Mac IIsi questions and experiences...
Posted 15 November 2011 - 12:21 AM
I've been using Macs since I first purchased a new Mac SE with two 800k drives (no hard drive). I have collected many Macs over the years, some bought some given to me, others that used to be my main system which are now used for other reasons or are stored in a closet in my home. I was an Apple Student Rep while in college (here in Texas) and ran The Mac Help Center on Hotline for five years. So, I'm pretty knowledgeable about Macs, but of course there is a lot I don't know, too
The system I'm writing about is a Mac IIsi configured in several different ways. I'll have to give a little history to make things clear:
I had a IIsi, but it got water damaged (and rusted) without me knowing it. It was stored inside in a closet, but an old water jug that was in a box over it had some water in it that leaked and it rusted the mobo (thank you, wife). So, I bought another one off ebay. I got it and put a Daystar Turbo 040 in it (from my old computer) that had been stored in another location, it was working fine. The adapter card has a pass-thru nubus slot, which remained empty (at first). The hard drive that came with it was only 80 megs, so I decided to put a larger drive in it. The old hard drive from the other IIsi was broken (which is why it was stored) so I put a Seagate Barracuda (ST3218N) 2.1 gig hard drive in it. I had it configured to get it's termination from the computer bus and the termination enabled (via jumpers on the hard drive).
The computer wouldn't recognize the drive, no matter how I configured the jumpers, until I put an external scsi drive that had active termination. Then the computer booted and recognized the drive. I erased the internal drive and loaded 7.6.1 on it. When I took off the external drive, it wouldn't recognize the internal drive. So I put a passive terminator on the external scsi port, and that worked, it recognized the drive. Everything worked fine, for a couple days. I have no external drives or anything else connected to the computer, except for the keyboard and a monitor. After a couple days, I was using the computer and suddenly I smelled "magic smoke" coming from it and quickly turned it off. I examined the board and the internal scsi cable had burned through two wires on the left side (looking at it from the front of the computer). Also, one of the tantalum caps right under the cable was leaking fluid, and one of the tiny tantalum caps was completely burnt out. I have included a photo of the blown board below, but I tossed the cable so I don't have a pic of that. You can see in the photo that it stripped the green coating from the trace between one of the tantalum caps and the tiny cap (or is that a resistor?) on the board. The cable was burned right where it had been resting on one of the tantalum caps, so perhaps the cap burnt the cable instead of it being burned just by the voltage or amperage demands from the hard drive. The only thing I could figure was the hard drive was pulling too much power and had fried the cable and caps. I checked the voltage requirements on the Seagate and it was 5 volts, 650-850 milliamps. The original drive pulled 250-500. It didn't seem right that a couple hundred extra milliamps would do all that damage, but what else could it have been? I checked the drive to make sure it wasn't supplying termination power to the bus, and it was not configured to do so. So, I can only assume that the drive pulled too much power and burnt things out. I spoke with a friend who used to be a Mac repair tech and he said it shouldn't have done that, and mentioned something about passive termination vs active termination, but I couldn't find any manual for the ST3218N to see if it was active or passive term.
I went on ebay and found someone that had 8 new IIsi boards, sealed in anti-static bags for $28 a piece (yeah, it was awesome). I bought two, and they raised the price to $89 each. I guess when I bought two they thought they were worth more and raised the price. I received both boards and put one of the new boards into one of the IIsi's I have. This time I put the original Apple (Quantum) 40 meg hard drive back in. I put the Daystar Turbo 040 back in and put in an old Cabletron nubus ethernet card so I could trade files via Appletalk over ethernet. The computer was fine for a day or so. Then, suddenly, the monitor blinked and put up some lines and the computer was stuck. I restarted, no "bong" no nothing. The hard drive read for a while, but nothing. I tried resetting the PRAM, I unplugged and pressed the power button a few times, I pulled the PRAM battery, I pulled the Daystar card, I put in the stock RAM, I did everything and the computer just sits there. Nothing. I still have the owner's manual from the '040 board and there is no caution about an additional nubus board drawing too much current. It is made to hold two boards, the '040 and an additional nubus card (in this case, an ethernet board).
So, here are my questions:
1. Could a hard drive pull too much power and burn out the caps and do that to the SCSI cable?
2. Could the Daystar card and the ethernet card be pulling too much power and blowing out the board?
If anyone has any idea what could be going wrong, I would sure appreciate any help. Sorry for the length, but I thought it would be best to give all the pertinent information rather than people having to ask me for more information over and over. Thanks.
[attachment=0]Blow IIsi Board.jpg[/attachment]
Posted 15 November 2011 - 04:28 AM
First off, if you're using the DayStar TurboCache and its "two slot" adapter, those are both 120 pin PDS slots, not the shorter 96 pin NuBus slots.
Just because the TuroCache/adapter combo has room for a second card in its passthru slot does not mean that a power hungry second PDS card would be fine.
Depending on the current draw of the NIC, you might very well be taxing the IIsi's weakly PSU, but the electron pushers will have to answer those questions . . .
. . . along with the freakish occurrence of blown apart board components and burned SCSI cables.
Have you tested with a second MoBo and a different PSU? My IIsi is running great with a Quadra 700 PSU sticking out the top of its lidless PSU Enclosure. If you have one of those or a IIci or IIx PSU available, they probably all supply enough power to rule that out as a cause.
Just from the sounds of it, I'm thinking bad Caps, a flaky PSU and wild power surges, but that's just a WAG, not an educated guess.
Trash Hauler: call sign: eight-ball
C.O. AC130H SpecOps 68kMLAAF
Posted 16 November 2011 - 06:28 AM
I have looked in the Daystar book and haven't found anything about current draw. I know you are right, there must be a limit, and I guess I'll have to assume I've reached it. It's sad though, because I'd like to be able to put the computer online and have the accelerator running, but I suppose I'll have to choose between the two.
I have another IIsi that has one of the MacCon ethernet boards in it that runs fine, I really don't want to try to put these boards in there and run the risk of blowing out - yet another - logic board. That would simply be too depressing and I don't have the money to keep buying IIsi's and/or logic boards.
Has anyone heard of a hard drive pulling too much power and blowing out a logic board? I have spoken to two knowledgeable Mac folk, who have been servicing and/or selling Mac parts for years and neither has heard of such a thing, but then they haven't seen everything.
The three boards, with the ethernet NuBus board at top, the '040 card in the middle and the adapter on the bottom:
The page from the Daystar user's manual showing the availability of a PDS or a NuBus pass-thru slot for the IIsi:
A close up of the Daystar adapter, with the NuBus pass-thru slot at the top and the PDS slot (for the '040 adapter) in the middle (note that the PDS slot in the middle matches the PDS connector at the bottom which connects to the computer):
Posted 16 November 2011 - 01:36 PM
Is there any chance that you could scan your manual so we can add it to the installation manual I've posted in the peripherals forum?
Any way to reduce current draw would be important:
__Check out the Developer Notes and User Manuals for the IIsi for maximum power draw, I'll check Designing Cards And Drivers for same.
__ Switch to as new an Ethernet as will run on the max OS that will run with your accelerator.
____I'd suggest watching eBay etc for one of the Futura cards that'll give you better Video and Ethernet at the same time.
______When you don't have a cable attached to the vampire video connector, you'll have full, unbuffered use of Bank A's measly 1 MB of RAM
____Any full length NuBus card designed before the IIsi was headed to a Mac with a much beefier PSU than the IIsi.
__Switch to a PowerBook SCSI Drive for booting to draw the least amount of power possible.
____Best bet, boot from an external HDD with none installed in the IIsi
____Run your drive in an external case or power it from a secondary PSU during your next round of tests.
____Take a gander at JDW's
thread, which I haven't read as yet, but will soon.
SE/30 booting off CF success
____Consider buying an inexpensive IDE or SATA/IDE <-> USB adapter just for the aux PSU included and wire it directly to AC inside the box.
____Consider Hacking an ATX PSU into the PSU shell of one of IIsi as I'll doing for my SuperIIsiHack™ (linkage missing or unposted?)
It looks like I need to pull together a lot of widely separated references into another Hacks topic!
ink to Q700 PSU running a IIsi thread
My aim is to mount at least the CP2IIsi and an Ethernet card underneath the IIsi MoBo . . .
. . . while running a Radius Rocket with MaxRAM in the NuBus adapter.
I haven't done the SubterraneanCardHack yet, gotta source some solder paste for that one.
Right after that's done, I'll be doing the NuBus/PDS Cards running simultaneously test.
I'm still in search of the ROM IMAGE of a later Mac for dougg3's SIMM Card Adapter that'll increase the IIsi's limit of three Pseudo Slot ID's.
gotta run to work, I'll edit in anything I think of later.
p.s. If anyone has a GREAT source for ChipQuick and Solder Paste, please PM me!
Trash Hauler: call sign: eight-ball
C.O. AC130H SpecOps 68kMLAAF
Posted 16 November 2011 - 06:27 PM
The 47uF caps mostly act as filters between the power supply (5V or 12V) plane and the Ground plane on the circuit board. So, if one develops an internal short, that's a short directly between 5V (or 12V) and ground and will generate a lot of heat.
Regarding your second board, that could be any of a number of things. If you're still using the same power supply, it's possible that the earlier short circuit damaged a component in your power supply which later failed.
However, it's also possible that the "brand new" IIsi board has leaky capacitors which caused a failure after a few days of use. Even though the boards you got were never used, they were still manufactured more than fifteen years ago and that's plenty of time for the electrolytic caps on board to fail.
Because you have two of them, and since there's memory on the IIsi logic board, I would pull the RAM, and all cards, disconnect the SCSI and floppy cables, and just power up the logic board with a monitor connected to the built-in port. See if you get to the flashy question mark floppy image.
If so, then your logic board is probably okay, and the problem is elsewhere.
If not, then try the other "new" logic board that you bought.
That should get you started on the trouble-shooting path.
0 user(s) are reading this topic
0 members, 0 guests, 0 anonymous users