top of page
  • Writer's pictureDominik

Omega Losses and Gap Effects

In addition to our previous blog articles about the electronics and mechanics of tape machines, let’s dive a little bit deeper into the theory of magnetic tape recording. The intention is to get a better understanding of specific machines or tape types and why these are often associated with certain sound characteristics. The post also aims at eliminating some uncertainties which can occur during the calibration process.

Legendary Klangfilm all tube tape electronic modules

Workhorse Versus Sound Refiner

In the past, the studio tape recorder was mainly a tool for preserving sound documents by means of recording as true-to-life as possible. On top of this, it provided the possibility of intervening in the timing of individual sequences by cutting and combining different recording passages. In the course of time, the number of tracks and the tape width were increased, and later the operation of the machines was facilitated by microprocessors.

Meanwhile, the development departments of the tape and equipment manufacturers were literally working at full speed on optimizing tapes, heads and circuits, always striving for a better signal-to-noise ratio and a frequency response that was as flat as possible.

And then came digital: Many analog devices went into storage or trash, and only a few enthusiasts who enjoyed spinning reels held on to tape technology.

Today, in the age of tapeless digital editing and storage, it is precisely this analog sound that is desired again, to add some endearing roundness to a digital production through tiny technical "shortcomings" from the past. These quite minimal shortcomings existed in magnetic recording for physical and manufacturer-specific reasons, and there were actually quite a lot of them.

So, let’s first have a look at the theory and then proceed to practice…

Tape Heads

In principle, the record, playback and erasing heads are all the same: they are so-called toroidal heads on which the sides facing the tape are interrupted by a narrow gap. The gap insert is a non-magnetic material such as copper or mica. The zone around the gap is finely ground in the shape of an arc. The tape with its magnetic layer slides past this mirror surface.

Record head. The iron packs and the gap are clearly visible.

Playback process (TFK laboratory book)

The record head is to transfer the audio signals supplied by a special amplifier as magnetization to the tape. The playback head, in turn, is to generate a voltage from the magnetic fields passing by. Finally, the erasing head demagnetizes the tape using a very strong HF magnetic field.


Because of the nonlinear characteristic of the magnetizable tape layer, pre-magnetization is necessary for the recording process, similar to the grid bias in tube amplifiers. Without this process, the nonlinear characteristic results in poor performance, especially at low signal levels. Biasing is achieved by adding a high-frequency signal to the audio signal to be recorded (a deeper look into the background of this process would go beyond the scope of the article).

This high frequency share – typically ranging from 80 to 200 kHz – is not audible during playback. It only serves to determine the correct operating point on the magnetization characteristic curve of the tape while recording. This operating point is at a different position for each tape type and is directly responsible for a certain sound – and also for a certain noise – depending on the individual setting.

Bias Magnetization (AGFA publication)

Hence the indication on many machines: "Calibrated to tape type LGR... or PER...". This setting can have a very strong influence on the quality of a recording. If you want to stay as true-to-life as possible, it is important to know to which tape the corresponding machine is calibrated to, because changing to a different tape type will have a direct effect on the sound quality. Pre-magnetization is set on the basis of the technical data supplied with each tape, indicating its individual operating point.

The Record Head

Another aspect is responsible for the recording quality as well: the iron from which the head is made. Here, a material with low re-magnetization loss must be used, such as soft iron or ferrite. Soft iron offers excellent noise ratios, but – as the name suggests – is soft and therefore wears out more quickly. Ferrite, on the other hand, is very hard but generates 2 to 3 dB more noise. To further reduce losses, the soft iron head core is layered from thin lamellas (similar to transformer construction).

The manufacturer must ensure that the gap is absolutely straight, because the recording quality does not depend on the width of the gap, as many assume, but on the flatness of the edge where the magnetized tape leaves the head.

There is also an additional air gap on the back of the head, which ensures a more even magnetization of the tape and a reduction of remanent magnetism, which is problematic as it creates additional noise. The winding of a record head has a very low resistance, usually in between 5 - 20 ohms.

The Playback Head

The playback head is similar in construction to the record head, but with two very important differences: The gap facing the tape must be exceptionally narrow, and it must have a very large magnetic resistance compared to the head core. The latter is achieved by tapering the profile to a few tenths of a millimeter.

Sectional view of a playback head (Christian, Magnettontechnik)

Due to the magnetism of the tape passing the head, a voltage is generated in the winding. Strictly speaking, it is generated by the magnetic short circuit of the force lines in the tape and the head core. In summary, the following criteria are important for the playback head:

• Very small gap width

• Low intrinsic capacitance, so that the interaction with inductance cannot cause audible resonances

• Good magnetic shielding against external interference fields

Omega Loss

When a tape with different frequencies but the same magnetism passes the head during playback, the induced voltage increases with rising frequency as defined by the law of induction – twice the frequency will approximately result in twice the voltage. However, for physical reasons, this does not proceed in a linear manner over the entire audible frequency range. Above a certain frequency, the magnetic poles on the tape move so close to each other that they disappear in the layer of the tape itself, so they cannot generate outward induction into the head anymore. The resulting voltage progression at higher frequencies is shown in the next picture.

Omega loss in playback heads at different tape speeds (AGFA publication)

However, it is possible to compensate for this effect with appropriate equalization in the playback amplifier, which helps to create a quite linear frequency response.

Gap Effect

But that's unfortunately not all as far as treble reproduction quality is concerned… Another phenomenon causes a limited reproduction of high frequencies: The gap effect. This is where wavelength comes into play, describing the dependency between the maximum recordable frequency, the tape speed and the gap width of the playback head:

Let’s say a tape runs at the speed of 9.5 cm/sec, so it has a wavelength of 6 µm at 16 kHz (95 mm: 16000 Hz = 0.0059 mm). If the playback head has exactly the same gap width, a complete wave fits into the gap – there is no voltage induction and consequently no playback as a result. The gap therefore needed to be made smaller or the tape speed needed to be increased.

Mirror Resonances

As another result of physics, there are also shortcomings in reproduction linearity in the low frequency range. Once again, the wavelength is getting in the way. This time, it is not the gap that causes ripple, but the entire head package which comes into contact with the tape.

Here is an example: Assuming that the tape touches the head on a length of 5 mm and runs past it with a speed of 19 cm/sec, the result is 190 mm: 5mm = 38 Hz. This means that the frequency response has a maximum at 38 Hz and a minimum at 76 Hz (which is where a complete wave equals the head package size).

Mirror resonances (Schröder, Tonbandgerätemeßpraxis)

In Practice

Some hints for calibrating a tape machine can be derived from the theoretical explanations given above. It is important to point out that although nothing can be destroyed by an incorrect calibration, you will quickly start turning circles if you ever leave the correct calibration sequence. And: without a reference tape, the exact calibration of a machine is unfortunately not possible.

Calibrating the Playback Section

The playback section comes first, as it will also be needed for adjusting the recording section later on. As seen, the calibration of this section depends on the available options (trimmers) provided by the manufacturer. Depending on the circuit design, sometimes two trimmers for treble and one for bass are offered.

Calibrating these can be quite tedious because the individual frequency blocks on the reference tape are usually very short.

When Is a Head Worn Out?

It’s particularly suspicious when the reproduction is very "treble-friendly" and this treble-friendliness can only be reduced with great difficulty: This usually indicates a more or less worn head. This can also be recognized on the head mirror, when the polished rounding has become a flat surface and the core profile remains maybe just a hundredth of a millimeter thick in the gap area. Further wear then inevitably leads to dull reproduction: the gap becomes wider, the head unusable.

Worn head

Calibrating the Record Section

If the playback frequency response is relatively straight, you can start calibrating the recording section. As mentioned, the most important thing here is the setting of the pre-magnetization, because this determines the operating point related to the tape type used.

While feeding in a frequency of approx. 10 kHz with a level of approx. 20 dB below the maximum playback level, the device is switched to Record and, while observing the output level in the corresponding channel, the available maximum is set with its corresponding trimmer. Once the maximum is found, the bias is further increased until the level has dropped by a certain amount which can be found in the documentation of the individual machine.

A general rule of thumb for the required amount of level drop is: 3 dB at 19 cm, 15 dB at 38 and none at 76 cm. Please do not try to adjust the level or the frequency response with this trimmer! There are different ones available exactly for these purposes.

After adjusting the operating point, the calibration of the frequency response can be started. This procedure is performed at a reduced level as well (20 dB below full scale). First, the corresponding playback level must be set at 1 kHz, after which the frequency response can be straightened by repeating the procedure with additional frequencies. Just like on the playback section, the calibration is performed by setting the corresponding treble and bass trimmers. At the end of this procedure, check the level at 1 kHz once again, and then get ready to record.

With multitrack machines, all calibration steps are repeated per channel. So, on a 24-track machine that's 24x playback, 24x record and additionally 24x synchro – which means 72 measurement runs... but I like it.

This article is presented with kind permission of its original publisher, the amazing Studio Magazin, enriching Pro Audio since 1978! The author, Uli Apel, is an incredibly versatile and experienced engineer as well as one of the most qualified experts in vintage broadcasting and audio technology around these days.



bottom of page