One of the most recognizable issues of the cassette is the background hiss. Almost any person in this world (30 years-old and above, though) whom you ask about cassettes will complain about this particular thing: the hiss.

Magnetic tape has many good things, but it also has drawbacks, and one of them is the hiss. Most cassette enthusiasts don’t like the hiss but at the same time, many of them don’t care at all. So how much hiss does the tape have and how to reduce it?
In this article I will go in deep and explain everything you need to know about the noise of your cassettes and magnetic tapes.

Why is it important to learn about noise?

Noise in a magnetic tape is where all their knowledge starts. The way you listen to your tapes and the way you record them will determine the amount and type of noise your tapes have. They will also help you decide whether you need a noise reduction system and which one to get.

The origin of the noise

The hiss in inherent to magnetic tape technology, although not all kind of tape formats have the same amount. The amount of perceived hiss depends mostly on four things:

• The formula of the tape: the pigment used as well as the finish of the tape surface.
• The equipment used: the playback equalization, the record and playback heads, and the playback electronics.
• The width of the track, and
• The speed of the tape

The actual tape hiss depends on the tape itself and its characteristics, but the perceived hiss depends also on the strength of the signal (the music). Two tapes with same hiss, one with a stronger signal than the other will cause the listener to perceive the hiss stronger in the tape with weaker recorded signal. That means if the listener sets both tapes at the same sound level, then the hiss is more apparent in the tape with the lower signal because the weaker music masks the hiss worse.

Once you have a deck or a walkman, it has a fixed head. There aren’t very big differences between two particular models. Some have better heads, some worse, but the main performance is somewhat similar between models in the same category. Bigger differences are found between categories. Walkmans and boomboxes have the worst performance, while three-head decks have better performance because each head is designed for a specific role in the recording/playback process
.
Usually the hiss is 55-60 dB below the 0 VU signal. This means that when you record a sound so it measures 0 in the deck’s meters and then you stop the sound (but keep recording), you’ll hear the hiss at about -55 or -60 in the meters. The problem is that many decks' meters don’t even go beyond -40dB. But the hiss is there.

It’s worth mentioning that the decibel (dB) scale doubles every time you go up or down by 6. I mean: 9dB is twice loud the signal than 3dB. And -60dB is twice quieter than -54dB.

There are many kind of heads out there (ferrite, permalloy, sendust, amorphous, hyperbolic, GX X’tal…) but the head is much less important than the playback circuitry in regards to the hiss.

The 2nd factor is the tape. There are four different kind of tapes. Read more about this here. Some kinds exhibit more noise than others, but also some can be recorded hotter. So in the end, what counts is not the absolute value under zero, but the difference between the noise level and the maximum recording level, which is called S/N ratio or dynamic range. And how can we know what S/N has a particular tape? Well, by reading about it: manufacturers always publish these data. They usually publish different measures for low and high frequency S/N which is called MOL (Maximum Output Level) and SOL (Saturation Output level). So you may look for it.

With cassettes, the width of the tape is fixed and defined by the format itself, so all them have exactly the same value. However, there are other formats with different characteristics. Reel to reel tape has ¼ inch typically, which is almost double that of cassettes. Anyway, what really counts is the width of the track and not the width of the tape itself. For example, a cassette used in mono (with a special deck with a mono head) uses double the width and the signal will be higher. But R2R will always have higher signal than cassettes. And remember: if the noise is the same and the signal is higher, the perceived noise is lower.

Also the speed counts: the higher the speed, the lower the noise* (up to a particular speed, I guess). Cassettes usually play at 1 7/8 inch/second, but they can go faster or slower. Some machines let you choose the speed of the tape, but it’s not a usual feature. Many double decks can duplicate cassettes at double speed, which leads in lower quality copies, but that’s another story.

Interestingly, although 2X speed copies made in double well decks and boomboxes are actually worse than those copied at 1X, this speed increase reduces the hiss as I explained…

*: The noise actually increases with a higher speed, but the signal does so in a greater factor. Also faster speed allows for a different playback EQ, which is what actually reduces the noise.

How we perceive noise

Noise is a relative concept. It’s a bunch of sound frequencies on which we found no interesting information or message. But what truly noise is depends on the subject who is listening. What is something interesting for us can be noise for a dog or a bee.

Measuring noise is also something that is not an absolute thing; it depends on the environment and the way we focus on the noise.
Because of that there are several ways to measure noise, but the most used are these two: A-weighted and ITU-R 468 weighting.

In magnetic tapes usually the A-weighted is used and the ‘unit’ is decibel (dB). The decibel is not a measuring unit really, but a ratio between two signals. In this case, the ratio between the absolute silence and the measured signal (the noise). It can also be the ratio between the maximum recorded signal (with acceptable quality), that can be measured as zero, and the noise itself (with negative values).

Umbrals of perceived noise

Our hearing system is very complex. It doesn’t work like a microphone and a recorder that records every sound that arrives to our ears. It is more sensitive to some frequencies than to others, and the environment affects our perception a lot.

E.g: if there’s a light random noise (such as a motor running) and then loud music starts, you probably will stop hearing the noise, because the music is louder in the same frequencies. But if the noise is focused in a particular frequency (like a ‘whirrrr’) and the music doesn’t have many frequencies in that particular range, then you will hear it.

Also, if you are in a noisy environment (such as in the street or an airplane) you probably won't notice the hiss, but if you are in a very quiet environment, you’ll hear it.

So, our perception changes a lot depending on the noise itself, the music and the environment.

Now I invite you to do a simple experiment. For this experiment we have two very different kinds of songs (a loud song and a ‘standard’ song) and two different noises (a dishwasher and a whirr).

You can hear every possible mix between songs and noises, so you can notice that:

• The sound of the standard song (and many other “non-loud” music styles) is very dependent on the noise. The dishwasher can be heard in many parts of the song but not all the time. The whirr becomes annoying from the very first moment.
• The sound of the loud song is much less dependent on noises. The dishwasher cannot almost be heard from begin-to-end, while the whirr can be slightly noticed in the background.

What this teaches you is:

• If you listen to loud music you don’t have any problem, even in noisy environments. But with ‘non-loud’ music, you will need good isolating headphones or the result won’t be pleasant.
• If you listen to loud songs in quiet environments you don’t have to be very picky with the tape's own noise (so you can pick even standard type I tapes). If a dish washer cannot be heard, you can imagine how inaudible will be the tape’s own hiss.
• But with non-loud songs you may try to use quieter tapes (good type II or IV) and/or noise reduction systems (which I'll be covering in a future article).

Noise sounds were got from Soundbible.com with Creative commons attribution.