Types of Audio Cables: What Each One Is Actually For
Audio cable names tell you the shape of a plug. They do not tell you what the cable does. Three questions decide almost every order. Is the run balanced or not. What signal level runs through it. And is the signal analog or digital. Answer those three and the connector choice falls out on its own.
Buyers usually start at the wrong end. They pick a plug from a photo, then ask a factory to match it.
That is how a headset arrives with the microphone wired to the wrong contact. It is also how a long microphone run picks up hum that no amount of shielding fixes.
This page works from the signal backwards to the plug. It covers the jack family and the wiring trap inside four-pole plugs. Then the level gap between consumer and pro gear, and the seven lines to write on an order.
Two Families People Keep Mixing Up
Every audio cable carries one of two things: a changing voltage, or a stream of numbers. The plug on the end is a separate decision.
Analog cables carry the waveform itself. The signal is a voltage that rises and falls with the sound. Anything the cable picks up along the way stays in the signal. There is no way to tell it apart afterwards.
Digital cables carry samples. The receiver rebuilds the waveform from numbers, so small noise on the line is discarded rather than amplified. That is why a cheap optical cable can sound identical to an expensive one.
The confusion starts because the same plug appears in both families. An XLR carries a microphone signal on one job and a two-channel digital stream on another. A 3.5 mm jack carries stereo analog on a phone and nothing digital at all.
So the plug is not the cable. It is the part you can see.
Balanced and Unbalanced, in Plain Terms
An unbalanced cable has one signal wire and a return that also acts as the shield. A guitar cable is the common example. The screen carries the return current and blocks interference at the same time.
A balanced cable has two signal wires of equal impedance plus a separate shield. The receiver looks only at the difference between the two wires.
Noise from outside hits both wires almost equally. Because the receiver ignores anything the two wires share, that noise is subtracted away. This is the whole trick, and it works because of the impedance match, not because one wire is inverted.
That last point contradicts a common belief. Many balanced outputs drive only one wire, then match the impedance of the other to ground. The result still rejects noise. Engineers call it impedance-balanced, and it is fully balanced in effect.
| Property | Unbalanced | Balanced |
|---|---|---|
| Conductors | One signal plus shared return | Two signals plus separate shield |
| Noise rejection | Shield only | Shield plus common-mode rejection |
| Typical plug | TS jack, RCA, 3.5 mm TRS | 3-pin XLR, 6.35 mm TRS |
| Sensible run length | A few metres | Tens of metres |
| Cost | Lower | Higher, plus extra circuitry at both ends |
Length is where the choice stops being academic. A microphone puts out a small voltage. A long unbalanced run adds noise before the preamp ever sees the signal. Balanced cable fixes that by rejecting the noise at the input.
Inside most professional products the circuitry is unbalanced anyway. The balanced connection exists at the socket, for the cable run between boxes. That is a design detail, not a shortcut. It is worth knowing when a factory quotes you a “fully balanced” product.
The Jack Family: One Shape, Four Wires
The round jack that started life on telephone switchboards is now the most common audio connector in the world. What changes is how many contacts it carries.
| Name | Contacts | Common job |
|---|---|---|
| TS | 2 | Mono instrument cable, unbalanced |
| TRS | 3 | Stereo unbalanced, or one balanced channel |
| TRRS | 4 | Stereo plus a microphone, as on a headset |
| TRRRS | 5 | Stereo, balanced, as on some high-end headphones |
Three diameters share the same shape: 6.35 mm, called a quarter inch in the trade, then 3.5 mm, and 2.5 mm. A fourth size exists for studio patch panels at 4.4 mm, sold under names like Bantam and TT. It looks similar and does not fit a 6.35 mm socket, which is worth knowing before a sample is built.
On a three-contact plug used as one balanced channel, the tip is the signal. The ring is the inverted copy. The sleeve is the shield. Plug a stereo source into that socket and the input subtracts one channel from the other. The listener hears left minus right, which on most music means the vocal disappears.
Three-pin XLR does the same job with three pins. Pin 1 is the shield, pin 2 the signal, pin 3 the inverted copy. The studio trade remembers it as ground, live, return.
The Wiring Trap Inside a Four-Pole Jack
This is the single most costly misunderstanding in small audio products. Most buying guides never mention it.
A four-contact 3.5 mm plug has to carry left, right, ground and a microphone. There are two ways to assign them, and they are not compatible.
| Position | OMTP wiring | CTIA wiring |
|---|---|---|
| Tip | Left audio | Left audio |
| Ring 1 | Right audio | Right audio |
| Ring 2 | Microphone | Ground |
| Sleeve | Ground | Microphone |
The two standards swap the last two contacts. OMTP keeps ground on the sleeve and the microphone on the second ring. That version was accepted as a Chinese national standard, YD/T 1885-2009.
CTIA, also called the American Headset Jack standard, puts the microphone on the sleeve. Ground moves to the second ring. Apple used it on the iPhone up to the 6S and SE. In Western markets it became the default.
Plug a CTIA headset into an OMTP device and there is no ground where the device expects one. The two speakers end up in series and out of phase. In practice the vocals vanish and the track sounds like a backing tape.
Now put yourself in a sourcing conversation. A factory that builds for the domestic market builds to YD/T 1885. An overseas buyer orders “a 3.5 mm headset with a microphone”. What arrives is exactly what the drawing said, which is OMTP. Nothing is defective. The rule was never written down.
The fix is one line on the purchase order: name CTIA, or name YD/T 1885 for the domestic standard. If the product must serve both, ask for a jack that detects the wiring and switches. Expect to pay for the extra contacts.
Three Levels and an 11.8 dB Gap
The second line to write is the signal level. A cable does not fix a mismatch at the ends.
Consumer gear works at minus 10 dBV, which is 0.316 volts root mean square. Professional gear works at plus 4 dBu, which is 1.228 volts. The reference points differ as well: 1 volt for the consumer figure, 0.7746 volts for the professional one.
Put the two side by side and the gap is about 11.8 dB. Feed a consumer output into a professional input and the signal arrives far too quiet. The operator raises gain to compensate, and the noise floor rises with it.
| Level | Nominal | Volts RMS | Who uses it |
|---|---|---|---|
| Microphone | Millivolts | Well under 1 | XLR microphones, needs a preamp |
| Instrument | High impedance | Low | Guitar pickups, needs 47 k to 1 M input |
| Consumer line | Minus 10 dBV | 0.316 | Phones, sound cards, RCA outputs |
| Professional line | Plus 4 dBu | 1.228 | Mixers, interfaces, XLR and TRS |
| Speaker | Watts into 4 to 8 ohms | Several volts | Amplifier output, Speakon or binding post |
Impedance matters here as much as voltage. A line output sits between 100 and 600 ohms and expects an input of 10,000 ohms or more. That mismatch is deliberate: it keeps the output from having to supply current.
Connect a 4 to 8 ohm speaker to a line output and the output is effectively shorted. The sound is faint, and the circuit may not survive. A headphone output is a different animal again, with a source impedance of a few ohms.
One more number that changes how you read a spec sheet. Real audio has peaks far above its nominal level, often 20 dB above. A quote of plus 4 dBu describes the average, not the peak, and the peak can reach 17 volts.
Which Plug Goes With Which Job
Five connector families cover nearly everything a buyer will be asked to quote. The right question is not which is better. It is which one the signal deserves.
| Plug | Signal | Typical product | Why this one |
|---|---|---|---|
| 3-pin XLR | Balanced analog or AES3 digital | Microphones, mixers, stage boxes | Locks in place, three contacts, long runs |
| 6.35 mm TRS | Balanced or stereo unbalanced | Monitors, interfaces, patch bays | Same three-contact logic as XLR, smaller body |
| RCA | Unbalanced consumer line | Home audio, turntables, TVs | One channel per plug, cheap to build |
| Speakon | Speaker level | Powered speakers, amplifiers | Twist lock, thick contacts, cannot be pulled out |
| 3.5 mm TRRS | Stereo plus microphone | Headsets, handsets, body packs | Four contacts in a phone-sized plug |
The XLR shell is the reason it survives on stage. A locking latch costs a few cents and removes a whole class of failure. RCA has no latch, which is fine for a shelf system and wrong for a stage box.
Speaker connectors are a separate world. A loudspeaker cable carries amps, not volts of signal, so it needs thick conductors and no shield. The aim is low resistance against a 4 to 8 ohm load. The connector must also stay in place when a cabinet is moved.
Digital Cables Care About Impedance, Not Gold
Once the signal is digital, two properties decide whether the cable works: impedance and bandwidth. Plating does not enter into it.
Coaxial S/PDIF expects 75 ohms. A professional digital link under the AES3 standard expects 110 ohms on twisted pair, usually with XLR ends. Those two figures are not interchangeable, and a cable of the wrong impedance causes reflections at the far end.
The reflections show up as errors, which the receiver either corrects or mutes. On a short run at a low sample rate nobody notices. On a long run, or at a high rate, the link drops out.
Optical links behave differently again. A plastic fibre carries light, so it cannot pick up electrical noise. It also breaks a ground loop by design. That makes it useful across boxes on different mains circuits.
Shielding follows the same logic in analog gear. A balanced cable’s shield is bonded at one end to the chassis. That keeps noise current off the signal wires. Bond both ends and you build a loop that hums. That detail belongs on the drawing, not in a hope.
The Failures That Come Back
Four faults account for most audio cable returns I hear about.
The first is the wiring order already described. A headset arrives with the microphone and ground swapped, and the buyer reports that “the vocals disappear”.
The second is a long unbalanced run. Someone saved money on a 10 metre RCA lead between a laptop and an amplifier. The result hums whenever a light dimmer is on.
The third is the wrong level. A consumer output feeds a professional input and the gain has to be pushed. The noise floor arrives with the signal. Nobody blames the cable, but the cable is where the mismatch becomes visible.
The fourth is a splitter used backwards. Two outputs joined into one input is never safe. Each output tries to drive the other, both current-limit, and the sound drops or the circuit is damaged. A passive splitter from one output to two inputs is fine. The reverse is not.
Each of these is cheap to prevent and expensive to discover. The cable assembly drawing is where prevention happens, because a factory builds to the drawing and nothing else.
Seven Lines for the Order
A plug name is a shape. These seven lines make it a purchase.
- Family. Say whether the cable carries analog audio or a digital stream. Never leave this to be inferred from the plug.
- Balance. Say balanced or unbalanced, and how many conductors the cable has. Two plus shield and one plus shield are different products.
- Connector and standard. Write the exact size and contact count. For a four-pole jack, write CTIA or YD/T 1885. For digital, write the impedance.
- Level and impedance. Name the nominal level at each end, and the input impedance the product expects. This is where the 11.8 dB gap gets caught.
- Length and loss. Give the length and, for analog, the maximum capacitance the run may carry. Capacitance rolls off high frequencies as the run grows.
- Shield and bonding. Name the shield type and the end it bonds to. Write it down, because “shielded” alone says nothing about where the shield connects.
- Test. Name the continuity and pinout check, the insertion count, and the pull force the plug must survive. Then ask for the test record with the first shipment.
Line three is the one that saves a recall. A drawing that says “3.5 mm 4-pole” describes a shape that two incompatible standards share.
The rest of the specification lives one level up. It covers how the whole product is documented before a factory quotes it. The specification sheet guide covers that layer.
Which Cable for Which Product
Match the cable to the job rather than the other way round.
For a microphone, specify balanced XLR. A small signal over any distance needs the rejection, and the shell needs the latch.
For a studio monitor or an interface, specify balanced 6.35 mm TRS or XLR. For a shelf system or a turntable, RCA is correct and cheaper.
For a loudspeaker, specify a locking speaker connector and thick conductors, with no shield. Do not reuse a shielded instrument cable for speaker duty. The conductor is thin and the shield is not there to carry current.
For a headset or a handset, specify a four-pole jack and name the wiring standard. If the same product ships to several markets, ask the factory which standard it builds to. Then test both before the tooling.
For a digital link, specify the impedance and the length. Coax at 75 ohms for consumer S/PDIF, twisted pair at 110 ohms for AES3, optical where a ground loop is the problem.
A portable speaker usually needs two of these at once. One is a four-pole jack for a wired input, the other a locking or spring terminal for the driver. The portable speaker notes cover how that pair is specified on one drawing.
The safety and emissions tests come after the cable choice, and they will not rescue a pinout error. Ports, plugs and cables change hands in the same districts as the rest of the trade. The component market tour is where most of these parts are first priced.
One Line on the Drawing Prevents a Whole Recall
Send me the product, the market and the plugs you plan to fit. I will tell you which standard the wiring has to follow, which level the two ends expect, and where the drawing is silent in a way a factory will fill in for you. That answer costs nothing before tooling.
Send the PinoutDoes a more expensive audio cable sound better?
On a digital link, no. The receiver rebuilds the numbers and discards small noise. On an analog link, only if the costly cable has lower capacitance, better shielding or a tighter connector. It also matters only over a long run.
Can I use an instrument cable as a speaker cable?
No. Instrument cable has a thin centre conductor and a shield meant for millivolts, not the current a loudspeaker draws. The resistance is too high and the cable can heat.
Can I use a speaker cable as an instrument cable?
No. Speaker cable has no shield, so it picks up hum along its whole length. It is also usually thicker and stiffer, which damages jacks over time.
What is the difference between a 3.5 mm TRS and a TRRS plug?
TRS has three contacts and carries stereo audio with a shared ground. TRRS has four and adds a microphone, which requires a standard for the extra contact. The two common standards disagree.
Why does a microphone cable hum?
Usually a ground loop or a shield bonded at both ends. A balanced cable bonded at one end only keeps noise current off the signal wires. A long unbalanced run will hum no matter how the shield is bonded.
Can I split one output to two inputs?
Yes, with a passive splitter. Never join two outputs into one input, because each output then drives the other and both can be damaged. Use a mixer for that job.
Questions & Comments
We read every one and reply within 24 hoursA plug is a shape. The wiring inside it is a decision. Tell me what the cable connects and which market it ships to, and I will say which standard the drawing has to name.