Print

[snuci]

As a check, just after booting, do you hear the Twiggy drives "grunt" as they activate the ejection mechanism in either drive? At least, this is what happens on my Lisa 1. The ejection mechanism is also activated when I shut the Lisa back down.

I did make one disk which I think should be okay but I am not sure if it is functionally okay.   To make the disk, I cut the plastic melted plastic "rivets" on one side of the floppy diskette and unfolded the left side flap of the diskette; taking out the cookie.  I then cut the appropriate holes and put it back and glued the flap back into place.  Structurally, it should be okay but I am not 100% sure the cookie was unscathed.  My top drive acts like the video I am pointing to below.  Inserting the disk for the first time does it's initial checking.  You will hear a click in the background (that's the mouse button I clicked loudly) and then the disk was attempted to be initialized.  That didn't work so I ejected it.  The bottom drive doesn't move properly but I will investigate.

http://www.vintagecomputer.ca/files/Apple/Lisa/Testing/Twiggy.mp4

Is this normal or abnormal for the top drive?  I checked an online video by Alker33 and it is different but he also had a known good disk so is my head movement normal for a bad disk?

Thanks for any help.

[stepleton]

The video sounds normal until the heads assume that central position and do that "chunka chunka chunka chunka" thing. That's nothing I've seen before.

It's interesting also to see that the "chunkas" didn't happen when you tried the initialisation---but the heads still refused to retract any further than they do.

And yet, the heads clearly can retract fully when it comes time to unclamp and eject the disk, so it's not like there's a mechanical blocker. I don't have a lot of Twiggy repair experience, but I can't think of an obvious reason for the drive to refuse to retract the heads during ordinary seeks like yours does. There's a sensor that tells the head carriage when it has been fully extended (look for the red, brown, and orange wires feeding into it at bottom-right of this photo), but there's no separate sensor to detect full retraction that I know of.

(My guess is that the head positioning is calibrated by the full-extension sensor, and then the disk drive controller just counts stepper motor "steps" after that. It seems unlikely to me that the stepper motor would still step but that its counting would go off, so I don't think it would be a problem with the computer losing track of where the heads are.)

Maybe it's what's on the media. Would it be possible to try formatting the disk in BLU? Perhaps the Office System is expecting an uninitialised disk to have been "more formatted" than yours is.

PS: Kudos on the precision of your cutout for the near head---it looks very nice!

[snuci]

Okay so I booted from the X/Profile/BLU and tried to format a floppy.

Upper drive: "Floppy drive error 1B" after some trying. When I pressed "F" to fail, I got "Operation Failed Error Code FA20"
Lower drive: still doesn't clamp down.  Got an error 16 on that one.

Is there  list of error codes and what they mean?

[stepleton]

While I don't know what FA20 means, the other two codes are described on page 6-37 (PDF page 206) of the Lisa Hardware Manual.

$16 is "Unable to Clamp", so nothing too newsworthy there.
$1B is "Unable to Adjust Speed".

As you probably know, Twiggy disks spin at different speeds depending on which track is being accessed. I guess this is why your heads refused to retract any further---they were supposed to change speeds once the heads passed a certain point, but they didn't. I don't know too much about how this is controlled, so an error like that for me would be the beginning of a schematics-n-probe adventure, I expect.

Perhaps first check the easy stuff. On your Twiggy drives... are all the connectors plugged in? Are all the chips well-seated in their sockets? Here are a pair of pictures of one of my drives to compare against:

https://photos.app.goo.gl/63moQ338i9PWuFUx6

Speaking of schematics, none of the Twiggy schematics I've seen online match my drives very well. This set on Ray's site is the same as the one on Bitsavers, and there are some big differences---for example, page 3 says that an MC3470-E5 read amplifier chip should be present somewhere, but there's nothing like that on my drives.

(I see you're getting some information over on Facebook---I'll tune in there, too and see what they come up with...)

[snuci]

I meant to post my pictures here too.  Here are my boards.  I will double check all of the connections and even test the components.  I can hear at least three distinct speeds when the upper drive tries to read the outer track.  Kinda cool but I don't think it finds the right speed so something.

http://vintagecomputer.ca/twiggy-pics/

[snuci]

Okay, so I remove all chips and connectors and used contact cleaner.  Now I have lower Twiggy drive working.  I was able to make two floppies and wrote Lisa Office System I and II from Bitsavers.  While my upper drive gives 1A (Unable to Find Calibration) errors, it did try to format three tracks when first accessed and then has given this error in BLU since.

I was able to install LOS on an X/Profile but it uses the upper drive for "disk 2" when you start LOS from "disk 1" and you attempt to install LOS on a Profile.  Oddly enough, the upper drive sure looked like it was working and the installation completed with no errors but there are no apps in LOS except for the clipboard and preferences.  Does Lisa Write, etc normally get installed?  Could my upper drive have worked or did it just silently give up copying files from that disk and complete with missing files? 

Trying to boot from the LOS disk 1 in the upper drive gives me an Error 45 (Bus error?) but I did turn off my X/Profile to try to boot from the floppy.  Any experience with this and advise would be appreciated.  I will try cleaning the drive and rechecking the ICs/connections.

[stepleton]

The Office System tools (LisaWrite, etc) come on separate disks. You can find them here:
http://bitsavers.trailing-edge.com/bits/Apple/Lisa/twiggy/

Note that these disks may be serialised to a different Lisa, and while the disk images are probably deserialisable using something like the usual editing procedure for 3.5" diskettes, I'm uncertain whether anyone has tried it.

I've never encountered any mismatch of upper and lower drives on my Lisa. Is it the case that the ROM correctly identifies the upper drive as Drive 1 and the lower drive as Drive 2, while the Office System gets it backward? (In other words: are you sure your cables are installed the right way round?  )

[stepleton]

Ah sorry, I misunderstood---the installation procedure was just referring to the other disk drive as the "second drive" or something. I think this is normal

[snuci]

With both Twiggy's I have one working and have narrowed down the other to having an issue with the bottom board.  If I swap bottom boards, the bad drive is fine.  I'll have to start swapping chips to see where the issues lies now. That is time consuming but needs to be done.  Perhaps I'll start by swapping all chips from the good bottom board to see if that fixes it and then take it from there.  It might also be the axial caps but I usually don't see those type going and they look visibly okay.

[stepleton]

As a computer scientist I have to recommend doing a binary search among your chips 

For your sake I hope that this is the problem, although there are plenty of other things that can go wrong. The good news is that virtually all the ICs on the digital board are commodity parts that are widely available even today.

PS: How difficult was it to remove the drive belt from the spindle? I've never attempted it.

[rayarachelian]

It might also be the axial caps but I usually don't see those type going and they look visibly okay.

There are in circuit capacitor testers out there for $50~$60, might be worth using one of those to save more time later on. Hopefully all the chips on the questionable board are socketted. From Tom's photo, at least those are.

PS: How difficult was it to remove the drive belt from the spindle? I've never attempted it.

And an even better question, how would one go about building a new belt? Not that I own any twiggy drives, but that's something that's bound to eventually dry out and fail.

[snuci]

As a computer scientist I have to recommend doing a binary search among your chips 

Well, swapping all chips on the bottom board swapped the issue so it is definitely one of the chips.  Thankfully, they are all socketed.  That is very good news but now I have to swap chips one-by-one or half at a time to narrow things down. I do not have replacements for most of them and the ones I do I have already tried replacing with no luck.

The problem is that inserting a disk or formatting will cause the head to go to the outer track on the disk and it cannot find the right speed as it tries several before it gives up and displays an error in BLU.  If I write a disk from memory, it starts and the inner track and writes out and it does several tracks until it fails as it travels towards the outer tracks.

PS: How difficult was it to remove the drive belt from the spindle? I've never attempted it.

Actually, it is very easy.  It's very much like a bicycle tire.  You pull up on one side of the belt and turn the center spindle until it comes off.  To put it back on, you put it on the motor side, then put the belt on one side of the center spindle side and start turning the center spindle and it belt eventually evens out and it's on.  The belt is made up of woven fiber rubberized material that is quite strong and will last another 30 years. It has not stretched at all which is a bonus.

Getting closer...

[blusnowkitty]

I can't say if it's applicable to your case with Twiggy disks, but my 2/5 had an issue where it would attempt to format a disk but fail to verify until I replaced P6A. Might be worth checking out if you have a spare one of those, unfortunately it's a custom Apple part.

[snuci]

The problem chip was the MC14015B 4-bit shift register on the bottom board.  Swapping this between drives makes them work/not work depending on which drive the bad chip is in.  I ordered an MC14015BCP so I should do some  final testing when it arrives.

I have a scratchy sounding drive that I tried to grease the rails but still seems scratchy.  I'll have to do my best to clean it while I wait for parts. Stepleton, have you had to clean your drives at all?

[stepleton]

I was cleaning the drive a couple of weeks ago, as a matter of fact---you can see where I was asking for tips beforehand here.

But I took a minimalist approach inspired by the assembly guide I mention in that post. I decided not to lubricate my drive rails and suspect that they weren't ever lubricated---I could see no sign of any kind of grease on them, and when I dusted them with a tissue there was no residue. I think the "teflon coating" referred to in the assembly document is a permanent coating.

I wound up cleaning the little channel in the casting where the "frog leg" roller goes---seems to me like it may have had a bit of graphite lube on it---and put down a bit of polyolefin grease there. Meanwhile, the manufacturing instructions specify Tri-Flow PTFE lubricant for the stepper motor lead screw, and while I don't have any to hand, I've ordered some and will add a few drops later on.

None of my ministrations so far have changed how my Twiggy drives sound---mildly suggestive of snapping themselves to bits during disk insertion and ejection, the same way they've done for the last 22 years