recapped LCII plays chimes of death, no startup chime

[bigmessowires]

My next victim is a Macintosh LCII logic board. The seller said it played chimes of death at power-up, possibly due to missing VRAM, and needed to be recapped. The board looks pretty clean, so I was surprised when during recapping I nudged one of the ICs and it fell off the motherboard (see the circled chip in the photo). I cleaned up the pad and resoldered the chip, and finished recapping. I added a VRAM SIMM stolen from a 475, and I'm using a working power supply from that same 475. It does not have any RAM besides what's soldered on the motherboard.





When powered-on, this LCII shows a solid gray screen and plays chimes of death after a few seconds. There is no startup chime - it goes directly to chimes of death. After the ascending death chime, there's a second chime of four more notes - is this an error code? I haven't encountered this before.

I'm thinking maybe the RAM is bad? Or the VRAM SIMM that I stole from the 475 isn't compatible?

 

[Phipli]

My next victim is a Macintosh LCII logic board. The seller said it played chimes of death at power-up, possibly due to missing VRAM, and needed to be recapped. The board looks pretty clean, so I was surprised when during recapping I nudged one of the ICs and it fell off the motherboard (see the circled chip in the photo). I cleaned up the pad and resoldered the chip, and finished recapping. I added a VRAM SIMM stolen from a 475, and I'm using a working power supply from that same 475. It does not have any RAM besides what's soldered on the motherboard.

View attachment 64383

View attachment 64384

When powered-on, this LCII shows a solid gray screen and plays chimes of death after a few seconds. There is no startup chime - it goes directly to chimes of death. After the ascending death chime, there's a second chime of four more notes - is this an error code? I haven't encountered this before.

I'm thinking maybe the RAM is bad? Or the VRAM SIMM that I stole from the 475 isn't compatible?

The long death chime is just the death chime on these. Same on the Mac II and IIx.

Sound is distorting - did you get the voltage after replacing the regulator that fell off?

Did you know most of these machine output error data through the modem port?

 

[bigmessowires]

Here's the VRAM SIMM that was stolen from the 475. If I'm reading this correctly it's a 256K SIMM.

 

[Phipli]

By the way, that is the tidiest LC II I've seen in years. Probably not much wrong with it. I'd start by spraying deoxit in the slots.

Are the ROMs soldered?

 

[Phipli]

Oh, another thing, the Egret chip (U10) often gets clobbered by cap juice, so worth cleaning and inspecting, perhaps refreshing the solder in situ.

 

[cheesestraws]

I may be misremembering, but I don't think missing VRAM on these causes a death chime, I think it just sits there not displaying anything in a deeply annoying fashion. So it is probably something else.

 

[bigmessowires]

Thanks for the ideas. Good to know about the VRAM. This board looks very clean, and I don't see any damage to traces or vias, so I wasn't expecting it to give me trouble.

The voltages all look good, including the 8V from the chip that fell off (measured 7.89V).

I fluxed the pins on the Egret chip and reflowed them all until they looked nice and shiny. Then for good measure I did the same to all the other chips in that corner of the board near all the capacitors. I went through them all under 10x magnification to verify the solder joints look OK.

Unfortunately I somehow made it worse, and now the board doesn't make any sound at all. The video output is the same as before: it shows a solid gray screen for about two seconds, then black for about half a second, then solid gray again.

Would I be correct in guessing that UB10 is the sound chip? I thought maybe I'd fouled up when I reflowed its pins, so I double-checked them all visually under magnification, and also tested to make sure none of them were accidentally shorted. I did find two adjacent pins that are shorted, but it's a supply voltage and I think it's supposed to be that way.

Maybe worth noting: the first time I powered on this board after recapping it, I heard the chimes of death at very low volume, almost inaudible. But after trying again, it started playing the chimes at normal volume, until this more recent cleanup job when it stopped playing the chimes.

 

[bigmessowires]

The LCII has started booting even though I didn't touch anything. Maybe I hadn't waited long enough for the alcohol to evaporate after my last cleanup job? I'd thought that alcohol wasn't conductive. This is very strange!

There's something wrong with the video though - black vertical lines through the picture. I tried two different VRAM SIMMs with the same result.



But when there's other stuff on the screen, or you drag a window around, the black lines begin to shimmer or even disappear from some areas of the screen.



So there's probably a data line at the VRAM socket that's flakey - at least that's my guess. If you look carefully, there are actually a few places where the black lines become white lines, like in the letter t in "Disk First Aid". Maybe two of the VRAM data lines are bridged.

 

[bigmessowires]

The lines repeat every 16 pixels on the 12th pixel (0-based). And it's a 16-bit data path, so that makes sense. It might be one of the data bus contacts or connections between VRAM and the CPU, or the other set between VRAM and the CLUT DAC. I will beep them out.

 

[cheesestraws]

There's something wrong with the video though - black vertical lines through the picture. I tried two different VRAM SIMMs with the same result.

One vertical line every 16 - sounds plausible.

edit: ahh, you beat me!

 

[bigmessowires]

I have a spare power supply I could use with this LC II motherboard, but no case. My punishment for fixing this thing is now I need to find a case for it. Guess who gave away a free LC case two weeks ago at Mactoberfest? Me!

 

[cheesestraws]

LC II logic boards in nondescript cases make great AppleTalk routers/LocalTalk-to-Ethernet gateways, just saying. ;-)

 

[bigmessowires]

I found and fixed a broken trace going to the CLUT DAC, and now this LCII is ready for... whatever one does with a 30-year-old computer. This was after I spent 15 minutes troubleshooting lack of video due to not having any VRAM installed, oof. I can't switch out of B&W mode at the moment, because there's no Monitors control panel on the only disk I have handy with the required system enabler extension, but the video works.

Another one saved from the scrap heap!

 

[LaPorta]

Another tip: if you run without a PRAM battery, it will revert to B&W no matter what you set each restart.

 

[Phipli]

: it shows a solid gray screen for about two seconds,

Showing grey is normal for an LC and a LC II. I don't know why, but they're the only 68k macs I know that briefly show grey instead of the half tone.

Not getting past that was weird thought.

 

[Phipli]

Would I be correct in guessing that UB10 is the sound chip?

Yeah UB10 "DFAC" is the sound chip, it will have an external OpAmp close by... UB9 probably? An LM3080?

DFAC is insanely sensitive to cap juice, flux or alcohol residue on the board.

Good to see you got it running.

They're not the best of Macs. The LC III was a huge step up and the first "good" LC. The LC and LC II were cheap and colour but... Slow. Very slow compared to their contemporaries (except the Classic)

 

[bigmessowires]

I think it must have been the 1% water in my 99% isopropyl alcohol that was causing trouble. I only waited a few minutes after cleaning the board before testing it, then it didn't work. When I came back 45 minutes later, it was working. The sound is better too, not distorted like before.

Yeah the LC and LCII are really hampered by their 16-bit bus. Are they the only color Macs with a 16-bit bus? I wonder if the narrower bus really saved very much money for Apple, or if these models were intentionally crippled to reach a particular price/performance target and avoid stealing sales from the other color Mac models. With the narrower bus Apple could eliminate a few buffer ICs, and two SIMM sockets, and the PCB could be simpler with fewer traces to route. But I wouldn't have guessed any of those were major cost drivers.

I think the LC and LCII should have nearly identical performance. The only important difference is having a 68030 instead of 68020, which should perform similarly under Mac OS. 68030 has a slightly bigger cache which should give it a small performance edge over the 68020.

I have a soft spot for the original LC, since it was my first color Mac, and the first Mac I bought entirely with my own money. It's also the computer that I used to develop Tetris Max. I bought the computer from a local store and bought an off-brand monitor from a mail-order company because it was slightly cheaper than Apple's monitor. The mail-order monitor took an eternity to be delivered, many weeks. I remember flicking the LC power switch on and off just so I could hear the chime while I waited forlornly for the monitor delivery to hurry up. When it finally arrived, a friend of mine received the package and told me "methinks it is thine monitor". Funny that I remember his exact words 32 year later, and I don't know why he lapsed into Olde English for that moment. I later spent a lot of money to add an FPU card to that LC, which turned out to be a complete waste since virtually no software used it anyway. I'll stop my old man ramblings now...

 

[cheesestraws]

Are they the only color Macs with a 16-bit bus?

The Colour Classic does too, but that's very very similar to the LC II.

 

[Phipli]

Yeah the LC and LCII are really hampered by their 16-bit bus. Are they the only color Macs with a 16-bit bus? I wonder if the narrower bus really saved very much money for Apple, or if these models were intentionally crippled to reach a particular price/performance target and avoid stealing sales from the other color Mac models. With the narrower bus Apple could eliminate a few buffer ICs, and two SIMM sockets, and the PCB could be simpler with fewer traces to route. But I wouldn't have guessed any of those were major cost drivers.

I suspect the big reason was to cut down on RAM SIMMs. Four take lots of physical space and RAM was expensive.

Besides, the full 32bits actually go to most of the things on the bus, it is more 16bit memory, on a 32bit bus.

I think the LC and LCII should have nearly identical performance. The only important difference is having a 68030 instead of 68020, which should perform similarly under Mac OS. 68030 has a slightly bigger cache which should give it a small performance edge over the 68020.

No, there is a noticeable difference, in benchmarks at least (which are sometimes just a "do I have cache" test). The differences between the 020 and 030 do lift it up a bit. Urgh... I... can't think where I can most easily get the figures, but I've forgotten the difference. I'll have a bit of a dig.

I later spent a lot of money to add an FPU card to that LC, which turned out to be a complete waste since virtually no software used it anyway. I'll stop my old man ramblings now...

You weren't doing enough audio and 3D! Excel though was the thing that people commonly used it for. Matlab, Mathematica and MathCAD were the things I'd have been using an FPU for if it had been my university machine.

 

[Phipli]

I think the LC and LCII should have nearly identical performance. The only important difference is having a 68030 instead of 68020, which should perform similarly under Mac OS. 68030 has a slightly bigger cache which should give it a small performance edge over the 68020.

No, there is a noticeable difference, in benchmarks at least (which are sometimes just a "do I have cache" test). The differences between the 020 and 030 do lift it up a bit. Urgh... I... can't think where I can most easily get the figures, but I've forgotten the difference. I'll have a bit of a dig.

Hahaha

No, half of nothing at all. About 1.06%. Found the benchmark scores I was thinking of :