When you hear stories about record producers who link up as many as three 24-track recorders because they 'keep running out of tracks', it makes you wonder how you can get any useful work done at all with less sophisticated equipment. Everyone has to start somewhere, even the Trevor Horns of this world, and what I am going to describe here is a method of 'multitracking' you can do with nothing more than a domestic hi-fi.

The technique is called 'sound on sound' and has been going since the proverbial year dot. I'll tell you straight off that you're not going to get master quality recordings, but it is definitely a good way of getting songs down onto tape in demo form. Detailed explanations will follow, but basically it works like this...

Suppose you had two tape (or cassette) recorders. As long as you have one yourself, and a willing friend who doesn't mind lending his (in a good cause!), you are in business.

The first step is to record a basic track on one machine. It could be a drum machine track or a simple bass line. Next, you replay this track into the second machine while adding another musical line on top of it. If you keep 'bouncing' the recording back and forth between the two machines, you can soon build up the basics of a song. You lose a little quality on each bounce, but five or six overdubs should be possible while still achieving an acceptable result. It sounds too simple to be true, but I can assure you that it does work and I shall tell you later about my own experiences with the technique. Time for a diagram...

CASSETTE-TO-CASSETTE

As outlined above, this method requires the least amount of specialised equipment - none. Diagram 1 shows the hook-up. It's important that you have an amplifier with a 'mono' selector switch, but if you don't (or if your mono switch doesn't work on the tape outputs) then you will have to mix-and-match from methods I shall explain later on.

The first thing you will notice is that only one track of the cassette tape is used. This means that the end result will be in mono. Sound on sound recording can be done in stereo, but you would need a mixer and I'm trying to keep the equipment list down to a minimum.

Given a choice, I would record a drum track first onto cassette machine 2. In fact, there is no reason why the drum machine and first keyboard track (I have to assume some instruments are available) shouldn't be recorded together to preserve as much audio quality as possible. The important thing to remember is that the two have to be mixed together onto one track of the cassette, therefore the mono button on the amplifier has to be pressed to do this. Now that you have drums and keyboard on cassette, take the cassette and put it in machine 1. Put a blank cassette in machine 2.

To record another track on top of what you have already, simply play cassette 1 and add keyboard (guitar, or whatever) to taste. The level of the second keyboard part will have to be adjusted on the instrument before you 'go for a take', but you can have as many attempts as you like to get your performance right. This will give you three instrumental parts on the cassette in machine 2.

It should be pretty obvious what the next part in the proceedings is: transfer the cassette from machine 2 into machine 1 and carry on as before. When you have enough instrumental backing recorded, plug a microphone into the amp instead of the keyboard and warble your vocal. This time it's better if you connect up both inputs of the cassette machine so that when you take the finished product to the A & R man's office, it comes out of both speakers - not that you can rely on both speakers working in a typical A & R office set-up!

Although what I have said so far is a complete explanation, I haven't mentioned any of the problems that may crop up. Well here goes...

Firstly, as I hinted before, there is a loss in sound quality as you transfer from cassette to cassette - build up the 'generations', to use the official jargon. It's important to keep the levels at their optimum settings so that you guard as much as possible against noise and distortion. It's a good idea to keep the drum track relatively high in level, and to record the vocal last. This way the most important elements of your music will come through clearly. And while I'm talking about level, I must tell you that once recorded, levels are fixed. You can't go back and push the bass up a notch, like you can with a proper multitrack set-up. The moral is to keep it simple and plan ahead.

Another problem might be that the microphone doesn't have a high enough output to compete with the level of the cassette. If this is the case, it should be possible to record the microphone into the 'Mic' input of the second cassette machine, while copying the rest of the track into the 'Line' input of the other channel. The snag is that you get instruments in one speaker and vocals in the other. Oh well, nothing's perfect.

STEREO REEL-TO-REEL

Many stereo reel-to-reel tape recorders can perform sound on sound recording all by themselves, such as the good old Akai 4000DS (it was never actually that good) and the famous Revox A77 or B77. Other stereo tape recorders can manage it with a little assistance. Diagram 2 shows a method I used to use as my home 'notebook' in those dark days before the portastudio was invented.

In this system I used a twin channel guitar amplifier which had a line-level output as well as its standard speaker output. The advantage of a reel-to-reel tape recorder is that it can (usually) record on one track at a time rather than necessarily both together, as with cassette machines. This means that you only need one recorder.

What you do now, is take the output from the left channel to your amplifier, together with the output of your keyboard, and mix them together onto the right channel of the tape recorder. To record the next track, reverse the connections and use the right channel as the output, the left channel as the input. Keep repeating until finished.

The limitations of the reel-to-reel technique are much the same as those of the cassette, except that the sound quality ought to be a little better because of the higher tape speed. I made up a little switch for my tape recorder so that I could easily swap from recording from left-to-right channels, to right-to-left channels. I found it a very quick way of recording, and because there was no possibility of the result being 'master quality' I tended to concentrate on musical ideas, rather than worrying about the sound.

4-TRACK

Moving up the scale a little into the world of multitrack cassette machines, which the world has TEAC/Tascam to thank for, the technique of sound on sound recording is still practised - except this time it is usually called 'track bouncing' or 'ping-ponging'. If you have a 4-tracker, then you will probably be doing this already, but if you have yet to take the plunge - here are the possibilities.

I must admit that I get fed up every time I hear a manufacturer of these machines remind us that 'Sergeant Pepper was recorded on a 4-track'. So it was, but they forget to mention that musical styles and sounds have changed an awful lot since then. Still, it's amazing what you can do with a little bit of thought and effort. Those simple four tracks can be stretched to ten musical layers without any track going through more than one generation of re-recording. This is how it's done...

Start off by recording three basic instrumental parts on the first three tracks of the cassette. Then mix these tracks plus an extra instrument (played live) onto the fourth track. Next, record two more parts onto tracks 1 and 2, then mix them with another live part onto track 3. Follow the same procedure to fill track 2 with two instrumental parts, then add a vocal onto track 1. If you add it up, it comes to ten tracks - like this:

Track 4 1 + 2 + 3 + live track
Track 3 1 + 2 + live track
Track 2 1 + live track
Track 1 live

In real life it's not quite as simple as this. Adding the live part is hard because you can't stop the tape and perform a 'drop-in' without spoiling the continuity of the other tracks. Also, since you will probably be developing the musical arrangement as you go, it may not fit neatly into this scheme. Never mind, there is always a way to do it. If in doubt, try recording a rough 'demo' so that you can work out a way of getting everything you want onto tape before you do it for real. You may have to extend to more generations of bouncing, but if this is kept to the less important parts, such as backing vocals, then it should not be too noticeable.

24-TRACK

Yes, sound on sound recording is used here too! Though sometimes you wouldn't think so, not all producers need a zillion tracks to doodle on before they make up their minds. Still, it's amazing how quickly those tracks can be filled.

The way to make more space on the tape is to group tracks together and bounce them down onto a stereo pair of tracks. Backing vocals, once again, are a prime candidate for bouncing down because they are frequently done as overdubs by one or two people onto a large number of tracks. They rarely need individual attention when it comes to final mixdown, so why not mix them as soon as they are finished? Doing this frees up tracks, and also frees the engineer from having more than the strictly necessary number of mixer faders to worry about.

So that's it - 'sound on sound'. It's a long way to go from a couple of cassette recorders up to 24-track operation, but this technique will go with you all the way. As a starting point, it's a way of developing recording skills without having a lot of equipment and also a way of trying out musical arrangements. It's nice to have all the latest and best gear of course, but it's more important to have the ability to use it and get the best out of it. This 'Studio Magic' series will tell you what is possible, and how to achieve it. Now it's up to you - get those reels spinning and make some music!