new system bootup woes

Discussion in 'PC Hardware' started by Adam, Dec 17, 2014.

  1. Adam

    Adam Guest

    At this point, the possibilities outweigh the impossibilities.
    But, I am "confident" that I did not do anything "callous".

    If anything, those fan assembly designs need much improvement to
    make CPU installation much less likely to damage the mobo.
     
    Adam, Dec 24, 2014
    1. Advertisements

  2. Adam

    Paul Guest

    Manual Early_BIOS Feature Serial?
    SABERTOOTH 990FX E6674 2011.06.07 C8xxxxxxxxxx = 2012/09
    SABERTOOTH 990FX R2.0 E7335 2012.06.25 DirectKey
    SABERTOOTH 990FX/GEN3 R2.0 E8041 2012.12.10 DirectKey

    Should the manufacture date on the first generation of board,
    occur after the R2.0 is released ? It sounds like somehow
    the motherboard was given a newer serial number. If
    the board was actually manufactured in 2011, the battery would be
    flat now.

    I guess when you RMA, you'll find out if there is a problem.
    This sounds like something Asus did.

    The computer I'm typing this on, the board was relatively
    old in terms of the release date. It was stale stock
    sitting in some small seller's store. But the manufacture
    date, was from the first batch. So someone who isn't selling
    a product all that well, they tend to get stuck with the
    first batch, rather than dumbly buying a whole bunch near
    the end of the manufacturing cycle and getting stuck with
    them. They know from their first batch sales,
    whether they have any chance of selling a late
    batch. And that's why a late serial number
    has me scratching my head, in terms of a
    justification.

    Paul
     
    Paul, Dec 24, 2014
    1. Advertisements

  3. Adam

    Godzilla Guest

    Look at your stupid name.
    Fixadent is faster.
    Nothing matters here, doofus.
    You break into Best Buy or something?
    Hmmm ... you surround yourself with computers.
    You use dozens of distros.
    Wonder which one of us is alone.
    I'm betting it's you, dork.
    --
    _____ _______ ____ __ __ _____ _
    / ____|__ __/ __ \| \/ | __ \ | |
    | (___ | | | | | | \ / | |__) | | |
    \___ \ | | | | | | |\/| | ___/ | |
    ____) | | | | |__| | | | | | |_|
    |_____/ |_| \____/|_| |_|_| (_)
     
    Godzilla, Dec 24, 2014
  4. Adam

    Jax Guest

    Good info! :)
     
    Jax, Dec 24, 2014
  5. Adam

    Adam Guest

    There are slight differences in feature between the versions.
    For example, my version still had FireWire/1394,
    which was removed from R2.

    Maybe they keep both versions available briefly to give customers a choice?
    Also, to make sure major problems do not surface (in the field) with the
    newer version?
     
    Adam, Dec 24, 2014
  6. Adam

    Adam Guest

    Yep, good info. Thanks!
     
    Adam, Dec 24, 2014
  7. Adam

    Dustin Guest

    DecadentLinuxUserNumeroUno <>
    Wed, 24 Dec 2014 00:15:50 GMT in
    alt.comp.hardware.pc-homebuilt, wrote the following message:
    I've seen them pop at around 24 volts or so. I've seen power supplies in
    the field providing that or more to the twelve volt rail... Cheap. low
    end junk. Think, best tec.
     
    Dustin, Dec 24, 2014
  8. Adam

    mike Guest

    Second notice! Do the math. I used one ohm so you wouldn't have
    to learn long division.

    I'm pretty sure my resistor, for as long as it lasts, will dissipate
    the same power as yours...spoiler alert: 144 Watts.
    So, you grab your 20W resistor really tight when you do the experiment.

    You continue your long history of finding (or manufacturing)
    something to
    nitpick/denigrate/bitch
    about and thrash the messenger...while you fail to comprehend what
    actually matters.

    When you're drowning in your own effluent, it's a good idea to quit
    insulting the guy holding the life preserver.
     
    mike, Dec 24, 2014
  9. Adam

    Adam Guest

    I submitted RMA request yesterday (Dec 26) via
    http://www.service.asus.com/#!rma-request/cope
    But, still have not heard a peep from them.
     
    Adam, Dec 27, 2014
  10. Adam

    Adam Guest

    Well, the new mobo has arrived...
    http://dlcdnet.asus.com/pub/ASUS/mb/SocketAM3+/SABERTOOTH_990FX_R2.0/E8042_SABERTOOTH_990FX_R2.pdf

    The CMOS battery measures +3V (although not when installed in mobo).
    I had to remove and place on Antec case to measure +3V.

    Should I measure the green/purple of the 24-pin EATXPWR before
    or after CPU/RAM installation (on cardboard of course)?
     
    Adam, Dec 28, 2014
  11. Adam

    Paul Guest

    You can "skip lots of checks" if you want.

    I certainly don't take baby steps with every
    build I do. Some of my baby step runs, are out
    of curiosity. That's why I've tested the "no components
    in motherboard, yes, it will turn on the PSU" test case.
    Just to prove that indeed, there is no dependence on
    installed components. But I don't insist on doing
    that test for every motherboard I buy.

    If you're going to do that test, I recommend adding
    a disk drive or two, to the PSU, to add some
    additional loading.

    *******

    For my last build, I put motherboard, CPU, heatsink/fan,
    RAM, and video card into system, connected keyboard and
    mouse, plugged in the two power cables, and did the whole
    thing in one test (look for BIOS screen). Being prepared
    to back out and pull stuff out of the motherboard if necessary.
    If testing a video card on cardboard, you have to be particularly
    careful to not pull it out of the slot. The heel clip on the card
    helps a bit, but still you have to be careful. You stress
    relieve the cable, so it's not tugging on the card.

    Cardboard testing on the kitchen table is not recommended
    if there are kids around... Because somebody is likely
    to bump the video card and cause something to fall over.

    If you test without a video card, the setup is more stable
    mechanically, but then it'll beep at you. Getting some level
    of beeping, is a good positive sign, because it takes BIOS code
    execution, to make it beep. Systems with build-in video offer
    the advantage of supporting "safe cardboard testing", but
    your gamer level motherboard won't have build-in video.
    Boards with lots of PCI Express lanes, they don't generally
    give you a chipset video as well. There are clearly defined
    "market segments" that prevent that sort of overlap. My newest
    system is like that, lots of PCI Express lanes, and no
    motherboard video. So I have to install a card to see the
    screen.

    Paul
     
    Paul, Dec 28, 2014
  12. Adam

    Paul Guest

    It shouldn't be thermal tape. It's probably phase change material.
    Solid at room temperature. Viscous at higher temperature.

    On a running system, the recommended sequence would be:

    1) Prime95 or CPUBurn for 10 minutes.
    2) Shut down and immediately start work on disassembly.
    (While the metals are still warm).
    3) Undo the clamp. Rotate the heatsink assembly slightly,
    while attempting to work loose the heatsink by lifting.

    That's to try to introduce some air into the gap, so
    the thing will come apart.

    Now, obviously, that procedure only works for a subset
    of all possible conditions. If the system never powered
    up yet, maybe the phase change material hasn't even had
    a chance to turn into cement :)

    It's pretty difficult to apply heat safely, after
    the fact. So warming it now wouldn't be quite as easy.
    Yes, you can get things like heat guns or an electric
    hair dryer. But you don't really know what peak temperature
    that will give, or where you'll be generating that temperature.
    You could melt part of the plastic fan, or exceed the safe
    temperature of the organic CPU packaging (the packaging
    actually limits the CPU temperature, and not the
    silicon die itself). So if you're going to heat the
    thing up, if you think that'll help, the source of
    heat can't be too strong or focused to burn or
    melt something.

    I've had to remove phase change here, and it's a bitch
    to scrape off. It doesn't clean up quite as well with
    isopropyl, and may need some other chemicals. The Arctic
    Clean solvent has orange oil in it (something like
    mono-terpene limonene), which might be in the right
    ballpark for the job.

    If I found an aftermarket product with phase change
    on it from the factory, and the odds were good I'd
    need to disassemble it a couple times over lifetime,
    I'd probably scrape it off when new. And replace
    it with something a bit easier to work with.

    It doesn't seem to hurt the ZIF sockets too mucn,
    to lift out a CPU with the thing locked. That's if you
    do it just the one time. If you do it over and
    over again, eventually you'll tear the socket apart.
    While in the past, there have been snapped off pins, that
    would only happen now if you were unlucky, and lifted
    at a bit of an angle, instead of straight up. If
    you pull it out, and only a couple of pins remain
    holding, and you pull on an angle, it could
    snap one.

    Paul
     
    Paul, Dec 28, 2014
  13. Adam

    Adam Guest

    Luckily, all the pins still look perfect (straight and intact). Whew!

    http://www.tomshardware.com/forum/301098-28-stuck-heat-sink-remove
    With a dropper, I soaked that phase change material with
    isopropyl rubbing alcohol 70% for some minutes.
    Pointed the hair dryer at the crack between the CPU and
    heatsink (where the phase change material is). Then,
    with a plastic stick, I pushed the CPU board along the heatsink,
    which was sitting upside down with the copper facing up.
    The CPU slid loose. Then, I cleaned the heatsink with
    lens cleaning wipes. Thinking about cleaning the CPU with
    lens cleaning wipes after it is installed in the new mobo to
    avoid bending the pins. Waiting for things to dry now.

    BTW, I saw how that phase change material oozed out the sides
    just a tiny bit. So, will look for that with the Arctic Silver 5.
     
    Adam, Dec 28, 2014
  14. I just use the same isopropyl alcohol to clean the heatsink and cpu. No
    real drying time.
     
    Jonathan N. Little, Dec 28, 2014
  15. Adam

    Paul Guest

    This isopropyl business, isn't because it's the perfect solvent.
    It's because if you get the isopropyl on any electronics, it
    won't hurt anything. The isopropyl is actually a piss poor
    solvent for this use, but it's commonly available and won't
    hurt stuff.

    There are people who visit the hardware groups, who use
    gasoline, and I can't imagine a worse choice. In terms
    of safety issues.

    *******

    Using the available MSDS sheets for ArctiClean cleaner kit for computers...

    (First solvent bottle, thermal material remover)
    D-Limonene <--- orange oil
    Methyl Esters of Soybean Oil
    Non-Ionic Surfactant <--- soap

    (Second solvent bottle, used as a surface purifier)
    Dipropylene Glycol Monomethyl Ether <--- no idea, but "miscible in everything"

    The difference there, is the Limonene is more suited
    to non-polar pastes than isopropyl would be. As you would
    expect with products like that, you use one of the solvents
    at a much higher rate than the other one, but they give you
    equal sized bottles.

    Paul
     
    Paul, Dec 29, 2014
  16. Adam

    Adam Guest

    "gasoline" ?!?!? Even the fumes (not to mention flammability) are not safe.

    I just used whatever I had around that made sense.

    The fans spin!!! :)

    purple +5.166V
    green +4.321V (short PWR-GND with screwdriver) => +0.12V

    That was with CPU and RAM (no video card) on cardboard.

    Will install in Antec case and then install video card.

    Thanks to ALL for your help.
     
    Adam, Dec 29, 2014
  17. I would not say piss poor. It works, it degreases, it does no harm to
    the PCB or components...and it wont kill you if you do not choose to
    drink it. This is in contrast to the tetrafluoro chloride that was used
    to clean up boards after wave-soldering.

    Make sure they are smoking when they do it ;-)
     
    Jonathan N. Little, Dec 29, 2014
  18. Adam

    Paul Guest

    See, it's magic :)

    Now, on with the testing.

    Paul
     
    Paul, Dec 29, 2014
  19. Adam

    Adam Guest

    Thanks, I was surprised that the pins were still okay.

    Actually, I was shocked to find the CPU stuck to the heatsink.
    Good thing I didn't put the heatsink down. :)
     
    Adam, Dec 29, 2014
  20. Adam

    Adam Guest

    Amazing magic is right.

    Any good burn-in testing ideas? I read something about 200 hrs?
     
    Adam, Dec 29, 2014
    1. Advertisements

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.