Jump to content

Reverse Engineering the Macintosh SE PCB & Custom Chips for 1:1 reproduction


Recommended Posts

@cheesestraws @PowerMac_G4 - I HAVE DISCOVERED A TRACE ISSUE!

 

Look here - you'll see i've connected stuff that has no business being connected. I accidentally pistook a trace going into a pad - turns out that i'd actually shoved the trace out the way while desoldering. I just verified against other SE boards - see the original, and the fix. Can cut the trace and patch direct with a bodge wire :)

Annotation 2020-08-15 165223.jpg

DSC_0021.JPG

Link to post
Share on other sites
  • Replies 448
  • Created
  • Last Reply

Top Posters In This Topic

Posted (edited)

....aaand there's a short circuit between the power planes in all the board revisions. CANCEL YOUR BUILDS, DESOLDER. I'll send out another batch of boards :(

 

FFS - its only on the split power plane at the back - can't locate it...c8 links to gnd and power, and even without a capacitor in, the planes are linked. 

Edited by Kai Robinson
Link to post
Share on other sites

Nice!

 

What did your plan for the PSU issues turn out to be?  Did you end up fixing your existing PSU, get a replacement one, or some other magic didn't require changes there?

 

Also, did you hear a reasonably sane startup chime?

Edited by quorten
replacement
Link to post
Share on other sites

Judging by the troubleshooting info around here at 68kMLA, that's definitely a logic board issue!  Tongue-in-cheek, check the address multiplexers, check the ROM, try with different RAM or different amounts of RAM, etc.

Edited by quorten
Troubleshooting joke
Link to post
Share on other sites

Cool, really glad to know that TechKnight was right.  I was also thinking similar based off of anecdotal evidence about what happened after I jammed a floppy disk into a Macintosh SE upside down and backwards over 20 years ago, but I thought it would sound too silly to try to mention that and say something similar could prevent the PSU from starting up.

Link to post
Share on other sites
10 hours ago, Kai Robinson said:

Second batch of prototypes have now been put up for order - hopefully they split the rear i/o power plane properly. 

I literally just logged in to let you know I'd found a few faults in your design and here you are two steps ahead of me!

I've not even got a donor SE on my bench yet so no rush...

 

p.s. I'd love to see some red PCBs some day. I love red PCBs and Apple prototypes all have red boards IIRC.

Link to post
Share on other sites
On 8/15/2020 at 2:36 PM, quorten said:

jammed a floppy disk into a Macintosh SE upside down and backwards over 20 years ago

I feel less alone for having read this sentence.

 

On 8/15/2020 at 10:22 AM, Kai Robinson said:

CANCEL YOUR BUILDS, DESOLDER. I'll send out another batch of boards

I'm still waiting for my 68pf capacitors to arrive, for some reason I do not understand, soooooo

Link to post
Share on other sites

Considering the point that all moderator decisions are final, I'd probably defer to an admin/moderator to fully answer the question (as it seems you're also assuming), but looks like there were no admins/moderators who were eager to add themselves as followers directly to this thread.  You could @-mention or PM Cory5412 and see what he thinks.  But, from my read of the rules, it doesn't seem like there's any restriction against this, just "don't be a spambot to generate ads revenue."

Link to post
Share on other sites

Oh the joys of PCB design. 

 

I dont know exactly what CAD package your using, but I always run DRC and ERC checks before I render out gerber files. it will tell you right away (as long as you have a valid annotated schematic and or netlist) if there are trace issues, or overlaps, shorts, etc. 

Link to post
Share on other sites

Sprint has a basic DRC function - it came in handy for moving a few traces out of the way - some were drawn a little too close, although JLCPCB's tolerances are pretty good for even small stuff. Nothing is now unconnected - i thought rather than mess with more bodge wires, the first prototype has done it's job and shown me that there is light at the end of the tunnel - hell, when doing the A500++ boards, theres a reason the production boards are v2.1 :p

Link to post
Share on other sites

yea the power planes could be shorted together at multiple Vias. only way to know is run a ton of current across the two planes and figure out which vias get hot. Drill them out. but thats the "not so fun but fun" solution. haha. 

 

Biggest mistake I ever made was designing one of my boards and I accidently crossed the 5V and 3.3V VCC planes in the middle layer of the board. and didnt catch it. Whoops. 

 

by the time I realized my mistake and why CPUs had such a short life time, it was too late obviously. So I got myself familiar with the Dremel and dug into the inner plane of the PCB and broke the short. 

Link to post
Share on other sites

Ideally for that to work, i'd need a FLIR camera - fingers crossed i won't need it this time. Using the pin test function in Sprint i've verified that everything that should be on ground, goes to it, everything that goes to vcc goes to it without a short to ground. 

Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...

×
×
  • Create New...