Apple Lisa questions

[stepleton]

That's longer than I've ever run a Widget continually I think my record is probably an afternoon or so.

It would be interesting to see if a few resets during that interval start to get the heads to be a bit more limber. You may not need all of that time...

 

[ScutBoy]

I'd focus on improving this situation before trying other, generic fixes. I still think the best thing to to is to get the drive to try and free up itself rather than moving the head servo on your own.

My advice: be patient.

If @stepleton gives that advice, I'd follow. Patience is not always my virtue.

 

[ried]

New CRT just arrived and should fix the burn-in from its past life.



Found a helpful YouTube video showing the replacement process.

Still no change in the Widget's behavior.

 

[LaPorta]

This is a great thread...keep up the good work!

 

[stepleton]

Hmm about the Widget --- so after repeated resets, or startups and shutdowns, the heads don't seem to advance any further than they did originally?

Don't forget the white ground wire that comes off of the CRT socket. It's easy to forget. (Ask me how I know.)

 

[ried]

This is a great thread...keep up the good work!

Thank you!

Hmm about the Widget --- so after repeated resets, or startups and shutdowns, the heads don't seem to advance any further than they did originally?

Don't forget the white ground wire that comes off of the CRT socket. It's easy to forget. (Ask me how I know.)

Correct. I plugged in the ground, and the new CRT's picture looks lovely.

However, one more issue. I added an Apple Parallel expansion card, so I can use a parallel drive emulator (while we try to fix the Widget). But adding that card to either slot 2 or 3, per instructions, results in an I/O error 52 on power up. Removing the card solves the issue. Did I get a bad card? Or should I be doing something else to prepare the system for the parallel card? No drives are attached (and it wouldn't boot anyway if they were).

 

[stepleton]

52 is an "I/O board COPS error", and I'm not sure why an expansion card should cause that kind of trouble. How soon does it come up --- right during the power-on tests? Does it happen during the I/O board test or later when the expansion cards are under test?

By pinpointing when it happens, it may be possible to deduce which part of the boot ROM code is producing the error and work backwards from there.

Lacking further info, it's not a bad idea to do the basics if you haven't done them yet. Examine the card edge sockets in your Lisa --- they haven't been used in decades, if ever. Same for the card edge connector on the parallel port card. Check that socketed chips are well-seated.

There's really no special trick to installing a parallel port card --- you just put it into the expansion slot and tighten the ZIF mechanism.

You might want to put the puzzle to the more accomplished hardware experts over at http://lisalist2.com

 

[ried]

52 is an "I/O board COPS error", and I'm not sure why an expansion card should cause that kind of trouble. How soon does it come up --- right during the power-on tests? Does it happen during the I/O board test or later when the expansion cards are under test?

It happens immediately upon hitting the power switch. Before the screen even illuminates it generates an error tone, and when the screen illuminates that error 52 is showing. It certainly does not get through the CPU and Memory test first, as it otherwise would.

Card edge connector and expansion socket appear clean, and I've applied Deoxit to facilitate the connection. No change. There is one socketed chip on the parallel card that I cleaned and re-seated with no effect. Hmmm...

 

[ried]

Well, good news. I pulled and re-seated the I/O board, which resolved error 52. But the it went through it's normal CPU, Memory and I/O startup checks before reporting error 93. I moved the card from slot 3 to slot 2 and the error is now resolved. The parallel card seems to be passing all tests.

 

[ried]

Well, crap. I broke 2 pins off one of the Widget hard drive's ribbon cable connectors. I can only find one of the pins, too. Soldering those back on will be quite the challenge.

I had purchased a replacement motherboard, controller, servo, and read/write boards in case one of those was the culprit.

Will probably have to use the hard drive emulator exclusively from here on out.

This is why I can't have nice things.

 

[ried]

Well, this solution has several nice features. Not the least of which is that it's dead silent, and leverages a lot of the work that @stepleton has done on Cameo/Aphid. Also, it's rather nice looking hardware.







Not giving up on the Widget, either. We'll sort that one out, one way or another.

 

[joshc]

Nice work so far, you are really making me want a Lisa.

 

[ScutBoy]

Well, this solution has several nice features. Not the least of which is that it's dead silent, and leverages a lot of the work that @stepleton has done on Cameo/Aphid. Also, it's rather nice looking hardware.

Not giving up on the Widget, either. We'll sort that one out, one way or another.

Yes - I have one of these, and just bought another.

In Widget news, I brought back a dormant Widget that would not come ready by swapping the interface and read/write board from another drive that would (but had other problems) and got the below. Good thing I kept the root PW blank back in the day!

 

[stepleton]

Well, crap. I broke 2 pins off one of the Widget hard drive's ribbon cable connectors. I can only find one of the pins, too. Soldering those back on will be quite the challenge.

This is a setback, but it may not be a very big deal.

The connector is an eight-pin connector for a flat-flex ribbon with what looks like only two conductors. It's possible that you still have up to three pins for each of them: the drive may still work as well as ever in this condition. Do you have a way of determining which pins connect to which of the lanes in the flat-flex? You may be able to tell just by looking.

There may be more elaborate workarounds for the problem, too.

Glad the hard drive emulator is serving you well!

 

[ried]

Excellent observation, @stepleton. It appears it still has two of the four pins that stem from the same conductor. Here's a close up:




I reassembled the full Widget sandwich using the replacement parts. It's back to generating error 81 upon startup. The head servo does seem to be more active than previously, but still not moving through its full sweep.

I am now seeing some green status light activity, however. That is new.

 

[stepleton]

Good. This is what it looks like when a Widget is waking up, in my experience. Keep at it.

 

[ScutBoy]

For what it's worth, with the drive I recently got going, the machine sat at the hourglass for a long time - 5-10 minutes - after the drive activity stopped before it booted into Xenix

 

[Paralel]

Well, crap. I broke 2 pins off one of the Widget hard drive's ribbon cable connectors. I can only find one of the pins, too. Soldering those back on will be quite the challenge.

I had purchased a replacement motherboard, controller, servo, and read/write boards in case one of those was the culprit.

Will probably have to use the hard drive emulator exclusively from here on out.

This is why I can't have nice things.

Condolences. I feel your pain. I've broken pins off of ribbon cable, and even torn what was basically an irreplicable ribbon cable, over the years. I am sure most of use here can commiserate.

 

[3lectr1cPPC]

Same here. Broke a very rare display cable on one of my PowerBooks, still haven't found a replacement after over a year. It happens.

 

[ried]

Question for @stepleton: I'm using NeoWidEx to rehab the drive, but have hit a roadblock. For example, trying to reset the servo (L), and eventually NeoWideEx says that it can't find a Widget and is asking me to pretend there is one. I press Y and Return to indicate yes, but that doesn't seem to register. After an hour or so I can press Q and abort the attempt just fine. Is there anything else I can do to force it to pretend there's a Widget and send the servo reset? The drive seems to be unresponsive to NeoWideEx, so the application doesn't actually pretend there's a Widget and send the command. Eventually, I just press Q to quit.




BLU also finds the disk unresponsive, so it's not an issue with NeoWideEx or with BLU. Just trying to figure out if I can force anything to move forward through software.

Warning: Loud video.