Connector reference: every plug and port you’ll actually meet in Australia.

USB, the complete guide

Nearly all USB confusion comes from treating three separate things as one. Every USB connection has a shape (the connector — the cards above), a speed (the USB version it talks) and a power level (how it charges). They are specified separately and they mix freely: a USB-C plug can be running a 25-year-old speed, and a 25-year-old USB-A plug can be on the end of a modern fast charger. Work out each one on its own and everything else falls into place.

1. Every USB version, in order

Each version is backwards compatible: a new device on an old port works, at the old port’s speed. USB 1.0 barely shipped; USB 1.1 was the first version people actually used, for keyboards, mice and the first flash drives. USB 2.0 is still what every keyboard, mouse, printer and webcam uses today, because none of them need more.

2. The naming mess, decoded

The speeds are simple. The names are not, because the 5 Gbit/s speed has been officially renamed three times and the 10 Gbit/s speed twice. All of these labels are in shops right now:

So “USB 3.0”, “USB 3.1 Gen 1” and “USB 3.2 Gen 1” are the same thing, and a box that says only “USB 3.2” tells you nothing: it could be 5, 10 or 20 Gbit/s. The current official names are just the speed — “USB 5Gbps”, “USB 10Gbps” and so on. Ignore the version number and look for the figure in Gbps.

3. What the speeds mean in practice

Speeds are quoted in bits; files are measured in bytes. Divide by eight, then take off some overhead:

A spinning hard drive cannot read or write much faster than 200 MB/s, so 5 Gbit/s is already more than any hard drive can use — paying for a 10 Gbit/s enclosure for one is wasted money. The jump that matters is from USB 2.0 to anything newer: if a backup to an external drive is crawling at 30 MB/s, it has fallen back to USB 2.0, and the cause is nearly always a black port, a non-blue cable or a tired connector.

4. Which connector can carry which speed

“Can carry” is the ceiling, not a guarantee. Plenty of USB-C ports — on budget phones, keyboards, and most cheap charging cables — are wired for USB 2.0 only.

5. Power and charging

Every USB port supplies 5 volts. What changed over the years is how much current a device is allowed to draw, and — with Power Delivery — whether the voltage can be raised.

• A bigger charger never forces more power in. With Power Delivery the device asks for a voltage and current and the charger agrees or offers less. A 100 W laptop charger is perfectly safe on a phone.
• The cable is part of the deal. An ordinary USB-C cable is good for 3 A (60 W). 100 W and 240 W need a cable with a chip in the plug — an “e-marker” — that tells the charger it can take 5 A. With the wrong cable a 100 W charger quietly drops to 60 W.
• Portable hard drives sit right at the limit. A 2.5-inch drive can want close to 1 A for a moment as it spins up — twice what a USB 2.0 port must supply. That is why older drives came with a two-headed “Y” cable, and why a drive that clicks and disconnects often behaves on a blue port or a shorter cable.
• Some chargers speak their own dialect. Qualcomm Quick Charge and various phone makers’ fast-charging schemes pre-date Power Delivery. Off-brand combinations still charge — at plain 5 V.

6. USB-C: one plug, many possibilities

USB-C has 24 contacts where USB-A has four, and what they are used for is decided by negotiation every time you plug in. A given port or cable may or may not support each of these, and nothing about the shape tells you:

• Data speed — anything from USB 2.0 to 80 Gbit/s.
• Power — from 2.5 W to 240 W, in either direction. A laptop can charge a phone, or a monitor can charge the laptop, down the same cable.
• Video — “DisplayPort Alternate Mode” sends a monitor signal over the same plug. This is how single-cable docks and USB-C monitors work. Many phones and cheaper laptops leave it out.
• Thunderbolt — a faster, stricter superset. See below.

Cable length is a real limit. A passive USB-C cable can be up to 4 m at USB 2.0 speed, about 2 m at 5 Gbit/s, 1 m at 10 Gbit/s and only 0.8 m at 40 Gbit/s. So a long, thin, cheap USB-C cable is almost certainly USB 2.0 and charging only — fine for a phone charger, useless for a dock or an SSD. If a USB-C monitor or drive “doesn’t work”, swap the cable for the short thick one it came with before anything else.

What to look for. Certified cables and ports carry a logo with the speed and wattage printed on it — “40Gbps”, “240W”. A small D-shaped DisplayPort logo beside a port means it does video. A lightning bolt means Thunderbolt. No markings at all usually means the basics: USB 2.0 and modest charging.

7. Thunderbolt and USB

USB4 was built on Thunderbolt 3, so the two have largely converged. The practical rule: a Thunderbolt port is always also a full USB-C port, but a USB-C port is not necessarily Thunderbolt. A Thunderbolt dock or drive plugged into a plain USB-C port will either run slowly or not at all. Thunderbolt 1 and 2 devices need Apple’s Thunderbolt 3 to Thunderbolt 2 adapter to reach a modern machine.

8. Hosts, devices and On-The-Go

Classic USB is lopsided by design: one host (the computer) runs the show and the devices answer. The connectors enforce it — A plugs go towards the host, B plugs towards the device — which is why you cannot join two computers with an A-to-A cable, and should never try: it shorts their two power supplies together.

USB On-The-Go (2001) let a phone or camera act as a host so you could plug a flash drive or keyboard into it. That is what the fifth pin in Mini and Micro connectors is for, and what the Micro-A plug and Micro-AB port exist to signal. USB-C made the whole idea ordinary: either end can be host, device, power source or power sink, and they sort it out between themselves.

One host can address up to 127 devices, through hubs chained up to five deep. A hub shares one port’s speed and — unless it has its own power supply — one port’s power between everything plugged into it, which is why drives and webcams misbehave on unpowered hubs.

9. Cable length

To go further, use an active extension cable or a powered hub part-way along; each one restarts the count. Plain passive extension leads beyond these lengths are the classic cause of devices that connect, drop out and reconnect.

10. Colour codes

Only blue is in the specification; the rest are conventions that manufacturers follow, bend or ignore, and plenty of well-made equipment uses no colour at all. Treat colour as a hint and the port’s printed label — “SS”, “10”, a battery icon — as the evidence.

11. What’s inside the cable

That is a complete USB 2.0 cable: four wires, with the two data wires twisted together. Mini and Micro connectors add a fifth pin, ID, used only for On-The-Go. USB 3 adds five more wires — a transmit pair, a receive pair and a ground between them — which is why USB 3 cables are thicker and stiffer and why the plugs have nine contacts. USB-C doubles up again so the plug can go in either way, and adds two CC wires that carry the negotiation described above.

A “charge-only” cable simply leaves the two data wires out. It will charge anything and connect nothing — so if a phone charges but the computer never sees it, try a different cable before blaming either device.

Which cable does my iPod need?

“Charging is not supported with this accessory.” Early speaker docks and car kits charged the iPod through the FireWire pins of the 30-pin plug, at 12 volts. From late 2008 — the 4th-generation nano, the 2nd-generation iPod touch and the iPhone 3G — Apple removed FireWire charging, so newer devices play through those docks but won’t charge. A small in-line adapter that converts the 12 V to USB’s 5 V fixes it.

Other iPod leads. Apple’s composite and component AV cables are 30-pin at one end and RCA plugs at the other, for showing video on a TV. The headphone socket is a standard 3.5 mm stereo jack on every model; the 3rd and 4th-generation iPods have a small extra slot beside it for Apple’s wired remote. And any 30-pin iPod will reach a modern Mac or PC with nothing more than a 30-pin to USB cable — the USB end is ordinary USB-A.

Which DVI plug fits which port

Single-link and dual-link are interchangeable in practice — nearly every port has the full set of holes — but if either end or the cable is single link, single-link resolution is all you get. DVI’s digital signal is the same as HDMI’s, so a passive DVI-to-HDMI cable works in either direction — picture only, as DVI carries no sound.

Putting a TV antenna on an SDR dongle

A roof antenna beats the little whip that comes with an SMA-socketed SDR dongle by 10 to 20 dB, and no single PAL-to-SMA adapter is sold here, so it is two cheap parts: a short PAL-to-F-type fly lead into the wall point, and an SMA-male-to-F-female adapter on the dongle. If the wall plate is already F-type, any ordinary F-to-F TV lead and the same adapter will do.

The impedance mismatch — 75 Ω TV coax into a 50 Ω receiver — costs under 0.2 dB and can be ignored on receive; a “matching transformer” is not needed. What does matter is a masthead amplifier: its power injector (a small plugpack with coax in and out, usually behind a TV) puts 12 to 17 volts on the coax on its antenna side. Connect the dongle only to the TV side of the injector, or to a wall point that doesn’t carry the power; most dongles have no DC block, and that voltage will destroy the front end. An inline F-type DC block costs a few dollars and is cheap insurance if you aren’t sure.

Reading an RF adapter kit

The adapter kits sold for SDR dongles are usually a set of short leads with an SMA plug on one end and a different connector on the other: BNC, F-type, N-type, UHF (PL-259), PAL, MCX and RP-SMA. Each card above shows what that far end fits. Before you buy, check two things on your own gear: whether the socket has a pin or a hole, and whether you need the plug or the socket version of each end.

The TV ends (F-type and PAL) are 75 Ω and the radio ends 50 Ω. For receiving, that mismatch is too small to matter. One you may meet that most kits leave out is TNC: a threaded version of BNC, used on some GPS antennas and older radio and mobile-phone gear.

RCA colour code

Because the plugs are all the same, a yellow-white-red AV lead works perfectly well as three separate audio or video leads, and any spare RCA lead will do for composite video in a pinch.

Mono, stereo, balanced: reading the contacts

Balanced means the signal travels on two wires, one a mirror image of the other, inside a shield. Whatever hum and interference the cable picks up lands equally on both, and the receiving end subtracts one from the other, cancelling the noise and keeping the signal. That is why microphone cables can run 50 metres across a stage and a guitar lead starts to buzz after six. A balanced connection needs three contacts per channel, which is exactly what XLR and TRS provide.

Unbalanced connections — RCA, TS jacks, ordinary headphone outputs — use one signal wire and the shield. They are fine over a metre or two between hi-fi components and increasingly noisy beyond that.

Plugging a balanced output into an unbalanced input (or the reverse) usually works, at a lower level and without the noise rejection. The one combination to avoid is phantom power: never switch on 48 V with an unbalanced or adapted lead plugged into that input.

Ethernet wiring order

Hold the plug with the latch facing away from you and the cable hanging down; pin 1 is on the left.

The two differ only in swapping the green and orange pairs. T568A is the convention for fixed cabling in Australian buildings; many factory-made patch leads are T568B. Either works, as long as both ends of the same cable match.

Australian plug ratings

The rule is the same all the way up: a plug fits a power point of its own rating or higher, never lower. Each step changes a pin so that a heavier appliance physically cannot be plugged into a lighter circuit.

IEC leads at a glance

Odd numbers are the socket on the end of the lead; the next even number is the matching inlet on the equipment.

Common barrel sizes

The laptop plugs with a centre pin use that pin to tell the laptop what wattage the charger is. A generic brick with the right voltage but no working ID pin will often power the laptop but refuse to charge the battery.

Reading a power brick label

Ignore the INPUT line — nearly every modern brick accepts 100–240 V. Everything that matters is on the OUTPUT line, and four things must line up with what the device asks for:

• Voltage must match. 12 V means 12 V. A higher voltage can destroy the device in a fraction of a second; a lower one usually just fails to start it.
• Current must be equal or higher. The amp figure on the brick is the most it can supply, not what it forces in. A 12 V 3 A brick is fine on a device that wants 12 V 2 A. The reverse — a 1 A brick on a 2 A device — overheats or cuts out.
• Polarity must match. Compare the symbols, as above.
• DC or AC. A solid line over a dashed line, or the letters DC, means direct current. A wave (~) or AC means an alternating-current output — rare, but some older modems, answering machines and audio gear use it, and the two are never interchangeable.

Wattage is just volts multiplied by amps: 12 V at 2 A is 24 W. And in Australia, a legitimate mains power supply carries the RCM mark — a tick inside a triangle. A brick without one, bought cheaply online, hasn’t been shown to meet Australian electrical safety rules.

External hard drive power

Whether a drive needs a brick at all comes down to its size:

• Portable (2.5-inch) drives and SSDs take all their power from the USB cable. No brick. If one clicks or disconnects, the usual culprit is a weak USB port or a long, thin cable — try a shorter cable or a rear port before assuming the drive has failed.
• Desktop (3.5-inch) drives need a brick, almost always 12 V DC, centre positive, somewhere between 1.5 A and 3 A, and most often on a 5.5 mm barrel plug. The inner diameter and the current rating vary between brands and generations, so match the old brick’s label rather than guessing.
• Some older and multi-drive enclosures use a round 4-pin plug that supplies 12 V and 5 V together. These are not interchangeable with barrel-plug bricks, and the pin wiring differs between makers.

If a desktop drive’s brick is lost, the specification you need is usually printed on the drive’s own case, near the power socket.

Power bricks by drive model

Drives we have had on the bench, with the numbers printed on the box and the case so you can match yours. The regulatory number (R/N) on the underside is often the quickest way to find the right replacement brick.