LisaList2

Advanced search  

News:

2022.06.03 added links to LisaList1 and LisaFAQ to the General Category

Pages: [1]   Go Down

Author Topic: Recent Lisa acquisition and repair - really bad NiCad leak.  (Read 7347 times)

rayarachelian

  • Administrator
  • Hero Member
  • *****
  • Karma: +101/-0
  • Offline Offline
  • Posts: 772
  • writing the code,writing the code,writing the code
    • LisaEm
Recent Lisa acquisition and repair - really bad NiCad leak.
« on: March 16, 2019, 06:09:58 pm »

I recently picked up another Lisa, this makes it four.
This one is an XL. It has the 800K Sony drive and I/O ROM, and Sun Rem hard drive inside that self-identifies as a Profile, but uses a ST-138 (which is a 30MB drive?). The drive is connected from the external motherboard parallel port via ribbon cable snaked back internally from the back to the drive cage.BLU identifies the drive as having f1b8 blocks or 61880 blocks, or 30940K, so it's a 30MB disk. Block size shows up 0x214 (512 + 20), revision as 390 spares as 0x20 (32), 0 spares allocated, 0 bad blocks.

What's interesting is that the drive shows up as PROFILE (but not PROFILE-30!) so perhaps I can manage to get LOS on it if I swap out the ROMs with H ones, not sure, unless the ProFile driver hard codes the size of the disk or has other limits, that might be an interesting thing to try. If it works, or if I can find where in the driver such limits exist, it would mean that we can get larger disks in LisaEm.

The drive isn't bootable unfortunately, it shows a single line of text saying "LOADING." and then shows more ...'s like "LOADING......" and eventually goes back to the boot ROM without starting up. Likely it has some version of MacWorks on it, but got damaged. Perhaps I can clone it with BLU and see what's there.

I wasn't able to boot NeoWidex as it got a disk error 47 after reading about 65K or so, so I'll need to dig up a floppy cleaning disk and some isopropyl alcohol, or will need to take the drive apart and use q-tips, etc. But BLU booted up fine.

The Video display also bleeds to the right, that is if there's a vertical line on the display, it smears to the right of that line, such as from the hourglass icon or text or the border of windows, so either the video board or the power supply I used needs a bit of work as well.

Some lessons about this Lisa.
1. The floppy cable was connected backwards into the LisaLite adapter, when this happens, the power light will light up for about half a second, but will go out before you let go of the button - so that's something to add to the LisaFAQ. This is different from the tick-tick-tick of a bad power supply.

2. There is corrosion not just on the I/O board but also on the CPU board and mostly ABOVE the batteries! Infact I wasn't able to get video out of it with its original CPU board at all, and had to use a spare.
There is a pin messed up on the I/O board connector of the motherboard, but it seems not to affect it - or maybe it does and I haven't noticed it, or is causing the floppy issues, not sure yet.
Other than this damage, this Lisa looks impressively clean and only slightly yellowed. I don't actually think it was cleaned. I suspect it wasn't used very much, and perhaps stored in a sealed environment, which then concentrated the corrosive gas from the NiCAD pack.
The back of the CPU board had some of this crud crystallized into weird patterns, but it was like a film of dust.

Besides the corrosion on the I/O boards and CPU board, the boards inside were very clean, the memory boards were pristine.

But the lesson here is that the NiCAD's don't just leak alkaline fluid, but also offgas corrosive gas which then damages other things. I found some slight corrosion on the inside corners of the case on the steel casing itself, and even some stuff on the outside of the CRT - now that might just be regular dirt, I don't know, but as it had the nylon screen protector and the dirt was between the nylon and the CRT, I'm kinda thinking maybe it is the stuff from the battery after all!
Incase you're wondering, I used a high power electric air blower meant to replace compressed air to clean the cruft off the boards. Next, I soaked a paper towel with some distilled white vinegar and RO filter water (not as good as distilled, but close enough) and wiped both the I/O and CPU board thoroughly and then used an antistatic electronics cleaning brush to scrub (it's much like a toothbrush but bigger).
I then used WD40 Electronics Contact Cleaner spray (no this is not regular WD40, do not use that!), and gave both boards a generous spraying on both sides until they were dripping with it, and then left them to offgas the alcohol and whatever else overnight in the garage.

That spray is very strong and gave me a bit of a headache even though I did that outside the house, so take that into consideration before using it. It did clean the boards, but looks like there's still a lot of corrosion on the solder points and on some legs of various ICs. Likely I'll have to reflow the solder on these and check for broken traces, but weirdly there's more damage on the CPU board than on the I/O board. Most of the damage is ABOVE the batteries, only a little is below. This is why I say these batteries offgas.
« Last Edit: March 16, 2019, 07:31:32 pm by rayarachelian »
Logged
You don't know what it's like, you don't have a clue, if you did you'd find yourselves doing the same thing, too, Writing the code, Writing the code

rayarachelian

  • Administrator
  • Hero Member
  • *****
  • Karma: +101/-0
  • Offline Offline
  • Posts: 772
  • writing the code,writing the code,writing the code
    • LisaEm
Re: Recent Lisa acquisition and repair - really bad NiCad leak.
« Reply #1 on: March 16, 2019, 07:32:06 pm »

photos 2
Logged
You don't know what it's like, you don't have a clue, if you did you'd find yourselves doing the same thing, too, Writing the code, Writing the code

rayarachelian

  • Administrator
  • Hero Member
  • *****
  • Karma: +101/-0
  • Offline Offline
  • Posts: 772
  • writing the code,writing the code,writing the code
    • LisaEm
Re: Recent Lisa acquisition and repair - really bad NiCad leak.
« Reply #2 on: March 16, 2019, 07:34:29 pm »

photos 3
Logged
You don't know what it's like, you don't have a clue, if you did you'd find yourselves doing the same thing, too, Writing the code, Writing the code

rayarachelian

  • Administrator
  • Hero Member
  • *****
  • Karma: +101/-0
  • Offline Offline
  • Posts: 772
  • writing the code,writing the code,writing the code
    • LisaEm
Re: Recent Lisa acquisition and repair - really bad NiCad leak.
« Reply #3 on: March 16, 2019, 07:36:02 pm »

corroded cpu board front
Logged
You don't know what it's like, you don't have a clue, if you did you'd find yourselves doing the same thing, too, Writing the code, Writing the code

rayarachelian

  • Administrator
  • Hero Member
  • *****
  • Karma: +101/-0
  • Offline Offline
  • Posts: 772
  • writing the code,writing the code,writing the code
    • LisaEm
Re: Recent Lisa acquisition and repair - really bad NiCad leak.
« Reply #4 on: March 29, 2019, 03:21:59 pm »

This 5 video youtube list: https://www.youtube.com/watch?v=OQIWmmxifow&list=PLOkflnR85gThmBTDaI1hAmYPGSeYlayys&index=1 is relevant to this discussion though you don't see the actual repairs in action.

These two links are useful as well:BTW, The fiber glass pens should be used with care and masks as you don't want to breathe that stuff in.
I'll wait for a warmer sunny, day and do that bit outside on the deck while wearing a facemask.
Logged
You don't know what it's like, you don't have a clue, if you did you'd find yourselves doing the same thing, too, Writing the code, Writing the code

jamesdenton

  • Administrator
  • Sr. Member
  • *****
  • Karma: +59/-0
  • Offline Offline
  • Posts: 142
  • ArcaneByte
    • ArcaneByte
Re: Recent Lisa acquisition and repair - really bad NiCad leak.
« Reply #5 on: May 07, 2019, 09:34:55 pm »

Good luck in your resurrection attempts. Thanks for sharing!
Logged

mactjaap

  • Full Member
  • ***
  • Karma: +6/-0
  • Offline Offline
  • Posts: 37
Re: Recent Lisa acquisition and repair - really bad NiCad leak.
« Reply #6 on: April 22, 2020, 11:51:28 am »

Long time ago this post, but I would like to know if you succeded in restoring this Lisa?
Logged

rayarachelian

  • Administrator
  • Hero Member
  • *****
  • Karma: +101/-0
  • Offline Offline
  • Posts: 772
  • writing the code,writing the code,writing the code
    • LisaEm
Re: Recent Lisa acquisition and repair - really bad NiCad leak.
« Reply #7 on: April 22, 2020, 02:44:56 pm »

Long time ago this post, but I would like to know if you succeded in restoring this Lisa?

Sort of, but I've cheated.

I did clean up the boards with baking soda and RO filter water - I didn't have distilled but RO water should be close enough, and then gave them a good spraying on both sides with 3M Electronics cleaner which is mostly alcohol and likely a bit of hexane and other things, I've been meaning to reflow the solder but haven't done so.

The motherboard has corroded CPU or I/O board connectors, I bought new edge connectors but haven't resoldered them.

Haven't been able to fix the corroded boards, but I've had a set of spares from a Lisa that died a long time ago, and it now works with the spare cards and motherboard from that, until the display starts bleeding, so I've got to either clean/replace the video board trim pots and/or possibly replace a cap or two there.

Also it seems the 30MB SunRem drive stopped working completely the last time I've checked. I don't know if it's the drive or the controller, but I'd like to get this one working as well. It never worked, but now it stopped displaying the "LOADING" banner.

(I've also got a 2nd Widget to repair, I managed to get one of the two working), and I've got two dead power supplies to recap as well - one is a spare, but the other is preventing one of the other Lisas from working.

These are all on my to-repair-list after I release 1.2.7, so plenty of hardware work to have fun with.
Logged
You don't know what it's like, you don't have a clue, if you did you'd find yourselves doing the same thing, too, Writing the code, Writing the code

mactjaap

  • Full Member
  • ***
  • Karma: +6/-0
  • Offline Offline
  • Posts: 37
Re: Recent Lisa acquisition and repair - really bad NiCad leak.
« Reply #8 on: April 22, 2020, 04:46:42 pm »

Thank you for the nice round up. These kind of stories are really interesting to read. I learn a lot from it and it shows that restoring a Lisa is not a job for the faint of heart!

I did similar job at a friend of mine. He never ever started his Lisa. And finally we got it working.
Logged

rayarachelian

  • Administrator
  • Hero Member
  • *****
  • Karma: +101/-0
  • Offline Offline
  • Posts: 772
  • writing the code,writing the code,writing the code
    • LisaEm
Re: Recent Lisa acquisition and repair - really bad NiCad leak.
« Reply #9 on: April 23, 2020, 04:32:19 pm »

I suspect this Lisa was bought from SunRem as an XL and likely never used. It came with the original boxes, so likely it just sat for 30 years somewhere unused, not sure why. Maybe it was bought by a business or as a gift for someone who didn't want to use computers and then just stored.
Not sure why the floppy cable was in backwards, but it kind of fits in. For whatever reason instead of repairing it or sending it back to SunRem the owner saw it wasn't working right and stored it for a very long time not realizing the battery time bomb that over the years would corrode it.

I can't say it was "new in box" by any stretch, but I'm pretty sure the corrosion happened after it was stored as there was some cruft on the CRT, likely the alkaline stuff evaporated and precipitated back inside the bagged Lisa. This makes sense as some of the corrosion was on the back of the CPU and I/O boards above where the batteries sit.
I suppose the photos can serve as documentation of what happens. I just hope the traces on the boards aren't corroded on the inside, but damage this bad is likely going to mean it's going to have lots of bad traces.
I hope the hard drive didn't get hit by the corrosion, and likely the metal around it would have caught some of the gas, and having looked at it with a flash light back then, I didn't notice any so likely it's fine. I suspect it'll just be a bunch of bad capacitors on there, or at least hope.

Logged
You don't know what it's like, you don't have a clue, if you did you'd find yourselves doing the same thing, too, Writing the code, Writing the code

rayarachelian

  • Administrator
  • Hero Member
  • *****
  • Karma: +101/-0
  • Offline Offline
  • Posts: 772
  • writing the code,writing the code,writing the code
    • LisaEm
Re: Recent Lisa acquisition and repair - really bad NiCad leak.
« Reply #10 on: May 17, 2020, 07:49:24 pm »

So since there was a recent thread on 68kmla about trying to transfer BLU images to LisaEm, I took this Lisa out and one of my ProFiles and an old dell E6500 laptop, I've upgraded to Ubuntu 20.04 and installed minicom on and tried to image it as I wanted to make some BLU images to play around with, and also since I haven't powered on any of my ProFile drives in at least 4 years and got worried that the lubricant would freeze.

This Lisa's not doing so good. I mentioned that the display is bleeding and it gets worse with time. By the end the display turned off and tweaking the brightness showed just a white screen with diagonal scan line reflections. I used a 1.8A known good power supply from another Lisa, so it must be the video board in addition to the I/O board.

I also saw all sorts of issues such as COP failures on the I/O board turning the Lisa on as soon as power was applied, random beeping, moving the mouse causing the COP to interpret mouse movement as keystrokes, and all sorts of other fun. I wasn't able to transfer the Profile it all.

If I remember correctly I had swapped out the I/O board with a spare, so this one is now bad too, and had swapped the motherboard too since the original one was corroded. But this spare was also from a Lisa that had leaky NiCads... so more work to do...

There was a very tiny bit of good news in that when I first plugged it in, I heard the 30M SunRem hard drive spin up, which when I put this Lisa away had stopped. I didn't want that to happen, but I'm glad it's still alive. But there's probably some capacitor issues there since it started somewhat working after an extended period of being left alone (almost a year now!)

I'm going to try to take another BLU image tonight, and I got the top lid unscrewed so I can get to the video board and catalog the capacitors. I think I should start creating Lisa capacitor spreadsheets as part of this. I'll start with this. Last time I fixed my power supplies, I sort of had a shopping list, but... I think I need a better way to track this.
Logged
You don't know what it's like, you don't have a clue, if you did you'd find yourselves doing the same thing, too, Writing the code, Writing the code

snuci

  • Full Member
  • ***
  • Karma: +9/-0
  • Offline Offline
  • Posts: 38
Re: Recent Lisa acquisition and repair - really bad NiCad leak.
« Reply #11 on: May 18, 2020, 11:25:34 am »

Hi Ray,

When I cleaned my badly corroded I/O board (and other boards), I used an Ultrasonic cleaner with 50% vinegar and 50% water and do this for about 50 minutes.  The traces where the corrosion was turned black but you could use a fiber pen to clean the black stuff off and you would get the copper traces underneath is some cases.  Note: I did have a couple of components fall off :)

Pictures of before:



Ultrasonic cleaner I use:


After:
Logged

rayarachelian

  • Administrator
  • Hero Member
  • *****
  • Karma: +101/-0
  • Offline Offline
  • Posts: 772
  • writing the code,writing the code,writing the code
    • LisaEm
Re: Recent Lisa acquisition and repair - really bad NiCad leak.
« Reply #12 on: May 18, 2020, 02:36:01 pm »

Is it this guy? https://www.amazon.com/VEVOR-Commercial-Ultrasonic-Capacity-Solution/dp/B01HGNYO0U/ 6L?

Fall off you say? I remember an old joke about a guy who gets his willy infected and the doctor wants to chop it off, so he goes to another one and another and they all say the same thing, then he gets to the last one, and the doctor goes "Cut, cut cut, that's all these butchers know how to do. Just give it a week and it will fall off by all by itself." :-D
Logged
You don't know what it's like, you don't have a clue, if you did you'd find yourselves doing the same thing, too, Writing the code, Writing the code

snuci

  • Full Member
  • ***
  • Karma: +9/-0
  • Offline Offline
  • Posts: 38
Re: Recent Lisa acquisition and repair - really bad NiCad leak.
« Reply #13 on: May 18, 2020, 04:21:05 pm »

Is it this guy? https://www.amazon.com/VEVOR-Commercial-Ultrasonic-Capacity-Solution/dp/B01HGNYO0U/ 6L?

Mine is also 6L and heated but it is probably a different manufacturer but that ought to do it.  I hadn't heard of anyone trying this before on a PCB let alone a Lisa I/O board so if you do decide to go this route, you can always do other boards you have that may have had corrosion just to make sure it is all off.  I was pretty amazed.  Good luck!
Logged

patrick

  • Sr. Member
  • ****
  • Karma: +85/-0
  • Offline Offline
  • Posts: 103
    • Patrick's Hardware Page
Re: Recent Lisa acquisition and repair - really bad NiCad leak.
« Reply #14 on: May 18, 2020, 06:04:13 pm »

Ultrasonic cleaning is fine for PCBs. However, you should remove crystals and ceramic cavity packages like EPROMs -- these might crack.
Logged
Pages: [1]   Go Up