How seriously does cable length affect USB 3.0 transfer speeds?

Recently I've been thinking of upgrading my computer with some USB 3.0 ports. However, the problem is my computer is some ways away from my desk so that would mean I would need a long cable, approximately 10ft or so.

My question is how badly does a long cable (10ft for example) impact the transfer rates of USB 3.0? At what point will the USB controller revert back to say, high speed mode as opposed to super speed mode? What have you guys experienced?

Thanks

Faken

Posted 2010-10-27T03:33:49.400

Reputation: 1 301

Answers

USB 3.0 boasts 4.8 Gbits/s as long as your cable is within the (approximate) 3 meter threshold. Anything longer and the full speed is not guaranteed. The length limitation is a small step down from USB 2.0's 5 meters.

For greater distances there are USB hubs or special signal extender cable. (Section: What improvements are made to USB 3.0 cables & hubs?)

John T

Posted 2010-10-27T03:33:49.400

Reputation: 149 037

So the speeds work in tiers? – Faken – 2010-10-27T03:46:47.553

11-1 This answer is incorrect. USB3.0 standard does not define fixed cable length. Cable for USB3.0 can be of any length as long as it can sustain SuperSpeed. First cables were around 3 meters. Already there are 5 m cables available which can work at SuperSpeed. In future, longer cables may be constructed. – AndrejaKo – 2010-10-27T08:54:50.177

@AndrejaKo I never said anything about the standard. Although maximum cable length is not specified in the USB 3.0 standard, the electrical properties of the cable and signal quality limitations may limit the practical length to around 3 metres when multi-gigabit transfer rates are desired. http://www.everythingusb.com/superspeed-usb.html

– John T – 2010-10-27T12:53:45.940

"USB 3.0 boasts 4.8 Gbits/s as long as your cable is within the (approximate) 3 meter threshold." to me looks different than your comment. – AndrejaKo – 2010-10-27T15:14:25.437

6@And, 4.8G/s is the max theoretical speed of USB 3.0 which has been tested with a cable 3 meters or less in length. As the article states, signal quality will degrade at greater lengths unless special devices are used. My point stands, I don't know how else to convey this simple concept to you. No - it is not written in blood in the standard -- however, theoretically, the cable itself has limitations. All I am trying to say. You are not "guaranteed" the 4.8G/s speed with longer cables. – John T – 2010-10-27T22:11:51.890

The USB 3.1 specification contains a table of wire thickness versus length:

11.4.7 Wire Gauge Table

Table 11-3 is a table of VBUS/Gnd wire gauges showing the relationship between gauge and maximum length in order to achieve the previously cited voltage drop values. The user should note that these lengths are the maximum length possible to meet the voltage drop budget, thus gauges smaller and lengths greater than the table values will fail to deliver the expected voltage value.

Table 11-3. VBUS /Gnd Wire Gauge vs. Maximum Length
American Stranded Wire Gauge │ Ohms Per 100 │ Meters Maximum Cable Length
(AWG) on VBUS /Gnd           │ (Maximum)    │ (Meters)
   28                        │   23.20      │   0.8
   26                        │   14.60      │   1.3
   24                        │    9.09      │   2.0
   22                        │    5.74      │   3.0
   20                        │    3.58      │   5.3

It also states:

5.5.7 Cable Assembly Length

This specification does not specify cable assembly lengths. A USB 3.1 cable assembly may be of any length as long as it meets all the requirements defined in this specification. The cable assembly voltage drop budget defined in Section 11.4.2 and the cable assembly loss budget defined in Section 5.6.1.3.2, limit the cable assembly length.

Myles Kelvin

Posted 2010-10-27T03:33:49.400

Reputation: 91

"Any length" is a little inaccurate. See Maximum length of a USB Cable?.

– fixer1234 – 2017-01-17T06:19:30.847