So often have I been asked to explain the workings and respective merits of various different types and makes of sequencers now available, that I felt that it was about time to devote two articles purely to sequencers. So this month I will be looking at sequencers as a whole; next month I'll review the most popular and noteworthy of the models currently available.

To many keyboard players the sequencer comes as the last straw, after having contended with electric this, that and the other — synthesisers, polyphonics, ribbon controllers... this is the end, the machine has taken over. But the player who can face up to puritan colleagues will find the sequencer a very powerful weapon to add to the musical arsenal. So...

What is a sequencer?

Well, firstly it is necessary to state what a sequencer is not: it is not a musical instrument. On its own a sequencer is totally useless, much akin to a record without a gramophone. To use a sequencer you must have a synthesiser — be it very primitive (even without a keyboard), or a wallpaper job (see photograph) — but the synthesiser must have voltage controlled parameters.

A sequencer, in simplistic terms, stores at least one bit of information for each note to be played (eg the pitch) and when activated feeds this information into the synthesiser which then produces the programmed 'tune'. The sequencer can be said to be playing the synthesiser; all the musician need do is activate the sequence.

As a synthesiser is a voltage controlled instrument, what in fact is happening is for each note to be played one or more voltages are programmed and then played back sequentially. Fig 1 will make things a lot clearer. Here we have a typical sequencer/synthesiser configuration. The sequencer shown is of a three-layer eight-step type; that is to say, for each note or step there are three control voltages. This is known as an analogue sequencer. Fig 1 shows control voltages 'A' and 'B' driving voltage controlled oscillators 1 and 2, while control voltage 'C' is used to change the filter characteristic for each note. So to set up the sequence it is necessary to step through each note adjusting the various controls to give the desired quality. A step button is found on most sequencers of this type to make setting up easier. So the sequencer is now able to present the synthesiser with the information necessary for each note — but it is also vital that the synthesiser is triggered at the start of each new note. This is accomplished by connecting an extra lead between the two machines to send a trigger pulse activating the envelope generator for each new note.

So we have eight pre-programmed notes at the touch of a button. 'Big deal,' I hear you say. Well, as it stands I suppose I must agree, but now the sequence can be automatically repeated when it gets to note 8. (In most cases the sequences of any length are obtainable.) This is where things get interesting because your sequencer can now be considered somewhat like a drum machine, but infinitely more useful.

In some cases you are limited in choice of control parameters by the type of synthesiser you have; some instruments have only a control voltage input to the oscillators and a trigger input, so it is wise to check your synthesiser before you go and buy too elaborate a sequencer.

Fig 2 shows a set-up which makes greater use of some of the analogue sequencer's facilities. The rate at which a sequence is played back is determined by a low frequency oscillator called the clock. In most analogue sequencers this clock is voltage controllable and therefore, as shown in Fig 2, one of the sequencer's control voltages (in this case voltage 'C') can be fed back into the machine to control the timing of each note. We can therefore programme the sequencer to play phrases made up of different length notes (ie crotchets, minims, quavers, etc). In the diagram control voltage 'A' is pitching oscillators 1 and 2, while control voltage 'B' is being used on the filter to give variations in timbre. In the set-up shown, the keyboard is also being used to control both oscillators; this is extremely useful, as it means that the sequence can be transposed into any key. Say, for example, that an arpeggio in C maj is set on the CVA when bottom C is played on the keyboard (if that is the bottom note), then the sequencer will play in C. If then an F is played, an F maj arpeggio sounds. Simple, isn't it?

Obviously the more control inputs your synthesiser has, the more you can do with an analogue sequencer. Don't forget, however, that you can always get your instrument modified by a qualified engineer to give you extra scope. Fig 3 shows one other interesting and often used configuration. The noise source in the synthesiser is fed through the filter, the cut frequency of which is controlled by control voltage 'A'. This enables a percussion line to be set up, the timing being controlled by CV3. CVB, in this case, is used on the voltage controlled oscillators, and can be set to give a bass line. Clever use of CVA on the filter can cause the notes to be filtered so much that they don't sound so you can have the percussion line coming in on certain beats of the bass line. The analogue sequencer is often apparent in disco music as a repetitive and machine-like beat.

Once you have your sequence running there is no reason why you shouldn't alter the control on your synthesiser. Gradual alterations of the filter cut-off frequency and changes in the envelope shaper can produce interesting effects without losing the beat.

Up till now I've only dealt with analogue devices, but there is another type of sequencer that has recently become very popular; the digital sequencer. The difference between analogue and digital devices is, in essence, quite simple — digital circuits count and analogue circuits sort of flow! A digital sequencer is heavily dependent on computer technology but can now be considered to act rather like a tape recorder — however, instead of recording audio signals, it appears to record dc voltages and trigger pulses. The truth of the matter is that the sequencer really only remembers numbers which are translated into voltages by devices called digital-to-analogue converters. Anyway, let's not get bogged down with the electronics involved, but look at the operation and application of the machine.

Fig 4 shows a very simple digital sequencer which is hooked up to a basic synthesiser. It is necessary for the synthesiser to have all the output/input sockets shown. The CV out (control voltage out) comes from the keyboard and normally has a one volt/octave output, eg when bottom C on the keyboard is played and zero volts is present at CV out, pressing the C one octave higher gives an output of 1 volt; the next C up, 2 volts, etc. The trigger output socket provides a pulse when a key is pressed, this tells the digital sequencer that a new note is being played.

So when a melody is played on the keyboard, all the required information (namely which note is being played — the control voltage — and the relative timings — the trigger pulses) is fed to the digital sequencer which, when in record mode, stores all this information. When the melody has been fed into the sequencer, the end-of-sequence button is pressed. If this control were not included there would be no way of telling the sequencer how long the last note would last.

The sequence can now be replayed by... no, there are no prizes... pressing the replay button; the same value voltage and trigger pulses are then fed back into the synthesiser via the CV in and trigger in sockets; when the sequence finishes it is normally repeated, and so on until the end of sequence button is pressed.

It may not be immediately apparent, but the digital sequencer allows the musician to program easily melody lines, bass lines, etc that don't necessarily have equal length notes; notes lasting up to ten seconds can usually be programmed. The digital sequencer (once the controls are mastered) can be a lot more straightforward to use than an analogue device as the notes are actually played into the machine. Like the analogue sequencer, the digital machines can be played back at various speeds, transposed into various keys, and used in similar ways to control the synthesiser. The digital sequencer is limited, by its memory capacity, in the number of notes that it can store. A standard model can usually handle at least 48 notes and can usually be expanded to hold more. It is necessary, therefore, to have some sort of position indicator that shows how many notes have been used or how much money is left. This, however, is a much larger capacity than that of most analogue models.

The sequencer shown in Fig 4 consists of the very basics. More complicated versions have several layers, transposition controls, larger memories, battery backups... getting more and more elaborate until we come to devices such as the Roland Microcomposer (the ultimate digital sequencer) which does away with the standard keyboard for programming. Instead the information is fed into a small computer terminal with a calculator-type keyboard, all the music being 'typed' in. This sort of device can store eight layers or independent lines of music, lasting about 15 minutes, and needs a modular system like the one shown in the photograph on p57 to play back on.

Experimentation with sequencers can be never ending and, I think, to get the most out of your synthesiser this accessory is vital. Anyway, next month I will be dealing with all (or nearly all) the sequencer units that are currently available and assessing the relative merits and demerits of each. So if I've convinced you that a sequencer is a necessary accessory to your synthesiser, don't miss next month's issue. (Don't miss it anyway — Ed.)

Dave Crombie is an experienced electronic design engineer specialising in keyboard instruments.