Mac IIfx Restoration Issues

[NeutralStraw]

Hey folks,

I bought what I thought was a Mac II with a pretty standard monitor a few weeks ago. When I opened it up I was surprised to find a Mac IIfx motherboard staring back at me. I replaced the batteries and pulled my unused floppyemu out of the box it came in last year. When I plugged it in and booted the mac, the lights came up on the floppyemu and then shut off. Rebooting the mac wouldn't bring the floppyemu back to life. Leaving it connected long enough does seem to create a bootloop on the mac, though. I thought maybe the CPLD on it got blown as sometimes happens with that board and poked at it with a meter trying to track down the issue. No concrete results. I finally broke down and ordered a bluescsi 2 and a bench powersupply i'd been meaning to pick up for a while.

The bluescsi 2 came first. I set up the sd card and got the board working after a few renames of various files but i'm getting a sad mac with 0000000F 00000003 every time I boot with it attached. I switched images a few times thinking maybe the image itself was a problem, but no dice. I have very little experience with macs of this vintage being a PC guy in HS, but googling, the internet seems to think this is either a bad scsi driver or an invalid instruction trap. I know that the motherboard isn't completely hosed because if I disconnect everything I get a flashing floppy icon on boot.

In the meantime, the bench powersupply arrived and I hooked it up to the floppyemu. The floppyemu booted immediately. No issues.
I put it back in the mac with the bluescsi removed and it does nothing. No life at all.

I'm thinking something blew when I first attached the floppyemu and now the floppy ribbon cables aren't getting power.
I ordered a bunch of stuff from Mouser to recap the IIfx motherboard, hoping that will fix the issue and maybe the bluescsi issue, too, if it's power related.

Is there anything else I should check or replace? I did open the power supply when I first got the machine, cleaned it thoroughly, and lubed the fan because it was really crusty and fairly loud. I didn't pull any of the circuit boards out because they seemed fairly difficult to remove (metal tabs maybe? can't recall). I don't see much online about these powersupplies dying. Should I recap it, too?

 

[olePigeon]

These old Macs are at the point where they should just all be recapped as a matter of course.

If you decide to do so, make sure to get a 20v or higher rated capacitor for C1. It likes to blow up and leave a little crater.

I haven't needed to recap my power supply, but you might as well check voltages before ordering a set of caps. Save on shipping.

 

[Phipli]

Do you have a photo of the logic board?

 

[NeutralStraw]

Here are some pics. I found a few suspicious spots I zoomed in on. But overall it looks good.

 

[joshc]

C24 and C9 need to be replaced with new capacitors, at the very least.

 

[Phipli]

C24 and C9 need to be replaced with new capacitors, at the very least.

@NeutralStraw
But not the rest, the black ones are ok, unless they're not.

 

[joshc]

That is what I meant, yeah. My mention of 'at the very least' meant there could be other things aside from capacitors to look at.

 

[GRudolf94]

The area around C9 will most definitely need rework. That's a good guess for the Sad Mac there. Have you tested the floppyemu elsewhere after the incident? (I'd understand not wanting to risk another machine though)

 

[jmacz]

Hard to tell from the pictures but it does look like you have some broken traces around R3 R4 and R5 near C9. A bunch of them. I would scrape those dark spots on those traces and see if there’s still copper underneath and then test continuity through those spots.

 

[NeutralStraw]

Thanks for the suggestions. I'm trying to clean up around C9 and i'll post once i've figured out if the traces are intact.
While I was looking at the back of the board, I noticed that a whole bunch of the really small smd caps look hand soldered rather than machine placed. Is that a thing with these boards? Or am I looking at someone with my smd skill's replacement work?

 

[Phipli]

Thanks for the suggestions. I'm trying to clean up around C9 and i'll post once i've figured out if the traces are intact.
While I was looking at the back of the board, I noticed that a whole bunch of the really small smd caps look hand soldered rather than machine placed. Is that a thing with these boards? Or am I looking at someone with my smd skill's replacement work?

I can't imagine why someone would be messing with the caps on the underside of the board? Do you have a picture?

 

[NeutralStraw]

Here’s a few examples

 

[joshc]

I think that's just where the pick and place machine places them slighty off (I think that must be a thing, as I've seen this on a lot of boards). Sometimes you do just get slightly wonky components, not often but I have seen it numerous times, on boards that hadn't been reworked.

These boards were not hand soldered and those don't look like they've been replaced to me.

 

[Forrest]

On boards with surface mount components on two sides, the components must be placed and soldered one side at a time. Typically the less complex side is placed and soldered first (bottom side in this case), then the board is flipped over and the more complex side is placed and soldered (top side). The bottom side components move around a little the second time thru the reflow oven. This is normal.

 

[GRudolf94]

The wonky caps are a result of Apple using the wrong footprint for the part, or the wrong part for the footprint. The pads are too small. Not a concern anyhow. (Bottomside components typically receive a dab of red-brownish glue so that they don't wander off inside the reflow oven (or other machinery) when the boards go in for the topside assembly, or selective, or wave soldering).

 

[NeutralStraw]

You folks are a font of knowledge. Thanks!

 

[NeutralStraw]

I spent some time last night cleaning up the suspicious traces around C9 and found all of them to be intact. That was positive.
What wasn't positive was noticing that there was what looked like a bare copper wire under the left floppy connector. I ended up desoldering the connector and found the power trace going to pin 11 to have been essentially melted into pieces. That explains why the floppy emu wasn't getting power, but doesn't explain why it melted in the first place. I bodged it, put the floppy connector back and got the floppyemu to boot and update it's firmware, but it doesn't look like it's reading anything. I'll check the rest of the traces tonight if I have a chance. I also made a mess of the board and replaced C9 and C24 with through hole capacitors while I wait for the smd versions to arrive.

With that change, along with some deoxit in the simm sockets and rom socket, the other larger caps and most of the small black rectangular caps replaced (I know, probably not necessary, but I ordered them from a list I found), I was able to get the bluescsi2 to boot off an iso cd image. Once booted, it was stable. In color mode there was an unusual line on the screen in a fixed position. Bad ram maybe? It doesn't seem happy with any of the populated hd images I tried, but I was able to install 7.6 TO a blank 80mb image... Oddly the drive/image wasn't present/available when I restarted after the install. Something else to look into. I may have prepped the drive wrong. There were a bunch of different partioning utilities and versions on the Legacy restoration iso I booted. Perhaps I picked the wrong one?

What's left? I guess I have to determine if any other traces got nuked when the pin 11/floppy power trace died. I'd LOVE to know what happened there. Struggling to explain it without there having been a short on the floppyemu. I'm almost wondering if I put it down on the metal powersupply or something without realizing it, bridging some of the few pins on the back of the board...? I hope the floppy controller on the IIfx motherboard didn't die.

 

[Phipli]

On boards with surface mount components on two sides, the components must be placed and soldered one side at a time. Typically the less complex side is placed and soldered first (bottom side in this case), then the board is flipped over and the more complex side is placed and soldered (top side). The bottom side components move around a little the second time thru the reflow oven. This is normal.

Yeah, as mentioned, both sides are done automatically. The glue stops things falling off. Another trick that some board use (don't thing it is the case here) is to use different melting point solders. If you solder the first side with 250° solder, it won't melt when you do the second side with 200° solder, or whatever.

I bodged it, put the floppy connector back and got the floppyemu to boot and update it's firmware, but it doesn't look like it's reading anything.

The floppy chip on the logic board might have been damaged, they're fairly sensitive.

Once booted, it was stable. In color mode there was an unusual line on the screen in a fixed position. Bad ram maybe?

Probably bad VRAM, try different numbers of colours and see how the pattern changes.

 

[NeutralStraw]

Looks like the SWIM chip is dead. I checked continuity from the floppy connector to all the should-be-connected pins on the chip and they all look good. So i'm going to try to pull a replacement off a IIci I have with a completely hosed case. Looks like getting case parts will be almost as expensive as getting a new complete unit so I might as well use it for other things. In the mean time, i've never desoldered a plcc chip like this with the pins bent under, so that should be fun. I have a hotair station but no end that fits around the chip as i've seen in a few videos... anyone have suggestions on how to make the process easier?

 

[NeutralStraw]

Also... is there any good reason to keep the RF cage inside the case at this point?