IIfx Won't boot

[protocol7]

I'm still a bit confused by this claim that active terminating drives don't need special termination on the IIfx. The dead IIfx hard disk is the only one that when connected up to the SCSI bus (with no power cable connected) gives me the SuperMac logo on-screen. Why can't the other drives do this?

I connected up the IIfx's floppy to the LC III last night to eject the 6.0.8L disk that was stuck in it. It actually tried to boot it (but stopped because of the system version). So I popped in a 7.5 Network Access Disk and it booted right to the desktop. Yet on the IIfx it doesn't even look for a disk. Could I maybe have bad RAM or a bad slot? It's going to be a little tricky to test as I can't put this RAM into anything else. I'll just have to go through it stick-by-stick and slot-by-slot.

 

[protocol7]

I took out the RAM and ROM and cleaned the contacts on them with q-tips and isopropyl alcohol. Still no improvement. The RAM slots are in two banks of four. If I leave two slots empty in one bank I get a death chime. Filling them up again it goes away.

Still flummoxed with this logo. Surely if the drives are not getting any power they shouldn't be terminating at all?

 

[trag]

The SuperMac image is resident in the firmware for the video card. That is, it lives on the EEPROM chip on the video card.

If you're not getting that screen with some configurations, then those configurations are probably hanging before it reaches that point in the boot process. It is also possible, I suppose, that the screen flashes by too fast to see it, but that seems far fetched.

RAM must be installed in groups of 4. And if RAM is installed in only one bank, it must be Bank (?). I can't remember. But there are only two of them. So you could try populating one bank only. Then try populating the other bank only.

Try booting with no RAM installed, no floppy connect and no hard drive connected.

There's 32KB of cache on the logic board, and the IIfx can get pretty far in the boot process just by using that 32KB cache. At least, that's what I remember. I can't remember how far, but I remember being surprised by how far along it got.

There are some jumpers on the IIfx logic board which must be configured properly. I've never seen any documentation about what they do. I'll try to remember to note the condition of each jumper (installed, not installed) on my machine at home, but my memory for things after I leave work is not very good. It could take a while.

 

[protocol7]

The screen takes a bit of time to kick in, so it's possible that the SuperMac logo appears before the grey screen with the other drives. By the time the screen is visible the logo is either there or it isn't. The one and only time it booted I think it popped up after the Happy Mac. The boot floppy started loading, then the logo appeared and the disk was ejected, leaving me with the missing system icon.

I tried a few different memory setups. It needs at least 4 sticks to boot (the display won't kick in). It won't go anywhere with no RAM installed. I've also tried filling each bank by itself. I think it's Bank A that should be filled first. It's the faster of the two (or something like that). I also tried mixing the 4 sticks between the two banks by filling the first slots in numerical order (they're numbered J30 - J37 but the slots aren't numbered in sequence). I spotted those jumpers but I haven't looked them up to see what they do. I'm assuming for now that they're set correctly.

I've also tried with no hard disk and no floppy. It just keeps hanging on this grey screen. Nothing will convince it to go any further. What could cause it to hang this early on? I don't think it can be the display card because even without it, it still won't boot either the floppy or the hard disk. And I know both of those are good. Would those tests with the hard disks/cdrom and the 7500's cable have ruled out SCSI termination issues?

I think the only thing left I can try for now is to wash the board. Perhaps there's something causing it to flake out. I can't see any apparent leakage but it is a bit dusty so there might be something there. Also I don't know what the underside of the board is like. The base of the chassis has some visible corrosion. My Classic II was slow to boot sometimes and eventually just packed up with the thick vertical stripes. Since washing the board it boots up straight away. I'll need to recap it at some stage but the wash worked well. I guess it can't hurt to try it anyway.

Thanks for the ideas. I'm sure we'll figure out a way to get it running again.

 

[protocol7]

I read back over the start of this thread again. Just wanted to add some pieces of info.

I too have two different brands of RAM. 4x Texas Instruments and 4x Accutek. I've tried each brand in Bank A.

Unlike Tempest's IIfx, the keyboard lights DON'T blink on startup. The only time they light up is when I power-on (which is usually from the keyboard's power key). After that I hear two pops from the speaker and then the display fires up to the grey screen.

As for the jumpers (partially documented here), J106 is open, J103 is closed. This seems to be the correct setup.

I tried it again with no RAM installed. It chimed but the speaker didn't initialise and the keyboard light didn't blink.

 

[trag]

As for the jumpers (partially documented here), J106 is open, J103 is closed. This seems to be the correct setup.

That is a great find. Nice to know. I tried googling and never found anything myself.

I tried it again with no RAM installed. It chimed but the speaker didn't initialise and the keyboard light didn't blink.

That may be as far as I got. Getting a chime with no RAM installed is unusual, and is probably because the cache is there to use as scratch space for the functions up to the chime.

As to your problem, I'm still inclined to believe SCSI problem, but I would expect you to get to the flashing question mark with no cable plugged in. It's conceivable that it needs the internal terminator even with no cable attached, but again, far fetched.

No, the 7300/Classic drive test doesn't necessarily prove anything, because this IIfx could be finicky and could really need the special terminator. But the 7300/Classic drive test was indicative. It suggests that the problem is not SCSI related, but it could still be.

Can you hear hard drive activity when those drives are booting in your other machines?

You migth try pulling the video card(s) and attach the SCSI drive with a usable operating system on it. Then listen for drive activity beyond the initial spin up and seek. If it sounds like the thing is loading the OS, then the problem is probably the video card hanging up the bus. Even better would be if you have LED activity indicator you can attach to the hard drive. You'd have to identify the pins for it on the drive though. Although, some drives have an LED built-on.

These are the sort of low probability suggestions at this point....

You're somewhat hampered by not having a different working NuBus machine on hand that you could use as a test bed for comparison.

 

[protocol7]

As to your problem, I'm still inclined to believe SCSI problem, but I would expect you to get to the flashing question mark with no cable plugged in. It's conceivable that it needs the internal terminator even with no cable attached, but again, far fetched.
No, the 7300/Classic drive test doesn't necessarily prove anything, because this IIfx could be finicky and could really need the special terminator. But the 7300/Classic drive test was indicative. It suggests that the problem is not SCSI related, but it could still be.

Can you hear hard drive activity when those drives are booting in your other machines?

You migth try pulling the video card(s) and attach the SCSI drive with a usable operating system on it. Then listen for drive activity beyond the initial spin up and seek. If it sounds like the thing is loading the OS, then the problem is probably the video card hanging up the bus. Even better would be if you have LED activity indicator you can attach to the hard drive. You'd have to identify the pins for it on the drive though. Although, some drives have an LED built-on.

I was away on holidays for a few days so I haven't been back to this yet. I'd tried it with the 7500's drive freshly-formatted with a universal 7.5 install and no gfx card but it just won't booting anything. The drives I tried are good in their host Macs. I should look up how to attach a LED but with SCSI disks you can pretty much hear them in action once the case is open.

I'm still suspecting SCSI voodoo on the internal bus and hope to be able to check this out in the coming days as I have an internal scsi filter/terminator block combo en-route.

 

[protocol7]

Well that didn't work out either.

SCSI filter on the 7500's hard disk didn't change anything. Filter and terminator (and no internal drive) also didn't boot. I added an external hard disk in case it still needed the external bus terminated as well. No change (though I'm only assuming it was new enough to have the required active termination). Pulled all the cards (gfx and network) as well.

I guess the final step is to get my friend up again with his multimeter and check for shorts or whatever.