What Are Those Codes on Cassettes? (December 2019)
When I first began purchasing cassettes decades ago, I’d look at the title, artist, track list, and artwork on the insert, load the cassette, and enjoy the music. I had no idea what the esoteric codes like “Type II” or “120 μs” meant (if I even noticed them) or what I was supposed to do with them. Since the music sounded OK, I never gave them a second thought. Eventually I took the time to learn about the codes, and that changed what I do in preparing to play a cassette. This article describes what I found out about cassette type, playback equalization, and noise reduction and the cassette deck settings I select before pressing “Play.”
Type
Magnetic tape consists of a mixture containing microscopic, magnetizable particles (called “domains”), which is affixed to a flexible polyvinyl chloride strip like Mylar. Domains are incredibly lethargic. It takes considerable electric current to get them to abandon their non-magnetized state and begin to accept a magnetic charge. They’ll eventually catch up to the electric signal’s amplitude, but every time the electric signal oscillates between positive and negative voltage, the domains will pause at the non-magnetized state before catching up again. The result is that the wave form represented by changes in the domains’ magnetic flux is a poor representation of the electric signal’s wave form—and the original sound’s wave form. Left unaddressed, tapes would sound terrible.
It was discovered by accident that adding a very high frequency sine wave on top of the sound’s signal would overcome the lethargy, resulting in significantly improved performance by the domains and a much better sound. This very high frequency sine wave is termed “bias.”
The frequency of the bias signal is optimized to reduce noise while retaining the best audio quality, and it differs depending on what the domains are made of. Various formulations were developed during cassettes’ heyday, and these formulations were eventually standardized by the Interglobal [“International” in some uses] Electrotechnical Commission (IEC) and labeled as Types I, II, III, and IV in the order in which they were released. Each successive type was capable of producing better audio quality than its predecessor. It was also more expensive. (See Figure 1 for a collage of cassette markings reflecting type.)
Figure 1. Several cassette markings reflecting type.
Type I, based on ferric oxide (Fe2O3), was the first formulation developed. It became and remained the most prevalent and least expensive cassette produced, and most prerecorded and blank cassettes intended for home use were of this type. These cassettes frequently bore no markings (“Low Noise” often appeared but represented little more than a marketing label and required no special treatment).
Type II cassettes, based on chromium dioxide (CrO2), required different electronic treatment during playback, and players were often equipped with a switch indicating “CrO2,” “Chrome,” or “Special.” Later Type II cassettes extended the record-protect notches on the “spine” of the cassette (the long edge of the cassette opposite the edge where the tape is exposed), and some latter-day players could determine the cassette’s type by detecting these extended notches (see Figure 2; the record-protect notches are the holes on both ends of the “spine”).
Figure 2. Type I (top) and Type II (bottom) cassettes (both with record-enable tabs removed).
Type III cassettes used a mix of Fe2O3 and CrO2 to improve performance, but they were available for only a short time in the 1970s.
Type IV, the most expensive formulation, employed metal domains rather than oxides. Later Type IV cassettes (sometimes labeled as “Metal”) added a pair of square holes on the cassette’s “spine” in addition to the Type II notch extensions, and these holes could be detected by some players. (Note in Figure 2 that both cassettes have two squares toward the center of the “spine” that, if punched out, would indicate a Type IV cassette.)
All cassette players can accommodate Type I tapes, and players lacking a selector switch may play Type II, III, and IV cassettes in addition to Type I, although the audio quality won’t be quite as good. Similarly, players with a Type II (or “CrO2” or “Chrome” or “Special”) selector switch can use the Type II setting to play Type III and IV cassettes, again with a slight reduction in quality. Better cassette decks also incorporated a setting for Type IV (or “Metal”) tapes in addition to Type I and II settings. I have not encountered a device with a Type III setting, likely due to the comparatively short commercial life span of these cassettes.
Playback Equalization
During playback, the electric current induced in the playback head increases in amplitude as frequency increases, and consequently so does the volume—about 6 dB/octave. Left uncorrected, this would yield unacceptable, treble-heavy audio. Fortunately, players contain built-in electronics that apply equalization to compensate for this artifact, and for the most part this compensation is transparent to the user.
The current and volume increases just described start off as linear, but the increases begin to drop below 6 dB/octave above a certain frequency. By convention, the equalization curves that need to compensate for the increases are described by a “time constant,” which is a function of the frequency at which the increase has dropped from 6 to 3 dB/octave (tc = 1/(2πf), to be precise). The time constant is 120 μsec (microseconds, sometimes indicated as “μs”) for Type I cassettes and 70 μsec for Types II, III, and IV. In order to ensure compatibility with cassette players lacking a playback equalization selector switch, prerecorded Type II cassettes were often released using the equalization curve with a time constant of 120 μsec, even though one with a time constant of 70 μsec would have been optimal. (Better cassette decks allow the user to select type and equalization values independently.)
Figure 3. Type I and II cassette labels reflecting playback equalization time constant.
Noise Reduction
Cassette tape is about 60% the width of reel-to-reel tape (0.15” vs. 0.25”) and is played at 50% or 25% the speed (1.875 inches/second vs. 3.75 or 7.50 inches/second). Consequently, cassette tape is trying to do the same amount of “work” as reel-to-reel tape, but with only 30% or 15% of the domains to carry the load as they pass across the record head. Upon playback, the current that cassette domains can induce in the playback head is weaker compared to reel-to-reel tape, the resulting sound is quieter, and the noise floor—primarily hiss—is more prevalent. Techniques were needed to increase the amplitude of the electric signal during recording while avoiding maxing out the magnetic charge (which introduces a separate kind of distortion known as “saturation”), to reduce hiss audible during playback, and to make cassettes competitive with vinyl records (see Figure 4).
Figure 4. Cassette labels reflecting noise reduction systems.
Many noise reduction (NR) systems employ a two-stage approach where the first stage is applied during recording and the second during playback. This is similar to the application of RIAA pre-emphasis equalization prior to cutting an LP master, coupled with de-emphasis performed by a pre-amp during playback (see Equalization). With tape, this process in known as “companding,” a term derived from “compression” and “expanding.”
Compression increases the amplitude of the electric signal before the signal reaches the record head and magnetizes the tape’s domains. At the same time, compression reduces the dynamic range (difference between the loudest and softest passages) so that signals that start out at high amplitude remain about the same (thereby avoiding saturation) while the lower amplitude signals receive greater increases (see Audio Compression for additional background). The amplitude of the hiss remains unaffected, but it’s at an amplitude that’s now lower than that of the electric signal.
Upon expansion during playback, the compressed signal’s higher amplitudes receive a small reduction, while the lower amplitudes receive greater reduction. Not only does this restore the original dynamic range, it also reduces the amplitude of the hiss. An important stipulation is that what’s done during expansion must correspond to what was done during compression. While mismatch expansion will produce better results than not expanding at all, the resulting audio quality won’t be optimal.
Dolby A was one of the early developments (1965), and since it was designed for professional use, I’ll simply mention it in passing. It independently compresses four fixed-frequency bands and can reduce noise by 10-15 dB.
Dolby B was developed for consumer use in 1968 and became a standard fixture. Though resembling Dolby A, the four frequency bands are variable, adjusting on the fly to the frequencies in what is being recorded. Moreover, softer passages in the higher frequency range (where hiss resides) receive a greater increase in volume, i.e., are compressed even more. Better cassette players incorporate Dolby B into their electronics to expand what is being read from the tape. As I mentioned above, hiss still occurs during recording, but after compression and the higher-frequency volume increase are applied. Expanding during playback reduces hiss volume at the same time the amplitude of higher frequencies is being reduced, and the result is significantly less noticeable hiss (about 9 dB less). Players without Dolby B circuitry can play cassettes with Dolby B compression, though the resulting audio will sound “brighter,” i.e., somewhat treble-heavy.
Dolby C is available on some playback devices. It literally provides a double whammy because it consists of two Dolby B expanders in series, yielding noise reduction of about 15 dB.
Dolby SR (Spectral Recording), introduced in 1986, is another professional noise reduction system and combines fixed and sliding frequency bands characteristic of Dolby A and B, respectively. It yields ~24 dB noise reduction with a dynamic range in realm of digital recording.
dbx NR applies high-frequency pre-emphasis plus high-level, full-band compression. Unlike Dolby systems, it does not make any adjustments for frequency or volume of the original signal. Though the high-level companding significantly decreases noise during playback (up to 40 dB), it’s unable to mask low-frequency, high-level background noise. Unfortunately, cassettes produced with dbx NR do not sound acceptable when played on equipment without the dbx expanding capability.
HX Pro (Headroom Extension) adjusts the amplitude of the bias during recording rather than that of the signal. Since high-frequency content keeps the domains moving the same way bias does, bias becomes redundant, and its level can be reduced. The net effect is to allow the electric signal to be recorded at a higher level, separating it from hiss. Upon playback, the user adjusts the volume down to a comfortable listening level, reducing hiss at the same time. Unlike the other systems’ two-stage approach, HX Pro does not require any special processing by the cassette player—just a volume knob.
How I Approach Playing Cassettes
In converting prerecorded cassettes, I’ll match the type and equalization settings on my tape deck to the markings on the cassette’s insert, selecting Type I and 120 μsec if the cassette bears no markings. If there isn’t anything indicating noise reduction, I’ll try turning Dolby B noise reduction on and off during preconversion setup to see if it makes a difference in what I hear. If the recording sounds better with Dolby B, I’ll then switch back and forth between Dolby B and Dolby C and pick whichever sounds better, or Dolby B if they sound the same (see Figure 5; “EX,” “SX,” and “ZX” represent Types I, II, and IV, respectively).
Figure 5. Playback control panel on my cassette deck.
I generally take the same approach with personal recordings. Because cassettes and their inserts may have been mixed up over the years, I’ll experiment with different type and equalization settings, though I generally end up selecting Type I and 120 μsec because of the prevalence of Type I cassettes.
Some personal recordings I’ve encountered exhibit excessive hiss, often occurring when tapes have been rerecorded and especially if the recording device lacked an erase head. In cases like these, Dolby noise reduction settings have proved useful in reducing tape hiss, even though the tapes may not have been created using Dolby compression.
Back to Paul’s Blog & Contents