• 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.

Diagnosing a Classic Mac in a weird situation (Early Analog Board)

woodapples

Active member
Hello folks! So i've gotten a Mac Classic with a dead logic board and im trying to replicate a project in which a raspberry pi 4 drives its CRT. Also when i plugged it in, i forgot that i have 220V system and that i imported a North American variant of the Classic... so it kinda went kaboom.. turned on for a few moments then died and never turned on again.. At that moment i figured i might as well replace the PSU.. but to my horror it was apparently built in to the analog board. I looked into the schematics and i figured out a way to make it work somehow (with the help of this wonderful thread).

Now my problem is that, i have verified my vsync and hsync timings to an old SE analog board before. but with this mac classic i have "repaired", im getting this kind of video:

29388DF3-9357-4412-AA64-E5DB3F97A9D8_1_102_o.jpeg

ive tried to invert the hsync/vsync polarity to no avail. I hope that analog board veterans here can help me in making this almost hopeless case of a mac work again.
 

woodapples

Active member
okay, i guess i didnt properly inverted the sync signals, but when i do, im just hearing a buzzing sound and the 5V regs i have are shorting out for some reason (probably from overvoltage/shorts/overcurrent).

Also seeing this picture when i disconnect the video and vsync pin:
IMG_4881.jpeg
 
Last edited:

woodapples

Active member
So after removing all components with *P* in its designator/anything that has to do with the old power supply circuit (except RP39 and CP31 coz it's for the neck board's video signal), it seems to be working now.. albeit with a shimmering video signal though.


IMG_4883.jpegIMG_4884.jpeg
 

mg.man

Well-known member
NO idea if it will help... but a little while back I modified a 220V Classic II Analog / PSU board for US voltage (I was restoring one for a buddy of mine in the US). I found a really, really old (1995!) Usenet post that detailed what needed changing :

Date: Thu, 2 Feb 95 19:09:26 GMT
From: tjfs@tadpole.co.uk (Tim Steele)
Subject: Classic Power Mods

Some tips on modifying Mac Classic Analog Boards!
-------------------------------------------------

If you want to modify a Mac Classic from 110v to 240v (or vice versa) use
these tables:

Late Revision Classic Analog Board (JP2 marked on PCB across CP13 location)

110v 240v
JP1 Wire Link -
JP2 - Wire Link
CP1 220uF 250V 220uF 400V
CP13 220uF 250V -
CP19 4.7nF 250V 10nF 250V <-- Must be Class Y Rated
CP38 Wire Link 10nF 250V <-- Must be Class Y Rated

Early Revision Classic Analog Board (JP2 not marked across CP13)

110v 240v
JP1 Wire Link -
CP1 220uF 250V 220uF 400V
CP13 220uF 250V 220uF 400V
CP19 4.7nF 250V 10nF 250V <-- Must be Class Y Rated
CP38 Wire Link 10nF 250V <-- Must be Class Y Rated
 

woodapples

Active member
Thanks @mg.man for that info! Right now i think i have completed the "transition" to my own voltage regulators :) I've eliminated practically everything that related to the old power supply circuits and i bypassed 12v, 12v to the neck board, 5v and 29V to the flyback.

Though im smelling something bad while it's running.. im not sure what it is though. so be warned if you want to try this out. The SE board is quite more safer and reliable for me to deal with it seems.
 

woodapples

Active member
Ooo... never mind - I see you're in the Philippines! Shipping would be an issue. :unsure: Good luck!
I have a mail forwarder in California so it should be no problem :D


If it smells like dead fish... it's old cap goo you've not completely cleaned off the 'board! :ROFLMAO:
God i hope it's that. Does the MOSFET in the flyback really gets quite warm? The smell has a hint of something burning. Not sure if it smells like dead fish though lol
 

cheesestraws

Well-known member
The SE board is quite more safer and reliable for me to deal with it seems.

The SE AB is much better built than the classic one.

albeit with a shimmering video signal though.

If you look at your second picture, note that the offsets of the lines are whole pixel widths. That suggests to me that this is a digital rather than analogue issue: have you checked that there isn't jitter on the sync pulses relative to the pixel stream?
 

woodapples

Active member
The SE AB is much better built than the classic one.



If you look at your second picture, note that the offsets of the lines are whole pixel widths. That suggests to me that this is a digital rather than analogue issue: have you checked that there isn't jitter on the sync pulses relative to the pixel stream?
yeah, im trying to adjust the front/back porches but these settings worked perfectly on my SE before it bit the dust, will have to try stuff out.

here's the relevant part of the config.txt of the raspberry pi in case it helps:

```
dtoverlay=vc4-kms-dpi-generic
dtparam=hactive=512,hsync-invert
dtparam=hfp=12,hsync=178,hbp=2
dtparam=vactive=342,vsync-invert
dtparam=vfp=1,vsync=4,vbp=24
dtparam=clock-frequency=15667200,rgb888

```

Right now im just concerned about the smell... Tis really quite concerning smell at that
 

cheesestraws

Well-known member
Just fiddling blindly with the front and back porches probably won't help very much. You will probably want to look at the actual signals being sent to the AB on an oscilloscope, I expect.
 

cheesestraws

Well-known member
That is probably what I'd do. Sorry, I'm ... not quite out of my depth here, but I'm not really in my depth either. So please bear this in mind when reading my advice...
 

woodapples

Active member
Just FYI, i was able to do 3d acceleration on this setup + SE CRT&AB.. kinda useless but at least it's less burden on the Pi4's CPU :)
 

woodapples

Active member
So yeah, when i try to display a full picture, the area around the MOSFET starts to smell quite bad... and only smells bad when i show a almost-fully white image. And it has a very bad shimmering effect as well.. not sure what to do next honestly

Also tried to adjust the width coil thingy but it barely changes the width...

View attachment IMG_4889(1).mov
 
Last edited:

woodapples

Active member
After some cleaning and removing some gunk, i've revived the LB with alternative power regulators, but unfortunately im still getting the same herringbone kind of pattern, the same as im getting with the Pi4, as well as the really wide output. Any ideas folks?

View attachment IMG_4892-2.movB5B0721F-408F-4A65-BEBD-E621D343938A_1_102_o.jpeg7C9C6705-6100-445C-BE7E-6CB076CC7DA4_1_102_o.jpeg
 

woodapples

Active member
Okay... Kinda feeling like an idiot now... So there's probably nothing wrong with the A/B.. I adjusted the HOT voltage, the BOMARC schematics suggested +31V but when i turned the voltage down the bad smell seems to have gone away and the horizontal width has reduced. There's still some noise but i suppose i can try making the data signals with a twisted wire or shorter length
 

woodapples

Active member
Okay, after some tweaking, it seems that the LM317 and its adjacent passives are important for limiting the current of the 31v rail. Now it's fixed, hopefully.
 

mogs

Active member
Cool project, curious to know how you've connected stuff around the LM317?

I think maybe in my final build I was injecting power into the lm317 input.
I'd have to check the schematic but from memory it has soft start circuitry on the adj pin.

I haven't played with my one since I ran out of working classic/classic II logic boards (also I got a Mac plus working).
 
Top