The Cable Connection: Balanced and Unbalanced

by Jim Tanenbaum, CAS

As mentioned earlier, there are two basic cable types: balanced and unbalanced. But there are many variations on these two themes.

BALANCED

Balanced cables, which can be used for either balanced mike-level or line-level signals (or unbalanced mike or line signals, for that matter), consist of two (or three) insulated conductors surrounded by a metallic shield and an outer jacket of rubber or plastic. They do not have a standard impedance, but are usually close to 110 Ω. The inner conductors are composed of many thin individual wires twisted around each other for flexibility. The outer shield is available in various configurations, two of the most common are twisted and braided.

Twisted shields have many thin wires wrapped spirally around the insulated inner conductors. While this arrangement is more flexible than braiding, with repeated flexing the shield wires tend to separate, creating gaps for interference to enter and/or the inner conductors to bulge out. Their initial 95% coverage can fall to 70% or less. One attempt to overcome this is to have two layers of spiral wrapping, in opposite directions. Still, separations manage to occur, with the same problems as the single-wound shields. To further enhance their flexibility, most twisted shield cables have thin outer jackets of PVC plastic. While more supple, PVC-jacketed cables are also more easily damaged by abrasion, cutting or crushing. And they get really stiff in cold weather.

Braided shields have many thin wires woven (in an alternately over-and-under pattern) into a tube that encloses the inner conductors. This type of construction is durable, but somewhat less flexible than a twisted shield. The effective coverage ranges from about 85% to 95%. With repeated use, the individual shield wires will break, eventually causing increased susceptibility to interference and static when the cable is moved, particularly when phantom power is present. The Belden Company offers a line of mike cables that have a more open braid for flexibility, and then underneath, a layer of cloth impregnated with a conductive carbon compound to provide almost 100% shielding. The only drawback is that the black goo sticks to the shield and makes it difficult to solder. Finally, many braided shield cables are offered with rubber jackets that do not get as stiff when cold. IMPORTANT: Natural rubber quickly cracks when exposed to oil or smog — be sure to buy synthetic rubber (e.g. EPDM, Neoprene, Hypalon) jacketed cables.

A third type of shielding involves a wrapping of aluminum foil or aluminized plastic film, with one or more bare ground wires running alongside to provide a means of connecting to it. This type of cable is limited to permanent installations, as it is not very flexible, and sharp or repeated bends in the same area can cause the aluminum shield material to tear.

As mentioned above, some balanced cables have a third inner conductor. This will be discussed later.

A special class of balanced cable has recently been introduced for digital signals:Unshielded Twisted Pair (UTP) cables (such as CAT-5 and the newer CAT-5e) that consist of four twisted pairs of conductors twisted around each other and covered in a plastic jacket, without an overall shield, because digital signals can tolerate much more interference. For really difficult EMI (Electro-Magnetic Interference) environments, shielded twisted pair (STP) cable is available. The insulation thickness and spacing of the conductors is rigidly controlled, and these cables have an impedance of 100 Ω ± 15%. Each of the four pairs has a different twist pitch to minimize crosstalk. They are nowhere as flexible as some of the cables described above and are chiefly intended for fixed installations. Nevertheless, many mixers use them in production for digital audio, timecode, and even video for their monitors. IMPORTANT: If you are working with an existing installation, or cables with preattached RJ-45 connectors, be aware that there are two different color-coding “standards”: T568A and T568B. They differ by interchanging the plug pin positions of the green and orange wires, and the white-w/green-stripe and white-w/orange-stripe ones. Since the conductors at the cable ends are connected to the plug contacts one-for-one, either type of cable may be used with either type of jack.the confusion arises if you try to wire a CAT-5 cable directly into a circuit board and use the wrong color-code chart for the plug pin connections at the free end of the cable.

UNBALANCED

Unbalanced mike cables consist of a single insulated inner conductor surrounded by a metallic shield. These cables do not have a standard impedance, but can range from about 50 to 250 Ω. Shielding may be spiral or braided, and jackets plastic or rubber.

A second type of unbalanced cable is coax (coaxial). Like the mike cable, it has a single center conductor surround by a metallic shield. Unlike the mike cable, however, the physical dimensions and insulation composition are rigidly controlled, in order to maintain a constant impedance along its entire length. This is necessary because coax is used for very high-frequency signals, and a change in impedance can cause a loss of power by reflecting some of it back down the cable. Even with a constant impedance, the high-frequency signals are attenuated significantly as they travel, so in addition to an impedance specification, coax is rated for signal loss, in dB/100ft at various frequencies. Coax cable is identified as belonging to various groups or ”Types” primarily by impedance and Outside Diameter (O.D.). Within a given Type, there are cables with stranded or solid center conductors, foam or solid dielectric (the plastic insulation surrounding the inner conductor), and braided or foil shielding. The first term in each of the preceding three pairs represents the more flexible construction.

RG-8 Type is 50 Ω, low loss (≈ 3 dB/100ft @ 450 MHz), and about 3/8-inch O.D.

RG-6 Type is 75 Ω, low loss (≈ 4 dB/100ft @ 400 MHz), and about 3/8-inch O.D.

RG-58 Type is 50 Ω, medium loss (≈ 7 dB/100ft @ 450 MHz), and about 1/4-inch O.D.

RG-59 Type is 75 Ω, medium loss (≈ 7 dB/100ft @ 400 MHz), and about 1/4-inch O.D.

RG-174 Type is 50 Ω, high loss (≈ 15 dB/100ft @ 450 MHz), and about 1/8-inch O.D.

RG-58 is sometimes used to extend radio mike receiver antennas, but its loss is often more than the inverse-square loss of the radio signal traveling the same distance through the air. For this application, RG-8 would be a better choice if more than five to 10 feet is needed.

IMPORTANT: There are subtypes: RG-58A/U Type is slightly different than RG-58 Type, and RG-8x is considerably different from RG-8. Read the manufacturerfs data sheets carefully for the particular cables you are considering.

EVERY CABLE HAS TWO ENDS

To use a cable, it must be terminated with some kind of connector. (Unless it’s soldered directly to a circuit board.)

BALANCED

The most commonly encountered (balanced) microphone connectors are 3-pin XLR (originally a model designation in the Cannon Brand, but now used generically). These connectors have a metal shell and three insulated contacts. The “standard” wiring is:

Pin 1 = Shield

Pin 2 = + Audio (a.k.a. Hi, In-phase)

Pin 3 = . Audio (a.k.a. Lo, Out-phase)

Shell = Ground (most models offer a way to connect the plug shell to Pin 1, with the exception of the old Cannon XLRs. If you use these, solder a length of bare busbar wire to Pin 1 and run it out the back, between the rubber strain relief and the U-clamp.)

This standard is based on Pin 2 of the microphone going positive with respect to Pin 3 when pressure on the front of the mike is increasing. A further standardization is that outputs are on male (having solid pin contacts) connectors (called gplugsh when on a cable) and inputs are on female (having hollow receptacle contacts) connectors (called gsocketsh when mounted on a panel).

Murphyfs Law ensures that things are not so simple. In the 1960s,the first Nagra recorders used male mike-input connectors, necessitating the use of mike cables with female connectors on both ends. (Extension cables were female-male.) Some European equipment manufacturers followed suit, with male in and female out. Other Euro devices have female inputs and male outputs. There also are places in the eastern U.S. (and elsewhere) where the functions of Pins 2 and 3 are reversed, so check carefully when using equipment not your own. (There also are pin-swapping issues with normal and gred doth T-power microphones, but that is beyond the scope of this article.)

More recently, the Switchcraft Company brought out the ”TA” line of miniature connectors, originally intended to follow the U.S. standard practice of male out and female in. But the panel mounted female TA connectors were so much larger than the male ones that radio mike manufacturers were forced to use male TAs for their microphone inputs, and put the females on the mike cable. WARNING: Since some brands also use male panel connectors for receiver outputs, the possibility exists for accidently plugging in an electret lavaliere microphone to a line-level output and destroying the mike.

UNBALANCED

Unbalanced microphones normally use .-inch mono phone plugs (TS, or Tip and Sleeve). They often are high impedance, and are not usually encountered in professional work, although you may have the occasion to tie into them when they are used as props, or if you have to make a field recording of a local small-town musical group or public speaker. There is no industry-wide standard, and some of the microphones may be quite high impedance. Impedance-matching transformers are available, and may include a housing with a .-inch phone jack in and an XLR male plug out.

Unbalanced line-level signals may also use .-inch mono phone plugs, or the smaller and flimsier phono (RCA) plugs. (There are some high-quality semi-pro phono plugs, but even they become unreliable after repeated insertions.) The RETMA (consumer) line-level standard is -10 dB at 47 KΩ, but many manufacturers ignore it. You can make up simple wired adapters to interconnect unbalanced and balanced devices, but using a balun transformer (see below) will allow longer runs of cable and block common mode interference.

IMPORTANT: Studio patch panels use a plug that resembles a standard ¼-inch stereo phone plug (TRS, or Tip, Ring, and Sleeve), but there are dimensional differences (particularly at the tip), and you can damage a patch panel if you attempt to plug a standard TS or TRS into it. You may also get an intermittent connection. Itfs a good idea to make up (or buy) several adapters so you will be able to tie in to a patch panel if the need arises. A good configuration is a TRS patch plug with its send circuit wired to a female XLR and the return circuit wired to a male XLR.

While on the subject of different types of ¼-inch plugs, if you need to patch into an aircraft pilot’s headset, their ¼-plugs are much shorter (about ¾ inch) and have two ring contacts (TRRS or Tip, Ring, Ring, and Sleeve). You will have to make or rent/buy an adapter in advance.

The wiring is:

Tip = + Mike

Ring 1 = + Headphones (mono)

Ring 2 = – Mike

Sleeve = – Headphones

Another dimensional problem involves 1/8-inch (3.5 mm) phone plugs. The mono plug is slightly larger in diameter than the stereo plug, especially the tip portion. Depending on the particular manufacturer, a mono plug may not enter a stereo jack, of if it does, it may bend the contacts so a stereo plug will no longer work properly. (The jacks on Comtek receivers are designed to accept either type.) The smaller 3/32-inch (2.5 mm) plugs do not seem to have this incompatibility. Both these sizes of stereo plugs are used with some cell phone headsets, and some of them use a double ring plug. Again, you will need an appropriate adapter to patch in.

Coaxial cables are usually terminated with BNC (Bayonet-style) connectors. IMPORTANT: Because of the relationship between size and impedance, BNC connectors for 50 Ω and 75 Ω cables are slightly different in dimension. Using connectors with one impedance on a cable with a different impedance cannot only cause signal reflections from the impedance mismatch, but also can be damaged when a 50 Ω connector is mated with a 75 Ω one. For limited space applications such as radio mikes, SMA and even smaller SSMA threaded-style connectors are used. IMPORTANT: Radio mikes use “normal” SMA connectors, with a male pin in the cable-mounted connector (the one with the threaded collet). The more common SMA connectors used on computer Wi-Fi equipment are “reverse,” with the cable-mounted connector having a female receptacle for the male pin in the panel-mounted connector. Therefore, you cannot use a Wi-Fi SMA cable to extend a radio mike antenna.

TRANSFORMERS (NOT THE MOVIE)

Transformers have many uses, but here we are concerned with only four of them: changing impedance, converting between balanced and unbalanced circuits, blocking some kinds of noise, and splitting signals. A particular transformer may be designed to perform one, two, three, or all four of these functions.

A basic transformer consists of two coils of insulated wire wound around the same (usually iron alloy) core. Laminated iron sheets are used for low (e.g. audio) frequency cores; powdered iron alloy (ferrite) for medium to high frequencies. Air cores (wound on a plastic bobbin if the wire is not stiff enough to keep its shape) are used for even higher (radio) frequencies. If both coils have the same number of turns, a signal fed into one coil (the primary) will appear at the terminals of the other coil (the secondary) relatively unchanged. The second signal will, however, be electrically isolated from the original circuit. This removes most C-M noise.

An isolation transformer is usually 1:1, and will have an additional layer of non-magnetic metallic shielding over the secondary winding to block capacitive coupling of the electrical field produced by the noise on the primary winding. The entire transformer may be mounted in a shielded enclosure, with input and output connectors. In this case, the shell of the input connector must be electrically isolated from the shell of the output connector to block transmission of the C-M noise by this route, because XLR (and many other type) connectors often have their metal shells connected to the cable shielding, and thus ground loop current could bypass the electrical isolation of the transformer by flowing through its metal housing.

A transformer designed to change impedance will have a differing number of turns on the primary and secondary. The formula is: √ZP/ZS = NP/NS (N is the number of turns, and the subscripts P and S denote the Primary and Secondary windings.) e.g. To change 600 Ω to 150 Ω, an impedance ratio of 4:1, the square root of 4 is 2, so the primary will have to have twice as many turns as the secondary. (The actual number of turns required is determined by the impedance, frequency, core characteristics, power level, and other factors, again beyond the scope of this article.) NOTE: Theturns-ratio is always defined as primary (input) turns divided by secondary (output) turns.

A balun (BALanced-UNbalanced) transformer is used to convert a balanced circuit to an unbalanced one, or vice versa. At the same time, it can also change impedance if required. A typical application is connecting a 75- Ω coax (unbalanced) to a 100- Ω CAT-5 twisted pair (balanced). It may also provide the functions of isolation and blocking C-M noise. Changing a circuit from unbalanced to balanced will not remove T-M noise that is already there, but may prevent more from entering. IMPORTANT: Always put the balun as close as possible to the unbalanced source, so the cable run is made in balanced format.

A simple balun will have the two ends of one winding connected to the two conductors of the balanced circuit, and the two ends of the other winding connected to the center conductor and shield of the unbalanced circuit. The shield of the balanced circuit may or may not be connected to the case of the balun and/or the shield of the unbalanced circuit. An even simpler balun has only a single winding, with a center tap. The balanced circuit is connected to the two ends of the winding and the unbalanced circuit has the shield connected to a center tap of the winding and the inner conductor also connected to one of the winding ends. Obviously, this type of balun does not provide any isolation or blocking of C-M interference, and is mainly used in antenna circuits.

A splitter transformer has a single primary winding and two identical 1:1 secondaries. Since a splitter transformer is a passive device, each output will be -3 dB down from the input. Similar to those of isolation transformers, the two outputs will be electrically isolated from the input and from each other, but only if the transformer’s XLR connector shells are insulated from its case. Many commercial units do not have this feature, but it is possible to remove the connectors, enlarge the hole if it contacts the protruding back part of the connector, place an insulating plastic film between the back of the mounting flange of the connector and the splitter case, and reattach it with plastic screws. NOTE: Many splitters have a “ground-lift” switch, but this breaks only the connection between Pins 1 of their input and output connectors. Unless you have cables with the connector shells floating, or insulate the splitter connectors as just described, the ground-lift switch will be ineffective.

WARNING: If you use a simple Y-cable instead of a transformer to split an audio signal to feed two other devices (e.g. a recorder and a Comtek transmitter), there will be no isolation, so signals from one can get into the other (RF in this case), and the audio may be completely corrupted.

All types of transformers have certain parameters that must match their intended application. Only the ones relevant to this article are discussed here.

Transformer parameters:

1. Level: Mike or line. Mike-level transformers will overload and distort if used with line-level signals because the core will be completely saturated with magnetic flux lines well before the input signal reaches its maximum voltage. Line-level transformers can be used with mike-level signals, but the higher winding impedance might cause loss of high-frequency response when connected to certain types of output circuits.

2. Impedance: ranges from low (50 Ω) to high (>10 K Ω). Impedance matching is more or less critical depending on the nature of the circuits involved.

Typical values are: Input/Output Impedance, isolation: Mike-level = 150 Ω /150 Ω. Line-level = 600 Ω/600 Ω

Input/Output Impedance, impedance matching: Hi-Z Mike to Lo-Z Mike input = 6 K Ω /150 Ω

Input/Output Impedance, balun: Twisted-Pair to Video Coax = 100 Ω /75 Ω.

NOTE: The “impedance rating” of a transformer does not refer to the actual impedance of the windings inside the transformer itself, but rather the impedance of the input and output circuits it is designed to work with. The input impedance of a transformer will be the actual impedance of whatever the output winding is connected to, divided by the square of the turns-ratio. The output impedance is the input-circuit impedance (such as a 150 Ω microphone) multiplied by the square of the same turns-ratio.

3. Power Handling: The higher the power, the larger the diameter of the coil wire and the larger the core cross section, in order to handle the larger magnetic flux.

4. Frequency Response: Transformers do not respond equally to all frequencies. To give good performance over a range of frequencies requires certain design parameters. The lower the frequency, the larger the core must be. The higher the frequency, the lower the winding inductance and distributed capacitance must be. These two factors oppose each other, so transformer design must of necessity involve trade-offs. The particular core material is also a function of frequency. Professional transformers easily are flat within ±0.5 dB from 20 Hz to 20 KHz.

5. Distortion: All transformers produce some amount of distortion, primarily because of core saturation, hysteresis, and signal phase shift. A “good” transformer will have 0.01% distortion, an “excellent” one will have 0.003% or less. Most audiences aren’t aware of even 0.1% distortion in a movie soundtrack so this is usually not a problem.

6. Isolation: Electrically shielding the secondary winding from C-M noise on the primary is tricky because the alternating magnetic field will induce “eddy currents” in the metallic shield. Any design features that reduce this will decrease the efficiency of the shielding. However, most isolation transformers you will encounter provide adequate isolation.

7. Shielding: Overall electrical and magnetic shielding to protect the transformer from outside interference is somewhat easier, because that shield can be placed far enough from the core to avoid most of the external flux lines. IMPORTANT: Most inline transformers (e.g. isolation) are not magnetically shielded, so be careful where you place them. Avoid motors and power transformers. If magnetic interference is a problem, rotating the transformer 90 degrees to the magnetic field may reduce it sufficiently, if not, move it farther away.

IMPORTANT: Remember that a transformer is a passive device; it cannot give out more power than it receives. The input signal is characterized by voltage, current, and its circuit’s impedance. You may chose any one of these to change at will, but then the others will automatically alter to compensate. e.g. You can raise the voltage of a 150-Ω mike-level signal a thousand times to that of linelevel, but now the output impedance will be so high (150 Ω x 1,0002 = 150,000,000 Ω) that a 600- Ω line input would effectively short-circuit it. You could use a so-called infinite-impedance deviceto “see” the full higher voltage, but now the extra power comes from its amplifier, not the input signal.

SO HOW DOES THIS ALL WORK IN THE REEL WORLD?

Let’s start with the XLR cables. I have most of mine 50 feet in length, with some 25-footers for shorter runs. Also, an assortment of 1-, 2-, 5-, and 10-footers. If a longer cable gets damaged in a single area, it can be cut there and turned into several shorter ones. When cables have been in use for some time, they will develop so many breaks in their shield wires that they become susceptible to picking up interference or creating static when they are moved. Discard them, even if the problem seems to be in just one or two spots.the rest of the cable will fail shortly thereafter. Whether or not to reuse their XLR connectors depends on how much wear and tear they have accumulated. One thing that can be done peremptorily to extend the life of cables is to periodically “circumcise” them, cutting off the connectors and about two inches of cable, and then reattaching the connectors. Cables tend to fail at the flex point where they enter the connector much sooner than elsewhere. As soon as two or three of your cables have gone bad at their plugs, it’s time to service the lot.

Actually soldering the cables to the plugs is a skill beyond the scope of this article, but Local 695 offers an excellent training class. One thing to keep in mind is that shrink-tubing does NOT make good strain reliefs, because when shrunk it is too stiff and simply transfers the stress point to the far end of the shrink-tubing. Use plain PVC tubing (available in many sizes from electronic supply stores) instead. It is much more flexible and will form a smooth curve to more evenly distribute the stress. I save the sections of the outer plastic jacket I strip off various cables while attaching plugs, and use them for strain reliefs on smaller diameter cables.

Some brands of microphone connectors offer a means of connecting to the metal shell and some do not. There is still a considerable controversy over whether to ground the connector shell (sometimes called body) or not, and if grounded, whether to ground the shell at only one end of the cable. There is no simple, always-correct answer.

Here are the possibilities (using 2-conductor cable with the balanced audio always connected to Pins 2 and 3 at both ends):

1. Shield connected to male and female Pin 1; male and female connector shells floating.

2. Shield connected to male and female Pin 1; male connector shell connected to Pin 1; female shell floating.

3. Shield connected to male and female Pin 1; female connector shell connected to Pin 1; male shell floating.

4. Shield connected to male and female Pin 1: both male and female connector shells connected to Pin 1.

If 3-conductor cable is used, there are three more possibilities:

5. Third wire connected to male and female Pin 1; shield connected to male connector shell; female connector shell floating.

6. Third wire connected to male and female Pin 1; shield connected to female connector shell; male connector shell floating.

7. Third wire connected to male and female Pin 1; shield connected to both male and female connector shells.

IMPORTANT: Some people advocate not connecting the shield (and/or the third inner conductor if present) to Pin 1 at both ends of the cable, but then differ among themselves as to whether the sole connection should be made at the male or female end. In the following discussion, I will assume the standard configuration in which a male plug will be connected to an input and a female to an output. To begin with, if the cable is to be used with phantom-powered mikes, there must be a current path between both Pin 1s, so any further discussion is moot. If phantom powering is never a consideration (WARNING: “never” is not a valid term in Hollywood), connecting the shield to the male’s Pin 1 will usually provide the greatest protection from EMI (e.g. radio station) pickup; but it can also increase the amount of T-M noise which had previously been C-M. Connecting the shield to the female’s Pin 1 will usually provide the greatest protection from continuing the transmission of C-M noise without converting it to T-M; but now increasing the susceptibility to EMI. I do not believe these purported “benefits” of breaking the Pin 1 interconnection outweigh the potential disadvantages, especially the lack of compatibility with phantom power, and in the production environment. But if you really, really must break the Pin 1 circuit, connect the shield to Pin 1 at the female (input) end.

What to do? Consult a Ouija board. Actually, you could do worse. Or you could make up cables in each of these configurations, and try them oneby- one.

Here’s what I do: most of my cables are 2-conductor, and wired as per Number 4. I have made up several 3-conductor cables wired as per Number 5. On those occasions when I have encountered problems with the 2-conductor cable, the first thing I try is replacing the T-power microphone (e.g. a Sennheiser MKH406) with a phantom-power one (MKH40), or vice versa. This usually eliminates the trouble. IMPORTANT: Sennheiser’s new aluminum-cased mikes have a problem that is often attributed to a bad cable: The case is grounded by a screw near the plug that tightens against a bare patch of aluminum. In about a year or so, the aluminum oxidizes and forms an insulating layer, destroying the integrity of this grounding function and ability of the case to intercept interference. Loosening and retightening the screw a couple of times restores the effectiveness of the connection.

On those occasions when swapping mikes didn’t remedy the problem, I have substituted a 3-conductor cable. But in only two instances was there any improvement. Most of the situations occurred in proximity to AM radio broadcast towers (antennas), and the signal strength was simply so high that nothing could keep it out. One time it was possible to move the recorder very close to the mike, and connect it with a much shorter cable. Interestingly, grounding the sound cart’s chassis to a nearby cold-water pipe made matters far worse. I haven’t used a full digital system in this environment yet, so I don’t know if it will be any more resistant.

NOTE: To help block AM radio or other high-frequency interference, inline 50 to 70 KHz low-pass filters are available that can be inserted next to the mixer’s or recorder’s mike input receptacle. Some sound mixers RF bypass the inner conductors with 0.01-0.02 ìF capacitors inside the male XLR plug. You need a disc ceramic type (low internal inductance) and to keep the two leads as short and straight as possible. Adding a 1/10-watt 50-Ω resistor in series with the 0.01 ìF capacitor will help match the cable impedance and reduce the amount of energy reflected back into the cable. Solder one capacitor-resistor combination between Pin 2 and Pin 1, and another between Pin 3 and Pin 1. If you can get “chip” capacitors and resistors (as used on SMT circuit boards), they have no leads at all, just tinned ends, and are even smaller. Using chip components will make it much easier to install the parts.

I have been able to conduct some experiments on stage with buzzes from H.M.I. lights, and found that both 2- and 3-conductor cables were almost equally affected. Crossing these power cables at right angles was of no help. Only separating the two cables with an apple box worked, but there is always the danger of having the mike cable pulled off the box to land back on the H.M.I. cable. It is better to re-route your audio cable to avoid crossing any electric cables if at all possible.

Another common problem occurs with outdoor cable runs. Electricity always takes the path of least resistance—literally. (However, some of the current will still flow through other paths that have higher resistance). For example, if lighting units are set on the bare ground, there may be a flow of leakage current through the soil between the lamp stands and the grounding point of the generator. Now, if you have a run of interconnected mike cables lying on the ground along this path, some of the AC current will leave the soil where one of the cable connectors is located and flow along the mike cable’s shield until it leaves at the connector at the other end of the cable, closer to the generator. This is a case where having the connector shells floating would protect you, but it is easy enough to cover the connectors with gaffer’s tape. (IMPORTANT: Be sure to leave a folded-over tab to make removal of the tape quick and easy.)

While damp ground can be dealt with by gaffer-taping the connectors, protecting them from actual liquid water requires more extreme measures. The best one is not to do it in the first place: if you know in advance that you will need a long cable run underwater, make up a single continuous length cable. (You can always make several shorter ones out of it afterward.) If only mud or dirt is the problem, Neutrix makes a line of heavy-duty mike connectors. The male has a stainlesssteel barrel which resists deforming when stepped on or run over, and the female has an external rubber boot that mates with the open end of the male shell and also covers the latch button. This combination keeps out non-liquid contamination, and if you apply some silicone sealant inside the cable strain relief, will handle liquid splashes as well (as long as the sealing lip of the rubber boot is not damaged).

For last-minute emergency waterproofing of a pair of mated connectors, “Rescue Tape” brand silicone self-fusing tape can be used (www. rescuetape.com). Start a spiral wrap around the cable, about six inches from a connector, pulling the tape until it is fully stretched (about three times its original length). Completely overlap the first turn, then be sure to overlap the remaining turns almost half the width (be careful to avoid bumps from creating a third layer). Wrap over the two connectors, being sure to maintain the almost 50% overlap. Continue wrapping six inches into the next cable. Finish with the last turn completely overlapping the previous one. Squeeze all the tape with your hands to ensure complete adhesion of the layers. If you’ve done this properly, the connection should be good for submersion under several feet of water, at least for a short time. WARNING: Test your technique in advance. Unfortunately, removing the fused mass afterward is difficult. Slice through it with a sharp blade, gradually going deeper with each pass, and being careful not to nick the cable jacket or connector shells.

Editor’s Note: The next installment will take up issues of interconnecting equipment and optimal sound cart wiring.