Need help repairing a defective X68000 Super

Started by wolpi, June 03, 2021, 12:47:20 AM

Previous topic - Next topic

wolpi

Hello,

I am currently very sad because my new X68000 Super does not want to run. The first thing, I've done when I got the machine was a complete recap (except the floppy drives) with high quality caps. Furthermore, I've done the PicoPSU mod to make the machine compatible with 230V. The caps inside the PSU were already leaking and corroding the board - the rest of the X68000 looked perfectly fine. The battery on the bottom board looked also fine - but I've also replaced the original CR2450 battery with a CR2032 (it is a model where the diode can already be found on the board).

After my maintenance work, I wanted to power up the machine. The sad result was a constantly green LED. I disassembled the machine again and performed some tests - and suddenly it seemed to work fine (the correct LED signals have been performed - red, blinking green, green - unfortunately only this one time in a disassembled condition). After I reassembled the machine happily, it didn't work again. After giving power to the machine the LED turns red. When pressing the power button, it immediately turns green (no blinking). When pressing the power button again, the LED turns off and the PSU keeps running. When I disconnect the mainboard the power supply does turn off again, when the power switch is pressed.

Can anybody help me diagnosing the problem? Are there any points where it makes sense to check the voltage? I've read something about this behavior occurring due to a defective crystal (but only on the original X68000 where this crystal is located on the bottom board) - but I definitely don't want to change it without knowing that it is indeed the problem.

Thank you so much

costa

Wolpi, watch out for bad placement of the PCB -  double check if, upon placing it back into the chassis, you are not bending it a bit - this might be causing some kind of short, and thus the behaviour you are experiencing.
Also verify if you are not causing some short with the shields, or maybe some cold solder in the capacitors that you replaced. If you can identify the Caps for the soft power line (that controls the power on/off), look closely for any problems there.

The super is, on my experience, the second best model (after the XVI) to repair - it's a robust machine. Hopefully you will identify the problem and it will be up and running very soon.

wolpi

Costa, thank you so much for your answer. I did disassemble the unit once more - and the power sequence was suddenly working again correctly. I assembled the unit step-by-step with multiple tests - and it kept working. I honestly don't know what the issue really was - I already paid huge attention not shorting anything with the shields. Now I am one step further - the X68000 seems to be working again (pure happiness!) but sadly the problems don't stop there :(

Instead of a nice, crisp, beautiful picture, I got this:
VideoProblem01.png

After fiddling around with the video box a little bit, I got this:
VideoProblem02.png

More fiddling sadly didn't change anything anymore.

Now I don't know, if my VGA-adapter is the issue, the recapping of the video board or if there is some other issue (VRAM or something like that). Can anybody identify what could be the error when looking at the pictures?

Thank you very much

wolpi

#3
I now disassembled the mainboard again and removed the shield for better access. Unfortunately I am no expert in electronics. I ordered an oscilloscope (but I don't have it yet - and I'm also not sure if my knowledge is enough to make any reasonable readings) and would appreciate any help diagnosing the cause of this fault.
The issue is:
1.) The noisy white vertical lines which are displayed either all over the screen or only on the left side
2.) The static blue vertical lines
I already discovered:
1.) The intensity of the white noise changes when changing the intensity of the green signal (through the potentiometer)
2.) The intensity of the blue lines changes when changing the intensity of the blue signal
3.) Changing the intensity of the red signal doesn't really change anything

When looking at the schematics of the mainboard:



The issue COULD be the video processor (IC50), the DAC (IC14), the VRAM or one of the transistors or resistors.

When looking at the schematics of the video output board:



The issue could be a transistor, resistor or one of the caps (which I replaced with 16V 470uF models instead of 10V 470uF). The issue could of course also be my amateurish designed VGA-Adapter which I realized through a custom PCB.

Could anybody help me with a more experienced perspective?
Thank you SO MUCH already!

[Edit/Update]
I just noticed that inbetween all the OKI M51C262-10 memory modules there is one slower single MN47464L-12 module, which seemed to be soldered in by hand. However, when I opened the X68000 everything seemed pretty much in its original condition. Only the PSU was defective due to leaking capacitors. Could this module be the problem - or could it be a VRAM problem in general? I don't want to perform a shot in the dark and order new parts which aren't the culprit.

costa

wolpi, have you recapped the RGB pcb?
As you verified yourself, after tweaking with it, the image improved.
I suggest you recap everything first, then you look after the remaining issues.

These machines had some durability issues, but most of them related with the capacitors, once they are replaced you got them working mostly 100% - believe me, I may have got about 20 or so of these, and I recall only giving up on ONE unit, all the others were functional after recapping.

Now, if you are using a VGA adapter, I suggest to bypass this if possible for now. Try a verified adapter (have oyu got anyone nearby that own a X68000?).
Maybe consider buying one from retrocables or off ebay.

Not the disk drives, that is another story, but this issue you seeing is not related to drives, so check for capacitors first in the RGB pcb.

Also, other than the bad video signal, is it working? can it boot off floppy , or SCSI disk?
Try to get it functional, to improve morale, then you work on the remaining issues. :)

wolpi

Hello Costa,
thank you so much for your reply. I recapped the whole machine except the floppy drives as they work fine.
I now removed the non-fitting VRAM chip and the problem persisted. I removed another VRAM chip and the problem got worse. So I did the hard work and desoldered the 16 OKI DRAM chips, added sockets (the shield bulges a bit now) and inserted brand new DRAM chips - and voila - the problem is gone.

However, my X68000 doesn't really like me :(

I now got everything running (including the X68000 booting from a SCSI2SD device) but following problems do occur:
- After some minutes the machine randomly freezes. Afterwards, a reset doesn't work anymore and it usually takes multiple attempts/power cycles to get the X68000 running again. The freezes occur basically everywhere - in the basic OS / command prompt or in a game (the graphic was shifted and the game totally frozen).
MTEST is running fine:
X68Mem.png
Could the freezing issue still be a memory issue?
- The speaker is extremely silent. When I maximize the volume knob, the speaker outputs a relatively loud noise/hissing. However, when I start a game afterwards, the game sound has a VERY low volume (however, it is there). I haven't tested the headphone output yet - but I will do that, soon.

Does anybody have an idea what could be wrong with my machine - and how to fix these two remaining issues?

Cyothevile

Quote from: wolpi on June 25, 2021, 02:03:58 PMHello Costa,
thank you so much for your reply. I recapped the whole machine except the floppy drives as they work fine.
I now removed the non-fitting VRAM chip and the problem persisted. I removed another VRAM chip and the problem got worse. So I did the hard work and desoldered the 16 OKI DRAM chips, added sockets (the shield bulges a bit now) and inserted brand new DRAM chips - and voila - the problem is gone.

However, my X68000 doesn't really like me :(

I now got everything running (including the X68000 booting from a SCSI2SD device) but following problems do occur:
- After some minutes the machine randomly freezes. Afterwards, a reset doesn't work anymore and it usually takes multiple attempts/power cycles to get the X68000 running again. The freezes occur basically everywhere - in the basic OS / command prompt or in a game (the graphic was shifted and the game totally frozen).
MTEST is running fine:
X68Mem.png
Could the freezing issue still be a memory issue?
- The speaker is extremely silent. When I maximize the volume knob, the speaker outputs a relatively loud noise/hissing. However, when I start a game afterwards, the game sound has a VERY low volume (however, it is there). I haven't tested the headphone output yet - but I will do that, soon.

Does anybody have an idea what could be wrong with my machine - and how to fix these two remaining issues?

Are you using authentic ITX pico?

Generally the very low audio even when audio is cranked up is audio amp has failed. At least that is exact problem on FMT/Marty. I noticed wiki for X68 also has this issue documented at least for compacts.

Jehuty

Could be a problem with -12V. Don´t know the shematics so good, but in the Amiga it needs -12V for the audio amp.
did you try another PSU ? try a normal ATX to test if the freeze problem comes from PSU too.

costa

Quote from: wolpi on June 25, 2021, 02:03:58 PMI now removed the non-fitting VRAM chip and the problem persisted. I removed another VRAM chip and the problem got worse. So I did the hard work and desoldered the 16 OKI DRAM chips, added sockets (the shield bulges a bit now) and inserted brand new DRAM chips - and voila - the problem is gone.

What memory have you used in the replacement?
Does the timings match the original memory?

wolpi

Thank you SO MUCH for your helpful answers.

Quote from: Cyothevile on June 25, 2021, 11:21:11 PMAre you using authentic ITX pico?
Quote from: Jehuty on June 26, 2021, 01:52:38 AMCould be a problem with -12V. Don´t know the shematics so good, but in the Amiga it needs -12V for the audio amp.
did you try another PSU ? try a normal ATX to test if the freeze problem comes from PSU too.
I am indeed using an expensive, authentic Pico-PSU because I thought, this would be a good choice. Now I noticed, that it only delivers 0.05A on -12V :( When running it without any load, it also delivers the -12V (I tested the modified PSU before using it). However, after putting it into my X68000 the Voltage on -12V "drops" to +0.7V. I now bought a much cheaper Inter-Tech 160W PSU which can deliver 0.15A on -12V and the sound was fine. Even the interfering noises on my SCSI2SD were gone with this PSU (the Pico-PSU managed to make noise when accessing the SCSI2SD). However, the PSU is slightly too big so that the case of the original PSU cannot be closed properly anymore. So the saga continues - I will try now the 120W Rev. 2 Inter-Tech PSU which only delivers 0.1A on -12V - I hope it will work fine.

Quote from: costa on June 26, 2021, 07:13:37 PMWhat memory have you used in the replacement?
Does the timings match the original memory?
I have been using M51C262-80 (instead of the M51C262-10) chips. So the guaranteed timings are faster than the original chips.

costa

Quote from: wolpi on July 01, 2021, 02:36:05 AMI have been using M51C262-80 (instead of the M51C262-10) chips. So the guaranteed timings are faster than the original chips.

I was not able to find the datasheet for this memory. But I belive the sufix "-80" is the access time, and therefore it is higher than the original memory at "-10".

Might not be this causing your computer to crash, but it might also be it.


wolpi

Quote from: costa on July 03, 2021, 09:45:12 AMI was not able to find the datasheet for this memory. But I belive the sufix "-80" is the access time, and therefore it is higher than the original memory at "-10".

Might not be this causing your computer to crash, but it might also be it.
Thank you again for your help! DRAM from this time indeed uses the suffix for its guaranteed access time. However, -10 usually doesn't stand for 10ns but for 100ns - therefore it is "slower" than the chips with the -80 suffix, which stands for 80ns. I attached the datasheet of this specific RAM.
I received the new Pico-PSU and the sound is now OK - however the machine still freezes occasionally. Btw. is it correct that in Bubble Bobble the first music tracks in the "Test Music" option stay silent - at least without a MIDI card (I think track 5 or 6 was the first one which did produce music)?