USB extender cable - can multiple non-active be used to exceed the USB length limit if used with a powered hub?

0

If multiple non-active USB cables are chained together such that the combined cable is longer than the normal USB length limit but are placed between two powered USB hubs, will that effectively create an active cable and be fine even though it's longer than it should be able to be?

What prompted this question in the first place is wanting to buy an active extension cable in a certain color, but not finding one for sale in that color so trying to explore alternative solutions. Putting a hub every 15 feet is not an option for aesthetics reasons.

g491

Posted 2017-02-26T17:24:18.663

Reputation: 162

1What do you mean "active cable" that "isn't sold as active". And what the color of plastic around USB cable has to do with Electrical Engineering or computing? – Ale..chenski – 2017-02-26T18:21:25.737

@AliChen I've updated the question to clarify – g491 – 2017-02-26T18:54:10.373

What You trying to do? AFAIK even passive (without separate PSU) hub regenerate signal and allow longer distance. Or may be USB extender over RJ45 like this http://www.ebay.com/itm/Monoprice-6042-USB-Extender-over-CAT5E-or-CAT6-Connection-up-to-150ft-/391011212329 ?

– Mikhail Moskalev – 2017-02-26T18:57:11.333

How about to spray-paint your active extender cable in any color you wish? – Ale..chenski – 2017-02-26T20:18:29.797

Answers

1

Connecting passive extenders in series is a very bad idea.

In USB 2.0, the length of cable is limited by signal degradation along differential twisted-pair cable. USB 2.0 does not have any sophistication of USB 3.0 as signal pre-emphasis and adaptive equalization of receivers during link training, and relies on plain quality of transmission line between link partners.

The quality of line deteriorates due (a) frequency-dependent losses (attenuation of high-frequency edges), and (b) inter-symbol interference coming from multiple reflections from line imperfections. Even a uniform twisted differential line has certain degree of imperfections, so the end-of line accumulates significant jitter, and receivers fail to decode the data stream and recover embedded clock.

Now, consider how USB connectors are made. First, the pair of mated connectors (receptacle-plug) has quite different geometry, and impedance matching is not always good. Then, the connectors are not designed for automated assembly. The cable must be cut, wire spread out to fit connector spacing, and then soldered by hand. Most of the cable assemblies inside the connectors moldings are horrible from impedance matching standpoint. So when you connect a single extender, you introduce four (4!) bad segments (cable split1 -> receptacle - > plug -> cable split2 into the transmission link. As result, horrible reflections are coming in all directions along the line, and signal deteriorates beyond acceptable. This is why USB specifications explicitly disallows any passive extenders. If you put more extenders, things will go really bad.

The above considerations apply for cables with two junctions, at host side, and device side. Active extenders usually use "captive cable", which is soldered directly at device end, and thus eliminates at least a pair of connector elements and one cable split, so captive cables can do a bit better in terms of signal integrity, and can have longer functional cables.

In short, making a USB cable out of several mechanically interchangeable segments is a really bad idea.

Ale..chenski

Posted 2017-02-26T17:24:18.663

Reputation: 9 749

It seems surprising that an active cable can go up to 30 meters as compared to the normal 5 meters (at least per http://www.yourcablestore.com/USB-Cable-Length-Limitations-And-How-To-Break-Them_ep_42-1.html#maxactive)

– g491 – 2017-02-28T06:10:31.453

@g491, yes, it is surprising, but I am not sure if these products really work. I know that optical cables can go 50m, see Corning Communication products. The permissible length of USB2 cable is limited by its quality AND 1.7 us packet response timeout limit, YourCableStore is completely wrong on their numbers. If one can find good materials and make an exceptionally uniform cable, the length can be 125 -150m, see this answer, http://superuser.com/a/1105099/620011

– Ale..chenski – 2017-02-28T06:43:11.633

0

The designed maximum length for USB 2.0 is 5 meters (~16 ft) and for USB 3.0 the suggested length is less than 3 m. To safely exceed this, use powered hubs and/or repeater cables.

If the goal is to make a multi-color cable, rather than to increase the length, use multi-colored cable or combine short colored cables.

Consider also that daisy-chaining (putting cables in series) increases the likelihood of a connector wiggling loose at an inopportune time and that it degrades signal quality, so transfer rates may be slower or error-prone.

DrMoishe Pippik

Posted 2017-02-26T17:24:18.663

Reputation: 13 291