Jump to content
man4mac

SE/30 recap – functional with checkered pattern

Recommended Posts

I've recently gotten my hands on an old SE/30 that was stored quite well for the last few decades. I booted the machine up 2 years ago, and at the time it worked like a charm. When I went to start it up recently it displayed a typical Simasima (Horizontal lines, no boot) symptoms. I've done the following things:

 

- Replaced the PRAM battery (it was totally dead)

- Cleaned board thouroughly (although have not given it a full bath)

- Replaced capacitors: C1, C3, C4, C5, C7, C8, C9, C10, C12, C13 (with 47µF, 16v, tantalum) and C6 (with 1µF 50V, electrolytic). I did not replace C11 or C2 (axial).

- Removed and cleaned ROM SIMM

- Removed and cleaned all RAM

 

The good news is – it works! The machine now boots up and functions well. However, there are a few issues I have't been able to figure out:

 

1. The screen has a checkered pattern, almost as if its off register. However, all the troubleshooting information I've found basically says the machine shouldn't boot with a checkered pattern.

 

2. There isn't a proper bootup sound, although sometimes it does make a really strange garbled boot tone

 

3. The floppy drive is dead (probably unrelated)

 

I'm looking for any suggestions people might have. Here are my best guesses, but I'm completely stabbing in the dark:

 

- C2 or C11 is bad and causing the issue?

- I have a cold solder(s) on my recap job (I would give my recap a B-)?

- There's some kind of issue with the monitor's calibration?

 

I've included a video of the strange boot. Any help would be appreciated!

IMG_7242.mov

Share this post


Link to post
Share on other sites

EDIT page 175 has your issues

 

It does indeed! I had read that issue (here: http://www.biwa.ne.jp/~shamada/fullmac/repairEng.html#CheckerFlag)and foolishly assumed when it said "No startup bong" that it meant it didn't boot up at all.

 

Ok so, my next question...I've already pulled the ROM-SIMM out 2 or 3 times and done my best to clean with isopropyl alcohol. It all looks mighty clean to my eye. So any further advice on how to test if it's a poor ROM-SIMM seat?

 

One other tidbit that may or may not be relevant on this hunt: After having booted the mac 2 years ago (successfully), when I got it out to boot up recently on the first boot attempt it displayed Simasima, on the second boot it actually showed this same checkered pattern, but it did boot up. From then on it only showed Simasima until I recapped yesterday.

Share this post


Link to post
Share on other sites

Also I did try removing and cleaning/reseating the ram but I didn't try without the ram in. Should I do that?

 

How about C2 and C11 which I didn't replace. Any chance those could cause this issue? (Sorry for being such a noob at this, is there a schematic that would show what system C2/C11 are part of?

Share this post


Link to post
Share on other sites

Completely overlooking my answer. 

 

Let me repeat myself: 

 

Broken traces between VRAM, MUX, and UE8. Let me elaborate a bit: Check connections from VRAM to Muxes, and then from Muxes to UE8, as well as checking the address lines at the muxes from the rest of the system bus as well as the VRAM. You will find your issue. 

You will need to study the SE/30 video schematic and go trace hunting.

 

that IS your problem... 

Edited by techknight

Share this post


Link to post
Share on other sites

@techknight, sorry to overlook your answer - I read it and I was admittedly hopeful it wouldn't be that...I'm a bit timid of checking the traces, never done that - do I just use a multimeter to check for continuity? Might be above my pay grade but I'll give it a shot.

Share this post


Link to post
Share on other sites

I would look at the SE/30 schematics, Video section and start there. Yes continuity you will cleanly hear if there is a connection or if its broken. Thats the best way to go about this before swapping ICs. 

Share this post


Link to post
Share on other sites

Ok – again apologies for fumbling around, I'm trying to learn on my feet. So I've gotten a hold of the schematics for the "Video Interface" and I've been looking it over trying to understand exactly what you think I should check. Just to be sure I'm thinking of the right things, and that I'm looking at the ICs the right way, I've taken a picture the area I *think* you're suggesting has a problem. Can you confirm I'm looking at the right things (VRAM, and UE8)? Also, am I numbering the pins correctly?

 

 

post-8206-0-85253200-1485637593_thumb.png

Share this post


Link to post
Share on other sites

Id follow the schematic and meter everything out. Use a highlighter and mark off everything you have checked. 

 

Thats how I would go about it if I was not experienced in it. 

Share this post


Link to post
Share on other sites

Ok – again apologies for fumbling around, I'm trying to learn on my feet. So I've gotten a hold of the schematics for the "Video Interface" and I've been looking it over trying to understand exactly what you think I should check. Just to be sure I'm thinking of the right things, and that I'm looking at the ICs the right way, I've taken a picture the area I *think* you're suggesting has a problem. Can you confirm I'm looking at the right things (VRAM, and UE8)? Also, am I numbering the pins correctly?

 

Yes, these are the chips plus the muxes UA8 - UD8. Numbering scheme is correct.

 

You need to check all connections systematically. When I did it I built a little spreadsheet with the expected connections (chip & pin <-> chip & pin) and then checked all of them.

Share this post


Link to post
Share on other sites

Update on my progress:

 

I mapped out all the pins for UC7, UC6 (VRAM), UE8 and most of the other muxes using schematics and a spreadsheet. Then I checked continuity for all of those pins/traces, they all worked flawlessly. I found no bad traces at all. I was doubly careful to check around the C7 problem area and UE8.

 

Additionally, I went through each capacitor that I had soldered on, and followed their traces to their respective through holes or IC pins, to test if they had continuity to the capacitor solder – just to make sure I didn't have any bad connections. All good there too.

 

So my problem is still there. I'm out of ideas, so my next step is to replace C11 and C2 since those are still original. Does anyone have any idea if those could be related to the video system? Other than that I have no other ideas.

Share this post


Link to post
Share on other sites

Completely overlooking my answer.

 

Let me repeat myself:

 

Broken traces between VRAM, MUX, and UE8. Let me elaborate a bit: Check connections from VRAM to Muxes, and then from Muxes to UE8, as well as checking the address lines at the muxes from the rest of the system bus as well as the VRAM. You will find your issue.

 

You will need to study the SE/30 video schematic and go trace hunting.

 

that IS your problem...

Another update: I've replaced C2. Still the exact same pattern and considering how many traces I've Che's in the video video system im starting to think either there are bad ICs in there, or it's not the problem.

 

@techknight, you obviously know what you're doing, so pardon my ignorance, but how are you so convinced it isn't the rom-sim as described in the scrolls?

Share this post


Link to post
Share on other sites

The pattern looks like 8 pixels wrong, followed by 8 pixels correct. In other words, 1 byte each: even numbered bytes wrong, odd numbered bytes okay. If it were UE8, you would probably see the same bit(s) in each byte faulty.

 

The question is whether the problem is writing the data into the VRAM, or reading it out again. Try taking a screenshot when the machine boots up, and see whether the stripe pattern shows up in the picture.

 

If the stripe pattern (or something else distorted like duplicated columns) appears in the picture, the problem is writing data into VRAM, and you should particularly check UA8-UD8. If the screenshot is okay, it could be either side but I'd look more at the video scanning circuitry: UE6, UE7, UG6.

 

If I had to guess one most likely culprit, it would be UD8. Specifically, check continuity from UD8 pin 4 to UK6 pin 10, and from UD8 pin 12 to UK6 pin 9. These are address pins A0 and A1 on the bus. A bad connection on A0 specifically might produce the pattern you see. If the continuity is okay, replacing UD8 might be a good idea since sometimes the pins rot internally.

Share this post


Link to post
Share on other sites

 

@techknight, you obviously know what you're doing, so pardon my ignorance, but how are you so convinced it isn't the rom-sim as described in the scrolls?

 

Because scrolls was written when the machines were new. not 25+ years old with leaking capacitors,  battery corrosion, environmental effects, etc.. Plus a ROM SIMM would cause a no-boot symptom, and would have absolutely nothing to do with the video unless the video driver itself was corrupt, and if it was, it would fail the CRC32 and again result in no chime no boot! 

Share this post


Link to post
Share on other sites

Thanks for the great theory and reply apm. So following your advice I booted the mac up (interestingly the floppy does work intermittently, so that's a slight change from my original post). However, using the key command Shift-Command-3 to take the screenshot resulted in the system crashing (3 times). The cursor would display the watch icon, and the system (other than the mouse) became unresponsive indefinitely.

 

Does that fit with the UD8 theory? I don't have the hot air equipment or expertise to change out an IC (although I'm willing to learn). Would it be worth giving the board a good wash/toothbrush scrub with IPA and water in case its some hidden residue from the capacitor leakage shorting something out? Grasping at straws here haha.

Share this post


Link to post
Share on other sites

Because scrolls was written when the machines were new. not 25+ years old with leaking capacitors,  battery corrosion, environmental effects, etc.. Plus a ROM SIMM would cause a no-boot symptom, and would have absolutely nothing to do with the video unless the video driver itself was corrupt, and if it was, it would fail the CRC32 and again result in no chime no boot! 

 

Gotchya, that is helpful. This is the first time I've tried to resurrect an old 68k so I'm entirely new at this, but I'm enjoying learning. Appreciate all the help you've given so far (also ran into some of your videos on youtube - awesome stuff!). I'll be sure to pin 100% of the muxes, is there any way to test if an IC is bad other than using an oscilloscope? 

Share this post


Link to post
Share on other sites

Thanks for the great theory and reply apm. So following your advice I booted the mac up (interestingly the floppy does work intermittently, so that's a slight change from my original post). However, using the key command Shift-Command-3 to take the screenshot resulted in the system crashing (3 times). The cursor would display the watch icon, and the system (other than the mouse) became unresponsive indefinitely.

 

Does that fit with the UD8 theory? I don't have the hot air equipment or expertise to change out an IC (although I'm willing to learn). Would it be worth giving the board a good wash/toothbrush scrub with IPA and water in case its some hidden residue from the capacitor leakage shorting something out? Grasping at straws here haha.

 

Yeah, that does sound like a problem between the CPU and VRAM. Perhaps when it tries to read back the data in VRAM, it generates some kind of bus error because the address line is stuck in the wrong place. I'm not exactly sure of the mechanism there, but if the problem was on the video scan side, the CPU would be blissfully unaware of it and the screenshot would come out fine.

 

So I would check continuity on:

 

UD8 pin 4 to UK6 pin 10 and 68030 pin A2

UD8 pin 12 to UK6 pin 9 and 68030 pin C4

UD8 pin 7 to UC6 pin 17 and UC7 pin 17

UD8 pin 9 to UC6 pin 16 and UC6 pin 16

 

I also just noticed that the horizontal stripes in the video look to be 4 pixels high each. That's 4x512 bits = 2048 bits = 256 bytes per stripe. Probably not coincidentally, the address line that toggles every 256 bytes is A8, which also attaches to UD8. That makes a problem near/in UD8 even more likely.

 

So also check these:

 

Pin 14 should connect on UD8, UC8, UB8, UA8, connected to UE6 pin 13

Pin 2 should connect on UD8, UC8, UB8, UA8, connected to UE6 pin 8

[edit to add] Pin 1 and Pin 15 should connect to ground on UD8, UC8, UB8, UA8

 

It is possible, I think, to get the SMD chips off without hot air, if you patiently heat and pry up the legs one by one. Others here might have good advice on what works.

Edited by apm

Share this post


Link to post
Share on other sites

Now, I wonder if there was a known bad ROM that came from apple that was compiled wrong or something, or had bugs in it and they swapped it out with a newer version. I cant recall ever running into that though. 

Share this post


Link to post
Share on other sites

Man. I'm blown away you guys know this much about these machines. Incredible. I'm out of the country for a few days but illl post an update as soon as I'm able to check those pins.

 

@techknight, I don't have another se/30 to swap rom-sim but I do have a working SE. That wouldn't work would it?

Share this post


Link to post
Share on other sites

techknight and apm – I owe you both beers. She's working!!!!!!!!

 

apm, I'm blown away by the specificity of your diagnosis. You NAILED it. I pinned out the entire UD8 chip, and sure enough, UD8/4 had no continuity with UK6/10. I didn't have any really fine gauge wire, but I patched it with what I had, and holy crap – it works. It freaking works. That was a blast...

post-8206-0-33905600-1487739677_thumb.jpg

post-8206-0-24950300-1487739690_thumb.png

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×