HDD LED always on

CaffeineMeCaffeineMe Cedar Rapids, IA
edited December 2004 in Hardware
Please help me figure out why my HDD LED is always lit. I've run Adaware and Spybot, always run A/V software, and for the life of me cannot figure out why the LED is always on. I've no clue when it started, but I know it hasn't always been like this. I havent't flipped any connectors for LED's on the mobo, and I have defragged, to no avail. Is there a tool that I can run that will tell me what file(s) or process(es) are thrashing my disk, or is it just a bad LED? The brightness of the LED seems less when it's always on, but when a data read/write occurs, it's noticeably brighter.

Maybe I'm just paranoid. Thanks!

Bill

Comments

  • ThraxThrax 🐌 Austin, TX Icrontian
    edited December 2004
    Try flipping the connector for the HDD LED over on the motherboard.
  • CaffeineMeCaffeineMe Cedar Rapids, IA
    edited December 2004
    That's not it Thrax...I haven't moved that connector in months. Thanks tho!
  • ThraxThrax 🐌 Austin, TX Icrontian
    edited December 2004
    You never know, dude. It might be a shorting contact, and a wiggle just may fix it.
  • ShivianShivian Australia
    edited December 2004
    I'd check task manager for active processes and see if clearing up as much as I could from there gets rid of the problem. If it doesn't then I suspect either foul play (spyware or trojan or something like that) or a dodgy LED.
  • edited December 2004
    You might have a dead drive on one of your IDE cables or one might have the power cable loose or un-plugged.
    I know this because I forgot to plug the power into one of my hdd's once and since it wasn't needed for anything other than storage my PC ran like normal but my hdd LED stayed lit all the time.
  • primesuspectprimesuspect Beepin n' Boopin Detroit, MI Icrontian
    edited December 2004
    Is it only lit after windows is booted? If you booted into a different OS, such as the tiny version of linux that comes on a memtest floppy, does it still do it?

    What about if you boot into DOS or the recovery console?

    The first step is to determine if it's a hardware problem or a windows issue.
  • CaffeineMeCaffeineMe Cedar Rapids, IA
    edited December 2004
    Still pushing Memtest on us all, eh Prime? ;D

    Resolved thanks to Madmat. I had a CD ROM drive unplugged (power), on a sep. IDE channel. Once I returned power to that drive, the LED went back to normal blinkin' functionality. Thanks guys!
  • edited December 2004
    :thumbsup: Very cool.
  • Straight_ManStraight_Man Geeky, in my own way Naples, FL Icrontian
    edited December 2004
    Yes, that will do it. What basicly, in plain english, happens is this:

    IDE uses one controller or controller subcircuit on a motherboard. In theory, this is a two channel controller. BUT, some things are in common. The LED lights tend to be this way. The power connector has two black wires. Those wires are reurn lines for voltage, going BACK to the PSU.

    So, leave power disconnected, or have bad return lines, the drive circuitry builds up a small charge if the drive data cable is plugged in, and the IDE controller on the motherboard gets a response from the drive controller that something is there due to the charge buildup. The BIOS cannot figure out what is there, the IDE controller cannot figure out what kind of device is there, but there is a completed circuit in the drive data cable and the onboard IDE controller on the drive has a slight charge built up in it due to the lack of power return provided by the black wires on the power cable. So, the IDE controller on the motherboard keeps polling the "unknown" drive.

    Thus, there IS activity on the IDE channel this unplugged (as to power) device is connected to. BUT, it is the IDE controller cicruit trying to figure out what is hooked up to it.

    I'm sticking this there because in fact you CAN use this to do a rough detect of what is bad as to device or power cable or data cable. When you replace the item that is bad, you get a sudden secasstion of the problem with the light. If you have a solid IDE LED, try this:

    One at a time you unplug both the power and data cables to each device (with power off). Power up, and if ANY device hooked to either connector on the data cable (and powered up with power cable connected) results in a solid LED light, then either you have a bad data cable or two or more bad devices.

    Odds favor bad data cable in that case, and you can confirm by plugging a known working IDE storage device onto the cable and seeing if the IDE LED goes solid adn stays that way. If yes the LED does stay solid, check power cable connectors, make sure they are not corroded or plugged with dust and that the female pins are locked into connector housing. If in fact the LED stays solid and the connectors are in good shape, you can confirm that power supply 5 and 12 volt power is going into the device with a back-probing of the circuits in power cable. Consider if you have cross-checked this way, that the data cable is bad.

    Try replacing big wide and flat data cable. IF you still have a problem, then the IDE controller circuit on motherboard might be bad, or if this only happens when the computer hasd been on a while, things (North Bridge, CPU, IDe chip if separate) might be overheating.

    Is it possible for a bad IDE device to yeild a bad data cable??? Sure, if the device overheated or ran excess voltage through a data cable, the conductors in that kind of data cable are tiny, intended for very low voltage, and can also be melted as they are aluminum for the 40 conductor kind of cable normally. Not only that, but fold an IDE cable very tightly or bend it SHARPLY, and you can break the aluminum conductors. THEN, you might have a situation where if you plug the data cable into the devices and connector on the motherboard for the other IDE channel, THOSE devices appear to fail.

    Basicly, this is a "divide and conquer" strategy, change ont thing at a time until only one thing is left. Another name for this is the "process of elimination" or "process of isolation." When you figure out what the bad part or parts is or are, you end up eliminating the bad, you have "cut it out" and divided the bad from the good. And, you end up conquering the problem.
  • ThraxThrax 🐌 Austin, TX Icrontian
    edited December 2004
    That's not plain english.
  • CaffeineMeCaffeineMe Cedar Rapids, IA
    edited December 2004
    Thrax wrote:
    That's not plain english.

    No, but it is hella informative! Thanks!!! :thumbsup:
  • floppybootstompfloppybootstomp Greenwich New
    edited December 2004
    Thrax wrote:
    That's not plain english.

    ;D

    That was my first thought ;)
    CaffeineMe wrote:
    No, but it is hella informative! Thanks!!! :thumbsup:

    Agreed.
Sign In or Register to comment.