Print

[jamesdenton]

Hello all,

Lord knows I've read dozens of threads on various Widget issues, but when the time came to diagnose my own, none were to be found. So with that said...

I've got a Lisa 2/10 with a Widget that has been somewhat unreliable since I got it. I recently pulled the drive out of storage last week, installed it in the Lisa, and was successful in installing MacWorks XL 3.0 to the drive. I was able to reboot without any problems. I shut the Lisa down, powered it back up, and now find that the drive is not recognized by the system since it appears to still be working its way through self-tests. The green LED on the front of the drive flashes every two seconds or so, and is in cadence with the green LED on the top of the drive. I assume this is a doing a surface scan, but don't know for certain.

My question is.... how long should I expect to wait for this to complete? I've left the thing running for 20+ minutes to no avail. I can successfully send a RESET command to the drive with BLU, which causes all lights to go out, the 'chirp' to be heard, and the same behavior repeats.

Is there any hope for this drive?

Here's a video that shows what's happening: https://www.youtube.com/watch?v=7KFqD1hHfg4

Thanks for your time!

[stepleton]

Twenty minutes is way too long. Also, the blink pattern doesn't look like the surface scan to me --- that's more of a steady, rapid flashing, at least with a healthier Widget.

Naturally my first tool for investigation would be NeoWidEx, although two things:

1. my guess is that the drive is not getting far enough through the boot process to want to talk to the Lisa yet, so NeoWidEx won't work
2. NeoWidEx is not an automated diagnostic tool: you really do need to read the "required reading" and then put on your detective/diagnostician hat

One thing the blinking reminds me of is times when I've turned on a Widget after years of idle storage. The lubrication around the drive head arm gets sticky, and when the Widget tries to swing it, it can't articulate it all the way. I've run into this a couple times, and each time I could tell by listening carefully that whenever the Widget would power on, the head actuator would be able to swing the arm a little further. Presumably the lubrication was warming up and gradually getting looser.

So, without more intensive diagnostic tools like UsbWidEx (which is really the most powerful Widget debugger I know about, since it can investigate some issues without the Widget being ready to talk to the Apple), you might try the following. Attach the Widget to the Lisa, boot into BLU, and send it reset pulses every few minutes for a little while. See if you can notice subtle changes in what the Widget does each time --- how long it takes to blink, or whether the arm actuation sounds any different. If things seem to be progressing, keep doing it; with luck the arm will eventually recover full travel.

Note that this suggestion may put data on your disk at risk --- if the Widget decides it can move the arm, but then for some reason it can't or it's still sluggish, it may behave in a way that the authors of its firmware failed to anticipate or control for. You might want to devise a different method if the data on the Widget is too important to lose.

(Finally --- even though I've seen a few Widget problems in my time, I haven't seen them all. I don't know how a Widget misbehaves if the optical track gauge breaks, for example, and since that also concerns the arm, maybe it can cause a similar behaviour.)

Good luck!

[patrick]

My question is.... how long should I expect to wait for this to complete? I've left the thing running for 20+ minutes to no avail. I can successfully send a RESET command to the drive with BLU, which causes all lights to go out, the 'chirp' to be heard, and the same behavior repeats.

After approx. 20 seconds, when the motor has reached its nominal speed, the servo board is reset by the controller. This servo reset pulse releases the head brake solenoid, you hear a loud "click". Now the drive performs a short self-test (you will hear "squeak squeak") before a full surface scan is performed. During the scan all sectors are read and checked, bad sectors are stored in the spare table. You will hear a ticking sound while the LED flashes, interrupted by a "squeak-squeak" for each sparing action.

The slow flash on the controller board indicates a timeout, so something failed during this sequence.

Do you hear the loud click? If not, the motor does not come up to speed, the power good signal from the PSU is missing, or your controller board failed during RAM/ROM check.

Do you hear the first "squeak squeak" after the click? If not, check the alignment of the head brake. With the solenoid engaged (push it back by hand with the drive upside down, but don't move the head axle!), there should be a small gap between the brake and the axle. Sometimes the brake pad sticks. In this case, remove the magnet, clean the pad and reinstall it properly.

If the drive starts with the ticking selftest and then goes into slow flash mode, you have issues with the HDA mechanism, the controller or the analog electronics. Use NeoWidEx, UsbWidEx, Apple WidEx on an Apple III, or something comparable to read the controller status registers. This gives an indication what happened.