The transition from “all at once” recording sessions to recording individual sections or musical parts happened gradually. Les Paul was among the first to produce recordings with multiple layers of his own guitar playing. But he wasn’t alone. Engineers and producers recognized the advantages of this new technology and the hardware designers were quick to deliver machines with more and more tracks.
The “sound on sound” capabilities of his custom designed machines gave way to ever more sophisticated analog tape machines. True multichannel audio recorders are based on a fairly simple concept: the ability of the same magnetic head to function as a playback and recording head at the same time on different tracks.
Professional multitrack tape machines have three heads. The first head that the tape encounters is the erase head, which can selectively erase any track by randomizing the magnetic domains (the areas of recorded polarity) as the tape passes over it.
The next head is the “record” head. As with all of the heads on a multitrack machine, there are individual layers for each of the tracks stacked on top of each other separated by area of mumetal (a non magnetic material that is supposed to minimize magnetic flux leakage). If you look at the front of the head, you can easily see the tracks and the vertical line down the middle of each track. This is where the north-south then south-north magnetic modulations occur that magnetize the area of the tape that is passing over the gap. This happens during the head is used to record audio signals.
The process of recording analog also involves adding an ultrasonic frequency to the audio signal. This signal is called bias and it was a major advancement in analog recording. It dramatically improved high frequency response. Bias frequencies ranged from 50 kHz to over 100 kHz.
The playback or reproduce head is constructed exactly the same way as the record (with the exception that the gap is optimized for playback and is slightly wider). As the tape passes over the repro head the magnetic domains that have been recorded cause a new alternating magnetic signal to be generated at the gap, which is converted by the head an alternating electrical current that matches the original frequency and amplitude. There are some standard equalization curves used during recording and playback that optimize the quality of the process but this is essentially the way analog recording works.
The process of multitrack recording involves laying down a limited number of tracks during an initial pass. This is usually called “basic tracking”. The drums, bass, piano, rhythm guitar and possibly a scratch vocal are recorded. These parts are added to the passing tape at the record head position.
The next step in producing a multitrack project is called “overdubbing”. This is when additional parts are recorded to any unused tracks or when existing parts are replaced or updated. To accomplish this step, the playback of any previously recorded tracks must occur from the record head position in order to maintain synchronization. The playback of all tracks happens from the record head not the reproduce head. In fact, the repro head is really only used during the final mixdown, which happens after all of the parts are recorded on the tape.
Any track that is placed in “ready record” mode will playback when the tape transport is in PLAY but switches to “input monitoring” when the RECORD button is pressed. The output of the track switches from playback or OUTPUT to the new incoming signal or INPUT. This happens automatically and allows engineers to “punch in” or “punch out” in the middle of a track. This functionality means that a section of a track can be replaced without having to redo the whole thing. The overdubbing phase of modern record production occupies a substantial portion of the overall schedule.
To be continued…