Sweet rig! Thanks for posting the details. I agree that after your
tests, the wire sounds fine. You might want try the same thing while
twisting and bending the damaged part.
On Mon, 2009-09-14 at 07:38 -0400, Pete Theisen wrote:
> draeath wrote:
> > Although not ideal, the shield can be connected to ground without
> > issue. D+, D-, and Vcc should all be completely isolated. You can
> > look up the specs and figure out which pin is what, based on your
> > connector type.
>
> Hi Draeath,
>
> The shield is connected only to itself within the cable. I didn't change
> the connection and the damage didn't change it either. What the signal
> does inside these old no-name P IIs I run, no telling.
>
> Connector type? Hmm, USB 1 something on the computers, USB 2 on the
> cable. I suppose they are sort-of the same. It is an extension cable -
> gives you the USB plug from the back of the box to somewhere 6 feet
> away, Saves the back.
>
> Good idea to look at the specs, thanks for that tip.
>
> The specs are 11.2 MB zipped and download into a folder with 19 items
> including two subfolders. This caught my eye: Title: 5V Short Circuit
> Withstand Requirement Change.
>
> The USB 2 *was* required to withstand a 5V short for 24 hours, the
> proposal is to remove the requirement and replace it with a
> "recommendation" that the equipment withstand a 5V short for 24 hours.
>
> An excerpt:
>
> "Lab tests were also conducted to comprehend the probability of a VBUS
> short to Dp or Dm when the cables were crushed or cut. No failures were
> seen under “normal,” unintentional,damage circumstance like rolling a
> chair across a cable, shutting it in a metal cabinet, or a cable
> cut with scissors. In order to see what shorts could occur, more extreme
> (though improbable) tests were conducted. Fifty High Speed cables were
> crushed in a vise until a short occurred. In every case, VBUS shorted to
> ground first. This is protected by the over-current mechanism. Any short
> that might have occurred to a data signal line would be masked by a
> short to ground. Another extreme test condition used diagonal cutters or
> a utility knife to cut the cable. In this case, when a short was
> observed, 713 out of 713 High Speed cable cut shorted VBUS to ground."
>
> Well, the screw damage pierced the insulation, frayed the shield and
> left the inside wires partially unshielded but undamaged. I re-shielded
> it with some conductive tape, re-insulated the package with black
> electrical tape then tested everything for continuity and isolation and
> it all passed. There is no short and no open circuit.
>
> If there ever is a short it is supposed to withstand a 5V short for 24
> hours with no damage in USB 2, the proposal is to downgrade the
> requirement to a "recommendation", but I don't know if that passed and
> if it effects my equipment, which is USB 2 as opposed to, say, USB 2x.
>
> The cable cost over $20 and it is brand new. I feel like risking it
> since the cable is worth more than the computer it is attached to and it
> tests OK with the meter. If I fry it I will post that to the list.
>
> After all that I suppose I should tell you what I am doing and how I
> came to put a screw through the cable. I am building a Virtual Organ.
>
> This Virtual Organ consists of an embedded system running Puppy Linux on
> a 4 GB USB 2 flash drive. The software bundle includes programs called
> jOrgan, Fluidsynth, LinuxSampler and some others. I was working on
> mounting stuff inside the console. The cable was under something I was
> mounting with a wood screw. A recent picture of the project.
>
> http://tinyurl.com/mom4b4
>
> >>>>>>> I ran a screw through a brand new USB 2 cable, long
> >>>>>>> story. Exploratory surgery reveals that the insulation is
> >>>>>>> penetrated and maybe half of the shield wires are broken,
> >>>>>>> but the copper wire and the aluminum wrapped thingie are
> >>>>>>> untouched.
> >>>>>>>
> >>>>>>> Is it shot or not? If it was a shielded audio or TV cable
> >>>>>>> I would just tape it up and not worry about it but I
> >>>>>>> don't know a lot about USB.
>
> >>>>>> There are (up to) four conductors, two for power and two
> >>>>>> for signal. The shielding's probably to keep the signal
> >>>>>> from leaking out and other RF noise from leaking in. If
> >>>>>> the wires are intact I'd just tape it up (preferably with
> >>>>>> some nonconductive tape, but you're only talking about
> >>>>>> 500mA @ 5V) and use it like normal, but don't put too much
> >>>>>> stress on that part of the cable.
>
> >>>>> Thanks, I think I'll do that right now.
> >>>>>
> >>>>> I wrapped it with some aluminum AC tape first to replace the
> >>>>> shield, now non-conductive tape over that. If my keyboard and
> >>>>> mouse both start getting flaky at once I will know why.
>
> >>>> Uh, wait, what? Highly conductive things next to
> >>>> possibly-frayed wire is a no-no. If you must use Al tape, use
> >>>> something else inside it.
>
> >>> Only the shield is frayed, I thought I could repair the shield.
> >>> On the other hand, I have a Sperry SP-5A multimeter, is there a
> >>> way to check if everything is OK?
>
> >> Testing now . . . done.
> >>
> >> I would expect the wires to test continuous with themselves,
> >> isolated from each other and isolated from the ground. I would
> >> expect the shield to test continuous with itself and isolated from
> >> the wires. This is the case. Am I home free?
-----------------------------------------------------------------------
This list is provided as an unmoderated internet service by Networked
Knowledge Systems (NKS). Views and opinions expressed in messages
posted are those of the author and do not necessarily reflect the
official policy or position of NKS or any of its employees.
This archive was generated by hypermail 2.1.3 : Fri Aug 01 2014 - 13:32:09 EDT