• Updated 2023-07-12: Hello, Guest! Welcome back, and be sure to check out this follow-up post about our outage a week or so ago.

SE/30 Woes

SlateBlue

Well-known member
I recapped my SE/30, and the repair failed to fix it. Prior to recap, the Mac had horizontal bars from top to bottom and no chime.

The attached photo shows the screen after recap was completed, and still no chime. I checked to make sure RAM was installed properly. Also, if I remove the RAM, the horizontal bar pattern returns. Any help would be greatly appreciated.

I apologize if the pic is oriented incorrectly. The uploader does this automatically and I'm not sure how to rotate the photo.

image.jpeg

 

bibilit

Well-known member
Did you washed the board before the recap ?

can you post some pictures of the Logic Board ?

 

aplmak

Well-known member
It's like a crazy version of "shuffle puck".... lol!!!! But bibilit is right... we need a pic of the board.. a good pic... and also it should be washed... I've said this numerous times.. in the bathtub with a toothbrush to get in and under the chips with 90% rubbing alcohol... and rinse it down with it too to wash away the scrubbed up residue...  then shake it a little.. and put it in front of a small heater with fan for a little bit to dry it up... or hair dryer... alcohol dries quickly...

 

aplmak

Well-known member
Check the RAM... make sure you have it installed in pairs I believe... and also make sure you have the ROM installed and seated properly...

 

SlateBlue

Well-known member
Yes, I did wash the board... In a large tub...With Simple Green... And a toothbrush. ???? I will post pic of board tonight, but I fear I'll be attempting to repair broken traces.

 

SlateBlue

Well-known member
Here are a few pics of the logic board with close-ups of what I think are trouble areas. Let me know if additional photos are needed. And, if you decide to criticize my cap job, please do so constructively. :)

IMG_8858.JPG

IMG_8857.JPG

IMG_8856.JPG

IMG_8852.JPG

IMG_8853.JPG

IMG_8849.JPG

 

EagleTG

Active member
Can you post a pic of the back side of the board, too?  Something overall like you did of the front?

Thanks!

 

Elfen

Well-known member
This is my 3rd attempt to post here. Damn it, every time I do this damn PC Crashes! GRRRR!!!!!!
 

Check the RAM... make sure you have it installed in pairs I believe... and also make sure you have the ROM installed and seated properly...
Pairs? No. Quads - 2 banks of 4; the SE\30 is a 32Bit machine, requiring 4 - 8 bit 32 pin SIMMs.

Yes, I did wash the board... In a large tub...With Simple Green... And a toothbrush. I will post pic of board tonight, but I fear I'll be attempting to repair broken traces.
Anytime an SE\30 turns on with Bars, its broken traces. You may need to remove the caps and check the traces and vias that goes under them. Also, did you clean the cap goo that is under the caps after removing the caps; preferably with a Q-Tip Swab and acetone? No matter how clean you get the board, there is still crap under the caps!

Check out the posts made by genie_mac and Cloaked Alien on their SE\30 repairs; Cloaked Alien has a few You Tube Videos of him fixing a "more than dead!" SE\30 board. It is hilarious as it is educational!

I see you are using the black Tantalum Caps. For me they make the board look a lot better. The Yellow Ones, to me, cheapens the look of the board for some reason.

 
Last edited by a moderator:

Elfen

Well-known member
Here are a few pics of the logic board with close-ups of what I think are trouble areas. Let me know if additional photos are needed. And, if you decide to criticize my cap job, please do so constructively. :)

/monthly_08_2016/post-6277-0-35285200-1470967654.jpg">View attachment 9755

*Raises finger* Uhmmm.... those RAM SIMMs with 2 or 4 chips on them are Iffy on the Mac II series. They may or may not work, depending on the machine. It may even work on one SE\30 but not on another. It is strange. Better to get the RAM SIMMs with 8 chips on them. Speed is another issues, though I can not see it from here, using 100ns to 70ns RAM SIMMs are best for the Mac II Series. 120ns is too slow and 60ns is too fast and both may not work on the Mac II Series.

Sometimes those 2 or 4 chip RAM SIMMs work when you have a set of 8 chip in Bank A and the others in Bank B.

But I still say you need to chase down broken traces and vias first.

 

SlateBlue

Well-known member
I don't have any other 30-pin SIMMs to test. This was the RAM that was installed when I received the Mac. I don't wanna go dropping money for RAM unless I know it's the cause of the issue. I'll upload a photo of the back of the board sometime tomorrow. Thanks to all of you for your responses.

 

aplmak

Well-known member
Elfen that is right... my bad... you have to fill bank A completely or bank A & B completely... I haven't cracked open my SE/30's in a while... still on powerbooks right now...

 

techknight

Well-known member
Well, I now have this board. 

This thing is going to be a tough dog, I got a feeling. Did some continuity checks on the address and databus to the ROM, and its good. 

It walks the bus properly without the ROM, but without the data side of the bus toggling, no way to know whats going on there. 

So with the ROM inserted back in, this thing has a serious case of bus contention on a few of the lines. 

So there is an IC that has suffered a catastrophic failure somewhere, but god only knows what IC and where... The saga continues. 

20160827_115559.jpg

Bus contention is when a Driver is trying to drive the bus high (5V) while another chip is trying to drive it low at the same exact time (0v), So you get something somewhere in the middle. 2.5V roughly. 

But the weird thing is, From what I see in the scope shot, is the bus is stable at certain points, and goes into contention only so often. That leads me to believe that another IC is activating at the wrong time, almost as if its being called on when in reality, its not. 

If one of the data pins was stuck low, or stuck high, it would go into contention on every rise or every fall. But it isnt. This is a strange one indeed. IC #1 is calling the bus to go low, while another IC is activated and still trying to drive it high, or vice versa. 

It almost makes me think that there is a failure in the addressing/chip select bus. A chip is being called upon when another chip is still active. Also the CoPro is getting quite warm, and I dont recall those are supposed to get warm idle. 

 
Last edited by a moderator:

techknight

Well-known member
With some probing, D30, and D31 was broken between SCSI/SCC/ASC, and SWIM. This is surprisingly common becuase the bus junction/interconnect sits right below those 3 caps near the power molex plug. 

But, not necessarily a show-stopper as far as booting and video issues. Sure you get no SCSI, Serial, or Sound. but itll still boot. Or, attempt to. 

And there was the fact of bus contention. So I sat back and started thinking, the contention was happening randomly and sequentially. That rules out a hard failure because the contention would be there all the time and wouldnt go away. Even with or without ROM. 

So that rules out a failed IC, So I was leaning more towards chip select/address bus because of the chip is talking, but at the wrong time. 

Sure enough, A0 has a double break between ASC and GLU/CPU, and also to video muxes. A1 is broken between ASC and GLU. A0/A1 is used in the 030 for upper/lower addressing and bus sizing. VERY important! (24/32 bit addressing, and the 16-Bit UDS/LDS for the peripheral ICs.)

This allows the ASC addressing to float around. So it thinks its being addressed when its not. and presto, Contention. Plus, with A0 and A1 being missing on the video muxes, that explains the 24 pixels of jail bars, and 8 pixels of valid video data. 2 address wires can count to 3. :) , and if you break down the video image in the posted picture above, you got 8 pixels of good data, and 24 pixels of bad data. Break it down into 8 pixel groups, well the count of 3 gets 24. That explains that. 

What was throwing me off was I had open address lines before on the ASC with boards in the past, and everything was working fine. But the A0/A1 line is very important, and those are both open. First time I encountered that. 

Now, to perform the repairs and hope it doesnt make an ass out of me. lol. 

 
Last edited by a moderator:

techknight

Well-known member
Ahh Life.... Aint it the greatest? 

Nothing better than hearing a chime and having a screen on an otherwise brink of dead board. I dont know why I find that so satisfying? Maybe its just the nerd inside of me. :-S

20160827_154319.jpg

20160827_154400.jpg

 

joethezombie

Well-known member
Excellent work as always.  Yet another techknight diagnostic post downloaded to my offline archives.  I'm going to print them all out and make a book called the "techknight restoration chronicles".

 

just.in.time

Well-known member
Wow techknight, that is some seriously impressive diagnostic work.  Good job on getting the board back to life.

Edited to add, I have an emergency startup disk with the exact same folders:  System and Lido.  Lido is extremely handy to have around.

 
Last edited by a moderator:
Top