Loading ...
Sorry, an error occurred while loading the content.

Re: [LinkStation_General] Re: Vexing LS1 hardware issue

Expand Messages
  • KeepIt SimpleStupid
    ... Could the fan be installed so the airflow is in the wrong direction? ____________________________________________________________________________________
    Message 1 of 6 , Aug 6 8:13 PM
    • 0 Attachment
      --- musikgoat <musikgoat@...> wrote:

      > --- In LinkStation_General@yahoogroups.com,
      > Christopher Eykamp
      > <chris@...> wrote:
      > >
      > > Hello,
      > >
      > > I replaced the fan in LS1 (running Gentoo) and am
      > experiencing a very
      > > vexing issue. When the machine is disassembled,
      > it runs just fine, for
      > > days on end, under heavy load, with no problems.
      > When I reassemble the
      > > machine, however, it will only run for a short
      > period (10 seconds to 45
      > > minutes), and will almost always shut down under
      > heavy load (bzipping
      > > mpeg files is particularly good at triggering
      > shutdown ;-). After it
      > > shuts down, I must unplug it briefly, and let it
      > sit for a little while
      > > before it will restart. The shutdown is sudden
      > and uncontrolled. No
      > > logging, no nothing, just a sudden, instantaneous
      > shutdown.
      > >
      > > This would seem to indicate a heat issue, though
      > I'm not sure enough
      > > heat can build up in just 10 seconds to shut the
      > machine down. The new
      > > fan is rated at 5CFM, and is rated at half the
      > power draw of the old
      > > one. It seems to be running fine, both when
      > assembled and when
      > > disassembled. The hard drive never gets above 45
      > C (based on hddtemp).
      > >
      > > Any idea what is going on?
      > >
      > > Thanks,
      > >
      > > Chris
      > >
      >
      >
      > Can you specifically correlate the issue to heat?
      > As in other parts
      > of the system. If you put the device near an A/C
      > or other very cool
      > area, can you get longer life?
      >
      > I'm not quite sure what the issue could be other
      > than that. There
      > isn't any loose wires or anything close to
      > ungrounded metal is there?
      >
      >
      Could the fan be installed so the airflow is in the
      wrong direction?



      ____________________________________________________________________________________
      Boardwalk for $500? In 2007? Ha! Play Monopoly Here and Now (it's updated for today's economy) at Yahoo! Games.
      http://get.games.yahoo.com/proddesc?gamekey=monopolyherenow
    • James Stewart
      ... That depends on if you let the unit completely cool off before restarting it. With a quick restart, the unit is probably still at an elevated temperature
      Message 2 of 6 , Aug 7 3:40 PM
      • 0 Attachment
        --- In LinkStation_General@yahoogroups.com, Christopher Eykamp
        <chris@...> wrote:

        > This would seem to indicate a heat issue, though I'm not sure enough
        > heat can build up in just 10 seconds to shut the machine down.

        That depends on if you let the unit completely cool off before
        restarting it. With a quick restart, the unit is probably still at an
        elevated temperature and the heavy load of booting could quickly get
        it past the limit.

        I'm a little concerned why your unit would then be so sensitive to
        overheating that the reduction to a 5CFM fan would cause a problem.
        Did you install a new hard drive that runs hotter much hotter (and
        takes more power from the power supply) than the old one. My LS-1 had
        a low-performing Samsung drive in it which ran quite cool and didn't
        take much power. I had to search around quite a bit to find a drive
        that had comparable low heat and power requirements. I finally ended
        up with a single-platter 160-Gig Seagate. Granted this wasn't much of
        an upgrade over the 120-Gig Samsung, but I thought the Samsung was
        getting tired after running for a few years.

        I know people successfully put in bigger, more power consuming drives,
        but I'm just being a long-term reliability nut with my LS.

        It is also possible that since the your LS-1 is probably getting a
        little old, the power supply might be getting weak as well, thus
        starting to drop out of regulation under heavy load. Perhaps you
        could monitor the voltage, preferably with an oscilloscope, worse case
        just replace the smaller sized (thus used at higher frequencies)
        electrolytic capacitors in the power supply. Make sure you use high
        quality 105-degree(F) ones designed for switch-mode power supply
        applications.

        Finally, maybe it is a mechanical connection problem with your LS-1.
        For example, maybe when assembled there is different stress on the
        circuit board that could be causing a marginal solder connection to
        fail slightly?

        Good luck with this one. It would be good to run some more tests to
        determine if it is really heat related or not. Try running the unit
        in your refrigerator!
      • Christopher Eykamp
        Thanks to everyone who has responded so far to my overheating problem. I say overheating because I am prepared to declare that is what the problem is. I again
        Message 3 of 6 , Aug 9 1:26 PM
        • 0 Attachment
          Thanks to everyone who has responded so far to my overheating problem.  I say overheating because I am prepared to declare that is what the problem is.  I again disassembled the unit and confirmed that I could run it for several days without issue.  Then I turned it off, and let it sit for 12 hours in a cool room (c. 20C).  I reassembled the unit, checking carefully for any loose wires or cracks or anything else that might cause the system to fail when it was in a different physical configuration (and found no clues there).  The reassembled unit ran for c. 18 mins under heavy load, then failed.  I let it sit for about 1/2 hour, removed the gray "sheathing" to expose some holes in the case, and aimed a room fan on the unit.  I could not make it fail again.  Heat is almost certainly the culprit.

          James: I did indeed replace the hard drive some time ago with one that may well run hotter.  However, I have kept the unit in a hot room and it ran without fail for months.  It was only when I replaced the fan that the failures started.  I noticed you mentioned that the 5CFM fan is a "reduction".  Do you have any idea what the original fan is rated for?

          I may need to replace the fan again with a more powerful unit.  Alternatively, perhaps I can find a way to augment the cooling, perhaps by installing some sort of a heat sink.  Unfortunately, the construction of the unit makes that challenging.  Has anyone found alternative ways to cool their Linkstations?

          Thanks,

          Chris

          James Stewart wrote:

          --- In LinkStation_ General@yahoogro ups.com, Christopher Eykamp
          <chris@...> wrote:

          > This would seem to indicate a heat issue, though I'm not sure enough
          > heat can build up in just 10 seconds to shut the machine down.

          That depends on if you let the unit completely cool off before
          restarting it. With a quick restart, the unit is probably still at an
          elevated temperature and the heavy load of booting could quickly get
          it past the limit.

          I'm a little concerned why your unit would then be so sensitive to
          overheating that the reduction to a 5CFM fan would cause a problem.
          Did you install a new hard drive that runs hotter much hotter (and
          takes more power from the power supply) than the old one. My LS-1 had
          a low-performing Samsung drive in it which ran quite cool and didn't
          take much power. I had to search around quite a bit to find a drive
          that had comparable low heat and power requirements. I finally ended
          up with a single-platter 160-Gig Seagate. Granted this wasn't much of
          an upgrade over the 120-Gig Samsung, but I thought the Samsung was
          getting tired after running for a few years.

          I know people successfully put in bigger, more power consuming drives,
          but I'm just being a long-term reliability nut with my LS.

          It is also possible that since the your LS-1 is probably getting a
          little old, the power supply might be getting weak as well, thus
          starting to drop out of regulation under heavy load. Perhaps you
          could monitor the voltage, preferably with an oscilloscope, worse case
          just replace the smaller sized (thus used at higher frequencies)
          electrolytic capacitors in the power supply. Make sure you use high
          quality 105-degree(F) ones designed for switch-mode power supply
          applications.

          Finally, maybe it is a mechanical connection problem with your LS-1.
          For example, maybe when assembled there is different stress on the
          circuit board that could be causing a marginal solder connection to
          fail slightly?

          Good luck with this one. It would be good to run some more tests to
          determine if it is really heat related or not. Try running the unit
          in your refrigerator!

        • Paul Wright
          I had a problem changing the fan in my LS1. It turned out the problem was using a two pin fan. You must use a three pin fan, and ensure that it is properly
          Message 4 of 6 , Jun 28, 2008
          • 0 Attachment
            I had a problem changing the fan in my LS1. It turned out the problem
            was using a two pin fan. You must use a three pin fan, and ensure that
            it is properly connected.

            P

            --- In LinkStation_General@yahoogroups.com, Christopher Eykamp
            <chris@...> wrote:
            >
            > Thanks to everyone who has responded so far to my overheating problem.
            > I say overheating because I am prepared to declare that is what the
            > problem is. I again disassembled the unit and confirmed that I could
            > run it for several days without issue. Then I turned it off, and
            let it
            > sit for 12 hours in a cool room (c. 20C). I reassembled the unit,
            > checking carefully for any loose wires or cracks or anything else that
            > might cause the system to fail when it was in a different physical
            > configuration (and found no clues there). The reassembled unit ran for
            > c. 18 mins under heavy load, then failed. I let it sit for about 1/2
            > hour, removed the gray "sheathing" to expose some holes in the case,
            and
            > aimed a room fan on the unit. I could not make it fail again. Heat is
            > almost certainly the culprit.
            >
            > James: I did indeed replace the hard drive some time ago with one that
            > may well run hotter. However, I have kept the unit in a hot room
            and it
            > ran without fail for months. It was only when I replaced the fan that
            > the failures started. I noticed you mentioned that the 5CFM fan is a
            > "reduction". Do you have any idea what the original fan is rated for?
            >
            > I may need to replace the fan again with a more powerful unit.
            > Alternatively, perhaps I can find a way to augment the cooling, perhaps
            > by installing some sort of a heat sink. Unfortunately, the
            construction
            > of the unit makes that challenging. Has anyone found alternative ways
            > to cool their Linkstations?
            >
            > Thanks,
            >
            > Chris
            >
            > James Stewart wrote:
            > >
            > > --- In LinkStation_General@yahoogroups.com
            > > <mailto:LinkStation_General%40yahoogroups.com>, Christopher Eykamp
            > > <chris@> wrote:
            > >
            > > > This would seem to indicate a heat issue, though I'm not sure enough
            > > > heat can build up in just 10 seconds to shut the machine down.
            > >
            > > That depends on if you let the unit completely cool off before
            > > restarting it. With a quick restart, the unit is probably still at an
            > > elevated temperature and the heavy load of booting could quickly get
            > > it past the limit.
            > >
            > > I'm a little concerned why your unit would then be so sensitive to
            > > overheating that the reduction to a 5CFM fan would cause a problem.
            > > Did you install a new hard drive that runs hotter much hotter (and
            > > takes more power from the power supply) than the old one. My LS-1 had
            > > a low-performing Samsung drive in it which ran quite cool and didn't
            > > take much power. I had to search around quite a bit to find a drive
            > > that had comparable low heat and power requirements. I finally ended
            > > up with a single-platter 160-Gig Seagate. Granted this wasn't much of
            > > an upgrade over the 120-Gig Samsung, but I thought the Samsung was
            > > getting tired after running for a few years.
            > >
            > > I know people successfully put in bigger, more power consuming drives,
            > > but I'm just being a long-term reliability nut with my LS.
            > >
            > > It is also possible that since the your LS-1 is probably getting a
            > > little old, the power supply might be getting weak as well, thus
            > > starting to drop out of regulation under heavy load. Perhaps you
            > > could monitor the voltage, preferably with an oscilloscope, worse case
            > > just replace the smaller sized (thus used at higher frequencies)
            > > electrolytic capacitors in the power supply. Make sure you use high
            > > quality 105-degree(F) ones designed for switch-mode power supply
            > > applications.
            > >
            > > Finally, maybe it is a mechanical connection problem with your LS-1.
            > > For example, maybe when assembled there is different stress on the
            > > circuit board that could be causing a marginal solder connection to
            > > fail slightly?
            > >
            > > Good luck with this one. It would be good to run some more tests to
            > > determine if it is really heat related or not. Try running the unit
            > > in your refrigerator!
            > >
            > >
            >
          Your message has been successfully submitted and would be delivered to recipients shortly.