, while we do not use the Philips-Miller system for recording, we maintain reproducing equipment at the station for those advertisers who wish to utilise it.54



The disc technology to which Angier is referring is the new 'Supercut' disc, introduced by MSS in 1938. Pawley describes this as to be "unbreakable, and to have a long playing life and an absolutely silent surface." 55 . An interesting further comparison in the two rivals' recording technologies is in the speed of disc recording. Pawley states that before the war, 33 1/3 direct recordings "had proved a failure" in tests by the BBC. 56 The speed was not introduced by the Corporation until 1941. The BBC's Engineering Division Training Manual states two main disadvantages in the use of LP recording, the first being the requirement for new recording and playback apparatus.57 If the implication here is one of capital investment in new technology, the second objection was a purely practical one:

Because of the greater number of 'words per groove' it is impossible, without further development, to lower the pick-up to the same degree of accuracy in, say, a continuous speech.58


On the other hand, where selectivity of material was not required, as in a variety show, this was less of an issue. Angier writes in 1939 that at I.B.C. recording sessions "electrical transcription discs are recorded on a machine which is in effect a precision lathe, the turntable of which rotates at the exact speeds of 77.9 r.p.m. or 33 1/3 r.p.m."59 It is easy to see how the LP, with its increased duration of 15 minutes, fitted the Commercial Companies programming strategy very well, in that the majority of their broadcasts were of precisely this length.



The German invasion of Holland put an end to supplies of Philips-Miller film, and the system was never to reassert itself fully again. A similar problem affected the Blattner/Marconi-Stille process in that the steel tapes were manufactured in Sweden and at £20 for a fifteen minute reel in wartime, extensive usage became impracticable.

Of the death of the Philimil process, Peter Copeland adds:

It limped along for its niche applications where you either needed word-perfect editing, or you needed high fidelity, until the BBC developed its own disc cutting system using Watts' Cellulose Nitrate discs in 1944.60


In addition running costs for the Philips-Miller system were ten times as great as that of disc cutting.61 The system was finally abandoned as a recording medium in October 1950, with the last four Marconi-Stille machines written off in 1952.62 In 1945 the BBC acquired a German army Magnetophon Tonschreiber B magnetic tape recorder, and subsequently a Tonschreiber HTS broadcast standard machine.63 Experimentation using these and subsequent machines led to the gradual adoption of tape as the prevalent recording medium.




As a matter of context, it is necessary to briefly sketch German developments in tape recording for radio broadcast prior to WWll. Edward Pawley remembers visiting Berlin in the 1930's to witness a demonstration of a Magnetophon using paper tape impregnated with iron dust. He gives no date for this event, but goes on to mention that the tape frequently broke and "would clearly have no application in broadcasting until this problem had been resolved. A magnetic tape with a cellulose acetate base was first produced in Germany in 1934, but it was not applied to broadcasting on a significant scale until some years later."64

The true revolution in tape recording came during the early years of the war, when the German engineers combined the technology already evolved, with a neglected discovery of some twenty years earlier - in 1921 - by two Americans, Carlson and Carpenter. The device - which became known as Ultra-Sonic Bias - enabled sound recordists to eliminate background noise whilst at the same time greatly reducing distortion and offering the potential of increasing the frequency range. Peter Copeland points out that, at the time of its original discovery, and for many years subsequently, there was little application for such quality enhancement:

No one could think of a technology for which it would be relevant. So the other media were leaping ahead and no one could think of a way of applying this invention to magnetic recording. The Germans did it.65



Listeners at the BBC's monitoring station at Caversham near Reading during the war found themselves on occasion hearing Hitler broadcasting from two geographical sites at virtually the same time; clearly one of them was a recording, but the two transmissions were indistinguishable in terms of quality. This was the first clue that Germany was using a new and sophisticated recording technology, although the full answer was not discovered until the war was in its final stages. At that time Allied forces invaded Luxembourg, and discovered Magnetophon tape in the studios of Radio Luxembourg. Meanwhile British troops over-ran the factory of the Deutsche Grammophon Gesellshaft in Hamburg, where recordings of Hitler's speeches were stored for the posterity of the Third Reich. 66 Nevertheless it was not until plans for the BBC's high cultural Third Programme were being developed in 1946, that the idea of using a Magnetophon reproducing channel was actively discussed.



I have sought to demonstrate that the development of recording for radio in Britain during the decade 1930 - 1940 was not solely along an arc of technical improvement. Each of the three technologies discussed answered certain specific needs of broadcasters. In this the various systems were the servants of corporate policy shaping the face of British broadcasting in both the Commercial and Public Service fields. Just as the Blattnerphone and Marconi-Stille machines enabled edited and 'bottled' programming, offering the possibility of transmission across differing time-zones, so requirements of higher quality recording for orchestral music favoured the Philips-Miller system for a time.

The lead taken by the Commercial companies in returning to disc recording in the last years of the decade reflected the practicalities of cost-saving, together with an improvement in sound recording and reproduction quality in this particular area. It was to be the coming of war which would develop the crucial element of portability of disc recording, making possible single presenter operation, and thus shaping much of the style of radio thereafter.





1 Robert Angus:"75 Years of Magnetic Recording", High Fidelity, US, March 1973.

2 The Electrical World, US, 8 September 1888

3 Edward Pawley, BBC Engineering 1922-72 BBC Publications 1972

4 Asa Briggs, The BBC: The First Fifty Years, Oxford, 1985 p.121

5 Anthony Askew, correspondence with author,1999

6 BBC Year Book, 1932, p. 101

7 J.W. Godfrey, 'The History of BBC Sound Recording', BSRA Journal, Vol. 6, No. 1, 1959 quoted in Pawley,p. 179

8 Ibid, p.182

9 Ibid., p.181

10 BBC Year Book 1934, p. 418

11 Holt Marvell was the pseudonym of Eric Maschwitz, originally an assistant in outside broadcasting, then editor of Radio Times who was moved by John Reith to become Head of the Variety Department, newly created in 1933 as a BBC response to the tide of populist programming coming from the Continent. Among other things, the charismatic and brilliant Maschwitz co-wrote the hit show, Good Night Vienna, and one of the great songs of the 1930s, These Foolish Things.

12 Val Geilgud and Holt Marvell: Death at Broadcasting House, Rich & Cowan, 1934, p. 41

13 Ibid. p. 44

14 Pawley p.182

15 Peter Copeland, Conservation Manager, British Library National Sound Archive, interviewed by this author, National Sound Archive, 5 January, 2000

16 Ibid p.183

17 BBC Year Book 1934 p.418-9

18 Ibid.

19 Pawley p. 183

20 Cecil Watts, interview, National Sound Archive Developments in Recorded Sound Oral History collection, rec. 20 January 1961, ref. LP26453

21 Agnes Watts, interview, National Sound Archive Developments in Recorded Sound Oral Histort collection, rec. 1984, no reference number available

22 Ibid.

23 Pawley p. 184

24 Ibid. p. 186

25 Ibid. p.188

26 Michael Biel, Professor of Radio and Television, Morehead University, Kentucky, in correspondence with this author.

27 Radio Craft October 1934, p. 238

28 Pawley p.188

29 BBC Annual 1937 p. 68

30 W.G. Stone, Inspection Report, February 1, 1938, U.S. National Archives Accession PG.279 - Job No. 38-103, quoted in Michael J. Biel, The Making and Use of Recordings in Broadcasting Before 1936, pp 1002/3, Northwestern University, Evanston, Illinois, pub., UMI Dissertation Information Service.

31 Roy Plomley, Interview c. 1985 for radio programme, Searching the Ether, AIRC

32 Roy Plomley, Days Seemed Longer, p. 147, Eyre Methuen, 1980

33 Ibid. P.148

34 Peter Copeland, in an interview with the author, 5 January 2000.

35 A full technical description of the process is given by M.J.L. Pulling in 'Sound Recording as applied to Broadcasting', BBC Quarterly, Vol.lll, No.2, July 1948.

36 Peter Copeland, NSA, author interview

37 Antony Askew, correspondence with this author.

38 Pawley, p.193

39 Engineering Division Training Manual, 1942, p. 198, BBC

40 BBC Annual, 1937, p.68, BBC

41 Howard Thomas, With An Independent Air, p. 34, Weidenfeld & Nicolson, 1977.

42 BBC Annual, 1937, p.68

43 Pawley, p. 384/5

44 Bob Danvers-Walker, Interview c. 1985 for radio programme, Searching the Ether, AIRC

45 Stephen Williams, interviewed by Roger Bickerton, Autumn 1994, pub. in three parts in The Historic Record and AV Collector, Issues 39,40 and 41, April, June & October 1996.

46 Article, 'New London Recording Studio: Specially Built for Recording Sponsored Programmes', Newnes Practical Mechanics, September 1937.

47 Ibid.

48 Ibid.

49 Howard Thomas, With An Independent Air, p.34, Weidenfeld & Nicolson, 1977

50 Ibid.

51 Peter Copeland, Interviewed by this author.

52 Pawley p.385

53 MJL Pulling, Interview in NSA Developments in Recorded Sound Oral History collection, August 1984, No catalogue Number

54 Norman Angier, "Recording Commercial Radio Programmes", in This is the IBC, p. 14, I.B.C., 1939

55 Pawley, p. 186

56 Ibid. p.187

57 Engineering Division Training Manual, 1942, p. 197, BBC

58 Ibid. p.198

59 Norman Angier, This is the IBC, p. 14

60 Ibid.,p.14

61 Pawley p. 384

62 Ibid. 385

63 Ibid. p.387

64 Ibid. p.193/4

65 Peter Copeland, interviewed by this author.

66 Ibid.