Hi all, I'm working on repairing a Lisa 2/10. The machine appears partially alive but does not produce any usable video output besides a raster and has the clicking speaker. I've removed the floppy and widget drive for testing.
I have already did a recap on the CPU and I/O board and have had no change in behavior. No damage to the boards as this does not have the battery pack.
Also plugged in an Apple /// monitor and it behaves basically the same as the internal one.
Looking for some advice as I'm wondering if this is video RAM related, I did reseat them but no difference.
Current symptoms
• CRT powers up and produces a stable raster
• Screen shows horizontal bands/interference but no text or graphics
• No startup message or errors on the screen
• Constant click when powered on
• Behavior is unchanged even with RAM removed, clicking stops when I/O board removed
Troubleshooting done so far :)
CPU
Reset line tested with oscilloscope:
Goes high at power-up and remains stable
Drops correctly when reset button is pressed
Clock signals appear present and active
ROM
Both ROM chips removed and verified using programmer
ROMs match known good Lisa ROM
Also reseated the video rom, when removed raster went away but speaker clicking continued
Progress update, after using some deoxit on the video board trim pots I was now able to get a stable image. Image is below but basically only a capital H but still the constant speaker clicking. I believe I should see the self test at this point but I don't get that far. I do at this point have a keyboard, mouse and floppy disk connected.
Congratulations! That's progress. So the machine powers on and begins executing its boot ROM, but freezes before the power-on self test and its CPU board icon appears...
The good news is that the power supply and the video board seem to be working fine. On that front, you will likely need to replace the RIFA film capacitors in the power supply if they are original, as they are likely to release the magic smoke at any time now.
As a next step, I would try ruling out any issues with the RAM boards. Try pulling them and testing them one at a time, each board in both memory slots. If the POST is stalling out due to a RAM issue, having one working RAM board in a working slot might enable you to make further progress during POST.
Quote from: pugy365 on March 08, 2026, 09:11:33 AMI was now able to get a stable image.
...
Image is below but basically only a capital H but still the constant speaker clicking.
Nice progress!
AFAIK, this is not a common failure mode. The usual constant clicking occurs when the PSU over-voltage crowbar is triggering constantly, but you don't get video in that case, so probably not your situation. (You could watch the 5V rail to be certain.)
Please be more specific as to the timing/context of "constant clicking" ...
eg. Is the video static at that screen with the H in the menubar and the speaker clicks every half-second?
or
Does the screen re-draw in some fashion, draw a blank menubar, click, add the H, wait 10 seconds, then repeat?
The SRAM on the CPU board is part of the MMU, a portion of RAM on one of the memory boards is used for the video bitmap.
Thanks All! To confirm behavior clicking multiple times a second, so very fast like click,click,click one after the next. Screen is static from what I can tell, I don't see any obvious redraw just get stuck with the H but clicking continues. Note that clicking stops for maybe a little longer than a second if I press reset but then as it boots it come back. I will check my 5v rail again, it seemed ok but possible issues as I haven't recapped that yet. Yes along with the RIFAs been waiting on the smoke haha. Now that I have some life those components are in the mail along with an ESProfile hard drive simulator.
Also I just tested with using either RAM cards in both slots and behavior is the same. I also did test with no RAM card installed and I get the clicking speaker but no video if that helps at all.
Quote from: pugy365 on March 08, 2026, 07:31:33 PMclicking multiple times a second, so very fast like click,click,click one after the next. Screen is static from what I can tell, I don't see any obvious redraw just get stuck with the H but clicking continues.
Looking at the Boot ROM listing, at $0B24, just after the click procedure has started, the ROM checks the I/O Board ROM version so it can adjust the output frequency according to what kind of I/O board it is.
At this point it hasn't fully loaded the VIA hardware for the click, but it may have done enough to cause a noise to be generated. However, if the I/O board doesn't respond when the 68K tries to check the version, a bus error will be generated. Normally such a bus error is expected and handled, rather than causing a loop, so if this is what is happening, I'd guess that the problem is actually further up the chain on the CPU board.
So, I suggest: check the bus timeout signal on the CPU board (the 556 at U15B-9) to see if it is toggling when the clicks occur.
Ok so I attempted to probe it, but have been doing most of my probing with the I/O board removed so I can get to the back of the CPU board. Noticed something new, with the IO board I get the beginning of the self-test procedure and image on the screen (but no clicking of the speaker) and error 41. I understand this error is I/O board related, since it's removed. But curious I get this with it removed an no farther than H with it installed. It also appears to be constantly doing a redraw, made a GIF to show here. Also pin 9 is constantly HIGH no toggle. I think I may be able to get to it from the top with the cover removed.
Quote from: pugy365 on March 08, 2026, 09:41:10 PMOk so I attempted to probe it, but have been doing most of my probing with the I/O board removed so I can get to the back of the CPU board. Noticed something new, with the IO board I get the beginning of the self-test procedure and image on the screen (but no clicking of the speaker) and error 41. I understand this error is I/O board related, since it's removed. But curious I get this with it removed an no farther than H with it installed. It also appears to be constantly doing a redraw, made a GIF to show here. Also pin 9 is constantly HIGH no toggle. I think I may be able to get to it from the top with the cover removed.
That constantly redrawing error 41 is the exact behavior that you should expect to see from a Lisa that's missing its I/O board. So don't worry too much about that.
As @sigma7 said, the issue with the I/O board is likely something to do with a bus error when trying to communicate with the floppy disk controller shared RAM. The Lisa does this when clicking the speaker to figure out the system type and adjust the frequency accordingly, but something seems to be going wrong during these communications.
My first guess is that the floppy controller simply isn't generating DTACK. And if not that, then DTACK somehow isn't finding its way back to the CPU.
Thanks thats good to know! I wondered if this helped narrow it down to more likely being an I/O board issue versus a CPU board issue. I am going to try to probe the 556 tomorrow through the top with the I/O board installed. If there is anything else you suggest me checking let me know. Relatively new to using a scope and this level of electronics diagnostics so appreciate the help from everyone so far.
Hi all, any suggestions on probing while the I/O board is installed? I thought about soldering a wire to the pin and testing outside the machine. I also found it suggested that the Keyboard and Parallel port 6522 interface chips could be bad, as a test I swapped them with no behavior change, and also removed them both completely which caused error 50 to briefly show but then the image was distorted and did not redraw.
Yeah, soldering a wire and probing it is probably your best bet unless you have right-angle risers (https://github.com/warmech/lisa-diagnostic-hardware/tree/main/Riser%20Cards/Gerber%20Files) that let you probe the CPU board directly.
But given your symptoms, I'd be willing to bet that the problem is on the I/O board and not the CPU board. If something were wrong with the 556 that was causing it to assert BERR too quickly (or constantly), the Lisa almost certainly wouldn't be making it this far through diagnostics. So I'd suggest looking at the floppy controller on the I/O board instead, which is luckily a lot easier to probe!
Page 4 of the I/O board schematics should show you everything you need to see when troubleshooting the FDC. I've seen this exact same problem several times (including on my FPGA-based Lisa) and it's always been because of DTACK never getting asserted by the floppy disk controller at the end of the 68K read/write cycle. This could be caused by a variety of different things, from the 16MHz FDC clock being completely dead to a bad 8T97 that buffers the FDC's DTACK onto the main systemwide DTACK line, but something in there is likely to be your issue.
Great suggestion. I think I may have found the issue. The 16Mhz FDC clock Y1 appears to be dead if im reading my scope correctly. Compared to the others there is zero activity.
QuoteGreat suggestion. I think I may have found the issue. The 16Mhz FDC clock Y1 appears to be dead if im reading my scope correctly. Compared to the others there is zero activity.
If you are probing the portion of the circuit connected to the crystal itself, its behaviour may be altered/stopped by the loading of the scope probe. Double-check the signal after it is buffered, eg. at U6A-3
Thanks for the tip, I probed U68-LS132 Pin 3 and see basically the same thing. I lowered the voltage to 1.0/div for more detail but basically a flat line as before.
QuoteThanks for the tip, I probed U68-LS132 Pin 3 and see basically the same thing.
Sounds like you've narrowed it down a lot!
Suggestions:
Check for physical damage to the transistors as they can be bent/broken by rough handling.
Check for +12V at R4.
Swap in another crystal (even a different frequency) to see if the rest of the oscillator circuit works.
The transistors looked ok but the Hfe values on my tester were very low on Q1 compared to the Q2. Since I had some that would work I replaced the Q1 & Q2 transistors and had no change in my probe measurements or behavior of the machine. I also confirmed 12v+ at the R4 resistor, Im not sure I have a crystal on hand, but going to look and I ordered a new 16Mhz one which will be here in a couple days.
More good news! I dug out a crystal in my parts bins and did a swap and the Lisa came to life and passed all its test besides the widget drive which is expected. I may start a new thread troubleshooting that. I will install the proper crystal when the part arrives tomorrow but really happy with this progress. Now to recap the power supply and wait for my ESProfile board to come in for a boot drive! I'll update once we I have Lisa OS booting but I think we are looking good. Again appreciate all the help here it has been indispensable for this more advanced repair for a beginner!
Congratulations! Welcome to the party 8)
Hey, that's awesome! Congrats!
Not sure if you know, but a rarely-used (or at least it seems rarely-used to me because very few people mention it) feature of ESProFile is its diagnostic mode, which might be pretty helpful in getting your Widget running. I spent far more time developing the diag mode than I did the emulation mode, so it's pretty full-featured, and supports nearly every command in the Widget's rather large command set other than the block read and block write commands. So you might be able to use your ESProFile not only as an emulator, but also as an aid to get your Widget running again!