• Please review our updated Terms and Rules here

Apple III Troubleshooting

sev

Experienced Member
Joined
Mar 30, 2014
Messages
132
Location
Los Angeles
So I picked up an Apple III today, serial number is in the 30k's. When I tested it at the buyer's house, when I first turned it on, the picture was jumbled (due to bad NTSC signal), so i Re-seated the connector and it booted up to the memory diagnostic, but I didnt let it run all the way and shut it down.

So I brought it home and booted it, and now during boot, and pretty much always, the screen is filled with P's from top to bottom. The drive tries to seek at boot, but all the test is garbled on the screen. I'll post a youtube video showing what I mean.

I took it apart and pushed down gently on every chip, some of the memory chips were a tiny bit loose, but not enough to lose contact. I tried booting it, but no difference.

I went ahead and pulled the ram board, which appears to be a 12v ram board, I think with 256k. I removed each ram chip, cleaned a few and replaced each one. I noticed there were several different manufacturer of ram chips, the ones that were from ITT had very clean pins. One was from Motorola and that was clean too. The chips from TI had this dark black corrosion on them, almost as if they had gotten too hot. I cleaned them as best as possible and replace. The system still does the same thing

Also to note, the power supply has a faint sound, not sure if this is normal.


I tried putting it into diagnostic mode but it just freezes.

i'd love to get this thing going... I'm very new to the AIII... Any help would be greatly appreciated!!

https://youtu.be/kjEG8cVpSOE
 
Last edited:
It looks like RAM. Earlier Apple III logic boards were really bad. Could have bad traces due to corrosion or the like. Also, the solder joints on the RAM sockets on the logic board might need some re-work. I say RAM because it appears the ROMs are working, and the disk drive looks like its controller are at least functioning. This may sound like I'm joking, but I'd actually try lifting the unit about a inch or three, let it drop to the table, then power up. This allows the chips to re-seat themselves.
 
If it were me, I'd rather try reseating the chips manually, one at a time.

It would seem to me like a bad connection to a whole bank of RAM, or glue.
 
If it were me, I'd rather try reseating the chips manually, one at a time.

It would seem to me like a bad connection to a whole bank of RAM, or glue.



I removed and re-seated every ram chip, also pushed down on every chip on the motherboard
 
I've had chips that were all the way down and not making good contact with the socket. I'd probably try prying up and pushing down the rest of the chips.
 
Pushing down can work under the right circumstances, but to better ensure contact, Wayne Stewart's suggestion is best.

I don't know much about the ///. It would need to have one RAM chip cover the whole 1k screen contents, in order for this to be specifically a RAM chip problem. Otherwise it's a glue chip, or a shorted trace somewhere in the data bus. In fact, I doubt it's a bad connection at all. It's either a short due to foreign objects or misaligned card edge connectors, or internal to a RAM chip (if the above is true) or a glue chip.
 
just to rule out the ram, can i kick it down to 128k and try to use the remainder of chips as a "spare"?


I'll try removing and replacing every chip


Another concern i have is the power supply. it's making a bit of a chirping noise. Wondering if it's about to give up the ghost. I fear power cycling many times might blow up the filter caps.... hmmm
 
First of all your primary RAM bank is working or it would not boot at all. The primary bank contains the video RAM and the stack RAM ... no stack no boot.

Having deduced your primary RAM bank is functional then iis a video issue. Specifically the character generator is being fed bad data bits.

Looking at the schematic the Apple III uses two 2114 static RAM chips as a programmable character generator, their data inputs are fed by two 74ls374's (octal latch) without looking deeper into the circuit I assume one of them is used to program the RAM with the character data and the other by is fed by the video RAM to display what is in video RAM.

I'd be looking closely at those 374's at E2 and F2

Edit: If the supply is chirping then yes, recap the supply. I took a look at the supply schematic and compared to typical power supplies of the day its quite different in its approach. They derive the reference voltage from a -5v regulator and regulate the +12v supply rail instead of the +5v. The +5v rail just follows the +12v which I find strange and disconcerting. Normally you regulate the +5v rail because it is the most voltage sensitive. With this design a heavy load on the 12v rail will drive the +5v right up and with the over voltage protection only triggering off +12v effectively leaves the logic chips unprotected.
 
Last edited:
First of all your primary RAM bank is working or it would not boot at all. The primary bank contains the video RAM and the stack RAM ... no stack no boot.

Having deduced your primary RAM bank is functional then iis a video issue. Specifically the character generator is being fed bad data bits.

Looking at the schematic the Apple III uses two 2114 static RAM chips as a programmable character generator, their data inputs are fed by two 74ls374's (octal latch) without looking deeper into the circuit I assume one of them is used to program the RAM with the character data and the other by is fed by the video RAM to display what is in video RAM.

I'd be looking closely at those 374's at E2 and F2

Edit: If the supply is chirping then yes, recap the supply. I took a look at the supply schematic and compared to typical power supplies of the day its quite different in its approach. They derive the reference voltage from a -5v regulator and regulate the +12v supply rail instead of the +5v. The +5v rail just follows the +12v which I find strange and disconcerting. Normally you regulate the +5v rail because it is the most voltage sensitive. With this design a heavy load on the 12v rail will drive the +5v right up and with the over voltage protection only triggering off +12v effectively leaves the logic chips unprotected.



Hi David, I discovered a neat little feature on my TL866 eeprom programer, it tests TTL Logic chips!! I found the schematic you were referring to and pulled all but 2 of the TTL chips and tested them. All tested good. The tester goes not test the ls151 and ls 399 chip. I'm wondering if the the rom/asics might be bad? either 341-032 or maybe 342-0030. Any possible way to check? Are these just rom chips or actual asics? I wish there were other ls151 or 399's on the board i could swap, but there doesnt seem to be :(

I really want to get this thing working. any suggestions would be much appreciated!!
 
Last edited:
Hi David, I discovered a neat little feature on my TL866 eeprom programer, it tests TTL Logic chips!! I found the schematic you were referring to and pulled all but 2 of the TTL chips and tested them. All tested good. The tester goes not test the ls151 and ls 399 chip. I'm wondering if the the rom/asics might be bad? either 341-032 or maybe 342-0030. Any possible way to check? Are these just rom chips or actual asics? I wish there were other ls151 or 399's on the board i could swap, but there doesnt seem to be :(

I really want to get this thing working. any suggestions would be much appreciated!!

Sorry for the delay, I only just saw this message. They are both ROM's and appear to be 2716 eprom compatible. You should be able to read them in your eprom programmer as a 2716 eprom, just dont try to write to it.
 
I wanted to revive this thread. Still trying to get my A /// working. I think I have narrowed down the issue to the character Srams. I have since acquired a donor apple 3 board (that has broken bus bars) but was sold to me as previously working. When I turn the machine on, it does boot, seek the floppy drive and seems to work, except for text output. It appears to seek the floppy drive and fail and write 'retry' except its garbled.

i've pretty much narrowed down the issue to what i think is a bad trace because I've replaced most every socket in the video logic circuitry and checked continuity. Yesterday I found that the SRAM on E5 was not getting power so I put a bodge wire from the VCC pin on E4 to E5 but the issue still persists. However, before the motherboard was sensitive to touch and characters would jump around if you touched the board while it was on, which went away after my repair. Now the screen is showing a bunch of X's from top to bottom, with garbage where the word 'retry?' appears

I checked the power supply voltage and at least on the 12v+ line, im seeing 11.65v.

I've also checked for continuity on several parts of the video logic circuit. Specifically all the connections between F3 to E3, Between E5 and E4, (the two srams), and also the two sram's to F4 and E3. All were good, except.

I feel like i'm getting closer to resolving this but still not there. I also swapped out the 3410-030 rom chip with the one from the donor board, no difference. Also swapped the Ls20 chip in B10, still no difference.

Any suggestions or help would be super appreciated!
 
i also just put the 341-0030 rom inside my reader and verified it against a bin file of the same chip and it passed. So the rom is okay there.
 
FINALLY!!!!!!!!! well over 100 hours and 3 years of trying to fix this thing and i got it! I went as far as to resurrect a board with dead bus bars, replacing them with wire and getting that board to work 95% of the way.

I was checking continuity on all address and data lines on the resurrected donor board and found nothing. Did the same on the original board and everything checked out right until the end. I was checking lines DV0-DV7. I got to DV4 which checked out. The probe accidentally slipped at hit pin 19 on e2 and it beeped. I checked again. Beeeeep!

turns out there was a short between pins 4 and pins 15 on the 2114 srams. I verified against the donor board and it didn’t have the same issue. It looked like DV4 and DV6 were shorter together.

After removing the sockets for e2, f2, e3, f4 and both 2114s, i found a whisker shorting two adjacent pins. Cleaned the through holes, soldered sockets back and the first pic was from the first power on!!!

To get to the point of finding the fault, I looked over everything, including timing, ram, ram connectors, unsoldering/resoldering, checking the RAS/CAS lines, changing every single chip with a donor board and seeing no difference. Checking the roms.




Here's some notes from all the lines/busses/timing I checked:


DA, DB, DC, DX, VA, C1M, C1M*, OE374, BT

board 1 and 2 checked for H* (h0-h5 and a few others)
board 1 and 2 checked v* (all V lines)
board 1 and 2 checked bksw1-3 and abk1-4
Board 1 and 2 checked A from 0-15 and a few others if you type A* in the search
Board 1 and 2 checked for AY*
board 1 and 2 checked for D0-7


If anyone ever comes across this thread and needs help resurrecting their apple ///, reach out to me via pm. Ive spent so much time on this and learned so much, i'm happy to help.

One other interesting item to note. I used a tool written by a guy on the Apple /// Enthusiasts group on facebook to help find traces from the schematics. It's a python script and was instrumental for me figuring out what was wrong, you can find it here:

https://github.com/ppieczul/apple-i...xvS06C1M3dP7oPFAz4-Es-KjXA-nlmh9MuYAwkr9mov-o

you will need to install python 3 and a few libaries and it has a bit of a learning curve, but its SUPER handy to use.
 
Last edited:
So now I got the system back together and I'm facing another issue. I can boot the confidence disk and the diagnostic disk just fine but if i boot anything that has SOS on it, it crashes. At first I thought it might be the disk drive so I tried loading the ADTPro Boot disk through ADTPRO over serial and as soon as it begins loading SOS, it crashes and locks up. usually with the drive stuck spinning.

Any ideas?
 
So we're all set and my apple 3 is finally FULLY Working! The boot issue was a pin that was accidentally bent in on the LS133 on J7. I discovered it removing every single chip to test in my tl-866. The ls133 didnt pass and upon closer inspection I found the pin bent in, put it back and it came right up.

3 years and countless hours.... finally I'm done!
 
@sev I'm located in North County (San Diego) and recently picked up an Apple /// that's having some issues similar to yours. Any possibility you'd be willing to take a look at mine if I brought it up to LA? Thanks in advance :)
 
Back
Top