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

Re: [midatlanticretro] Need 3B2/310 Help!

Expand Messages
  • B Degnan
    Check the large rice capacitor on power supply. -- Sent from my Android phone with K-9 Mail. Please excuse my brevity. epsilonprocess wrote:
    Message 1 of 10 , May 3 4:38 AM
    • 0 Attachment
      Check the large rice capacitor on power supply.
      --
      Sent from my Android phone with K-9 Mail. Please excuse my brevity.

      epsilonprocess <als@...> wrote:
      Hi all,Well, if I'd powered up my 3B2 at VCF I'd have won the "First to Smoke" award.I finally decided to power up the 3B2/310 I got back in late 2009. It had been sitting in my garage still in its shipping box until a few months ago when I finally got everything unpacked.The person who sold it told me everything was working fine when he dismantled it and had it boxed up for shipment (famous last words.)Anyway, finally got around to setting it up and testing it yesterday, plugged everything in, it powered up, and a few seconds later smoke started coming out of the floppy. I disconnected everything, took the whole thing apart and took some measurements with a multimeter.Rather than go through the measurement details for everyone, if anyone with 3B2 experience can offer some help, I'd like to hear from you.Luckily part of the package he sent me included a backup main system board. But before I even try it, I would like some help from anyone with experience servicing 3B2's. Better yet, if anyone has any spares, especially a power supply, a service manual (or links to a PDF) or heck, even a complete system, I would be interested.Any help appreciated.-Anthony
      Yahoo! Groups Links<*> To visit your group on the web, go to: http://groups.yahoo.com/group/midatlanticretro/<*> Your email settings: Individual Email | Traditional<*> To change settings online go to: http://groups.yahoo.com/group/midatlanticretro/join (Yahoo! ID required)<*> To change settings via email: midatlanticretro-digest@yahoogroups.com midatlanticretro-fullfeatured@yahoogroups.com<*> To unsubscribe from this group, send an email to: midatlanticretro-unsubscribe@yahoogroups.com<*> Your use of Yahoo! Groups is subject to: http://docs.yahoo.com/info/terms/
    • David Gesswein
      ... The 3B1 have a failure mode where the connector will overheat without any other fault due to contact resistance increasing. I don t know if 3B2 s have the
      Message 2 of 10 , May 3 5:05 AM
      • 0 Attachment
        On Tue, May 03, 2011 at 04:57:26AM -0000, epsilonprocess wrote:
        >
        > However, I did find where there was OBVIOUS damage on the system board
        > power supply connector. While I can't be 100% certain this wasn't there
        > before, I would go with the fact that it happened as a result of whatever
        > caused the smoke. Unfortunately, I can't find any sign of a short...yet,
        > anyway.
        >
        > See photo: http://home.prog.fm/3b2-ps1.jpg
        >
        The 3B1 have a failure mode where the connector will overheat without
        any other fault due to contact resistance increasing. I don't know if
        3B2's have the same problem or you had a short on your board.

        https://groups.google.com/group/comp.sys.3b1/browse_thread/thread/39a65e00c6aed4ae?hl=bs
      • Mr Ian Primus
        ... I m going to doubt there is actually a problem with the floppy drive itself. More than likely, that was just a convenient exit for the smoke. ... That s
        Message 3 of 10 , May 3 5:18 AM
        • 0 Attachment
          --- On Tue, 5/3/11, epsilonprocess <als@...> wrote:

          > The cover was on when it started smoking, so the smoke I
          > saw came out of the floppy drive opening but I can't be sure
          > it was the floppy that was the issue. What I might do is try
          > to fire up the floppy drive using a standard ATX supply to
          > see if it still works.

          I'm going to doubt there is actually a problem with the floppy drive itself. More than likely, that was just a convenient exit for the smoke.

          > However, I did find where there was OBVIOUS damage on the
          > system board power supply connector. While I can't be 100%
          > certain this wasn't there before, I would go with the fact
          > that it happened as a result of whatever caused the smoke.
          > Unfortunately, I can't find any sign of a short...yet,
          > anyway.
          >
          > See photo: http://home.prog.fm/3b2-ps1.jpg

          That's odd - and it's not burned in the way that connectors like this usually fail. It's almost as if something conductive fell there and burned away.

          > What I found odd was that measuring the resistance between
          > the red wire bundle and the blue (or ground) there is only
          > 20ohms of resistance, while measuring the yellow bundle to
          > blue (or ground) is almost an open. Is that normal even
          > though I got a clean +12v and +5v? Oh and, I measured while
          > the system board power was still connected BEFORE I noticed
          > the damaged connector.
          >
          > I also didn't see any obvious caps blown on the power
          > supply. Where is the square cap located approximately?

          It should be fairly near the power input. These caps of which I speak are AC line safety capacitors. They're this clearish yellow color, and look a bit like a board mount relay. When they fail, they fail shorted, and blow open, spewing TONS of smoke - and they tend to fail the first time a machine has been powered on in a long time. I'm not even 100% sure that power supply has them, it's just that your description of the failure sounds so much like what I've seen in several other machines.

          Now, you say you get clean +5 and +12... does the computer work? Does the power LED come on? The diagnostic LED should come on until it's finished with it's internal self test, which it won't pass if the battery is flat.

          That burn on the board is odd. Have you checked between the power pins on that connector for shorts on the board? Also, if you have been testing the power supply unloaded (disconnected from the board and drives), you might be getting false readings. If there is something wrong with the power supply, it's possible to have it put out the right voltages (or nearly right), until you put a load on it, at which point, it goes into shutdown. I usually use a couple of junk PC IDE hard drives as a load - plug the power connectors in and turn on the supply, and the drives should spin up, then check your voltages.

          -Ian
        • Mr Ian Primus
          ... The capacitor of which I speak is directly in front of the AC line filter, and between the transformer in the middle and the heat sink on the left. Check
          Message 4 of 10 , May 3 5:21 AM
          • 0 Attachment
            --- On Tue, 5/3/11, epsilonprocess <als@...> wrote:

            >
            > See photo: http://home.prog.fm/3b2-ps1.jpg

            The capacitor of which I speak is directly in front of the AC line filter, and between the transformer in the middle and the heat sink on the left.

            Check to see if it's blown - but you should have noticed it if it was, there would be a lot of soot marks on the board.

            -Ian
          • epsilonprocess
            ... itself. More than likely, that was just a convenient exit for the smoke. ... That s what I was thinking. I ll give it a quick test at some point. ...
            Message 5 of 10 , May 3 6:45 AM
            • 0 Attachment
              --- In midatlanticretro@yahoogroups.com, Mr Ian Primus <ian_primus@...>
              wrote:
              >
              > --- On Tue, 5/3/11, epsilonprocess als@... wrote:
              >
              > > The cover was on when it started smoking, so the smoke I
              > > saw came out of the floppy drive opening but I can't be sure
              > > it was the floppy that was the issue. What I might do is try
              > > to fire up the floppy drive using a standard ATX supply to
              > > see if it still works.
              >
              > I'm going to doubt there is actually a problem with the floppy drive
              itself. More than likely, that was just a convenient exit for the smoke.
              >

              That's what I was thinking. I'll give it a quick test at some point.


              > > However, I did find where there was OBVIOUS damage on the
              > > system board power supply connector. While I can't be 100%
              > > certain this wasn't there before, I would go with the fact
              > > that it happened as a result of whatever caused the smoke.
              > > Unfortunately, I can't find any sign of a short...yet,
              > > anyway.
              > >
              > > See photo: http://home.prog.fm/3b2-ps1.jpg
              >
              > That's odd - and it's not burned in the way that connectors like this
              usually fail. It's almost as if something conductive fell there and
              burned away.
              >

              The red connector that has all the wires is undamaged. The white
              connector on the board itself took all the damage and there's some
              bubbling underneath on the PCB. Not sure if the traces are broken. I'll
              have to test with a meter.


              > > What I found odd was that measuring the resistance between
              > > the red wire bundle and the blue (or ground) there is only
              > > 20ohms of resistance, while measuring the yellow bundle to
              > > blue (or ground) is almost an open. Is that normal even
              > > though I got a clean +12v and +5v? Oh and, I measured while
              > > the system board power was still connected BEFORE I noticed
              > > the damaged connector.
              > >
              > > I also didn't see any obvious caps blown on the power
              > > supply. Where is the square cap located approximately?
              >
              > It should be fairly near the power input. These caps of which I speak
              are AC line safety capacitors. They're this clearish yellow color, and
              look a bit like a board mount relay. When they fail, they fail shorted,
              and blow open, spewing TONS of smoke - and they tend to fail the first
              time a machine has been powered on in a long time. I'm not even 100%
              sure that power supply has them, it's just that your description of the
              failure sounds so much like what I've seen in several other machines.
              >

              I'll take a closer look, but I didn't see any damage on the power supply
              itself. That's the first place I looked. Would the power supply even
              work and give anything close to +5 and +12 volts with that cap blown?


              > Now, you say you get clean +5 and +12... does the computer work? Does
              the power LED come on? The diagnostic LED should come on until it's
              finished with it's internal self test, which it won't pass if the
              battery is flat.
              >
              > That burn on the board is odd. Have you checked between the power pins
              on that connector for shorts on the board? Also, if you have been
              testing the power supply unloaded (disconnected from the board and
              drives), you might be getting false readings. If there is something
              wrong with the power supply, it's possible to have it put out the right
              voltages (or nearly right), until you put a load on it, at which point,
              it goes into shutdown. I usually use a couple of junk PC IDE hard drives
              as a load - plug the power connectors in and turn on the supply, and the
              drives should spin up, then check your voltages.
              >

              I took the readings with the floppy and hard drive disconnected, but
              with the system board connected. I didn't see the damage on the
              connector until I dismantled the system board/expansion cage, but there
              was no further smoking. I did later plug in the hard drive with the
              system board disconnected, and it did spin up, though I didn't take any
              readings with it connected. I'll have to do that.

              With the system board connected, the green LED lights up. If the reset
              button is pressed, the green LED flashes then goes solid again. There's
              only a red LED next to the green one that I figured was the hard disk.
              Is that the diagnostic LED? If so, that one never lit up. Also, the
              rocker switch on the power supply has "diagnostic" position that
              doesn't seem to do anything. The power supply only seems to work in the
              "on" position. Is it supposed to do anything in "diagnostic" ?

              At no point did the system POST anything to the terminal. It's a 610
              terminal set to 9600/8/n/1. I didn't check the battery level. I'll have
              to do that as well. Hopefully, I'll get a chance to get to the shop
              tonight or this week. Let you know what else I find.

              Thanks,

              Anthony
            • Mr Ian Primus
              ... OK, then it really sounds like your power supply is OK. Check carefully the burned area, and see if any of the traces were damaged. Also see if you can
              Message 6 of 10 , May 3 8:47 AM
              • 0 Attachment
                --- On Tue, 5/3/11, epsilonprocess <als@...> wrote:

                > I took the readings with the floppy and hard drive
                > disconnected, but
                > with the system board connected. I didn't see the damage on
                > the
                > connector until I dismantled the system board/expansion
                > cage, but there
                > was no further smoking. I did later plug in the hard drive
                > with the
                > system board disconnected, and it did spin up, though I
                > didn't take any
                > readings with it connected. I'll have to do that.

                OK, then it really sounds like your power supply is OK. Check carefully the burned area, and see if any of the traces were damaged. Also see if you can figure out what might have caused it - if there is a burned screw rattling around in there or something. That's not the way power connectors usually burn - they usually darken and discolor in one spot, and burn the associated wiring harness plug too. That looks like something fell in there and shorted it out then burned.

                > With the system board connected, the green LED lights up.
                > If the reset
                > button is pressed, the green LED flashes then goes solid
                > again. There's
                > only a red LED next to the green one that I figured was the
                > hard disk.
                > Is that the diagnostic LED? If so, that one never lit up.

                There are two LED's on the front, the left one is green (power), and the right one is amber/red (Diagnostic). There is no hard disk light.

                > At no point did the system POST anything to the terminal.
                > It's a 610
                > terminal set to 9600/8/n/1. I didn't check the battery
                > level. I'll have
                > to do that as well. Hopefully, I'll get a chance to get to
                > the shop
                > tonight or this week. Let you know what else I find.

                Are you sure the cable is wired correctly? Do you have a "christmas tree box", with the little lights that you can put on the RS232 line to see if there is any activity at all?

                -Ian
              • epsilonprocess
                Latest on the 3B2... I exhaustively checked the power supply and everything seemed good. So I swapped system boards and voilĂ , the new one POSTed to the
                Message 7 of 10 , May 8 12:22 AM
                • 0 Attachment
                  Latest on the 3B2...

                  I exhaustively checked the power supply and everything seemed good. So I swapped system boards and voilĂ , the new one POSTed to the terminal with a NVRAM error.

                  Tested the battery and it's completely dead. Looks like a pretty easy to find 3.6V lithium backup battery.

                  So apparently something went horribly wrong with the original board. Would still like to troubleshoot it at some point.

                  In any case, after that, couldn't get any of the hard drives to boot. Went into firmware mode and tried to boot one of the floppy disks, but I get nothing but CRC errors.

                  Soooo...

                  In my box of spare parts, seems like I was also given a SCSI HA board, so on ebay I grabbed a 327MB SCSI drive, a spare 4MB memory board, and to be safe, a replacement floppy drive, from apparently someone involved with the SDF 3B2's out in Seattle. Everything is supposed to be tested working, so fingers crossed. He told me the HD will have 3.2.3 pre-installed with the SCSI drivers so I should be good to go. I got so many floppies with the system, so hopefully I can find the 5620-related stuff. No idea if the floppies are still readable after all this time.

                  In any case, if I do manage to bring all this stuff down to VCF, it might end up being a live restoration.

                  Also, wanted to ask what the differnece was (other than the obvious visible differences) between the two boards. The one that ended up working has the power supply connector on the other side and one missing screw hole at the bottom (meaning I have to insulate the standoff so it doesn't short). From what I read, the /310 has the WE32100 while the /300 had the WE32000. Couldn't really read the chips on the swapped board, but the one that fried clearly has the WE32100.

                  Old broken board: http://home.prog.fm/3b2-sb2.jpg
                  New board: http://home.prog.fm/3b2-sb1.jpg

                  BTW one of the things I tried was swapping the PROM's but that didn't make a difference. So you're seeing the old PROM's on the new board and vice-versa.


                  --- In midatlanticretro@yahoogroups.com, Mr Ian Primus <ian_primus@...> wrote:
                  >
                  >
                  >
                  > --- On Tue, 5/3/11, epsilonprocess <als@...> wrote:
                  >
                  > > I took the readings with the floppy and hard drive
                  > > disconnected, but
                  > > with the system board connected. I didn't see the damage on
                  > > the
                  > > connector until I dismantled the system board/expansion
                  > > cage, but there
                  > > was no further smoking. I did later plug in the hard drive
                  > > with the
                  > > system board disconnected, and it did spin up, though I
                  > > didn't take any
                  > > readings with it connected. I'll have to do that.
                  >
                  > OK, then it really sounds like your power supply is OK. Check carefully the burned area, and see if any of the traces were damaged. Also see if you can figure out what might have caused it - if there is a burned screw rattling around in there or something. That's not the way power connectors usually burn - they usually darken and discolor in one spot, and burn the associated wiring harness plug too. That looks like something fell in there and shorted it out then burned.
                  >
                  > > With the system board connected, the green LED lights up.
                  > > If the reset
                  > > button is pressed, the green LED flashes then goes solid
                  > > again. There's
                  > > only a red LED next to the green one that I figured was the
                  > > hard disk.
                  > > Is that the diagnostic LED? If so, that one never lit up.
                  >
                  > There are two LED's on the front, the left one is green (power), and the right one is amber/red (Diagnostic). There is no hard disk light.
                  >
                  > > At no point did the system POST anything to the terminal.
                  > > It's a 610
                  > > terminal set to 9600/8/n/1. I didn't check the battery
                  > > level. I'll have
                  > > to do that as well. Hopefully, I'll get a chance to get to
                  > > the shop
                  > > tonight or this week. Let you know what else I find.
                  >
                  > Are you sure the cable is wired correctly? Do you have a "christmas tree box", with the little lights that you can put on the RS232 line to see if there is any activity at all?
                  >
                  > -Ian
                  >
                Your message has been successfully submitted and would be delivered to recipients shortly.