Announcement

Collapse
No announcement yet.

Intel D865GBF has bad caps

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

    Intel D865GBF has bad caps

    About 7 bad caps, half of which are leaking brown goo from the top. Only problem is with the X ones, all the K ones seem ok. $25 to call Intel motherboard support, deciding what to do.

    #2
    you are f**king kidding right?

    please we MUST have photos this is a rare case. Please provide also more details about the caps which are leaking, read what is written on the side of them, you might have to blow the dust from them and use a flashlight. Easier if the board is out of the case.

    I think the X ones are nichicon but if you get more info we can tell you.

    I think Intel doesnt like to deal direct with customers, you must get RMA through your reseller. It does specify option to contact reseller or intel customer support on the warranty page.

    You have 3 year warranty on your board, any other manufacturer i would say dont bother to rma it, just repair it yourself but since it is Intel i would demand a new board.
    capacitor lab yachtmati techmati

    Comment


      #3
      I'll have a camera later tonight so I'll post some pics then. They are all NICHICON ones, 7 total. I did a search in the forums and it looks like you guys like them - you even ranked them #3 in your top 5 good capacitor list.

      They are copper colored with:
      O O
      820uf6.3v
      HN(M)
      nichicon
      105 C
      A0323

      Three are bulging on the top, 3 are bulging with visible brown stuff, and one is bulging and the entire top is covered in brown.

      Comment


        #4
        What kind of environment was it in? Most likely they overheated unless they're Nichicon imposters, which I doubt Intel would use.

        Please do post some pics!

        Thx
        <--- 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


          #5
          I called Intel and the $25 charge is refunded if they decide the board was defective so I should be getting that back. I demanded a replacement board, as you suggested, and they said they would send me one as soon as they receive mine, no cross-shipment. Oh well, I have other computers I can use. This is from their e-mail:

          Replacement Products from Intel*:

          Line Item# Product Name Part# Qty Status
          ------------ -------------------- ------------ --- ------------
          100240476-1 BD, MB, D865GBFL, C25843-404 1 In Process
          SKU2, FAB3, WITH LAN


          On anandtech.com someone said even though it is Intel, the OEM is Asus.

          Comment


            #6
            HN series is "Ultra Low Impedance, High Reliability"... even at full ripple current (1.3 amps at 100khz for those) at 105C, they should have lasted 2000 hours with less than a 30% change in value... seems you'd have to pretty completely overheat them to make them fail quickly.

            I can only think they're fakes, they're a defective batch, they're installed backwards, the 5V line went over 6.3V (really bad power supply?) or had huge ripple, or they got really, really overheated... which is hard to do to every one of them simultaniously.


            Definitely need pics.


            --Randy

            Comment


              #7
              i think the oem for intel is Foxconn which make also serverboards for HP. It doesnt really matter about the oem factory because they are really hitech facilities. What matters is the price the boards are ordered at which affects the choice of components. Foxconn makes their own boards but with low quality caps. Intel always specifies high quality caps on their boards. I have a D875PBZLK and am installing only Intel boards on new computers i build for people.

              KC8 of these forums reported one case of Nichicon failing but this was seen due to the heatsinks being clogged up with dust and not being effective.

              If nichicon were bad we would have seen it already all over these boards. Either like topcat said there was overheating or there was a bad batch of the nichicons. I hope your rma is a success.

              yeah bring on the pics.
              capacitor lab yachtmati techmati

              Comment


                #8
                I set up a site with some pics:

                http://www.poccs.com/caps

                I build/fix computers for a living. This motherboard was from my main computer which receives a lot of tlc. The hardware/software monitors never triggered an alarm. I've had the Intel board for about 6 months - mostly in the fall/winter, so temps were good. The power supply is an Antec True 430. Supposed to be pretty solid. Everything is plugged into an APC Surge Arrest Professional. The only thing I don't have is a battery backup.

                preview:
                Attached Files
                Last edited by willawake; 03-02-2006, 04:24 PM. Reason: offsite pic upload

                Comment


                  #9
                  Thanks for the great pics. I was thinking a too large heatsink contacting a row of caps, but you have failures in the nichicons even by the ram. I was thinking a power surge but that would probably have taken something else out on the board as well.

                  I am going to go with either a quality control issue at Nichicon (somebody mixed up the ratio of additives to water in the electrolyte) or counterfeit Nichicons.

                  I do have the following thoughts about your setup though :

                  I wouldnt rely on the APC surge arrestor to be doing that great a job. I would prefer the APC Line-R for better protection. Anyway you should look into getting a BackUPS or SmartUPS. I got a 1000VA and run 4 computers/tft/crt/switch off it. a good investment.

                  Dont rely also on Antec to be without issue. We have seen posts on this board for failures of Antec PSUs due to badcaps. Actually most of the high end brand name PSUs use the same bad choice of caps, all the units are manufactured in china. I have seen only server psus to have good caps. Perhaps you should check that Antec out under load using a clamp multimeter.

                  Bushytails : please advise what you think from the photos with regard to counterfeit nichicons.

                  Are you sure that the caps are rated 820uf?

                  Digit Life notes :

                  The 2-channel switching voltage regulator incorporates 4 capacitors of 2200 uF and 10 of 1200 uF.

                  http://www.digit-life.com/articles2/...gbf-i865g.html

                  Has there been a f**k up on your board or is it an early revision? what is the revision anyway?
                  capacitor lab yachtmati techmati

                  Comment


                    #10
                    Originally posted by willawake
                    Bushytails : please advise what you think from the photos with regard to counterfeit nichicons.

                    Are you sure that the caps are rated 820uf?
                    What makes people think I'm any good at identifying fake capacitors?

                    I've never seen a fake capacitor... a couple Rulycons, but the writing on the side of those appears to say nichicon...


                    Yes, they're 820uf. it's visible in some of the photos.

                    And, yes, the polarity is correct, as least in terms of the negative stripe being on the opposite side than the + marking on the board.

                    pauloncall: do you still have the board? Do you have a 'scope? what's the voltage and ripple on those capacitors?

                    Underrated caps failing due to excessive ripple? seems unlikely. Are the caps on the 12v line or something useful? If they're 5v/3.3v filters, could the power supply have open filter caps and be dumping the switcher directly into these? (without actually looking at the board, I can't tell what the caps are doing very easily...)

                    A bad batch of nichicons seems the most likely, but if they made very many of them, you'd think we'd have heard...


                    --Randy

                    Comment


                      #11
                      What makes people think I'm any good at identifying fake capacitors?
                      who bought 150 nichicons then?

                      This one kinda hit close to home. i installed many intel boards telling people they will never have a problem with badcaps. I will be keeping an eye on the situation in case any more crop up.
                      capacitor lab yachtmati techmati

                      Comment


                        #12
                        A few interesting posts from Geek.com on the same subject. Looks like we will see more intel board failures from the nichicons.

                        Intel 865 PERL (2:01am EST Fri Jan 28 2005)
                        I have the same with Intel 865 PERL, 4 capacitors are bulge and leaking. - by Martin Ritchie

                        Intel 865 Perl (3:44pm EST Fri Feb 04 2005)
                        I have two of these. One has at least 5 bulgers, two with a yellowish brown stuff coming from the top. I have not inspected the second one but I fear it will feature the same problem with its caps. Neither has given problems so far but with leaky and bulgin caps the problems are around the corner. - by EMC
                        capacitor lab yachtmati techmati

                        Comment


                          #13
                          http://www.intel.com/cd/channel/rese...tes/193414.htm

                          Products Affected

                          Intel® Desktop BoardsD865GBF, D875BZL, D865GLC, D845EPI, D865PERC, D865GRH, D865PERL, D845GVSR, D865PESO, D845GLVA, D865GWV, and Intel® Entry Server Board S875WP1-E

                          Issue Description

                          During inspection and routine cleaning, it was discovered that some capacitors had residue on their surface. The safety vent releases built up vapor, leaving a residue on the surface of the capacitor. These electrolytic capacitors are used for bulk decoupling of VCCP. No functional failures were observed. Intel believes that the capacitors will not short out and there are no safety concerns.


                          ha ha ha

                          Comment


                            #14
                            Have any bad PL,PW, or PM caps been spotted? they're supposed to be good quality... and I own quite a collection of them, and will be very annoyed if I just recapped four boards with new bad caps.


                            --Randy

                            Comment


                              #15
                              well done for finding that Pauloncall. It is strangely written though.

                              what the hell to do now, i have to find 7 boards on monday and there is no availability on the good old d875pbzlk

                              randy : there is only reports on the HN, HM series. your caps are prolly good but forget about nichicon in the future, panasonic/rubycon have clean record.
                              capacitor lab yachtmati techmati

                              Comment


                                #16
                                another D865GBFL reported on abxzone with HN (M) caps. the board has not been used though so no fail as yet. I just report here for completeness sake.
                                capacitor lab yachtmati techmati

                                Comment


                                  #17
                                  The day the world ends is the day Rubycon produce shit capacitors.

                                  Comment


                                    #18
                                    I just received my replacement board from Intel. It's the same board/chipset but I'm guessing a newer revision b/c the capacitor arrangement is different (I can't seem to find a rev #).

                                    The exact same nichicon caps are on the board except there are only 5 instead of 7 and they are in different positions. There are some aluminum caps (blue/silver) near the cpu that replaced some other caps that didn't have any problems on the old board.

                                    So the types of caps that had problems are still there. And the caps that didn't have a problem were replaced by aluminum ones. I'll keep an eye on it and let you guys know if anything happens down the road.

                                    Comment


                                      #19
                                      D865grh

                                      I have one that has a pair of bulging Nichicons. All other caps are fine. Also, these 820's may really be significant as they are used on the DELL 400SC motherboard. I have one that has six leakers. I will get specifics on these soon.

                                      Comment


                                        #20
                                        pauloncall : sorry they used the same nichicon caps. I hope it works out. You can find the revision number of the board with the model number like C25843-401. the 401 is the revision for that model number.

                                        billger : can you post some pics?
                                        capacitor lab yachtmati techmati

                                        Comment

                                        Working...
                                        X