Announcement

Collapse
No announcement yet.

Dell Optiplex GX620 with Oozing Rubycon caps

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

    Dell Optiplex GX620 with Oozing Rubycon caps

    Unit is a Dell Optiplex GX620 tower with an Intel Pentium-D945 3.4GHz, 3.5GB RAM DDR2 and a 250GB SATA drive.
    Units does not boot up. MoBo (see pic 'mobo') has two clusters of oozing caps, one around the CPU/GPU (see pic 1) and one around the extra 4-pins 12V power plug (see pic 2). All caps are Rubycon, see pic 3.
    Attached Files

    #2
    Re: Dell Optiplex GX620 with Oozing Rubycon caps

    Yep, nothing new here. Bad Chemicon KZJ and RUbycon MCZ in a Dell mini-bake oven PC. Those Pentium D chips put out a lot of heat, and in a small case, just about any regular electrolytic cap will cool. So if you do recap it, you might want to use polymers (i.e. polies). And maybe see if you can add an extra fan in the system or otherwise optimize the cooling. Generally, those Dell cases have decent cooling, but it's just that the P4/PD CPUs they use produce too much heat for that case.

    Comment


      #3
      Re: Dell Optiplex GX620 with Oozing Rubycon caps

      The OP specified that it is a tower, which sounds like a mid tower case to me, not a SFF or slim (thin) desktop.

      Originally posted by momaka View Post
      Yep, nothing new here. Bad Chemicon KZJ and RUbycon MCZ in a Dell mini-bake oven PC. Those Pentium D chips put out a lot of heat, and in a small case, just about any regular electrolytic cap will cool.
      Don't you mean cook?

      And maybe see if you can add an extra fan in the system or otherwise optimize the cooling.
      Or to optimize the cooking?

      Just giving you a hard time. Not trying to make fun of you.

      I'd say what accounts for the failures is that Rubycon MCZ simply don't handle heat well, and Chemi-con KZJ are unstable even on the shelf.

      Comment


        #4
        Re: Dell Optiplex GX620 with Oozing Rubycon caps

        Originally posted by Wester547 View Post
        Don’t you mean cook?
        Hmm, I did. But I was holding a cool beer while typing this, so maybe that's why. And also because it was my second beer for the night instead of having a proper dinner. Oops.
        I bet I will wake up tomorrow to an inbox full of badcaps.net emails/threads with people correcting my posts from tonight, lol.

        Comment


          #5
          Re: Dell Optiplex GX620 with Oozing Rubycon caps

          The case should of had better fans and airflow, to avoid that happening.
          http://www.hse.gov.uk/pubns/indg305.pdf

          Comment


            #6
            Re: Dell Optiplex GX620 with Oozing Rubycon caps

            Originally posted by Wester547 View Post
            The OP specified that it is a tower, which sounds like a mid tower case to me, not a SFF or slim (thin) desktop.
            Looking back at the O/Ps pictures, it does appear that this is an SFF or USFF desktop and not a full tower. Look at the motherboard - it has only a PCI-E and a PCI slot. The full tower has 4 slots.

            Nevertheless, those MCZs would have likely blown even in the full tower. I just got two of these OptiPlex GX6xx and GX7xx machines (I think one was a GX620 and the other a GX760), and one of them has similarly-blown Rubycon MCZ caps. When I eventually get around to it (might be in the early spring due to cold weather right now, and I need to do a full clean-up and de-dusting of these towers outside), I'll see how the cooling is in these towers. I'm thinking that reversing the front 120 mm CPU/case fan (so that the tower blows hot air from the front and takes cool air from the back) might be better for the motherboard's cooling. But that's just a theory in my head right now.

            Originally posted by Fast Alpha View Post
            The case should of had better fans and airflow, to avoid that happening.
            Yes, indeed.

            Comment


              #7
              Re: Dell Optiplex GX620 with Oozing Rubycon caps

              Originally posted by momaka View Post
              Looking back at the O/Ps pictures, it does appear that this is an SFF or USFF desktop and not a full tower. Look at the motherboard - it has only a PCI-E and a PCI slot. The full tower has 4 slots.
              You're right, it's a SFF machine (not USFF). No wonder they failed. Even the slim desktop version has 4 slots IIRC. Part of the problem is an inadequate cooling design, but the other part of the problem is that MCZ are just really sensitive to heat (not “plague” capacitors though). For an example (I'm sure you know this), in the HP DC5000 and D530 SFF computers, the MCZ in the VRM out are almost always bulging and leaking, often before any of the Teapos bulge, but the FL are never bulged.

              But yes, improved cooling would certainly help the capacitors in these machines last longer.

              Comment


                #8
                Re: Dell Optiplex GX620 with Oozing Rubycon caps

                Originally posted by Wester547 View Post
                Part of the problem is an inadequate cooling design, but the other part of the problem is that MCZ are just really sensitive to heat (not “plague” capacitors though).
                Yes, they are sensitive indeed. But when well-cooled, they will last a very long time and not give problems at all, unlike Chemicon KZG and KZJ, which will fail on their own, with or without use.

                Originally posted by Wester547 View Post
                For an example (I'm sure you know this), in the HP DC5000 and D530 SFF computers, the MCZ in the VRM out are almost always bulging and leaking, often before any of the Teapos bulge, but the FL are never bulged.
                Similar ordeal in the HP DC7700 (and possibly DC7600 and DC7800) boards. I got a board from a DC7700 "mini" tower (not SFF or USFF)... i.e. mATX board. All of the Rubycon MCZ caps on the high CPU side along with the many elsewhere on the board were bulging and leaking. The Rubycon MBZs next to them were fine, though. However, I'm pretty sure it wasn't a bad batch of MCZ caps, because the board had quite a bit of darkening of the solder mask on the underside of the chipsets and also around the VRM area, implying that the tower had either poor ventilation or there was a Pentium D fireball in the CPU socket (or both ).

                Comment


                  #9
                  Re: Dell Optiplex GX620 with Oozing Rubycon caps

                  Originally posted by Fast Alpha View Post
                  The case should of had better fans and airflow, to avoid that happening.
                  ...or used polymer caps from the start... Those little toaster boxes don't usually get the best of a good set of polymer caps!
                  <--- Badcaps.net Founder

                  Badcaps.net Services:

                  Motherboard Repair Services

                  ----------------------------------------------
                  Badcaps.net Forum Members Folding Team
                  http://folding.stanford.edu/
                  Team : 49813
                  Join in!!
                  Team Stats

                  Comment


                    #10
                    Re: Dell Optiplex GX620 with Oozing Rubycon caps

                    Polymer caps are they easy to solder on and are better at cool.
                    http://www.hse.gov.uk/pubns/indg305.pdf

                    Comment


                      #11
                      Re: Dell Optiplex GX620 with Oozing Rubycon caps

                      Originally posted by Fast Alpha View Post
                      Polymer caps are they easy to solder on
                      Not a whole lot. But if you have a crappy iron, then yes, because they do have slightly thicker leads usually.

                      Originally posted by Fast Alpha View Post
                      and are better at cool.
                      You mean do they withstand heat better?
                      - It depends. Polymers still have a finite life/endurance, just like regular "wet" electrolytic caps. But because the electrolyte in polys is dry, high temperatures may affect them a little less. So for example, a 105C-rated poly cap may have the same 2000 hour endurance rating as regular wet electrolytic cap. But for every 10C drop in temperature, the endurance life increases 10-fold for certain polymer series. Thus, at 100C, a poly cap could last up to 20000 hours. Whereas for a regular 'lytic, the endurance life usually only doubles (i.e. 4000 hours for a 2000-hour electro cap) for every 10C drop in temperature.

                      Comment


                        #12
                        Re: Dell Optiplex GX620 with Oozing Rubycon caps

                        Originally posted by momaka View Post
                        You mean do they withstand heat better?
                        - It depends. Polymers still have a finite life/endurance, just like regular "wet" electrolytic caps. But because the electrolyte in polys is dry, high temperatures may affect them a little less. So for example, a 105C-rated poly cap may have the same 2000 hour endurance rating as regular wet electrolytic cap. But for every 10C drop in temperature, the endurance life increases 10-fold for certain polymer series. Thus, at 100C, a poly cap could last up to 20000 hours. Whereas for a regular 'lytic, the endurance life usually only doubles (i.e. 4000 hours for a 2000-hour electro cap) for every 10C drop in temperature.
                        Actually, for every 20ºC drop, the lifespan of polymers increases 10x. See here, page 12. So it would be 20,000 hours at 85ºC if it was rated up to 2,000 hours at 105ºC.

                        The disadvantage of polymers is that because they rely upon electrons to achieve conductivity, they are more sensitive to excessive bias and inrush currents. The mobile ions in a liquid electrolytic solution are relatively insensitive to overvoltage and surge currents (provided that the cap hasn’t been sitting in storage for too long, of course).

                        Comment


                          #13
                          Re: Dell Optiplex GX620 with Oozing Rubycon caps

                          Originally posted by momaka View Post
                          Not a whole lot. But if you have a crappy iron, then yes, because they do have slightly thicker leads usually.


                          You mean do they withstand heat better?
                          - It depends. Polymers still have a finite life/endurance, just like regular "wet" electrolytic caps. But because the electrolyte in polys is dry, high temperatures may affect them a little less. So for example, a 105C-rated poly cap may have the same 2000 hour endurance rating as regular wet electrolytic cap. But for every 10C drop in temperature, the endurance life increases 10-fold for certain polymer series. Thus, at 100C, a poly cap could last up to 20000 hours. Whereas for a regular 'lytic, the endurance life usually only doubles (i.e. 4000 hours for a 2000-hour electro cap) for every 10C drop in temperature.

                          Yes I meant are they better at withstanding the heat.
                          http://www.hse.gov.uk/pubns/indg305.pdf

                          Comment

                          Working...
                          X