Image Map Image Map
Page 4 of 4 FirstFirst 1234
Results 31 to 40 of 40

Thread: USB internal from GCT-8IT not work

  1. #31

    Default

    So the NEC USB controller seems to be working fine.

    The Intel USB controller is recognized but nothing works on that.

    With Linux running I would rule out BIOS faults as Linux does pretty much all on its own.

    Could it still be a wire swap problem (D+ from first/left channel and D- from second/right channel for example) ?
    The +5V and GND pins can be interchanged without much problems (although newer USB chipsets can switch off ports which are overloaded for example).

  2. #32

    Default

    i have already swap D+ and D- on linux and on win98, it don't work.

    I tested it with a multimeter, otherwise I would have had -5 volts on the multimeter, and the mouse, the peripheral lights up well, if I invert the +5 and the GND it does not work and it can burn out the peripheral , it's an old motherboard and there is no protection, I burnt 3 mice on the ps / 2 port because the +5 v was reversed, I solved the problem of the ps / 2 port by rewiring it correctly.

    I think is a bug from this chipset.

  3. #33

    Default

    Code:
    Intel PIIX4 (82371AB)
    
        The PIIX4, used in the Intel 430TX chipset, is USB 1.1-compliant, despite the fact that its datasheet claims only USB 1.0 compliance.
     The PIIX4 supports one USB 1.1 UHCI and two root hub ports. Although this chipset should in theory support modern USB 1.1 devices,
     it is quite buggy and we therefore recommend not attempting to use it. Again, replace the motherboard or install a modern add-on USB adapter.
     If for some reason you must use the 82371AB, you may be able to resolve some problems by substituting the updated 82371AB/EB driver.
    
        	
    
        For the Intel PIIX4 and later Intel chipsets, we suggest as a first step that you visit http://developer.intel.com/design/so...atform/inf.htm
     and download the most recent Intel Chipset Software Installation Utility. Heed all cautions listed.

  4. #34

    Default

    Quote Originally Posted by andromeda92 View Post
    I plugged an adata usb key on a pci usb2 card so it works correctly, it's normal.

    On the internal USB 1.10 port I plugged in a mouse and an old USB key.
    The logs show a USB keyboard/mouse HID thingy on the EHCI/OHCI PCI card, and nothing on the UHCI ports. If I understand correctly, we should have seen another mouse and mass storage device, which we don't. That is concerning.

    Quote Originally Posted by andromeda92 View Post
    On internal USB 1.10 i have 4.89V, when i connect USB data key, i have 4.86 V.
    That is acceptable, not power supply issue here.

    Quote Originally Posted by gertk View Post
    Could it still be a wire swap problem (D+ from first/left channel and D- from second/right channel for example) ?
    No. If D+/D- are swapped, then the speed detection fails (i.e. it will try to detect a full-speed device as low-speed and vice versa). That causes severe communication errors and Linux will repeatedly reset the USB bus. We don't see any of this in this log. Overcurrent events also appear in the logs and cause USB bus resets.

    Quote Originally Posted by andromeda92 View Post
    I think is a bug from this chipset.
    I think this is a broken chip. The host controller is obviously somewhat USB compatible and can handle some USB devices, so there should at least be some reaction when plugging in devices. Since there is absolutely nothing, I suspect the USB data lines to not be connected at all - could be a cable problem, broken or burnt traces, chip bonding issues. Something along those lines.

    Edit: The USB device detection is already triggered by having a pullup/pulldown resistor on the data lines. No actual device is necessary. This should cause lots of messages, both in Windows and Linux. We don't see this, therefore I suspect the data lines to not be connected correctly.

  5. #35

    Default

    I tried several cables (2 * 5 pin) and it didn't work, the connections are correct, the internal connector is the same as that of the GCT-8ITB card (+ 5v, data-, data +, gnd).

    May be that the circuit is cut at the level of the printed circuit, a cut track for example.

    Otherwise there is a mouse connected but on the PS / 2 ports connected to the motherboard using a 1 * 4 pin connector, this is what we see in the log, but which is not has nothing to do with the usb port.

  6. #36

    Default

    Quote Originally Posted by andromeda92 View Post
    I tried several cables (2 * 5 pin) and it didn't work, the connections are correct, the internal connector is the same as that of the GCT-8ITB card (+ 5v, data-, data +, gnd).
    If you have used the wrong cable in the past and accidentally destroyed something, the cable pinout won't matter anymore.

    Quote Originally Posted by andromeda92 View Post
    Otherwise there is a mouse connected but on the PS / 2 ports connected to the motherboard using a 1 * 4 pin connector, this is what we see in the log, but which is not has nothing to do with the usb port.
    To be honest, I don't understand your writing. It is not sufficiently clear to me what is connected, how it is connected, what I should expect when reading the logs and what you specifically did. I did not look at PS/2 devices, but USB only.

    Anyway, assuming that what you said you did was done correctly, this whole thing smells like a hardware defect. So it doesn't matter.

  7. #37

    Default

    I've had success with a Intel 430HX motherboard on a IBM PC330. You must use NUSB and make sure you uninstall the standard windows USB root hub driver first and make sure you use Windows 98SE. USB mice and keyboards worked fine, but only once windows has loaded of course since the BIOSes on many of the classic Pentium motherboards do not support USB keyboards.

  8. #38

    Default

    I've used external hard drives of up to 1Tb with the Win 98SE 48 bit LBA patch. Formatted FAT 32 on a USB 1.0 port. The header was already integrated on the motherboard though. It was a IBM PC 330 i430HX ,BIOS date Dec/1996. USB is just an enhanced serial port and it's not worried about the capacity on these old motherboards since the BIOS itself does not have support for USB disks.

  9. #39

    Default

    I wonder if it is a standard usb pinout? I had a Compaq Presario 5000 with front mounted USB ports, and when I replaced the oem board with a Gigabyte board, the connector was made for a different pin out.

  10. #40

    Default

    @Moondog: Yes the pinout is standard, i show in the documentation.

    since I dropped, as indicated in the intel documentation, this chipset has a bug and it recommends changing the motherboard, but there is a version of this chipset that works, I do not remember, it seems to me that it is is an N version.

    I have installed PCI card USB2 with chipset NEC and it work fine under win 98Se

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •