IIci rework: fate hangs on a single pin. Save, or salvage for chips?

[Addicted]

In a search for a transplant RAMDAC to fix another IIci, I landed a moderately corroded board as a donor. I now think the board might be salvage-worthy as a whole - but for one critical flaw.

MDU pin 96, /CAS , is either detached at the package or hanging on by the barest of threads. I doubt I can clean the board without this pin falling free. Any hope for a save?

Photo taken after a quick first pass of initial cleaning. I suspect this board sat unprotected somewhere and took some scrapes; a 74-series chip also has been lost, for example. Capacitor leakage for certain, but also plain old humidity I think; the battery area looks good.

 

[finkmac]

you might be able to scratch some of the plastic package away, but it's risky

 

[Addicted]

Started on that approach, @finkmac .. it will take patience.

While doing other restoration work on that board yesterday, I found the '030 had several leads that were just resting on oxidized solder, no bond remaining at all. I desoldered and reflowed all 132 pads, one at a time. No more opens, no shorts created. I mention this becase it made me wonder if perhaps I could transplant an MDU from a donor IIsi or ci.

But for now, modifying the existing chip. I don't dare try a Dremel. And an X-acto knife blade barely scratches the package. So, first step is to find the right tool. Are there X-acto file (abrader) blades? If I can expose enough to solder on, I will seal it with conformant coating. That should be durable and humidity-proof. After that cures, solder the free end of the patch to the pad.

 

[zigzagjoe]

#11 Scalpels are what i typically use. Better than xacto blades and if it even gives a hint of dullness, new blade time. 100 blades for $10 makes that an easy call. Somewhat less tolerant of abuse but plenty strong for anything reasonable. I have saved quite a few 0.4mm pitch tqfps this way so that giant by comparison chip wouldn't worry me.

Two things to worry about - multiple bad leads and corrosion having penetrated deeper into the package. You may want to remove the chip to clean the pcb underneath as well as the non soldered portion of the leads. More dodgy legs are expected, better to find that now and gauge overall effort before doing one just for it to not work due to other issues.

 

[Addicted]

Spot on @zigzagjoe .. for example, I couldn't flush the gunk out from under RP14 (see photo above) and decided to remove it.. depressing amount of reddish-brown mud under there. The traces test fine, but the pads were so oxidized that I am having trouble getting solder to flow onto them.

And that same gunk can be seen under the MDU, the RBV, and the 030. This may be futile. But no one said restoration was easy.

It must be cap ooze, or perhaps more than that.. there's just so much of it for those tiny caps to have oozed. I wonder if this board has been rinsed before, if this is old dust and ooze rinsed together into the deepest crannies and left to age - not just ooze.

I have rinsed with 99% IPA as well as diluted vinegar>distilled water flush. Maybe I need to expand my stock of cleansers.

I'm holding off on putting fresh components (caps, TTL logic) down until this issue is concluded one way or the other. I reflowed the 030 because I initially decided to lift it - but chickened out after thinking, what if the pads come with it..

If I had an MDU on a donor board, I'd opt for scrapping this MDU and transplating any MDU with all of its leads. It seems the more likely road to success, and the opportunity to mop out under any of the large chips would be welcome.

 

[zigzagjoe]

Spot on @zigzagjoe .. for example, I couldn't flush the gunk out from under RP14 (see photo above) and decided to remove it.. depressing amount of reddish-brown mud under there. The traces test fine, but the pads were so oxidized that I am having trouble getting solder to flow onto them.

And that same gunk can be seen under the MDU, the RBV, and the 030. This may be futile. But no one said restoration was easy.

It must be cap ooze, or perhaps more than that.. there's just so much of it for those tiny caps to have oozed. I wonder if this board has been rinsed before, if this is old dust and ooze rinsed together into the deepest crannies and left to age - not just ooze.

I have rinsed with 99% IPA as well as diluted vinegar>distilled water flush. Maybe I need to expand my stock of cleansers.

I'm holding off on putting fresh components (caps, TTL logic) down until this issue is concluded one way or the other. I reflowed the 030 because I initially decided to lift it - but chickened out after thinking, what if the pads come with it..

If I had an MDU on a donor board, I'd opt for scrapping this MDU and transplating any MDU with all of its leads. It seems the more likely road to success, and the opportunity to mop out under any of the large chips would be welcome.

That's a battery bombed board you're looking at, for the record - having brought a few back from the grave; I'd recognize that corrosion anywhere. Naaaasty stuff. Probably someone has already had a go at it, reading between the lines, at a guess: they gave it a topical cleaning, maybe solder a new holder in, it was still very dead (obviously) so they gave up. May well have introduced other issues....

I love a good basket case but gauging the effort long and hard before you go deep is 100% the right strategy.

 

[Addicted]

That's a battery bombed board you're looking at, for the record

I think you're right. At first glance, I saw a solid battery holder with no battery carcass in it and gave that a pass. But that much crud, and where it's found .. unit may have been stored vertically, floppy-end high.

So this is a battery blast scene. With some cap goo around the PSU logic area, too. Doesn't mean it's out of reach, but.. battery issue narrows the odds.

I'll sleep on this and maybe commit to removing the MDU next week. If it comes up cleanly, then back on the shelf it goes until a donor surfaces. Else, scavenge all the chips and connectors I can, over time.

(If I'm going to get back into restoring periodically, maybe it's time to buy one of the pad & trace repair kits.)

 

[Addicted]

I thought about this, and while thinking did extensive continuity checks on the logic board. Which came out surpassing well. So I decided to try and save this board.

The RTC, which in the IIci is adjacent to the battery, had some gunk under it and between the pins that I could not just wash out with IPA or weakened vinegar, no matter how many times I tried. So I lifted the chip. And wicking away the old solder revealed really good pads, and the chip came away very easily.

Except.. the power pin had corroded at the edge of the package. I didn't notice it had remained on the board for a few seconds.

So, damn the torpedeos, full speed ahead. This would be a rehearsal for patching the MDU's missing pin.

Photo history is a lot shorter than me describing it. One thing to point out is that the metal inside the package was 'shiny'; the corrosion stopped outside.



I need to attach a battery and then scope the pins on the RTC to see if it is alive - to see if my patch made contact with the internal metal. (I presume the Iic_ RTC generates interrupts at all times. Not sure which RTC pin is the interrupt.. pin 2?)

Then, on to the MDU. Literally, strike while the iron is hot. Here's the outcome. I used a pin's head worth of solder paste and hot air to try and bridge the patch wire to the internal lead. Only way to test the connection is to put some SIMMs in Bank A and power it all up with a scope on /CAS.a (RP14 pin 4).