LCIII Repair

[3lectr1cPPC]

LCIII Recap - Apple design fault -47uF reversed!

Uniserver said Can some one please take a nice picture so people can visually see how it should be and post it in this LC-III Caps! thread? It's no use me taking a photograph of the board I recapped correctly because I put the tantalums on upside down and soldered fuse wire to the pads. So...

68kmla.org

Found this. So… do I leave it be? That thread is quite old so it would nice to hear if anyone’s looked into it further in the last 10 years, especially with the reverse engineering efforts

 

[cgp]

LCIII Recap - Apple design fault -47uF reversed!

Uniserver said Can some one please take a nice picture so people can visually see how it should be and post it in this LC-III Caps! thread? It's no use me taking a photograph of the board I recapped correctly because I put the tantalums on upside down and soldered fuse wire to the pads. So...

68kmla.org

Found this. So… do I leave it be? That thread is quite old so it would nice to hear if anyone’s looked into it further in the last 10 years, especially with the reverse engineering efforts

The debate about C22 crops up from time to time. Here's a thread I started earlier this year after seing a failure: https://tinkerdifferent.com/threads/lciii-c22-polarity.2708/

The silk-screen and the original manufacture is wrong. No fault seems to occur using aluminum caps. But tantalums will cause trouble when reverse-polarized.

Since you're using an aluminum cap, I'd leave it.

 

[3lectr1cPPC]

Alright, guess I’ll leave it. Lucky that I didn’t put a tantalum cap there. I used whatever I had around to recap this - including a few radials and higher voltage rated parts, as you can probably tell from my photo.

 

[bigmessowires]

Yow! Do you know of any other common Macs with wrong silkscreens like that? I just recapped an LCII recently, and I simply followed the silkscreen. I know the IIcx has some caps where no polarity is marked, and the actual polarity is backwards from the caps immediately adjacent, but that's not quite as bad as this.

 

[3lectr1cPPC]

I've run into PC desktop boards before where the stripe meant POSITIVE, which led to some confusion when I was recapping.

The main funny thing here seems to be that apparently, the cap itself was on wrong too from the factory.

 

[Juror22]

This is SO timely, so thank you! I just pulled an old LCIII board out of storage that I had re-capped (badly) several years back, when I was less experienced than I am now. At that time, I was doing two LCIII's at once and only had one working power supply between them, so the fact that one did not turn out and was missing several pads and could have other issues, was not really worth following up on at the time, so I packed it away for later. Much later as it turns out.

Now that I have been having some success repairing boards with damage, I thought I would revisit a couple of earlier failures and see if I could get them to turn out this time around, so I removed the old-new caps, my original attempts at bodge wires, cleaned off the pads and I'm currently cleaning up the board (wow - lot of cap residue remained, so sloppy).

Last night I checked the locations that had missing pads (Did I mention the original re-capping did not go well?) and confirmed where to run the new bodge wires if needed, ordered some ceramic 47uF caps, for another project and included a few extras in the order. Imagine my surprise when I read through this post and saw that one of those would be really handy to use for C22. Interesting discussion on that, so I went and checked the other LCIII that I have been using for years and it was recapped with tantalums, following the polarity direction on the silkscreen, but lo, it has not popped (exploded).

Why? Isn't that the expected result?

 

[cgp]

... Interesting discussion on that, so I went and checked the other LCIII that I have been using for years and it was recapped with tantalums, following the polarity direction on the silkscreen, but lo, it has not popped (exploded).

Why? Isn't that the expected result?

In this case, the reverse-polarized tantalum draws a large current and pulls the -5V rail down (typically to around -2V). So there's not enough power for a pop, but you'll find that printer and modem ports may not work since they rely on the -5V supply.

 

[Juror22]

...and since I have not used those ports on that machine, I would not have seen an issue. Thanks for the response cgp. I have reversed the cap for now and if I remember when I get the other caps in, I will replace it with one of the ceramics.

 

[jmacz]

For what it's worth, I had a tantalum in my LCIII (which I'm in the process of selling) in the wrong direction (ie. following the silkscreen) but I didn't notice any issues. But after discovering that thread (perhaps like 2 months or so back?) I fixed it and put a new tantalum in the way described in the thread.

@3lectr1cPPC congrats!

 

[max1zzz]

Dang these were some fragile pads, or maybe I went too quickly. Turns out this one broke too. Luckily some of it is still intact so I’ll just solder to that… so dang. Might have to reconsider my methods if this keeps happening. But it hasn’t before, so I don’t know…
Neither of these are fatal fails, but still makes this more of a pain.

Could just be that the board was just fragile, but it could also be that not enough down pressure was applied to the cap when it was being twisted
(not trying to criticise the way you did it, just noting it because not applying down pressure is a common mistake)

 

[3lectr1cPPC]

I definitely think that was part of it. I think I went a little too fast and got a bit careless as a result.

 

[JC8080]

I just look my first L with the twist method. Not sure exactly what happened, but now I have to bodge to what’s left of the right pad, the resistor on the left, and best of all, that tiny via.
Fun days!

I generally use my hot air station to remove caps, but when one is in a bad place for hot air I will use a different quick-and-dirty method. I use a pair of side-cutters to cut the aluminum cap off the base, as close to the base as I can get. I do this while applying downward pressure. Once you snip the aluminum can you can lift it off the base. You will be left with two small leads still attached to the pads, you just use an iron to heat them up and remove them. I think this method puts far less stress on the pads than the twist method.

Edit: my hot air station was under $50 on Amazon, a worthy investment for this type of work.

 

[lobust]

Lately I've been using hot tweezers (Hakko FX8804) to remove them. Not quite as easy as hot air, but no risk of heating something you didn't mean to.

 

[3lectr1cPPC]

I do have a hot air station, but I don’t like using it for this application as it melts the bottoms of the caps, will produce the awful smell, and can even make them explode sometimes. That and the melting risk of things around them. The cutting method seems like a better alternative, I just don’t have that type of cutters currently.