Le mardi 18 avril 2006 à 11:47 +0200, Petr Kocmid a écrit : > On Tuesday 18 April 2006 10:14, Frédéric Grosshans wrote: > > > Any idea left ? > > So, you definitely have a hardware problem. Digging for the id of your device > 1043 8006 reveals a linux kernel mailing list archive thread: > > http://www.ussg.iu.edu/hypermail/linux/kernel/0410.0/0023.html > > with the identical problem: the device worked in some computer and not in > other.
That's a bad news, since the goal of an USB stick ist precisely to move data from one computer to another. <reading the complete thread, at http://marc.theaimsgroup.com/?t=109656092700004&r=1&w=2 > what is strange, is that the 1043:8006 was precisely the identity of the stick on the working computer. Futhermore, this usb stick has worked on the same computer. > I see two possibilities: a slave controller chip incompatibility or > insufficient power problem, device wants to sink more current than port > actually provides at +5V. Port should provide up to 500mA, while many > notebooks are very weak at usb power and do not keep up the standard. > > To eliminate the first one, you should seek for the computer (or maybe an > external usb hub) which will work with that chip. That's currently impractical for me. I may try several computers when I'm back home. > To eliminate the power issue on your equipment, you can try to measure > consumption at the +5V with some prepared usb cable or even try to feed the > device from an external power source. You will need laboratory equipment to > do it (A regulated laboratory power supply with current limitation). Ask some > electronics engineer. "never give up" is actually your motto ! I could look for such equipement here (I'm in the physics department of a Chinese university), but I do not feel like it (I already destroyed to much electronic equipment during my PhF thesis !) > > As a first aid, try an external usb hub with it's own power supply. I'll try when I'm back home (in may) > 1G flash chips require a lot of power to operate, and not having enough is > consistent with your symptoms of "no media". I didn't know. I naively thought "the bigger / the better" and I don't really need something so big... Could that be that the power consumption of the stick increases with the volume of the files on it ? Which would explain the correctness of the first tests (with small files) and the problem when the key holds a bigger file... By the way, if it's a hardware problem, would it be useful to get the stick replaced by another of the same model ? Or is that a model problem ? Fred -- gentoo-user@gentoo.org mailing list
