• Updated 2023-07-12: Hello, Guest! Welcome back, and be sure to check out this follow-up post about our outage a week or so ago.

Gutless Lisa 2/10

mst3k

Well-known member
Ok. A new day. I've tried moving the arm with drive running and it gets stuck on the down stroke. I'm not sure what is happening in there... Is this just a mechanical spindle? Might a micro drop of WD-40 help? One other conundrum, I've been told to move the arm while the drive is running and NOT to move the arm while the drive is running.
I'm still getting the 81 error code. This definitely the current problem. Until this arm moves freely I doubt if I can get to the next one. :)
 

mst3k

Well-known member
OK. New day. I've tried moving the arm into the 'sticky' zone and then pushing it out. I'm not confident that anything has really changed. There is another conundrum too. One suggestion was to move the arm only when the drive was off and the other was to move it only when the drive is on. I'd really like to give it a work out, but I'm not sure which path to follow (or if it matters).
Any thoughts on a micro drop (and I need micro drop) of WD-40 at the base of the spindle? I'm not sure how sealed up this is and the last thing I want is to get some odd lubricant seeping thru to the platter.
Anyway, man versus machine continues. So far the widget machine is winning. :)
 

Berenod

Well-known member
The spinning of the platter generates airflow, which basically keeps the head from actually touching the platter (with a very small separation).
When the disk is not spinning, so without the airflow between head and platter, moving the arm might drag the head over the platter inflicting tiny bits of damage.

It is the most likely reason why they designed the brake, together with the pick-up, which measures the rotational speed, so the drive will only move the head when the drive is spinning sufficiently fast.
 

stepleton

Well-known member
Never, never, never move the arm unless the spindle is turning!

In Widget and most other hard drives, the heads never touch the disk surface: they ride on a cushion of air built up by the rotational motion of the disk. When the disk stops, the heads land somewhere (ideally on a "landing track", a specific cylinder dedicated to this purpose) and don't move.

If the disk is not turning, you will scrape the head along the disk surface, which is good for neither of them.
 

stepleton

Well-known member
Servo trim pots are a set of potentiometers on the Widget motherboard used for adjusting the optical system that's part of the coarse servo positioning system. You can read how they are used starting from PDF page 44 of this document:
(Whatever you do, don't read the very last bit of the very last page. That information is only for Walt Webber!)

Anyway, we haven't got that far yet --- for now we still need to free up that head.

So far I've never heard of anyone attempt any sort of lubrication, and it's not clear that WD-40 would be the lubricant of choice (more likely some kind of light machine oil, or porpoise jaw oil if you're a time traveller and not a vegan). You would be a test pilot if you attempted it.

Instead of doing that right now, I might first recommend continuing to exercise the arm manually, smoothly and steadily, aiming to see if the stiffness improves --- the workout you're talking about. If there's no sign of progress after a few minutes of this, then I think it's time to think about other options.
 

mst3k

Well-known member
Is there any greater feeling than solving an old timey computer problem? There probably is, BUT NOT TODAY!!
After wiggly and jiggly and nothing happening, I wondered if the arm was binding somehow on the spindle.
So, I removed the three screws from the cover and gently rotate the metal cover. I then move it back in position and refastened the screws.
Power on and VOILA! Boots to desktop. I've been running it now for a half hour with no problems.
I have NO actual knowledge of what solved the problem, but I am betting on some kind of gunky stiction against the face plate.
I did not attempt to remove the plate, just rotated it a bit.
Anyway, thanks to all of you and especially Stepleton and Berenod! Yay!!
 

stepleton

Well-known member
Congratulations! This is a fascinating development, and it may illuminate a new way to deal with this common-seeming problem.

Can you be more specific about what you mean by "rotat[ing] the metal cover"? Did you lift it all the way off? If so, did you look underneath the cover? If you did that, what did you find?

When you rotated it, how much did you rotate it? Did you wiggle it back and forth or just give it one good twist?

Did you place the cover back down in the same configuration it was in originally, or did you leave it rotated by some amount? If so, how much?

When you replaced the screws, did you try to tighten them as tight as you found them, or did you leave them a little loose so that the cover wasn't pressing down quite as hard?

Thanks for any answers --- I think there's a lot to learn here!
 

mst3k

Well-known member
My original thought was to take a peek at the arm by removing the cover, however, after loosening the three screws, I discovered that the cover does not 'remove'. It does, however, rotate. I'm assuming that if you rotate it far enough something gets released or the cover comes off or something, but as I turned it, the friction actually increased. So, like the frightened newbie tech I am, I rotated it back and screwed the screws back in...
I had rotated the cover approximately through 2/3 of a turn (moving past two of the screw holes) counterclockwise and then clockwise back to it's original position. I tightened the screws but couldn't honestly say if they were as tight as previously.
After finishing, I checked the arm and it didn't move (held by the brake). As soon as I powered on the machine (assuming of course that I had totally destroyed the arm mechanism) the brake released and the computer promptly booted to the MacWorks desktop.
I did a jig, played around with the desktop, and then began restarting and rebooting to insure this wasn't a fluke.
Something was obviously interfering with the free movement of the arm and moving the cover freed it, but, let's be honest, I have no idea what actually occurred.

I'm letting the beast cool down completely and will try and boot again tomorrow AM to see if the magic is real.
Thanks again for the help!
 

mst3k

Well-known member
OK. Fired up the Widget today. Brake failed to disengage.
Fiddled with the brake, using a .012 feeler gauge, with no result.
I moved the brake completely out of the way to check on the drive arm and boot sequence.
Everything works perfectly. Drive boots right up and there are no problems.
Went back to work on the brake and NO RESULT.
Does anyone know, are the brake pads on the brake and the arm supposed to line up exactly or are they staggered to some degree?
I can't imagine that the brake problem is related to the stuck arm solution, since it all worked fine yesterday.
Anyway, will fiddle with it. It's just nice to have a working Widget.
 

mactjaap

Well-known member
Congratulations!!! Really cool story!
Could you make a picture of the three screws? I’m not sure which ones you mean. I have a widget with a little bit the same problems….
 

mst3k

Well-known member
Amen brother. If I tried the trick again, I would start with a much smaller twist. Maybe just a quarter turn or less, fasten it up and try again.
 

mactjaap

Well-known member
Amen brother. If I tried the trick again, I would start with a much smaller twist. Maybe just a quarter turn or less, fasten it up and try again.
Ok. Thanks for the tip. Will have to wait to try but goes on my to do list!
 

Berenod

Well-known member
OK. Fired up the Widget today. Brake failed to disengage.
Fiddled with the brake, using a .012 feeler gauge, with no result.
I moved the brake completely out of the way to check on the drive arm and boot sequence.
Everything works perfectly. Drive boots right up and there are no problems.
Went back to work on the brake and NO RESULT.
Does anyone know, are the brake pads on the brake and the arm supposed to line up exactly or are they staggered to some degree?
I can't imagine that the brake problem is related to the stuck arm solution, since it all worked fine yesterday.
Anyway, will fiddle with it. It's just nice to have a working Widget.
As we mentioned above, the brake is a rather important part of the drive.

Because of the damage that can be inflicted by moving the head over de disk while it is not turning, the widget only releases the brake when it is turning sufficiently fast.
On the widget the head is moved with a voice coil, meaning there really is no friction when the voice coil is not powered and the head is sort of "free moving".
Unlike drives where a stepper motor is used to move the head, very unlikey the head would move due to the high friction the steppermotor gives, basically being a brake when being passive.

So without the brake on, be very careful moving the drive, the head could move away from its "landing strip", can create damage!

Possibly the pick-up, which detects the speed of the drive, is not positioned correctly anymore, something to check maybe?


widget pick up.JPG

widget pick up2.JPG
 
Top