The name bundles three promises: a minimum lane rate, a floor on the tunneled PCIe payload, and a charging level on the same port. Folding the bus into USB4 did not change what the copper can carry, so the figure on the jacket still has to be read alongside the length of the run.

Two constructions answer the same question, and the wider picture is set out in our notes on where USB 3.2 and USB4 sit in the same family. One is a passive assembly that simply links the two ends; the other carries active electronics that recondition the signal and draw a little power to do it.

The first decision is passive or active, and the run length settles it. The rest are the PCIe floor, the charging level, and the documents that travel with the shipment.

What does a Thunderbolt 4 cable carry on one plug?

Inside the assembly are the high-speed pairs, a low-speed sideband channel and the power path. The host negotiates a contract over the configuration channel, then hands the link its lanes, so data, tunneled PCIe and video end up sharing one budget rather than three.

The 40 Gbps figure is the raw lane rate. With 128b/132b encoding the payload lands near 38.8 Gbps, about 4.85 GB/s, and that is the number to plan a drive or a capture card against rather than the number on the box. Thunderbolt 4 also sets a floor of 32 Gbps for the tunneled PCIe payload, near 3.9 GB/s after the same encoding, which is roughly double the floor the previous generation set.

Where the bench needs the full lane rate at a length past what copper will hold, a Chinese-made Thunderbolt 4 cable built as an active assembly is the part to ask for; where the run is short, the passive version is cheaper and simpler. Eilinks quotes both, because the answer depends on the run rather than on the plug.

Where a Thunderbolt 4 cable fits and where it does not

The natural fits are docking stations, multi-screen desks and bench rigs that move large files to an external array. A single-screen office desk at a lower rate rarely needs what the assembly costs, and a laptop that only charges is served by a plainer lead.

The poor fits are the ones where length is the problem. The USB Type-C specification caps a 40 Gbps passive assembly at 0.8 m, so anything longer is built with active electronics, and that is the point where a purchase either matches the listing or comes back.

Choosing a Thunderbolt 4 or USB4 cable usually comes down to the tunneled PCIe floor rather than to the shape of the plug. A bench that wants one lead to a dock should start from the Thunderbolt 4 cable listing with the 40G and 240W ratings, because the listing states the construction as well as the length. Eilinks usually asks for the run length and the dock rating before quoting, since those two answers decide whether the lane rate is what the bench needed or only what the box said.

How much charging can one plug pass?

Charging is negotiated separately from the data. USB Power Delivery defines a 20 V, 5 A contract, so 100 W, and the extended power range reaches 48 V at 5 A, so 240 W, on the same connector.

A full-featured plug carries 24 contacts, and the power path uses a separate set from the high-speed pairs, which is why a cable can hold its lane rate while the current on the power path changes under load.

What a bench actually receives depends on the charger rather than on the lead, so the rating to write on the drawing is the contract the host will accept. Our notes on reading the power and speed marks on a cable show where that figure is printed on the jacket.

What to confirm before ordering

Six items decide whether a 40 Gbps assembly survives its first week on the bench. The first three are physical; the last three are the ones that come back as a support ticket rather than as a part.

A drawing that names the length, the lane rate and the charging contract closes five of those six before a sample is requested. Eilinks reads the same six off every enquiry, and a one-line request tends to come back as another question.

Documents and conformity requirements for export orders

Documentation is where an otherwise sound purchase stalls. CE marking requirements for Thunderbolt 4 cable sit with the finished product placed on the market, so the importer carries the file rather than the component maker. USB-IF certification requirements for Thunderbolt 4 cable apply where the logo or the certified name is used on the pack.

The lane rate and the passive length cap behind those figures are published in the USB4 specification at usb.org, and that is the reference a drawing should point at when the label on the box says 40 Gbps and the schedule says 0.8 m.

Who decides what on a 40 Gbps cable order

Confusion here costs more than the part does. The facilities team measures the run and the mounting, the IT team owns the driver policy, and the sourcing team owns the drawing, and our note on how far a passive 40Gbps run reaches is the comparison most of them start from.

A Thunderbolt 4 cable manufacturer can only quote what those three have written down, which is why the first reply is usually a question about the length rather than a number. A rate drop in a Thunderbolt 4 cable traces back to the run more often than to the plug, and the same three parties are the ones who can answer it.

Where the run crosses a wall or a rack, the length on the drawing has to be measured rather than estimated, because a metre of difference is the gap between a passive part and an active one. A 40Gbps or 20Gbps cable choice changes that boundary as well, and the lower rate is far more forgiving of distance.

Dongguan Eilinks works from the drawing rather than from a listing, and a Thunderbolt 4 cable factory in China will ask about the run before asking about the quantity.

Questions buyers ask

Is a passive assembly enough for a 2 m run?

At 40 Gbps it usually is not. The Type-C specification caps a passive 40 Gbps assembly at 0.8 m, so a 2 m run at that rate needs active electronics in the cable, and the listing should say so in as many words.

Will the cable charge a laptop at 100 W?

With a Power Delivery contract and a capable charger, yes. The 20 V 5 A contract is negotiated over the configuration channel, and the power path sits on contacts that are separate from the high-speed pairs.

Why does a long run fall back to a lower rate?

The signal at the far end loses margin with distance, and the link retrains at a rate it can hold. That is a length problem rather than a plug problem, which is why the run is the first question on the drawing.

What belongs on the drawing for a repeat order?

The run length, the lane rate, the construction type, the charging contract, and which party holds the technical file. Those five decide whether the second batch matches the first sample.

Where the checks lead

A Thunderbolt 4 cable factory in China is judged on the same answers, and the useful test is whether the reply describes the part or repeats the listing. A Thunderbolt 4 cable for bench setups is specified by length first, because length decides passive or active.

Send the run length, the lane rate and the charger rating to the Eilinks Thunderbolt 4 cable desk and the assembly can be matched to the bench.

The name causes some of the trouble. USB 3.0 is the retail name for what the USB-IF now calls USB 3.2 Gen 1, and both describe the same 5 Gbps rate on the wire.

Nothing about the plug changed when the name did, so the questions are the ones that apply to any extension: how long the run is and what the device at the far end accepts. Our notes on USB-C cable types from USB 2.0 to USB4 set the wider frame.

Where the run is short, the extension is a convenience. Where it is long, it becomes part of the signal path.

What does a USB 3.0 extension cable carry?

A USB 3.0 cable keeps the four USB 2.0 conductors and adds five more. Two shielded differential pairs carry the SuperSpeed traffic and a drain wire returns it, which is why a full-size USB 3.0 connector has 9 contacts rather than four.

Those extra contacts are why a USB 3.0 extension cable has to be built for SuperSpeed rather than adapted from an older lead. A USB 2.0 extension carries four wires and no path for the second pair, so it caps any USB 3.0 device behind it at High Speed.

The arithmetic is worth seeing once. 480 Mbps works out at roughly 60 MB/s before protocol overhead, while 5 Gbps with 8b/10b encoding leaves close to 4 Gbps of payload, near 500 MB/s. That is the gap a buyer notices when a transfer slows down.

Current is the other half of the story. USB 2.0 allows five unit loads of 100 mA, so 500 mA; USB 3.0 defines a 150 mA unit load and allows six of them, so 900 mA. A bus-powered drive that stalls on an old lead can behave differently on one built for SuperSpeed.

Where the run itself matters, a Chinese-made USB 3.0 extension cable sized to the measured distance usually costs less than reworking the layout after the kit is built.

How far does a USB 3.0 extension cable reach?

Length is where a USB 3.0 extension cable most often disappoints. The USB 2.0 specification fixes a 5 m maximum for a single passive run, while the USB 3.x specifications publish no single number and instead specify an insertion-loss budget, which in practice puts a passive 5 Gbps assembly near 3 m.

The budget is what the two SuperSpeed pairs have to satisfy end to end, so every extra metre of copper and every extra contact pair spends part of it.

Eilinks Eletronics holds the conductor geometry and the shielding layers on one drawing, which is what lets a run be quoted by measured length rather than by a nominal part number copied from a listing.

Where a run has to go past the passive limit, the honest options are an active assembly with a repeater or a shorter device lead at the far end.

Why does the link fall back to USB 2.0 speeds?

A fall back to High Speed usually means the SuperSpeed pairs did not train. The host tries 5 Gbps, fails to get a channel it can trust, and settles at 480 Mbps without raising an error most users ever see.

The same question appears on a bench where the lead is handled every day. The pairs need a channel they can hold every time the plug is touched, and a run that trains once on a clean desk can retrain differently once the lead is fixed alongside a mains cable.

A braided USB 3.0 extension cable with a shielded pair changes the mechanical picture: the braid takes the abrasion and the foil holds the pair geometry when the lead is bent near the plug.

Eilinks quotes these assemblies by the run length the installer measures, because a length taken from a catalogue page is the single most common reason a USB 3.0 extension cable falls back to 480 Mbps in service.

USB 3.0 extension cable vs USB 2.0 extension

The two look close to identical from the outside and behave very differently on the wire. A USB 2.0 extension has four conductors and no SuperSpeed pairs, so putting one in front of a USB 3.0 device removes the 5 Gbps path entirely.

A USB 3.0 extension cable is thicker for a reason: the two added pairs, the drain wire and the shielding all take space inside the jacket.

For wall plates and cabinet panels the panel side matters as much as the run itself: a USB 3.0 keystone jack extension for a wall plate puts the socket where the installer needs it and keeps the pigtail short.

How to check a USB 3.0 extension cable before ordering

Four checks settle most of the risk: the measured run length, the current the device draws, the shielding the environment demands, and the strain relief at each plug. Each one maps to a failure that only shows up after the kit is in service.

Two of the four are documentation rather than hardware. USB-IF certification requirements for USB 3.0 extension cable describe what the assembly has to do, and the body that publishes them keeps the USB 3.2 specification as the reference for the 5 Gbps rate, while CE marking requirements for USB 3.0 extension cable sit with the finished product placed on the market.

Sampling is the other half: a shielded assembly can only be judged in the run it was bought for.

Where these extension cables are still specified

A USB 3.0 extension cable for bench rigs is the most common request, followed by a USB 3.0 extension cable for kiosks where the host sits behind a panel. Dongguan Eilinks supplies both with a moulded strain relief so the plug is not loaded sideways by the weight of the lead.

On a bench the failure mode is abrasion against a metal edge; inside a cabinet it is heat and a fixed bend close to the port.

Where the layout is still on paper, our notes on extension planning for fixtures, benches and enclosures cover the same questions one stage earlier, before the panel is cut.

China USB 3.0 extension cable supplier checks

Three questions separate a partner who can support a repeat programme from one who only ships boxes. Can they state the conductor count from a drawing? Can they hold the shielding across batches? And can they document the assembly without claiming a certification the part does not carry? Eilinks is a USB 3.0 extension cable manufacturer and answers all three from the drawing rather than from a catalogue page.

A China USB 3.0 extension cable supplier and a USB 3.0 extension cable factory in China are judged on the same three answers, and the useful test is whether the reply describes the part or merely repeats the listing.

A USB 3.0 extension cable built in China for an export programme still has to satisfy the same channel budget, so the country of origin changes the document trail rather than the physics of the run.

Questions buyers ask

Can a USB 3.0 extension cable run 10 Gbps?

No. USB 3.0 is USB 3.2 Gen 1, which is 5 Gbps on the wire. A 10 Gbps link needs USB 3.2 Gen 2 hardware at both ends, and an extension rated for 5 Gbps does not become faster because a device asks for it.

Does a longer USB 3.0 extension cable slow the transfer down?

Length does not lower the rate by itself, up to the point where the channel budget is spent. Past that point the link retrains at 480 Mbps, which looks like a slow transfer but is a change of rate.

Why does the packaging say USB 3.0 when the spec says USB 3.2?

USB 3.0 is the retail name that stuck, and the USB-IF later renamed it USB 3.2 Gen 1. The same 5 Gbps assembly appears under both labels, so a drawing that names only one invites the wrong part.

What belongs on the drawing for a repeat order?

The measured length, the conductor count, the shielding layers, the jacket material and the strain relief profile. Those five items decide whether the second batch matches the first approved sample.

Where the checks lead

Almost every USB 3.0 extension cable problem traces back to three points: the measured run, the current the device asks for, and the shielding the environment demands. Settle those three and the extension stops being the part that gets blamed first.

Send the measured run length and the device model to the Eilinks extension cable sample desk and the assembly can be matched to the bench rather than the bench rebuilt around it.