Fixing my Milan 040 (or not) blog

[agranlund]

The sourcecode for the Milan boot code can be had here:
https://github.com/ps68060/MilanBootblock

May or may not come in handy for comparing expected vs gotten behaviour, or just for knowing what the machine attempts to do at startup :)

[exxos]

The sourcecode for the Milan boot code can be had here:
https://github.com/ps68060/MilanBootblock

May or may not come in handy for comparing expected vs gotten behaviour, or just for knowing what the machine attempts to do at startup :)

Thanks :)

[dml]

Qwe said:

That seems to be a fairly detailed guide he provided as a startpoint. Hopefully you can figure something out from here.

[exxos]

That seems to be a fairly detailed guide he provided as a startpoint. Hopefully you can figure something out from here.

Yeah, I will have to hardwire the address bus low, and /AS etc to see if that triggers the flash. I know the flash is good, so if I can get all zeros on the address bus, the flash should output all zeros... thats going to be fun :)

[agranlund]

I found my programmer can read the flash..

Might be worth verifying if the address lines shows that it jumps to address $90 after the initial read of the values in $0-$7 ?
just to see if it even gets that far?

I’d expect it to then do at least some basic hw setup before it touches ram for the very first time.

[exxos]

On the rear of the board it does look like I must have suspected some damage before and scraped away the solder resist... But it looks like there is a gash right across the tracks above it as well. so I will have to scrap that away a bit to see if there is any damage...

IMG_3413.JPG

[exxos]

So nothing wrong there...

IMG_3414.JPG

IMG_3415.JPG

IMG_3416.JPG

[exxos]

Measuring all depends on the flash to 5V give some weird results.. Most major 4.7K as you would expect..

However,

D0 (which was the only one actually toggling) measures 250R to VCC.
D1 4.7k to vcc
D2 190R to VCC

D0 87R to GND
D1 4.87K to GND
D2 1R to GND

So D0,D2 seem wrong here.. But why would D0 even toggle in that state in the first place.. I guess it could explain why the voltage was a lot lower than the address bus. But its about 3V anyway.. The address bus measures 6V.. (but bad scope grounding so take with a pinch of salt) .

This is with the flash and CPU still in place. So I will remove those and retest...

EDIT:

So no CPU or flash and the "faults" are still there.. This isn't looking good :(

I can only find D0 going to the Intel PCIset chip and another big chip next to the PSU connector :( I assume D0 is on its own little bus to the big chips at this point.

I guess it's also possible that when the guy originally said he dropped a screwdriver, maybe he actually dropped it on a ISA card.. which are connected to the PCI set IC..

[dml]

I guess it's also possible that when the guy originally said he dropped a screwdriver, maybe he actually dropped it on a ISA card.. which are connected to the PCI set IC..

o_o

well it does seem like you might have found a lead...

[exxos]

well it does seem like you might have found a lead...

Yeah maybe..

The MFP seems to be the only thing getting warm. But don't know offhand if they get warm anyway.

Though finding the big chips (and ones which are actually new) is likely not simple.. but also, I don't know how fragile these Milan boards are. I could end up ripping pads up if they are not great quality boards. So changing chips id consider a last resort. But there isn't much else to look into at this point.