Every month in these pages you read of the virtues (or otherwise) of the latest gadgetry, and are exhorted to explore the wilder regions of MIDI. Now I want to take you down to earth and deal with some basic nuts and bolts problems — those of studio wiring.

There are many ways to construct a home recording set-up. Whether you have a humble Portastudio or a full-blown multitrack/sequencing set-up, many of the problems encountered are likely to be the same, due in no small measure to the 'lowest tech' link (literally) in the chain. Atari users should already know that all is not well with the MIDI Out on their computer, and will know that using the wrong type of DIN connector will result in horrendous Thru problems.

It is rumoured that in the rarified atmosphere of some very expensive recording studios all audio communications are made via digital links. A likely story. Surely any facility worth its salt would have a few Urei compressors lying about? And as far as I'm aware, no-one has yet invented the digital microphone. What I don't doubt, however, is that those of you with a home or semi-pro recording or sequencing set-up will be dealing in analogue audio, and it is on this area that I want to focus attention. Having run a 24-track studio for a number of years I have learned the hard way how to hire, fire, and most importantly, wire.

Let's start with a painfully simple example. Jimmy buys a keyboard, takes it to a rehearsal, plugs it into an amplifier, and before he even plays a note he notices a humming noise. So long as the hum isn't too loud when the band start playing he accepts it as an immutable fact of life. I have met this person time and time again when recording demos for local groups, and have concluded that what for me is a simple problem is a complete mystery to others.

AC/DC

Jimmy has encountered a mains grounding problem called an 'earth loop'. To fully understand how to minimise (effectively eliminate) hum from your audio environment, it is necessary to know something about what comes out of the 3-pin socket in your living room wall: alternating current.

The batteries in your Walkperson supply it with direct current, which flows whenever a less-than-infinite resistance (eg. the circuits and motors in this case) is shown to a potential difference (across the battery terminals). A potential difference is also, and slightly less correctly, called a 'voltage'.

Any piece of active sound equipment without batteries must have a power supply to convert AC 240V mains down to a variety of relatively low DC voltages which then go on to provide the power for the circuits and motors contained therein. Two main factors dictate how likely it is that the AC/DC conversion process will cause noise problems: how smooth is the DC (ie. how low is the AC component — ideally nil); and how well screened is the AC supply side from the rest of the circuitry.

Those annoying little external power supplies that look as though they will (and probably do) break the first time you trip over their cable may have proliferated as a result of their cheapness, but in a studio environment they can help to keep the noise as well as the cost down, as AC can then be kept at arm's length from the equipment. All serious mixing consoles have external PSUs for this reason.

To all intents and purposes, all audio equipment requires DC power, so it is unfortunate that only AC is available via the mains. (The choice of AC arises from transmission efficiency and safety requirements, not those of audiophiles!) Worse still, there is an inherent problem with AC: the electromagnetic field created by a mains current oscillates, as does the current itself, at 50 cycles per second (Hertz). Any length of wire in the presence of an oscillating electromagnetic field will suffer the induction of a potential gradient, and hence a current, varying in time and direction with the phase of the field — in practice this means hum, variously known as mains hum or RF (Radio Frequency) interference, depending on the culprit. Why radio frequency? Well, in addition to mains-induced fields, we and our equipment are constantly being bombarded with radiation from radio and TV transmissions, domestic appliances, cars, and a host of other natural and unnatural sources.

BACK TO THE WALL

The three pins of your living room mains socket are connected to the live, neutral and earth wires of the mains respectively. In some ways neutral is like earth, in that nothing comes 'out' of it as such; live is the odd one out in this respect. On the other hand, equipment with a 2-core mains lead (brown and blue only) does not necessarily need to distinguish between live and neutral; you may be able to connect the leads to live and neutral either way around. Confusing isn't it? Furthermore, there is nothing in theory stopping you from drawing power between live and earth. However, if you do this, expect hefty repair and medical bills to follow, since all metal surfaces will become 'live'. When plugs are ill maintained this can happen all to easily, and not a few guitarists have found themselves completing the circuit between a live guitar and an earthed microphone casing. Not recommended. Usually fatal.

Let's look at neutral and earth. Earth is literally that: the surface potential of the planet in your neighbourhood. A copper water pipe usually provides a good connection to earth (gas pipes are not recommended, for obvious reasons!). Neutral provides a route 'back' to the electricity generating station via the mains, and is intended to carry the resultant AC current that flows when a resistance is shown between it and live. Thus neutral is utilised for current flow, earth is utilised for screening purposes.

THE BIG SCREEN

Earth is a potential 'sink'. When your CD player is exposed (as it must be) to the various electromagnetic and electrostatic fields around it, its metal case intercepts those fields, and any resulting potential created on the surface of the box immediately drains away to earth rather than interfere with the audio circuitry. (The metal case is acting as a screen, or 'Faraday Cage', and even without the connection to earth would offer significant electromagnetic shielding). Aha, I hear you say, my CD player doesn't have an earth lead! True, but your amplifier almost certainly does, and by connecting the two together with the ubiquitous phono lead the amplifier's earth is offered to the CD player.

Phono leads, like all audio leads (except those for speakers) have a central core surrounded by a braided or twisted sheath. The core carries the musical signal, while the sheath screens out any stray electromagnetic radiation — provided the lead is properly connected to ground.

EARTH LOOPS

Jimmy was unlucky in that his keyboard had a mains earth lead. Many don't, and without it the hum that followed him around would not have existed. Why should this be? Surely the better your earth, the better screened you should be? Not quite.

The current induced in a straight piece of wire affected by an oscillating electromagnetic field is nothing compared to the current that this field will induce in a closed loop. Many aerials, specifically designed to receive electromagnetic transmissions, take the form of a loop. By plugging his keyboard into an amplifier, Jimmy is constructing such an arial, one half of the loop going via mains earth, the circuit being completed by the screen of the jack lead between keyboard and amp. The 'earth current' induced around this loop is the culprit, and whenever there are two earth connections between pieces of audio equipment, the problem may arise!

The obvious (and correct) solution to this problem is to break the chain, commonly achieved by disconnecting the earth from the 13A plug serving the keyboard. However, if the keyboard has any external metal surface (which will almost certainly be connected to mains earth), and should the 13A plug suffer a (hardly uncommon) failure involving the earth wire becoming disconnected and wandering over to the live terminal, that metal surface will show 240V AC potential, and Jimmy's hands might get a bit hot. A better solution is therefore to break the loop through the audio link by using a modified jack lead in which the screen has been disconnected at one end.

"Because AC can be transmitted by proximity, it is always advisable when laying out a studio to try to keep the mains and audio cables as far away from each other as possible."

Let's take another example. Fred has a multi-timbral keyboard with multiple outputs. The keyboard has no external metal parts, and a mains earth is not provided. Obviously the instrument picks up earth from the amplifier or mixer via the audio jacks, but as each output is connected to a separate mixer channel, Fred notices a cumulative hum. The trick here is to disconnect the screen at one end of all but one of the audio jack leads. Best of all, make up a loom with the correct configuration for that instrument. (See later: Tidy Up Your Loom.) Some well-designed equipment sports separate mains earth and audio screening. Unfortunately, there is usually no way to tell in advance whether such a provision exists, and wherever equipment has a mains earth connected it is always good practice to try it out with both a continuous and a discontinuous screen for the audio link: whichever configuration gives rise to the least hum is almost certainly correct.

STAR WIRING

In a studio environment, many audio links have to be made, and the possibilities for inadvertently forming hum-inducing earth loops are legion. However, by following a set of fairly simple rules, it should be possible to avoid them while maintaining electrical safety.

Think of a tree. Every bit of that tree is connected to every other bit, but not by more than one route; this obviously avoids earth loops. Your earth 'tree' should follow the same scheme, even to the extent of having a central 'trunk' which then roots to the ground.

Let's start with the trunk. In the studio there are three contenders for 'earth base': the mixer, the amplifier, and the patch bay. Both mixer and amplifier(s) require mains earth. Therefore the audio connections between them should each have a discontinuous screen. The patchbay is a passive device, therefore it is sensible to define the trunk earth as being between it and the mixer. This may take the form of a heavy duty copper braid. Now any audio links between the patchbay and mixer need to have the screens, as usual, disconnected at one end. For tidiness (and a marginal screening efficiency benefit) it is best to connect all the screens at the same end. I prefer the patchbay.

From the patchbay (or the back of the mixer if there is no patchbay) an ever increasing array of audio leads fan out to instruments and effects units. All of these leads should have discontinuous earths, except where a piece of equipment has no mains earth, or a separate mains earth and screen earth (see earlier). Here one of the audio connectors should carry earth from the 'trunk' via the screen, by (my) convention to the left input of the device in the case of an effects unit, or from output 1 or 'left' in the case of an output-only device such as an instrument.

GOOD TRIP?

Before I leave the subject of earthing equipment, I must draw your attention to a detail overlooked in many venues and studios: mains earth safety. There are a range of mains trip switches available. These inexpensive devices are designed to switch off the mains AC supply whenever a current in the milliamp region is detected to earth. Think of it as a vaccination against electrocution. (Tip 1: make sure that the one you buy has sufficient current carrying capacity, and don't put a fan heater on the same circuit. Tip 2: don't use it as a master power on/off switch.

Instead have an ordinary mains switch wired in series, so you won't need to replace the trip switch every two years due to contact wear.) If (if?) you use a computer you may want to install a mains filter, to prevent mains voltage spikes crashing out your programs. Call me lucky, but I don't have one and it has never happened to me. Better still is an uninterruptable power supply (UPS), which helps in the event of a total power failure — the UPS will provide temporary power for your computer, for at least enough time for you to be able to save all your work properly.

PLUGS FOR PLUGS

The most common cause of failure among mains and audio plugs is inadequate cable strain provision. Hold the plug in one hand and the cable in the other. Give it a good yank. If an audio cable survives this test, move on. If not, you have just saved yourself a future problem. If it's a mains cable, be sure to open up the plug and check the connections even if it has passed the test — if you've just broken the earth connection you could be in trouble. Incidentally, within the plug the earth lead should always be the one with the most 'slack', so that it is the last connection to be broken if the cable is stressed.

The most common types of connectors are: 13A plugs, IEC mains plugs, 1/4" jacks, phono plugs, and 3-pin XLR connectors. In addition to the above, various multipin formats are used, and for audio signals ELCO types are the current standard. In particular, I would recommend that you avoid mains plugs with screw-down cable restraints, since on these the collar threads inevitably get stripped. I prefer MK plugs with a 'V' collar, although some people feel that screw-down restraints are less likely to impose stress on the cable, and are therefore preferable. Whichever type of plug you use, don't use twin core lighting flex with them (or with anything else, come to think of it!)

TIDY UP YOUR LOOM

...as Mr. Sony's mum used to say. Because AC can be transmitted by proximity, it is always advisable when laying out a studio to try to keep the mains and audio cables as far away from each other as possible. If cable runs are intended to remain in place permanently, the use of cable ties and well designed paths can transform an anal retentive's worst nightmare into a work of art. For example, if you have a rack of effects units, run the mains down one side and the audio down the other. (Tip: when laying down a fixed loom, it is advisable first to lay the cables along the intended route, and then to tie by working along from one end, forming bends and junctions as you go, since a completed loom can be very inflexible.)

Looms which will have to be moved, eg. those for multi-output synths, can be constructed from multicore cable, but I prefer using an expanding, flexible neoprene sheath, which gives a very smart appearance when bonded to its attendant bunch of cables at either end using self amalgamating tape. You could also try heat-shrink sleeving. The use of coloured, numbered rings for cable identification is another neat touch.

Where cables run behind walls or under floors, it is advisable to first install a plastic pipe conduit. Cables can then be easily passed through, and future alterations to the wiring need not necessarily mean re-decorating your studio. (Note: the various bits and bobs referred to in these paragraphs can all be purchased from a single source, a supplier such as Studio Spares or Canford Audio).

BALANCING ACT

In a truly professional audio environment such as a multitrack studio, and especially in the sound reinforcement business where long mic cable runs and nasty RF fields generated by stage lighting controllers abound, much use is made of the virtues of the balanced line. I have talked about screened cables being used to prevent external fields impinging on audio signals: in some situations this screening can be less than perfect, eg. where the external field is particularly strong, or the signal level is particularly weak. The balanced line offers an elegant solution to this problem.

All audio signals comprise AC waveform(s). To use an analogy with mains (this is just an analogy; mains and audio signals should be kept totally separate) the audio cable has a 'live' core with a 'neutral' screen completing the circuit. With a balanced line, two cores (usually twisted) are used instead of one: two cores carry the signal, and the screen is separate. Via XLR-type connectors this cable is used to feed a balanced output (eg. microphone) into a balanced input (eg. a mic input on a mixing console).

A balanced output offers three signals:'+', '-', 'earth'. Earth is always carried on pin 1, but for some reason manufacturers can't agree on a standard for the other two pins, and as a result you should check phase (see later) when connecting together equipment made by different manufacturers. Even microphones and chassis speakers suffer from this ludicrous lack of conformity!

"Meters come in various shapes and sizes, and even the least expensive are a zillion times better than nothing, if only to test for continuity. Spend a little money and save yourself a fortune in fault-finding."

An input (or output) can be balanced electronically (using transistors), or by the use of high quality audio transformers. A balanced input works like this: it inverts the phase of one of the two signal cores, then adds it to the other one. In the process it does exactly the same to any interference on the line. However, since this interference has been picked up approximately equally by each of the two twisted cores, this inversion/addition process actually amounts to a net subtraction of one half of the total interference by the other half. Resultant interference? Nil!

Where a balanced signal needs to be fed into an unbalanced input, one half (or phase) of the signal should be shorted, at the input end, to the screen. That half of the signal is effectively 'lost', thus reducing the available signal-to-noise by 3dB, another disadvantage of using an unbalanced rather than a balanced line.

KEEPING THE PATCH AT BAY

Simpler home studio set-ups will probably have all audio connections made directly to the back (or top) of the mixing desk. Once the system expands beyond a certain point, (usually when there are more FX than aux. sends), this arrangement becomes unwieldy, which is why any serious studio will centre around the patchbay.

The patchbay is a passive device with no electronics. Simplicity itself. Why then do so many people, (and some big-money studios), get it totally wrong? Like MIDI patchbays, they very often complicate an otherwise perfectly manageable situation.

Patchbays (or 'fields') can come in many shapes and sizes, but all work in much the same way. The simplest would merely duplicate all inputs and outputs of your mixer and effects units in one location, either within the mixer chassis itself, or in a 19" rack format close to the effects. Any output can be patched to any input by means of a patch cord. Of course, this would be possible without a patchbay, but the leads would end up all over the studio instead of being neatly held in one place. Plus you'd have to get out of your comfy swivel chair, an unacceptable practice I'm sure you'll agree.

However, a patchbay is capable of much more than this. Most patchbays have paired rows of 'make or break' sockets. Understanding how these work can save you a small fortune in unnecessary patch cords, since all common connections can be essentially 'hard wired', the connection only being unmade when a jack is plugged into the input of the in/out pair in question (the output is unaffected by plugging in a jack). This is the basic underlying principle behind the 'normalising' patch field, and can be applied in all sorts of situations. Here are some examples:

1. Most mixers have 'insert' points on each input channel. Let's say that there is a row of (socket) pairs dedicated to these. The outputs are normalised to their respective inputs. Making a connection via a patch cord from 'Insert out 1' to 'Insert in 2' will leave the audio signal path through channel 1 unaltered, while feeding channel 1's signal into channel 2. This type of connection is often used when mixing a tape track that contains at different times two completely different instruments. You can then set the EQ etc. for one instrument on one desk channel, and that for the other on a second channel. Thus, altering the settings for each instrument becomes a simple case of channel switching, as opposed to the impossible task of manually resetting the EQ and so on while the mix is playing.

2. It is common practice to utilise channel inserts when introducing dynamic processing. A dynamic processor will have its inputs and outputs supported at the patch bay, but they will not usually be normalised to anything. A couple of patch cords will introduce a compressor, say, onto a vocal track. If an A/B comparison is required, you can route the compressed signal back in through another channel (as in 1).

3. Time Domain FX frequently have their outputs normalised to spare desk channel(s), or 'FX returns', and their inputs normalled to one of the mixer's aux send busses. They may, however, be used at the insert point just like a dynamics processor. Effects can be daisy-chained simply by connecting the appropriate ins and outs together.

4. On most mixers, 'groups' feed tape record inputs, or are used to 'subgroup' related instruments for easier handling. Having the group outputs available on the patch bay allows them to be configured as extra aux sends during mixdown.

NOT OUT AND NOT IN

Patchbays are mainly made up of pairs of rows corresponding to the ins and outs of a particular set of functions (eg. inserts). At the design stage it pays dividends to work out which FX are likely to be normalled to aux sends, which inputs are usually assigned to tape tracks, and so on, so as to develop the most logical layout for your studio. I stack the 'out' row over the 'in', and colour code them red (for 'hot') and black, respectively. However there are (and should be) some holes which are neither. These correspond to tie lines and links.

Tie lines are used to connect one 'junction box' (the patch bay), with another (eg. a wall box). Handy for keeping the keyboard player out of harm's way. Links tie together several adjacent and otherwise unused holes on the patchbay (typically in blocks of four). Let's say you wanted to mix a single track through four channels. (Maybe there are four different instruments on that track.) Take the insert out of the source channel, patch into 'Link X 4', then patch from the remaining three holes on that link back into three more channels. (Note: While it is possible to link a single output into several inputs with a fair degree of impunity, the reverse is certainly not true.)

Some patch fields (certainly all professional ones) utilise 'stereo' type jacks (the professional types are small, expensive, and called 'bantam' plugs). I have seen people (and even a certain music store) get into considerable bother by wiring these to carry a stereo signal. They are in fact designed to carry mono balanced lines. A comprehensive patch bay will be capable not only of carrying line level signals such as those passed by mixers and effects units, but will also accommodate tape recorders and microphones. Mies especially need a balanced feed, due to their low signal output and long cable runs. Furthermore, the majority of high quality mics derive their power from a 48V DC offset, sent down the core pair from the mixing desk, and known as 'phantom power'. (Note: Carelessly inserted jacks can cause shorting of the phantom power supply line: keep a stock of spare fuses!)

Wiring up a patch field is one of the most delicate operations the studio constructor has to contend with, since many solder connections have to be made within a relatively small area, and a myriad of cables routed tidily away from this nerve centre to the far corners of the studio. Patchbays with 'tie bars' can give mechanical support to the cables emanating from each row or pair of rows, thus reducing the possibility of cable failure due to strain when moving them about. Check for shorts as you go along with a meter. Meters come in various shapes and sizes, and even the least expensive are a zillion times better than nothing, if only to test for continuity. Spend a little money and save yourself a fortune in fault-finding.

THE END

...of the audio chain is the loudspeaker (some would say it's the ear!), and I've not a lot to say about wiring these up. Just make sure you use good heavy duty cable, twin-core or mains rather than screened, since the latter has significant capacitance which can interfere with the response of a low impedance (typically 8 ohm) loudspeaker. Also, before you settle into your first session, check the stereo phase of your system. If mono-ing the sound from the desk causes sound from the centre of the stereo image to recede, you've got phase problems (usually one side reversed).

Next check to see if your speakers are in phase by playing a mono signal through them. If it's not clear whether or not there is midimage (most notably bass frequency) recession, try it out with the two speakers facing each other. The bass will positively disappear if phase is reversed. If you're still not sure, try accountancy as a career. (Note: Take care that you don't have the misfortune to have a double phase reversal, leading to a normal result at the loudspeakers, but something really nasty in the mix.)