From Berlin to Hollywood: how German magnetic tape remade the sound of the twentieth century
A burnt exhibition hall, a lucky oscillation and two machines shipped home by a US officer: how the AEG Magnetophon and BASF tape became Ampex, Bing Crosby's radio show and the modern studio.
Every recording studio of the twentieth century, from Abbey Road to Motown, worked on a principle that left Germany in the luggage of an American army officer in 1945. The story of magnetic tape is usually told as an American triumph: Ampex machines, 3M tape, Bing Crosby and Les Paul. That version is true as far as it goes. It simply starts a decade too late and several thousand kilometres too far west.
Paper, iron powder and a patent
Magnetic recording itself was not German. The Danish engineer Valdemar Poulsen recorded sound on steel wire at the end of the nineteenth century, and wire and steel tape machines followed. What was missing was a cheap, light medium that could be cut and joined. In the late 1920s Fritz Pfleumer, an Austrian-born engineer, began coating thin paper strips with iron powder. German patent 500 900 was granted to him in 1928. An American, Joseph O'Neill, had patented a paper strip in 1926 but never pursued it commercially.
Pfleumer licensed his idea to AEG, the Berlin electrical group, and AEG turned to the chemists of IG Farben's Ludwigshafen works, today BASF. The choice was not random. Ludwigshafen had made extremely fine carbonyl iron powder since 1924 for coils in telephone cables, and knew how to disperse magnetic particles evenly in a binder. Paper was soon replaced by cellulose acetate film. According to BASF's corporate chronology, the company delivered the first 50,000 metres of magnetic tape to AEG in 1934.
A debut that went up in flames
AEG unveiled the Magnetophon K1 at the Berlin radio exhibition in August 1935. It ran tape at one metre per second and came in heavy suitcases.
What most people don't know: the launch was nearly a catastrophe. A fire during the show destroyed the hall where the machines stood. The German collectors' database Radiomuseum records that all four K1 units built were lost; the tape historians Friedrich Engel and Peter Hammar add that AEG built two replacements and exhibited them on a makeshift stand in the show's final days.
The next landmark came in November 1936, when Sir Thomas Beecham brought the London Philharmonic Orchestra to the Feierabendhaus, the works concert hall of IG Farben in Ludwigshafen, during a German tour. BASF engineers recorded the concert on tape. Fragments of Mozart's Symphony No. 39 and Delius survive and have been issued on CD by the orchestra's own label. It is widely described as the earliest surviving tape recording of a symphony orchestra; we have found no earlier claimant, but readers should treat "earliest" with the usual caution.
The machine America turned down
The Americans had a chance to adopt the technology before the war. According to IEEE Spectrum, an AEG affiliate in Schenectady, New York, received a Magnetophon in November 1937, and engineers at General Electric dismissed it. In Germany, meanwhile, the K4 of 1938 became AEG's first commercially successful model and the broadcasting standard. The United States was not idle. Marvin Camras at the Armour Research Foundation in Chicago designed wire recorders that the US military used to a limited extent during the war, and they too used AC bias. But wire could not be edited with scissors, and its sound was no match for German tape.
The happy accident of 1940
Early Magnetophons used direct current bias and sounded only slightly better than discs, with audible hiss and distortion. The leap came at the laboratory of the Reichs-Rundfunk-Gesellschaft, the state broadcaster. In 1940 the engineer Walter Weber, working with Hans Joachim von Braunmühl, noticed that a faulty amplifier oscillating at high frequency produced dramatically cleaner recordings. Adding a high-frequency signal on purpose, known as AC or high-frequency bias, lowered noise and distortion so far that tape could rival a live broadcast. The pair demonstrated their machine in Berlin on 10 June 1940, and the related patent was published in December 1943.
What most people don't know: the principle had been discovered before. Wendell Carlson and Glenn Carpenter in the United States filed an AC bias patent in 1921 for wire recording, and Kenzo Nagai's team in Japan published on it in 1938. Weber's contribution was to combine it with plastic tape in a broadcast-quality machine. Sources also differ on the year: most give 1940, some 1941.
Stereo, decades early
Few listeners realise that German engineers recorded in stereo during the war. Engel and Hammar document test recordings with three Neumann condenser microphones in 1942 and stereo demonstrations of AEG's K7 in 1943. Production never followed; stereo tape did not become commercial until the 1950s. The microphones came from the same Berlin firm whose U 87 later became a studio standard.
The midnight orchestra
Tape also served propaganda. Hitler's speeches could be recorded and aired from a different city to conceal where he was. Allied monitors noticed something stranger: orchestral music broadcast at night that sounded live, without the surface noise of discs, at hours when no orchestra would plausibly be playing. One of the listeners was John T. Mullin, a US Army Signal Corps engineer stationed in Britain. As IEEE Spectrum recounts, he concluded that the Germans had a recording method far better than anything known in the West.
Two machines in a mailbag
After the German surrender Mullin was attached to a technical intelligence team. Near Frankfurt he heard a Magnetophon in use, and at the radio station in Bad Nauheim, already in American hands, he acquired two AEG machines and around fifty reels of IG Farben tape. He shipped them home in pieces. The Allies, Wikipedia's Magnetophon history notes, had known of pre-war Magnetophons but not of high-frequency bias or the improved tape.
Back in California, Mullin and his colleague Bill Palmer rebuilt the machines with American electronics and demonstrated them to the Institute of Radio Engineers in San Francisco in May 1946. Among the audience were engineers from a small firm in San Carlos founded in 1944 by the Russian émigré Alexander Poniatoff. Its name, Ampex, was formed from his initials plus "ex" for excellence. Ampex set out to build an American Magnetophon.
Bing Crosby's problem
Crosby had a commercial reason to care. He disliked performing live radio shows twice for different time zones and wanted to record in advance, but disc recording degraded with every edit; one show, Mullin later recalled, used more than forty discs. After a demonstration at MGM in Hollywood, Crosby's team hired Mullin. On 10 August 1947 he recorded the season opener of Philco Radio Time on the captured German machines, and it was broadcast on ABC on 1 October 1947. Accounts differ on detail here: some say the October broadcast was played from tape, others that it was transferred to disc as a precaution against tape breaking.
Crosby invested 50,000 dollars in Ampex. In April 1948 the company shipped its first Model 200A recorders, serial numbers 1 and 2, to Mullin for the Crosby show, together with 3M's Scotch tape. Within a few years most major studios owned one.
What most people don't know: two habits of modern broadcasting date from this German hardware. Editing by cutting and splicing let Mullin assemble words from different takes, and he kept a file of audience reactions to insert later. He believed this may have been the origin of the laugh track. Crosby also gave an early Ampex to the guitarist Les Paul, who added an extra head and invented sound on sound overdubbing, the root of multitrack recording.
From sound to pictures and data
Les Paul's experiments kept pushing the machines. According to Mix magazine, he commissioned Ampex in 1954 to build the first eight-track recorder, the ancestor of the multitrack consoles on which rock and pop records were made for the next forty years. Ampex's place in music was finally recognised with a technical Grammy in 2008. The same lineage ran on. Mullin demonstrated a prototype video recorder in 1951, and Ampex launched the first practical videotape recorder, the VR-1000, in 1956. Magnetic tape went on to store computer data, and in large data centres it still does, as archive storage.
Why it matters
The Magnetophon story is a clean example of a pattern seen across German engineering. The core inventions, plastic-based magnetic tape and practical high-frequency bias, were made in Ludwigshafen and Berlin. The commercial scaling happened in California, catalysed by war, technical intelligence missions and a crooner's dislike of live radio. Credit belongs to both, and the record should say so.
Sources
- BASF: Chronology 1934, magnetic audio tape
- Friedrich Engel and Peter Hammar: A selected history of magnetic recording
- StorageNewsletter: AEG debuts Magnetophon, 1935
- Radiomuseum: AEG Magnetophon K1
- Wikipedia: Magnetophon
- Wikipedia: Walter Weber (engineer)
- Wikipedia: Tape bias
- MusicWeb International: LPO Formative Years review
- IEEE Spectrum: The Magnetophon and the laugh track
- ETHW: Bing Crosby and magnetic recording
- ETHW: First-hand, the evolution of the Bing Crosby radio show
- Wikipedia: Jack Mullin
- Wikipedia: Ampex
- Mix: Bing, Jack and Les
- KQED: How Bing Crosby and Silicon Valley revolutionised radio and TV