Whats the problem with this Lisa widget drive?

[BEU]

I have a Lisa 2/10 H/88 that was defect when I bought it 10 years ago. There was a problem with the static RAM but with excellent help from this forum ( stepleton and mactjaap) its now working, except for the widget harddrive. Now its time to try to repaire the harddrive. When starting I get the error code 82 from the harddrive. The harddrive has never started to rotate before as long as I have own it.
I adjusted the solenoid as described in Larry Pinas Macintosh Repaire and upgrade secrets p 264. The actual clearance was 0.050 and was adjusted to 0.012.

I opened the harddrive and pictures as shown.



After starting the Lisa I rotated the harddrive and for the first time it started to spin. First slowly but increased in speed. The sound was similair to the old harddrive in Macintosh II. After 15-20 sec it stopped. Some popping from the widget and smell of smoke. I turned it of. Not good.

The front of the widget was warm and the Q3 on the motor board was hot. I started it again but no action in the drive.
The Q3 is not working when tested, see pictures




Probably Q3 was burned and the motor stopped. Its easy to replace Q3 but should I also replace U1 , ULN2074, the quad Darlington chip?
https://www.applefritter.com/content/schematic-motor-control-widget-050-5023-b
When I look at the pictures of the hardrive now I noticed that the glass platter is probably displaced, no screw as the pictures in https://68kmla.org/bb/index.php?threads/apple-lisa-widget-drive-repair-need-some-advise.35420/

Should it be screwed or glued?
Thankful if anybody has any suggestions what to do.

 

[stepleton]

I knew this day would come

First, a criticism. It's always interesting to see the insides of a Widget, but I don't think it was wise to open this one yet. Hard drives want to be kept clean inside, and while the risk of catastrophe for an open drive is sometimes overstated, it's certain that the odds of a head crash or other contamination-related problems are much greater now than they were previously. With the equipment you are likely to have, this hazard is not practically reversible.

When troubleshooting, it's a good idea to avoid taking steps that could create new problems!

It's also good to avoid inventing new problems for yourself as you go along .

It's possible that your coarse positioning graticule (the glass thing) is loose, but to me it seems to be in the right position. Apple may have had several ways of attaching the graticule to the frame during the time when they manufactured the Widget, and perhaps yours is attached to the frame in a different way to MacNo84's (e.g. with glue only). Later on, once your disk spins again, we can verify whether coarse positioning works without opening the drive. But let's also remember:

When troubleshooting, it's a good idea to solve just one problem at a time!

So let's set aside the graticule for now and focus on the first problem only: your drive won't spin. Take these steps:

1. Please close up the drive already, and leave it closed.
2. Gently try to rotate the spindle manually, stopping immediately if something seems "stuck" or "caught".

Does it still spin? If so, how well?

If it does still spin, then let's carry on with fixing the problem you've spotted: the burned-out driver resistor Q3. My general philosophy is to make the fewest changes possible, so I'd investigate replacing that transistor first and then applying power briefly, just to see if the motor begins to turn again. I would not leave the drive running for more than a few seconds because I'd want to think carefully about how to deal with the sticky spindle that's probably caused all of the trouble you've encountered so far. We can save that for a next step.

It probably wouldn't cause any harm to replace the quad-darlington IC, but other options include (a) doing nothing and including it in future troubleshooting steps if the spindle won't turn, and (b) taking out the IC and testing it outside of the circuit to see if it's obviously broken. I'd probably recommend (a), but I might still buy the IC and have it handy in case replacing Q3 didn't make any progress.

I seem to recall that you have an oscilloscope --- is that correct?

 

[mactjaap]

I think it is wise to follow stepleton's advice. He is one of the experts when it comes to widget drives.

I also had a stuck drive and got is spinning by powering up and rotating by hand. I managed to revive it. But a few years later it failed. But is it is stuck to much you could burn the motor.
I will follow this thread with the utmost interest. Widgets are "my thing". And if possible I will contribute, but I doubt if I can add more then stepleton.
You could also have a look at the lisalist2.com site. There are many interesting post about widgets. You can also look at my YouTube channel. I have made many widget video's: https://www.youtube.com/channel/UCWbS5zkXeiV_WW-s_-HGh6Q

 

[BEU]

Thank You both for advice. I have realized that You are true experts on widget drives. I will follow your advice and report later. The drive was open for 5 min so I hope no harm was done.

 

[BEU]

I have studied the thread https://68kmla.org/bb/index.php?threads/apple-lisa-questions.43118/ and there is a lot of good information about the widget drive. In the picture of my widget the "ruler scale" is clearly visible. I compared to MarNo84 widget citied above. its obvious that the glass platter is not parallel to the metal case.



If I draw a line from center of the arm to the glass platter its perpendicular to the glassplatter and the ruler scale, and not to the metal case.
This is only to convince myself that the glass platter is in good position.

 

[BEU]

Gently try to rotate the spindle manually, stopping immediately if something seems "stuck" or "caught".

Does it still spin? If so, how well?

Here is a video when I manual rotate the drive. Its rotating 8-9 turns with some mechanical noise.


View attachment 20230502_180539 (1080p).mp4

I have disconnected Q1, Q2 and Q3 and they are tested OK. I have measured the resitance in the motor from J2. Resistance from pin 1 (12V) to the motor pin 7, 8, 9 and 10 is 2 ohms. I hope this is a sign that the motor is not burned.
Is it possible in any way to lubricate the drive without opening it??

I will soon make a test when I get the new Q3.

 

[stepleton]

Hmm. I think that motor lubrication may not be the problem here; I think I've encountered spindles that were harder to turn than that.

Meanwhile, I didn't realise that Q3 was good --- I thought it was bad!

We would expect the resistance across any of the windings to be quite low anyway, and I think that if there had been damage, we would see different measurements across pins 7-10, since some windings would probably be more burnt than others. I think your motor is probably OK.

Let's see what happens when you replace the transistor. If that doesn't do anything, we'll want to see if we can find out whether the fault is on the motor control board at all.

One thing you might do in the meantime is take resistance measurements across each of the four darlington transistor arrangements in the quad-darlington IC, with the motor unplugged. As they're all probably hooked up to different motor phases in the same way, we would expect these measurements to be similar to each other for each darlington. You'll want to measure across base/emitter, collector/emitter, collector/base for each, and in both directions since the transistors can behave like diodes. If there's a big difference between any of the darlingtons, that's probably suspicious...

 

[BEU]

Sorry stepleton, my fault. Q1+Q2 +Q4 are all OK. Q3 is burned and will be replaced. I have ordered a new Q3 and a new U1 ( ULN2074) just in case. I will test the U1 as You described.

 

[BEU]

Testing the U1, ULN2074: I think its damaged. Disconnected it from the board and got the following result.

The resistance between B(T) and E was 11.5 to 11.9 kohms. That is pin 2-3, 6-7, 10-11 and 14-15. That should be OK according to the schematics.




but when I tested it in a transistor tester i got this picture from transistor unit 1, 3 and 4 . pin 1 on the tester is pin 1 (C) in U1. 2 is E , 3 is B. Im not sure thats this OK but the same picture for this 3 units.


This is from unit 2, pins 6, 7 and 8.


Probably damaged.
I have ordered a new U1, but the question is if more components are bad?
COP 421??

 

[stepleton]

This is retrocomputing There's always a chance that more components are bad. If the motor continues to avoid spinning, then we'll keep working to find the problem.

But there's not a lot around the quad darlington array: there's the motor and there's the microcontroller, with its current-limiting resistor pack.

As you've desoldered the component, why not try it out on a breadboard and see if the darlingtons are capable of switching a current. If they work, that doesn't mean that the part is OK, but if one doesn't work, then you know it's bad.

I've asked earlier, but do you have an oscilloscope? Once you replace components, we can use it to probe the motor controller to see if it's behaving correctly.

 

[BEU]

Yes, I have a 4 ch oscilloscope and a logic analyzer. Probably not only Q3 thats defect. The HD didnt spin att all first. Alter adjusting the solenoid it started and burned something (Q3?).

 

[Berenod]

Here is a video when I manual rotate the drive. Its rotating 8-9 turns with some mechanical noise.




I have disconnected Q1, Q2 and Q3 and they are tested OK. I have measured the resitance in the motor from J2. Resistance from pin 1 (12V) to the motor pin 7, 8, 9 and 10 is 2 ohms. I hope this is a sign that the motor is not burned.
Is it possible in any way to lubricate the drive without opening it??

I will soon make a test when I get the new Q3.

Oooh, that does not sound very good.
Almost like a mechanical issue, like shot bearings!

My (working) widget is out of my Lisa, so could quickly do a video like yours, and on mine the rotation is totally silent.

If on yours the alignment of the axle gets out of whack due to worn out bearing, might be bad news.

The way the widget works, at a high enough rotational speed the heads build up a small distance (like few thousands of mm) between them and the disk due to airflow generated by the spinning disk.
Which is why there is a "speed pick-up" on the motor, and only when the speed is high enough, the brake gets released so the heads can be moved.

Also, the motor has very little torque, even the slightest resistance can keep it from starting to rotate.
Even without resisance, due to that very small torque, it takes like 10 seconds or more to spin up to full speed!


View attachment VID20230505102035.mp4

 

[Berenod]

Complete spin-up of my widget, till the brake releases at the end with the typical click sound.


View attachment VID20230505105734.mp4

 

[BEU]

Oooh, that does not sound very good.
Almost like a mechanical issue, like shot bearings!

Thank You Berenod for that video. Quite different sound from Your working drive. The problem is that I have not found any description if its possible to repair och replace those bearings. Or otherwise I will look for a substitute for the widget drive.

 

[Berenod]

Thank You Berenod for that video. Quite different sound from Your working drive. The problem is that I have not found any description if its possible to repair och replace those bearings. Or otherwise I will look for a substitute for the widget drive.

Eventhough mine is working, I too am looking around for a replacement, as sooner or later the widget will eventually fail.

In below post, and the one right after you'll see some reccomondations I got over here for a replacement...

The slow process of coaxing my Lisa (2/10) back to life

You’ve got the widget running. Unbelievable! You are very lucky. Congratulations. You can run both Lisa OS and Macintosh OS on it, although I would be reluctant to to any formatting and installing tasks on the disk… you never know if these old disks can handle this….

68kmla.org

 

[BEU]

I have looked at the IDEfile and maybee thats the best alternative.

 

[ScutBoy]

Complete spin-up of my widget, till the brake releases at the end with the typical click sound.


View attachment 56132

That is exactly what my working Widgets sound like. At this point I don't think there's much to do to make them any quieter. Mine do tend to get less noisy the longer they run.

I do admit I like the little chirps the drive makes as it seeks! I wish I could get all my vintage drives to do that.

 

[Berenod]

That is exactly what my working Widgets sound like. At this point I don't think there's much to do to make them any quieter. Mine do tend to get less noisy the longer they run.

I do admit I like the little chirps the drive makes as it seeks! I wish I could get all my vintage drives to do that.

Oh, noise wise I prefer the Miniscribe over the Widget
Just with that noise alone you can visualize the movement of the head!

 

[stepleton]

Mine do tend to get less noisy the longer they run.

Mine too. While it wouldn't be something I'd count on, it wouldn't surprise me too much if the Widget we're trying to fix here spun well enough after some exercise.

@BEU , glad to know you have an oscilloscope. Do give us an update if you try out the quad darlington on a breadboard!

 

[CC_333]

As DIY technologies improve, it might one day be possible to replicate these and affordably manufacture batches of brand new copies of historical spinning drives. Then everyone can have Widgets (and Miniscribes) without going broke!

Or, at least, maybe a way to replicate replacement parts, such as motor bearings, heads and maybe even platters (for those that suffered scratching from a catastrophic head crash).

At the very least, the electronics are probably copy-able, so drives that are mechanically sound but with faulty electronics boards that are damaged beyond repair can be restored to working condition.

c