Amiga 1200 2B Mainboard - Data bus high

[highpuff]

Got an issue here that is harder than I expected to fix. The problem is that after a while, depending if the PCB is cool or not, all data bus'es gets high after a while. No picture is ever displayed.

Things to know:
- The PWR-led stays dimmed.
- This happened when the owner tried to upgrade from 3.0 to 3.1.
- I'm using the Dialoga 2.0 ROM's that I verified a few times during my tests that they work.
- The Caps button works a few pushes then stays off/on.

I have done the following:
- All clocks are checked and has the correct frequencies
- Checked all traces from U6A and U6B
- Checked for shorts
- Removed both U7 and U8 (CIA A-B)
- Removed the 68EC020 CPU and used a working Blizzard 1220

Any ideas? I cannot see any damage at the ROM sockets but will remove them tomorrow and see if there is something there. Maybe I will try to replace the MPU as well.

For a short period of time I was suspecting that it could not read from from the ROM's. The exact same behaviour occurs when I remove U6A or both of the ROM's and power on the Amiga.

Thanks in advance

[Zetr0]

@highpuff

I usually start from the power input and work my way to the CPU (figuratively)

If you are comfortable with power input, decoupling, VIO levels then I would start looking at the Keyboard resonator - ensure thats its pumping - this caught me once - couldn't figure out why the whole thing was kept in a reset state.

Next on my list would be to inspect the solder of the CHIP set and sundry components - paying close attention to the native video socket (particularly the +12/+5)

I have one board here that turned out to have a LISA fault - that would only present itself after 20/30 minutes of power up - that had taken a while to isolate - initiailly I thought it was a bad solder joint and a reflow would likely solve the problem - sadly no - I had to replace the IC - which I got from a donor motherboard that had some serious (non-repairable PCB Damage)

[utri007]

Does it still have a 3.1 roms? My 2b didn't work at all with 3.1 roms without timing fixes. Could it be that timing fixes are made wrong?

[highpuff]

Thanks for the reply Zetr0.

I have only checked the clocks etc. But I will check power input etc as you suggested. Ordered a 16ch logic analyzer so I can dig some more in dept also.

[Zetr0]

@highpuff

16ch Logic Analyser - Very Nice :

Sadly I keep putting off getting a proper scope as the one I want is £300 - and that is just too big to fit under wifie-radar

Now utri007 has reminded me of something - you could have an anomaly in the power supply so give this a check over - ensure voltage levels are good =)

Good luck with your quest sir!


@utri007

Originally Posted by utri007

Does it still have a 3.1 roms? My 2b didn't work at all with 3.1 roms without timing fixes. Could it be that timing fixes are made wrong?

I had a very similar 2B issue - however, I found that replacing the power input / VIO filter and decoupling capacitors solve most issues - it would be quite stable for some time - however struggled with my Apollo 040 and 060 - after the timing fixes this was alleviated for the most part - but it could of been an anomaly in the power supply at the time since if memory serves I didn't check that part of the loop out.

In regards to that though - it worked flawlessly on the plethora of other A1200 motherboard revisions.

[highpuff]

Utri007:

No, it is back to 3.0 but I'm also using Dialoga 2.0 ROM's and the issue is the same.The board has been modified with timing fixes but I have temporary reverted it to standard to avoid any issues.

Zetr0:

I have a RIGOL DS1052E that is good for 50MHz but is supposed to be upgradable to 100MHz with simple firmware. Still, it is around 300€ I think.

I have tested with two power supplies and one also with a new Meanwell RT-65B inside of it (recommended). So that can be ruled out. I have 2 working A1200 PCB as reference but sadly they are not version 2B.

Next step(s):
- Check voltage levels
- Check resistors on the board
- Remove the ROM sockets (replace with new ones)

Thanks so far

[highpuff]

Update 2016.08.16

I have done some checkup's on traces and also VCC for I/O's. Also De-soldered the DIL-sockets to check if I could find anything below them that could be hidden, but nothing.

Checked the keyboard resonator as well and it is working well.

Tonight I found that R103 is 28Ω instead of 470Ω as specified. Since the resistor is not put in a parallel circuit it should show 470Ω.

Will remove it tomorrow and measure it isolated and replace it. The R103 is the terminator for _AVEC (Autovector).

Let's see how this goes :P